Evidence review for non-invasive cardiac output monitoring

National Guideline Centre (UK)

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Evidence review for non-invasive cardiac output monitoring

Perioperative care in adults

Evidence review J

NICE Guideline, No. 180

Authors

National Guideline Centre (UK).

London: National Institute for Health and Care Excellence (NICE); 2020 Aug.

ISBN-13: 978-1-4731-3827-8

1. Non-invasive cardiac output monitoring

1.1. Review question: What is the clinical and cost effectiveness of non-invasive cardiac output monitoring during surgery in adults?

1.2. Introduction

Cardiac output monitoring has been a part of perioperative practice for a number of years, primarily used to achieve fluid optimisation and guide the use of vasoactive and inotropic drugs for patients undergoing major surgery. This section looks at the evidence for the most clinical and cost-effective strategies for the use of non-invasive cardiac monitoring, with consideration of the benefits and risks of the various available monitors being considered.

1.3. PICO table

For full details see the review protocol in appendix A.

1.4. Clinical evidence

1.4.1. Included studies

Twenty-three randomised controlled trials were included in the review³^, ²³^, ²⁵^, ²⁹^, ³⁷^, ⁴²^, ⁴³^, ⁴⁶^, ⁵⁶^, ⁶²^, ⁷¹^, ⁷⁵^, ⁸⁰^, ⁸⁴^, ⁸⁵^, ⁸⁸^, ⁹²^, ⁹⁸^, ⁹⁹^, ¹⁰³^, ¹⁰⁶^, ¹⁰⁷^, ¹¹⁴ these are summarised in Table 2 below. Evidence from these studies is summarised in the clinical evidence summary below (Table 3). See also the study selection flow chart in appendix C, study evidence tables in appendix D, forest plots in appendix E and GRADE tables in appendix F.

One study compared oesophageal Doppler monitoring to pulse contour analysis and the remaining twenty-two studies compared cardiac output monitoring to conventional clinical assessment. Non-invasive cardiac output monitoring interventions were grouped for this comparison to assess the overall efficacy of non-invasive cardiac output monitoring interventions. Non-invasive cardiac output monitoring was used as an umbrella term to encompass interventions measuring stroke volume / cardiac output / central venous pressure for the purposes of evaluating volume status of a patient. This measurement would guide clinician decision making regarding fluid replacement therapy. Subgroup analysis would explore differences between intervention methods if heterogeneity in outcome data was observed.

1.4.2. Excluded studies

See the excluded studies list in appendix I.

1.4.3. Summary of clinical studies included in the evidence review

See appendix D for full evidence tables.

1.4.4. Quality assessment of clinical studies included in the evidence review

See appendix F for full GRADE tables.

1.5. Economic evidence

1.5.1. Included studies

Six health economic studies were identified with the relevant comparison and were included in this review.⁵^, ⁵⁰^, ⁵⁴^, ⁶³^, ⁶⁶^, ⁸⁷ These are summarised in the health economic evidence profile below (Table 7 - Table 11) and the health economic evidence tables in appendix H.

1.5.2. Excluded studies

No relevant health economic studies were excluded due to assessment of limited applicability or methodological limitations.

See also the health economic study selection flow chart in Appendix G:.

1.5.3. Summary of studies included in the economic evidence review

1.5.4. Health economic modelling

Model methods

A previous NICE medical technologies guidance (MTG3) assessed the clinical and cost-effectiveness of the CardioQ-ODM oesophageal Doppler monitor and recommended to consider the use of monitoring in people undergoing high risk or major surgery. Since the publication of this medical technology guidance in 2011, there have been improvements in the perioperative care pathway, which have resulted in reductions in complications and length of stay. More surgeries are being performed in a minimally invasive way instead of as open procedures, which can lead to a quicker recovery. Also, a recent audit of ambulatory surgery has showed that there has been an increase in the number of procedures undertaken as day-case surgery in the NHS. For example, there has been a steady increase in the rate of mastectomies carried out as day-case surgeries, with the rate rising from 3.8% in 2011/12 to 10.8% in 2016/17.⁹⁴ UK audit data has also shown there has been a decrease in length of stay following major surgery, for example, the National Emergency Laparotomy Audit 2017/18 showed that patient’s average hospital stay decreased from 19.2 days to 16 days from 2013 to 2018.⁷⁰ Also, the National Oesophago-Gastric Cancer Audit 2018 report highlighted that the average length of stay reduced from 10 to 12 days to 7 to 9 days in five years.³⁹ Although six published economic analyses were assessed and included in the review, methodological limitations meant there was still uncertainty about cost effectiveness. Additionally, variation in current practice and improvements in perioperative care and outcomes since MTG3 meant that the savings in MTG3 might not be as significant as previously demonstrated. For these reasons, alongside the fact that the monitors have a high cost, this area was prioritised for original economic analysis.

A cost-utility analysis was undertaken with lifetime quality-adjusted life years (QALYs) and costs from a current UK NHS and personal social services perspective. Both costs and QALYs were discounted at a rate of 3.5% per annum in line with NICE methodological guidance. An incremental analysis was undertaken.

The population was adults having major or complex or high risk surgery and high risk adults undergoing any surgery. Due to this population typically having worse health than the general population, using general population data such as mortality for the baseline data wasn’t felt appropriate. The committee agreed that that a large proportion of major surgery in England is for treating people with cancer, and some of the studies included in the clinical review were for adults undergoing major bowel or gastrointestinal surgery therefore adults with bowel cancer was chosen as the base case population.

The first part of the model consisted of a 30-day decision tree which modelled the probability of experiencing complications or death up to 30 days post-surgery, with these probabilities being taken from the clinical review. This found that COM reduced complications and also led to small reduction in mortality. Those that experienced complications were further broken down in to those having minor (Clavien-Dindo grades 1 and 2) and major (Clavien-Dindo grades 3 and 4) complications. The decision tree applied a different cost and utility to those that experienced complications. For minor complications, a quality of life decrement associated with a chest infection was applied and the NHS reference costs associated with a chest infection was also applied as a one off cost. For those experiencing major complications, the NHS reference costs and quality of life associated with being admitted to ICU was applied. Those that did not experience complications did not have any costs applied in the decision tree and the quality of life associated with bowel cancer was applied.

The second part of the model was a Markov model to capture costs and outcomes over a lifetime. A one year cycle length was used. Adults alive at the end of 30 days entered the Markov model. The Markov model was made up of 3 health states: ‘alive with no complications’, ‘alive with complications’ and ‘dead’. People that experienced major complications in the decision tree entered the ‘alive with complications’ health state, as it was agreed that they would experience long-term health implications which resulted in higher mortality and lower quality of life, compared to those alive with no complications. Those with minor complications entered the ‘Alive with no complications’ health state as it was assumed that their minor complication would be dealt with within 30 days. Those who experienced no complications in the decision tree also entered the ‘Alive with no complications’ health state. The probability of transitioning to the dead health state was based on bowel cancer related mortality rates. Those in the ‘alive with complications’ health state had a higher probability of death for the first 3 years, at which point it returned to the baseline. This was based on a cohort study conducted in England which showed people experiencing complications 15 days after surgery had a higher probability of death for 3 years, and then it returns to baseline.⁶¹ The committee highlighted that once a person has been cancer free for over eight years, their mortality returns to that of the general population. Therefore, bowel cancer mortality rates were applied for 10 years. The bowel cancer mortality data supported this as it showed that at 10 years there was no change in net excess mortality. General population mortality was then applied for the following years until the end of the time horizon. The quality of life associated with having bowel cancer was applied for 10 years in the model, and then it returned to the age-related general population quality of life. Costs associated with living with bowel cancer were also applied for 9 years and were obtained from a study conducted in England.⁴⁷ Those in the ‘alive with complications’ health state had quality of life associated with ICU survivors applied for 3 years as well as additional costs associated with ICU survivors taken from a study conducted in Scotland.⁵²

Results

The base case results showed that cardiac output monitoring was associated with additional costs and higher QALYs with an ICER of £25 per QALY gained, which is considered cost-effective at the NICE threshold of £20,000 per QALY gained. Table 12 shows the 30 day and lifetime results.

Various sensitivity analyses were conducted to test to the robustness of results. Firstly, different treatment effects were used and showed that COM was dominant when: excluding non-UK studies; excluding cardiac and emergency surgery; and excluding studies that were conducted before the publication of MTG3. Other sensitivity analyses were conducted to make inputs conservative towards COM, and in all analyses COM remained cost-effective. In one analysis, the ICER increased to £7,924, and this was where: the general population was used as the baseline population; assuming there was no 30 day mortality difference; and using the upper confidence interval value for complications treatment effect. Sensitivity analyses showed that the model was sensitive to the treatment effects and the mortality rates used in the Markov model. For example, when using the general population mortality rates instead of the bowel cancer mortality rates, there was a smaller QALY difference between the two comparators. Also, some of the costs used in the model had some impact on results, such as removing the cancer related costs. This resulted in COM being dominant, as when cancer costs are included then more people are alive in the COM arm to accrue expensive healthcare costs from cancer, making the COM arm more expensive, and therefore omitting these made COM cheaper compared to CCA.

Limitations of the model included the use of the proxy bowel cancer population. Although the committee agreed that this was more representative of the major surgical population, it could have overestimated the mortality in the model. This was tested in a sensitivity analysis by using general population mortality rates and this did not impact conclusions. Treatment effects were based on the guideline clinical review, and the committee highlighted some issues with this data. Firstly, there were only a small number of studies published since MTG3, which was a limitation as current practice has evolved since 2011 and since a lot of the included studies were conducted. Some of the randomised controlled trials would have included central venous pressure as part of conventional clinical assessment, which is no longer considered standard practice. Also, there has been a trend towards administering fewer fluids in recent years. Another limitation involved the assumptions made regarding complications. The data used in the model to represent a minor complication was based on a chest infection. This was very specific, and in reality people can experience many different types of minor complications. Also, the assumption that a minor complication would not impact health after 30 days could vary in real life, however the committee highlighted that there was no available evidence to indicate how long the impact would last. In addition, major complications were associated with a long-term cost and health impact of 3 years. Although this was based on published evidence, the committee highlighted that this could vary between different types of surgery and different people.

1.6. Evidence statements

1.6.1. Clinical evidence statements

Oesophageal Doppler monitoring versus pulse contour analysis

Complications

One study showed a clinically important benefit of Oesophageal Doppler monitoring for the number of patients experiencing complications at 8 days compared to pulse contour analysis (1 study, n=21, moderate quality evidence).

Evidence not suitable for GRADE analysis

One study showed no statistically significant difference in length of hospital stay between Oesophageal Doppler monitoring and pulse contour analysis (1 study, n=21, low risk of bias)

Cardiac output monitoring versus conventional clinical assessment

Mortality

Twelve studies demonstrated no clinically important difference in mortality between cardiac output monitoring and conventional clinical assessment (12 studies, n=2012, low quality evidence).

Complications

Thirteen studies found a clinical benefit of cardiac output monitoring for the number of patients with complications compared to conventional clinical assessment (13 studies, n=2049, moderate quality evidence).

One study found no clinical difference in complications (POMS ≥1) at 3-days between cardiac output monitoring and conventional clinical assessment (1 study, n=220, moderate quality evidence).

One study found no clinical difference in complications (POMS ≥1) at 5-days between cardiac output monitoring and conventional clinical assessment (1 study, n=220, low quality evidence).

One study found no clinical difference in complications (POMS ≥1) at 8-days between cardiac output monitoring and conventional clinical assessment (1 study, n=220, moderate quality evidence).

Length of hospital stay

Eight studies showed no clinically important difference for length of hospital stay between cardiac output monitoring and conventional clinical assessment (8 studies, n=941, high quality evidence).

Length of ICU stay

Two studies found no clinically important difference in length of stay in ICU between cardiac output monitoring and conventional clinical assessment (2 studies, n=214, high quality evidence).

Readmission

Five studies showed no clinically important difference in readmission rate between cardiac output monitoring and conventional clinical assessment (5 studies, n=707, moderate quality evidence).

Evidence not suitable for GRADE analysis

One study found no statistically difference in mortality between cardiac output monitoring and conventional clinical assessment (1 study, n=114, low risk of bias).

One study found no notable difference in quality of life at 4-6 weeks between cardiac output monitoring and conventional clinical assessment (1 study, n=128, high risk of bias).

Four studies showed a trend to benefit with cardiac output monitoring for total number of complications compared to conventional clinical assessment (4 studies, n=404, high risk of bias)

Fifteen studies showed overall no statistically significant difference in length of hospital stay between cardiac output monitoring and conventional clinical assessment (15 studies, n=2197, high risk of bias).

Two studies showed no statistically significant difference in length of ICU stay (days) between cardiac output monitoring and conventional clinical assessment (2 studies, n=315, high risk of bias).

1.6.2. Health economic evidence statements

One original cost–utility analysis found that COM was cost effective compared to CCA in adults having major or complex or high risk surgery and high risk adults undergoing any surgery (ICER: £25 per QALY gained). This analysis was assessed as directly applicable with minor limitations.
One cost–utility analysis found that in adults 50 years and over undergoing major gastrointestinal surgery cardiac output monitoring was dominant (less costly and more effective) compared to conventional clinical assessment. This analysis was assessed as partially applicable with potentially serious limitations.
One comparative cost analysis found that oesophageal Doppler monitoring was cost saving compared to conventional clinical assessment in adults undergoing moderate and major risk surgery and high risk adults undergoing any surgery (cost difference: £1,091 per patient). This analysis was assessed as partially applicable with potentially serious limitations.
One cost–utility analysis found that in adults 80 years and over undergoing surgery for hip fractures COM was dominant (less costly and more effective) compared to standard care. This analysis was assessed as partially applicable with potentially serious limitations.
One cost-effectiveness analysis found that in adults undergoing intermediate and high risk abdominal surgery cardiac output monitoring (ODM and PCA) was dominant (less costly and more effective) compared to CCA. This analysis was assessed as partially applicable with potentially serious limitations.
One cost-effectiveness and cost-utility analysis found that in adults undergoing colorectal resection cardiac output monitoring (ODM with CCA and CVP) was dominant (less costly and more effective) compared to CCA. This analysis was assessed as partially applicable with potentially serious limitations.
One cost–utility analysis found that [in adults undergoing high risk surgery ODM was cost-effective at a threshold of £30,000 compared to CCA. This analysis was assessed as partially applicable with potentially serious limitations.

1.7. The committee’s discussion of the evidence

Please see recommendation 1.4.5 in the guideline.

1.7.1. Interpreting the evidence

1.7.1.1. The outcomes that matter most

The committee agreed that cardiac output monitoring is used within perioperative practice to achieve fluid optimisation and guide the use of vasoactive and inotropic drugs with the goal of reducing the metabolic impact of surgery on patients undergoing major surgery. As such, the committee considered health related quality of life, mortality and perioperative complications as critical outcomes to decision making. Length of hospital stay, length of stay in the intensive care unit and hospital readmission were also considered to be important outcomes.

1.7.1.2. The quality of the evidence

The quality of evidence that was suitable for GRADE analysis ranged from low to high. The majority of the evidence was graded at moderate quality. This was mostly due to imprecision of data. The committee felt that the evidence was of sufficient quality and quantity to support the recommendations made.

Outcomes which were not suitable for GRADE analysis were considered to be at low and high risk of bias.

1.7.1.3. Benefits and harms

The committee discussed the evidence on cardiac output monitoring in adults having major or complex or high risk surgery and high risk patients undergoing any surgery.

The committee noted evidence from one small study with 21 participants showing a benefit of fewer complications with Oesophageal Doppler monitoring when compared to pulse contour analysis. This study also showed no clinical difference in length of stay. The committee agreed that this evidence was insufficient to support any recommendation.

In a comparison of cardiac output monitors to conventional clinical assessment, the committee agreed that there was no clear benefit of one type of monitor over another. As such, interventions of COM were grouped for an overall comparison with conventional clinical assessment. From this dataset, the committee agreed that there was a benefit of COM with fewer total complications compared to conventional care. The committee also noted a trend towards a benefit for length of stay with COM, but highlighted a variation in results due possibly to the heterogeneity in populations included in the analysis. The committee discussed a possible increase in rate of readmissions with COM but noted the low quality of evidence caused by serious imprecision. The committee considered the possibility of increased readmissions with COM being linked to a shorter length of stay with the intervention, but highlighted that the noted differences between groups were not clinically important. No difference was found between COM and conventional care in mortality. The committee considered that the noted benefits in a reduced complication rate and shorter length of stay were significant and on balance with low quality evidence of increased readmission rates demonstrated an overall positive effect with the use of COM.

1.7.2. Cost effectiveness and resource use

Six published economic studies were included that compared cardiac output monitoring to conventional clinical assessment. Three of these were from a UK NHS perspective. One of the three being the manufacturer submission for the NICE medical technologies guidance 3 (MTG3), on CardioQ-ODM. This was a cost-comparison that involved six strategies, comparing oesophageal Doppler monitoring (ODM) in addition to conventional clinical assessment (CCA) with: CCA alone, central venous pressure (CVP) + CCA, pulse pressure waveform analysis (PPWA) + CCA, CVP + ODM + CCA, and CVP + PPWA + CCA. The analysis showed that ODM with CCA was cost-saving when compared to all other interventions. This study was rated as partially applicable with potentially serious limitations. This was for reasons such as not having any health outcomes. Some of the RCTs included in the analysis were excluded from clinical review due to starch boluses being used, and the time horizon was only ‘in-hospital stay’ thereby potentially omitting any long-term impact on costs and quality of life. The cost savings were largely attributable to the length of hospital stay savings associated with ODM. The analysis assumed that CardioQ-ODM was associated with a reduction in length of stay of 1.92 days, which was based on a combination of randomised controlled trials and audit data. The committee highlighted issues with this assumption as length of hospital stay data from randomised controlled trials can vary based on the country they are conducted in, and are not always reflective of current UK practice.

The second UK analysis was a cost-utility analysis for high risk surgical adults and compared ODM + CCA to CCA alone, as well as a second comparison which added CVP to both arms. A meta-analysis was conducted and the outcomes that fed in to the model were mortality and length of stay. This study did not give a breakdown of the costs or QALYs for each intervention but concluded that ODM was cost-effective at a threshold of £30,000 per QALY. No results were presented for a threshold of £20,000 per QALY. This study was rated as partially applicable with potentially serious limitations, as some of the RCTs included used starch boluses and the assumption that adults would only survive an average of five years post-surgery was not considered a reflection of what happens after surgery.

The third UK analysis was a cost-utility analysis with a lifetime horizon based on a single RCT (OPTIMISE), which is included in the clinical review. This study looked at pulse contour analysis (PCA) versus CCA in adults undergoing major gastrointestinal surgery. This analysis found that PCA was dominant. Results at six months were reported as well as lifetime results, and the intervention was dominant in both scenarios. Cost-savings were based on the reduction in hospital length of stay that was seen in the trial. The committee felt that as this was a UK study, the length of stay data was more reliable. However, limitations included: it only looked at one type of surgery and not the whole surgical population, it was based on a single RCT, standard care involved central venous pressure in some cases, cost sources were unclear and costing methods to avoid double counting may have impacted results. This study was given an overall rating of partially applicable with potentially serious limitations.

One study conducted a cost-effectiveness analysis from a French healthcare perspective on adults undergoing intermediate and high risk abdominal surgery. The study compared ODM + CCA, PCA + CCA and CCA alone. A meta-analysis was conducted which identified 13 RCTs and the model incorporated death and major complications. The study found that both types of cardiac output monitoring were dominant when compared to CCA, in terms of being less costly and reducing the number of complications and death. This study was rated as partially applicable with potentially serious limitations due to it being non-UK and only looking at abdominal surgery. Also, the time horizon was until hospital discharge which is too short to fully capture costs and outcomes and some of the RCTs included in the analysis used starch boluses.

One study from a Spanish healthcare perspective conducted a cost-effectiveness analysis for adults undergoing colorectal resection. They also conducted a cost-utility analysis as part of a sensitivity analysis. ODM +CCA was compared to CCA alone. Another analysis looked at adding CVP to both arms. Treatment effects were obtained from a meta-analysis of three RCTs. The study concluded that ODM increased health benefits (in terms of survival rate and reduction in complications) and reduced costs, which made it dominant. This study was rated as partially applicable with potentially serious limitations. Reasons for this rating included the Spanish healthcare perspective; the analysis only looked at one type of surgery and used treatment effects from other types of surgery to inform the analysis. Also, one of the RCTs included starch boluses and the analysis incorporated length of hospital stay from one RCT which was conducted in 2005 and may not be relevant to current practice.

The final analysis was a cost utility analysis from a Swedish healthcare perspective, that looked at COM compared to CCA in adults over 80 years old undergoing surgery for a hip fracture. A five-year time horizon was used to model longer term impacts of complications such as cardiac complications and stroke. Cardiac output monitoring resulted in less costs and additional QALYs over the five-year time horizon. The committee agreed that in emergency surgery cardiac output monitoring may be used more and is probably more likely to be cost-effective as the adult undergoing surgery may already be at a higher risk than someone undergoing elective surgery. This study was rated as partially applicable with potentially serious limitations. Reasons for this rating included the Swedish healthcare perspective may not be relevant to current UK practice, the analysis focuses on one type of surgery and it is unclear what tariff and population was used to obtained quality of life weights. Also, treatment effects were obtained from various studies looking at cardiac output monitoring that were not directly relevant to the surgery and population in the analysis.

After reviewing the published evidence, the committee considered there to still be uncertainty about the cost effectiveness of cardiac output monitoring versus conventional assessment in the current NHS setting, and prioritised this area for new analysis. Reasons for this uncertainty included: the studies relevant to the UK NHS were out of date or based on only a few studies for treatment effect. On a related point, committee opinion was that CCA has improved in the last decade, and therefore the relative cardiac output monitoring benefits may not be as large compared to previously, therefore the committee agreed there was likely to be new clinical data capturing this that could be used in a model. CCA improvement is based on a number of reasons such as central venous pressure no longer being used in current practice. Additionally, certain surgical techniques have also improved, for example the use of laparoscopic surgery instead of open surgery. The introduction of enhanced recovery programmes also means there are many processes as part of the surgical pathway which have reduced overall complications and length of stay. Also, some of the published evidence was in a specific population or only looked at one type of monitor, and the committee agreed that it was useful to analyse all of the data together for all surgeries and all monitors combined, and use this more up to date pooled data in a model, to see if COM was still considered cost effective.

A decision analytic model was constructed to compare COM to CCA. The committee highlighted that the population being modelled would be higher risk than the general population, therefore bowel cancer was chosen as a proxy population. The model structure consisted of a 30-day decision tree capturing the hospital period, followed by a lifetime Markov model with one year cycles. Treatment effects were taken from the clinical review to inform the decision tree, which had branches of death, complications, and no complications. Complications were broken up into minor and major complications. Intervention costs were based on a weighted average of the costs of the most commonly used monitors. No costs were attributed to the CCA arm, as the only difference in costs would be use of the monitor. After 30 days people that were alive entered a three-state Markov cohort model. The health states were death, alive without complications, and alive with complications. Those that experienced no or minor complications in the decision tree both entered the ‘alive without complications’ state. Those that experienced major complications were assumed to have long term health implications and entered the ‘alive with complications’ state. Mortality associated with bowel cancer was added to the general population mortality and the cancer mortality only applied for 10 years. Costs and quality of life associated with having bowel cancer were applied in the model. For those that experienced major complications, hazard ratios were applied to the mortality rate for three years post-operatively and they had additional costs and lower quality of life associated with ICU survivors applied for three years.

Results showed that the upfront cost of cardiac output monitoring was offset in the short term by the reduction in complications, as the 30-day results showed that COM was dominant. The lifetime results showed an ICER of £25 per QALY when comparing COM to CCA. Various sensitivity analyses were conducted to test the robustness of the results. Treatment effects were tested by: excluding trials that were not conducted in the UK, excluding trials in cardiac and emergency surgery, and excluding studies that were conducted before the publication of MTG3 (2011) – this left 5 studies. All showed COM to be dominant. Various sensitivity analyses were also conducted which assumed no 30-day mortality in the decision tree, as the committee did not believe that the type of haemodynamic monitoring would impact mortality. All of these analyses did not impact conclusions, however the ICER increased to £7,924, and this was where: the general population was used as the baseline population; assuming there was no 30 day mortality difference; and using the upper confidence interval value for complications treatment effect (SA30).

Various other inputs were varied such as the cost of complications, the cost of the interventions, and inputs related to the population such as assuming the population was the general population (and using general population mortality and no cancer costs). Age-specific costs were also incorporated in another analysis. Some sensitivity analyses varied various inputs to make the analyses conservative to COM to see if it would still be cost effective (for example, making adverse events cheaper alongside using the upper confidence interval of the relative risk of complications). In all these analyses COM remained cost effective with ICERs below £20,000 per QALY gained.

Limitations of the model included the assumptions about the base case population. As the population of interest was very broad, a proxy population of bowel cancer was chosen for the base case analysis. However, not everyone having major or complex surgery would be undergoing surgery for cancer. Also, the data that was used to inform the cancer mortality was taken from all adults diagnosed with bowel cancer in England and Wales and not everyone would have undergone surgery. There were also assumptions made regarding the type of complications in the model which can vary greatly between adults and different types of surgery. In addition, it was assumed that minor complications did not have any long-term impact on health but this could also vary. The committee agreed that although some minor complications could have long-term impacts, there was no evidence to support this. Extensive sensitivity analysis was undertaken and the conclusion was considered robust.

The committee discussed the clinical evidence and agreed that there was a signal of clinical effectiveness of COM with regards to avoiding complications in particular when complications were combined. The committee agreed that there were uncertainties in the clinical evidence used to inform the model, as there had been a limited number of studies published since MTG3 and there was uncertainty around mortality. Although their interpretation of the model was that the conclusions were robust in favour of COM, even when considering only complications and not mortality. The model was robust to inputs varying in sensitivity analyses. They discussed the many improvements that had been made in CCA since the recommendation from the NICE medical technologies’ guidance, such as the introduction of enhanced recovery programmes and a general trend towards administering less intravenous fluids, which led to the committee feeling that although there was evidence of effectiveness from the review, they were not entirely convinced that there would be additional benefit from COM. They agreed that clinical judgement was an important indicator as the adult’s health state and type of surgery can determine whether or not to use cardiac output monitoring. As a result, the committee agreed to recommend that cardiac output monitoring should be considered for use during major complex or high-risk surgery. This would give flexibility to clinicians who are already using COM, but also to those who are not. It would allow consideration about whether COM could be beneficial to specific cases.

The committee discussed that recommending cardiac output monitoring would not lead to a significant change in practice as most hospitals already have some cardiac output monitoring machines. The committee indicated that since the publication of MTG3 the uptake of cardiac output monitoring was significant especially for the use of the oesophageal Doppler monitor. Despite the large uptake of the machines, there is variation in practice as some anaesthetists may use the machines more than others.

1.7.3. Other factors the committee took into account

The committee agreed that the conventional clinical assessment as reported by the included research papers may not reflect the current practice within the NHS and recognise this as a limitation.

The committee noted that as surgical techniques have developed so have approaches to fluid management. As such, the observed benefit of cardiac output monitoring may be lessened in contemporaneous medicine. The committee added that central venous pressure monitoring is no longer used in contemporaneous clinical practice to evaluate patient fluid status, and may contribute towards improved conventional clinical assessment.

The committee noted that for laparoscopic and less complex surgery, COM is not standard in current practice, but is more common and more likely to demonstrate benefit for complex, emergency and tertiary patients.

The committee agreed that COM is now generally used as part of multimodal patient monitoring and therefore assists as a component in informing decisions about intravenous fluid requirements. COM is however less likely to be used as a singularly didactic indicator for the administration of intravenous fluids. COM may be considered as an adjunct to help anaesthetists determine the cause of reduced blood pressure (i.e. low CO vs low vascular resistance)

The consensus was that current practice is more bespoke when considering monitoring for complex, emergency and tertiary patients and might include COM in such situations. Further evidence on how cardiac output monitors should be used to haemodynamic optimisation for both elective and emergency abdominal surgery is being gathered through large clinical trials

References

1.: Abbas SM, Hill AG. Systematic review of the literature for the use of oesophageal doppler monitor for fluid replacement in major abdominal surgery. Anaesthesia. 2008; 63(1):44–51 [PubMed: 18086070]
2.: Bahlmann H, Hahn RG, Nilsson L. Agreement between Pleth Variability Index and oesophageal doppler to predict fluid responsiveness. Acta Anaesthesiologica Scandinavica. 2016; 60(2):183–92 [PubMed: 26373826]
3.: Bartha E, Arfwedson C, Imnell A, Fernlund ME, Andersson LE, Kalman S. Randomized controlled trial of goal-directed haemodynamic treatment in patients with proximal femoral fracture. British Journal of Anaesthesia. 2013; 110(4):545–53 [PubMed: 23274782]
4.: Bartha E, Arfwedson C, Imnell A, Kalman S. Towards individualized perioperative, goal-directed haemodynamic algorithms for patients of advanced age: observations during a randomized controlled trial (NCT01141894). British Journal of Anaesthesia. 2016; 116(4):486–92 [PubMed: 26994228]
5.: Bartha E, Davidson T, Hommel A, Thorngren KG, Carlsson P, Kalman S. Cost-effectiveness analysis of goal-directed hemodynamic treatment of elderly hip fracture patients: before clinical research starts. Anesthesiology. 2012; 117(3):519–30 [PubMed: 22814385]
6.: Benes J, Chytra I, Altmann P, Hluchy M, Kasal E, Svitak R et al. Intraoperative fluid optimization using stroke volume variation in high risk surgical patients: results of prospective randomized study. Critical Care. 2010; 14:R118 [PMC free article: PMC2911766] [PubMed: 20553586]
7.: Benes J, Zatloukal J, Simanova A, Chytra I, Kasal E. Cost analysis of the stroke volume variation guided perioperative hemodynamic optimization - an economic evaluation of the SVVOPT trial results. BMC Anesthesiology. 2014; 14:40 [PMC free article: PMC4041635] [PubMed: 24891837]
8.: Bisgaard J, Gilsaa T, Ronholm E, Toft P. Haemodynamic optimisation in lower limb arterial surgery: room for improvement? Acta Anaesthesiologica Scandinavica. 2013; 57(2):189–98 [PubMed: 22946700]
9.: Bisgaard J, Gilsaa T, Ronholm E, Toft P. Optimising stroke volume and oxygen delivery in abdominal aortic surgery: a randomised controlled trial. Acta Anaesthesiologica Scandinavica. 2013; 57(2):178–88 [PubMed: 22897633]
10.: Bock M, Sturm T, Motsch J. Non-invasive extended hemodynamic monitoring. Reduction of circulatory risk situations. Der Anaesthesist. 2007; 56(7):656–664 [PubMed: 17476469]
11.: Bonazzi M, Gentile F, Biasi GM, Migliavacca S, Esposti D, Cipolla M et al. Impact of perioperative haemodynamic monitoring on cardiac morbidity after major vascular surgery in low risk patients. A randomised pilot trial. European Journal of Vascular and Endovascular Surgery. 2002; 23(5):445–451 [PubMed: 12027474]
12.: Brandstrup B, Svendsen PE, Rasmussen M, Belhage B, Rodt SA, Hansen B et al. Which goal for fluid therapy during colorectal surgery is followed by the best outcome: near-maximal stroke volume or zero fluid balance? British Journal of Anaesthesia. 2012; 109(2):191–9 [PubMed: 22710266]
13.: Buhre W, Weyland A, Kazmaier S, Hanekop GG, Baryalei MM, Sydow M et al. Comparison of cardiac output assessed by pulse-contour analysis and thermodilution in patients undergoing minimally invasive direct coronary artery bypass grafting. Journal of Cardiothoracic and Vascular Anesthesia. 1999; 13(4):437–40 [PubMed: 10468257]
14.: Bundgaard-Nielsen M, Jans O, Muller RG, Korshin A, Ruhnau B, Bie P et al. Does goal-directed fluid therapy affect postoperative orthostatic intolerance?: A randomized trial. Anesthesiology. 2013; 119(4):813–23 [PubMed: 23756453]
15.: Calvo-Vecino JM, Ripolles-Melchor J, Mythen MG, Casans-Frances R, Balik A, Artacho JP et al. Effect of goal-directed haemodynamic therapy on postoperative complications in low-moderate risk surgical patients: a multicentre randomised controlled trial (FEDORA trial). British Journal of Anaesthesia. 2018; 120(4):734–744 [PubMed: 29576114]
16.: Cecconi M, Fasano N, Langiano N, Divella M, Costa MG, Rhodes A et al. Goal-directed haemodynamic therapy during elective total hip arthroplasty under regional anaesthesia. Critical Care. 2011; 15:R132 [PMC free article: PMC3218998] [PubMed: 21624138]
17.: Challand C, Struthers R, Sneyd JR, Erasmus PD, Mellor N, Hosie KB et al. Randomized controlled trial of intraoperative goal-directed fluid therapy in aerobically fit and unfit patients having major colorectal surgery. British Journal of Anaesthesia. 2012; 108(1):53–62 [PubMed: 21873370]
18.: Chytra I, Pradl R, Bosman R, Pelnar P, Kasal E, Zidkova A. Esophageal Doppler-guided fluid management decreases blood lactate levels in multiple-trauma patients: a randomized controlled trial. Critical Care. 2007; 11:R24 [PMC free article: PMC2151901] [PubMed: 17313691]
19.: Colantonio L, Claroni C, Fabrizi L, Marcelli ME, Sofra M, Giannarelli D et al. A randomized trial of goal directed vs. standard fluid therapy in cytoreductive surgery with hyperthermic intraperitoneal chemotherapy. Journal of Gastrointestinal Surgery. 2015; 19(4):722–9 [PubMed: 25595308]
20.: Colbert S, O’Hanlon DM, Duranteau J, Ecoffey C. Cardiac output during liver transplantation. Canadian Journal of Anaesthesia. 1998; 45(2):133–8 [PubMed: 9512847]
21.: Concha PM, Mertz KV, Cortínez FL, Zúñiga DA, Pinedo MG. Transesophageal echocardiography to monitor fluid administration during the perioperative period. Revista Medica de Chile. 2011; 139(9):1157–62 [PubMed: 22215394]
22.: Conway DH, Mayall R, Abdul-Latif MS, Gilligan S, Tackaberry C. Randomised controlled trial investigating the influence of intravenous fluid titration using oesophageal Doppler monitoring during bowel surgery. Anaesthesia. 2002; 57(9):845–9 [PubMed: 12190747]
23.: Correa-Gallego C, Tan KS, Arslan-Carlon V, Gonen M, Denis SC, Langdon-Embry L et al. Goal-directed fluid therapy using stroke volume variation for resuscitation after low central venous pressure-assisted liver resection: a randomized clinical trial. Journal of the American College of Surgeons. 2015; 221(2):591–601 [PMC free article: PMC4926263] [PubMed: 26206652]
24.: de Verteuil RM, Hernandez RA, Vale L, Aberdeen Health Technology Assessment G. Economic evaluation of laparoscopic surgery for colorectal cancer. International Journal of Technology Assessment in Health Care. 2007; 23(4):464–72 [PubMed: 17937835]
25.: Dhawan R, Shahul S, Roberts JD, Smith ND, Steinberg GD, Chaney MA. Prospective, randomized clinical trial comparing use of intraoperative transesophageal echocardiography to standard care during radical cystectomy. Annals of Cardiac Anaesthesia. 2018; 21(3):255–261 [PMC free article: PMC6078029] [PubMed: 30052211]
26.: Donati A, Loggi S, Preiser JC, Orsetti G, Munch C, Gabbanelli V et al. Goal-directed intraoperative therapy reduces morbidity and length of hospital stay in high-risk surgical patients. Chest. 2007; 132(6):1817–24 [PubMed: 17925428]
27.: El Sharkawy OA, Refaat EK, Ibraheem AE, Mahdy WR, Fayed NA, Mourad WS et al. Transoesophageal doppler compared to central venous pressure for perioperative hemodynamic monitoring and fluid guidance in liver resection. Saudi Journal of Anaesthesia. 2013; 7(4):378–86 [PMC free article: PMC3858686] [PubMed: 24348287]
28.: Elgendy MA, Esmat IM, Kassim DY. Outcome of intraoperative goal-directed therapy using Vigileo/FloTrac in high-risk patients scheduled for major abdominal surgeries: A prospective randomized trial. Egyptian Journal of Anaesthesia. 2017; 33(3):263–269
29.: Feldheiser A, Pavlova V, Weimann K, Hunsicker O, Stockmann M, Koch M et al. Haemodynamic optimization by oesophageal doppler and pulse power wave analysis in liver surgery: A randomised controlled trial. PloS One. 2015; 10(7):e0132715 [PMC free article: PMC4505861] [PubMed: 26186702]
30.: Forget P, Lois F, de Kock M. Goal-directed fluid management based on the pulse oximeter-derived pleth variability index reduces lactate levels and improves fluid management. Anesthesia and Analgesia. 2010; 111(4):910–4 [PubMed: 20705785]
31.: Funk DJ, HayGlass KT, Koulack J, Harding G, Boyd A, Brinkman R. A randomized controlled trial on the effects of goal-directed therapy on the inflammatory response open abdominal aortic aneurysm repair. Critical Care. 2015; 19:247 [PMC free article: PMC4479246] [PubMed: 26062689]
32.: Gan TJ, Soppitt A, Maroof M, el-Moalem H, Robertson KM, Moretti E et al. Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology. 2002; 97(4):820–6 [PubMed: 12357146]
33.: Giglio M, Dalfino L, Puntillo F, Rubino G, Marucci M, Brienza N. Haemodynamic goal-directed therapy in cardiac and vascular surgery. A systematic review and meta-analysis. Interactive Cardiovascular and Thoracic Surgery. 2012; 15(5):878–87 [PMC free article: PMC3480598] [PubMed: 22833509]
34.: Gomez-Izquierdo JC, Feldman LS, Carli F, Baldini G. Meta-analysis of the effect of goal-directed therapy on bowel function after abdominal surgery. British Journal of Surgery. 2015; 102(6):577–89 [PubMed: 25759947]
35.: Gómez-Izquierdo JC, Trainito A, Mirzakandov D, Stein BL, Liberman S, Charlebois P et al. Goal-directed fluid therapy does not reduce primary postoperative ileus after elective laparoscopic colorectal surgery: A randomized controlled trial. Anesthesiology. 2017; 127(1):36–49 [PubMed: 28459732]
36.: Gurgel ST, Do Nascimento Jr P. Maintaining tissue perfusion in high-risk surgical patients: A systematic review of randomized clinical trials. Anesthesia and Analgesia. 2011; 112(6):1384–1391 [PubMed: 21156979]
37.: Hand WR, Stoll WD, McEvoy MD, McSwain JR, Sealy CD, Skoner JM et al. Intraoperative goal-directed hemodynamic management in free tissue transfer for head and neck cancer. Head and Neck. 2016; 38(S1):E1974–E1980 [PMC free article: PMC7179726] [PubMed: 26829494]
38.: Harten J, Crozier JE, McCreath B, Hay A, McMillan DC, McArdle CS et al. Effect of intraoperative fluid optimisation on renal function in patients undergoing emergency abdominal surgery: a randomised controlled pilot study (ISRCTN 11799696). International Journal of Surgery. 2008; 6(3):197–204 [PubMed: 18424200]
39.: Healthcare Quality Improvement Partnership Ltd. (HQIP). National Oesophago-gastric Cancer Audit 2018. London. 2018. Available from: https://www.nogca.org.uk/content/uploads/2018/09/NOGCA-2018-Annual-Report-1.pdf#page=53
40.: Joyce WP, Provan JL, Ameli FM, McEwan MM, Jelenich S, Jones DP. The role of central haemodynamic monitoring in abdominal aortic surgery. A prospective randomised study. European Journal of Vascular Surgery. 1990; 4(6):633–6 [PubMed: 2279574]
41.: Kapoor PM, Kakani M, Chowdhury U, Choudhury M, Lakshmy, Kiran U. Early goal-directed therapy in moderate to high-risk cardiac surgery patients. Annals of Cardiac Anaesthesia. 2008; 11(1):27–34 [PubMed: 18182756]
42.: Kapoor PM, Magoon R, Rawat R, Mehta Y. Perioperative utility of goal-directed therapy in high-risk cardiac patients undergoing coronary artery bypass grafting: A clinical outcome and biomarker-based study. Annals of Cardiac Anaesthesia. 2016; 19(4):638–645 [PMC free article: PMC5070323] [PubMed: 27716694]
43.: Kaufmann KB, Stein L, Bogatyreva L, Ulbrich F, Kaifi JT, Hauschke D et al. Oesophageal Doppler guided goal-directed haemodynamic therapy in thoracic surgery - a single centre randomized parallel-arm trial. British Journal of Anaesthesia. 2017; 118(6):852–861 [PubMed: 28575331]
44.: Kawahito S, Kitahata H, Kimura H, Tanaka K, Oshita S. Recurrent laryngeal nerve palsy after cardiovascular surgery: relationship to the placement of a transesophageal echocardiographic probe. Journal of Cardiothoracic and Vascular Anesthesia. 1999; 13(5):528–531 [PubMed: 10527219]
45.: Krishnamurthy B, McMurray TJ, McClean E. The peri-operative use of the oesophageal Doppler monitor in patients undergoing coronary artery revascularisation. A comparison with the continuous cardiac output monitor. Anaesthesia. 1997; 52(7):624–9 [PubMed: 9244018]
46.: Lai CW, Starkie T, Creanor S, Struthers RA, Portch D, Erasmus PD et al. Randomized controlled trial of stroke volume optimization during elective major abdominal surgery in patients stratified by aerobic fitness. British Journal of Anaesthesia. 2015; 115(4):578–89 [PubMed: 26385666]
47.: Laudicella M, Walsh B, Burns E, Smith PC. Cost of care for cancer patients in England: evidence from population-based patient-level data. British Journal of Cancer. 2016; 114:1286 [PMC free article: PMC4891510] [PubMed: 27070711]
48.: Laupland KB, Bands CJ. Utility of esophageal doppler as a minimally invasive hemodynamic monitor: a review. Canadian Journal of Anaesthesia. 2002; 49(4):393–401 [PubMed: 11927480]
49.: Lee S, Lee SH, Chang BC, Shim JK. Efficacy of goal-directed therapy using bioreactance cardiac output monitoring after valvular heart surgery. Yonsei Medical Journal. 2015; 56(4):913–20 [PMC free article: PMC4479857] [PubMed: 26069111]
50.: Legrand G, Ruscio L, Benhamou D, Pelletier-Fleury N. Goal-directed fluid therapy guided by cardiac monitoring during high-risk abdominal surgery in adult patients: Cost-effectiveness analysis of esophageal doppler and arterial pulse pressure waveform analysis. Value in Health. 2015; 18(5):605–13 [PubMed: 26297088]
51.: Lewis SR, Butler AR, Brammar A, Nicholson A, Smith AF. Perioperative fluid volume optimization following proximal femoral fracture. Cochrane Database of Systematic Reviews 2016, Issue 3. Art. No.: CD003004. DOI: 10.1002/14651858.CD003004.pub4. [PMC free article: PMC7138038] [PubMed: 26976366] [CrossRef]
52.: Lone NI, Gillies MA, Haddow C, Dobbie R, Rowan KM, Wild SH et al. Five-year mortality and hospital costs associated with surviving intensive care. American Journal of Respiratory and Critical Care Medicine. 2016; 194(2):198–208 [PMC free article: PMC5003217] [PubMed: 26815887]
53.: Lopes MR, Oliveira MA, Pereira VO, Lemos IP, Auler JO, Jr., Michard F. Goal-directed fluid management based on pulse pressure variation monitoring during high-risk surgery: a pilot randomized controlled trial. Critical Care. 2007; 11:R100 [PMC free article: PMC2556743] [PubMed: 17822565]
54.: Maeso S, Callejo D, Hernandez R, Blasco JA, Andradas E. Esophageal doppler monitoring during colorectal resection offers cost-effective improvement of hemodynamic control. Value in Health. 2011; 14(6):818–826 [PubMed: 21914501]
55.: Maheshwari K, Khanna S, Bajracharya GR, Makarova N, Riter Q, Raza S et al. A randomized trial of continuous noninvasive blood pressure monitoring during noncardiac surgery. Anesthesia and Analgesia. 2018; 127(2):424–31 [PMC free article: PMC6072385] [PubMed: 29916861]
56.: Mayer J, Boldt J, Mengistu AM, Rohm KD, Suttner S. Goal-directed intraoperative therapy based on autocalibrated arterial pressure waveform analysis reduces hospital stay in high-risk surgical patients: a randomized, controlled trial. Critical Care. 2010; 14:R18 [PMC free article: PMC2875533] [PubMed: 20156348]
57.: Mayer J, Boldt J, Poland R, Peterson A, Manecke GR, Jr. Continuous arterial pressure waveform-based cardiac output using the FloTrac/Vigileo: a review and meta-analysis. Journal of Cardiothoracic and Vascular Anesthesia. 2009; 23(3):401–6 [PubMed: 19464625]
58.: McKendry M, McGloin H, Saberi D, Caudwell L, Brady AR, Singer M. Randomised controlled trial assessing the impact of a nurse delivered, flow monitored protocol for optimisation of circulatory status after cardiac surgery. BMJ. 2004; 329(7460):258 [PMC free article: PMC498021] [PubMed: 15242867]
59.: McKenny M, Conroy P, Wong A, Farren M, Gleeson N, Walsh C et al. A randomised prospective trial of intra-operative oesophageal doppler-guided fluid administration in major gynaecological surgery. Anaesthesia. 2013; 68(12):1224–31 [PubMed: 24116747]
60.: Michard F, Giglio MT, Brienza N. Perioperative goal-directed therapy with uncalibrated pulse contour methods: Impact on fluid management and postoperative outcome. British Journal of Anaesthesia. 2017; 119(1):22–30 [PubMed: 28605442]
61.: Moonesinghe SR, Harris S, Mythen MG, Rowan KM, Haddad FS, Emberton M et al. Survival after postoperative morbidity: a longitudinal observational cohort study†. BJA: British Journal of Anaesthesia. 2014; 113(6):977–984 [PMC free article: PMC4235571] [PubMed: 25012586]
62.: Moppett IK, Rowlands M, Mannings A, Moran CG, Wiles MD. LiDCO-based fluid management in patients undergoing hip fracture surgery under spinal anaesthesia: a randomized trial and systematic review. British Journal of Anaesthesia. 2015; 114(3):444–59 [PubMed: 25500940]
63.: Mowatt G, Houston G, Hernandez R, de Verteuil R, Fraser C, Cuthbertson B et al. Systematic review of the clinical effectiveness and cost-effectiveness of oesophageal Doppler monitoring in critically ill and high-risk surgical patients. Health Technology Assessment. 2009; 13(7) [PubMed: 19200414]
64.: Mythen MG, Webb AR. Perioperative plasma volume expansion reduces the incidence of gut mucosal hypoperfusion during cardiac surgery. Archives of Surgery. 1995; 130(4):423–9 [PubMed: 7535996]
65.: National Institute for Health and Care Excellence. Developing NICE guidelines: the manual, updated 2018. London. National Institute for Health and Care Excellence, 2014. Available from: https://www.nice.org.uk/process/pmg20/chapter/introduction-and-overview [PubMed: 26677490]
66.: National Institute for Health and Clinical Excellence. CardioQ-ODM oesophageal doppler monitor. NICE medical technology guidance 3. London. National Institute for Health and Clinical Excellence (NICE), 2011. Available from: http://guidance.nice.org.uk/MTG3
67.: NCT. Echocardiography management for patients requiring care for non-cardiac surgery [NCT01050361]. 2010. Available from: Https://clinicaltrials.gov/show/nct01050361 Last accessed: 05/04/2019
68.: NCT. Fluid management based on Pleth Variability Index (PVI) monitoring during high-risk surgery. 2013. Available from: Https://clinicaltrials.gov/show/nct01788293 Last accessed: 05/04/19
69.: NCT. Trial comparing routine intraoperative transesophageal echocardiography during radical cystectomy to standard of care. 2017. Available from: Https://clinicaltrials.gov/show/nct03058250 Last accessed: 05/04/19
70.: NELA Project Team. Fifth patient report of the national emergency laparotomy audit. London. 2019. Available from: https://www.hqip.org.uk/wp-content/uploads/2019/12/Ref-132-The-Fifth-Patient-Report-of-the-NELA-2019-Full-Patient-Report.pdf
71.: Noblett SE, Snowden CP, Shenton BK, Horgan AF. Randomized clinical trial assessing the effect of doppler-optimized fluid management on outcome after elective colorectal resection. British Journal of Surgery. 2006; 93(9):1069–76 [PubMed: 16888706]
72.: Organisation for Economic Co-operation and Development (OECD). Purchasing power parities (PPP). Available from: http://www.oecd.org/sdd/prices-ppp/ Last accessed: 17/06/2019
73.: Owall A, Stahl L, Settergren G. Incidence of sore throat and patient complaints after intraoperative transesophageal echocardiography during cardiac surgery. Journal of Cardiothoracic and Vascular Anesthesia. 1992; 6(1):15–6 [PubMed: 1543845]
74.: Pavlovic G, Diaper J, Ellenberger C, Frei A, Bendjelid K, Bonhomme F et al. Impact of early haemodynamic goal-directed therapy in patients undergoing emergency surgery: an open prospective, randomised trial. Journal of Clinical Monitoring and Computing. 2016; 30(1):87–99 [PubMed: 25851818]
75.: Pearse RM, Harrison DA, MacDonald N, Gillies MA, Blunt M, Ackland G et al. Effect of a perioperative, cardiac output-guided hemodynamic therapy algorithm on outcomes following major gastrointestinal surgery: a randomized clinical trial and systematic review. JAMA. 2014; 311(21):2181–90 [PubMed: 24842135]
76.: Peng K, Li J, Cheng H, Ji FH. Goal-directed fluid therapy based on stroke volume variations improves fluid management and gastrointestinal perfusion in patients undergoing major orthopedic surgery. Medical Principles and Practice. 2014; 23(5):413–20 [PMC free article: PMC5586912] [PubMed: 24994571]
77.: Pestana D, Espinosa E, Eden A, Najera D, Collar L, Aldecoa C et al. Perioperative goal-directed hemodynamic optimization using noninvasive cardiac output monitoring in major abdominal surgery: a prospective, randomized, multicenter, pragmatic trial: POEMAS Study (PeriOperative goal-directed thErapy in Major Abdominal Surgery). Anesthesia and Analgesia. 2014; 119(3):579–87 [PubMed: 25010820]
78.: Phan TD, D’Souza B, Rattray MJ, Johnston MJ, Cowie BS. A randomised controlled trial of fluid restriction compared to oesophageal doppler-guided goal-directed fluid therapy in elective major colorectal surgery within an enhanced recovery after surgery program. Anaesthesia and Intensive Care. 2014; 42(6):752–60 [PubMed: 25342408]
79.: Picard J, Bedague D, Bouzat P, Ollinet C, Albaladejo P, Bosson JL et al. Oesophageal doppler to optimize intraoperative haemodynamics during prone position. A randomized controlled trial. Anaesthesia Critical Care & Pain Medicine. 2016; 35(4):255–60 [PubMed: 27085221]
80.: Pillai P, McEleavy I, Gaughan M, Snowden C, Nesbitt I, Durkan G et al. A double-blind randomized controlled clinical trial to assess the effect of Doppler optimized intraoperative fluid management on outcome following radical cystectomy. Journal of Urology. 2011; 186(6):2201–6 [PubMed: 22014804]
81.: Polonen P, Ruokonen E, Hippelainen M, Poyhonen M, Takala J. A prospective, randomized study of goal-oriented hemodynamic therapy in cardiac surgical patients. Anesthesia and Analgesia. 2000; 90(5):1052–9 [PubMed: 10781452]
82.: Poso T, Winso O, Aroch R, Kesek D. Perioperative fluid guidance with transthoracic echocardiography and pulse-contour device in morbidly obese patients. Obesity Surgery. 2014; 24(12):2117–25 [PubMed: 24902655]
83.: Ramsingh D. Evaluation of clinical impact of non-invasive hemodynamic monitoring to optimize perioperative care. 2016. Available from: Https://clinicaltrials.gov/show/nct02950649 Last accessed: 26/07/18
84.: Ramsingh DS, Sanghvi C, Gamboa J, Cannesson M, Applegate RL, 2nd. Outcome impact of goal directed fluid therapy during high risk abdominal surgery in low to moderate risk patients: a randomized controlled trial. Journal of Clinical Monitoring and Computing. 2013; 27(3):249–257 [PubMed: 23264068]
85.: Ratti F, Cipriani F, Reineke R, Catena M, Paganelli M, Comotti L et al. Intraoperative monitoring of stroke volume variation versus central venous pressure in laparoscopic liver surgery: a randomized prospective comparative trial. HPB. 2016; 18(2):136–144 [PMC free article: PMC4814622] [PubMed: 26902132]
86.: Ripolles J, Espinosa A, Martinez-Hurtado E, Abad-Gurumeta A, Casans-Frances R, Fernandez-Perez C et al. Intraoperative goal directed hemodynamic therapy in noncardiac surgery: a systematic review and meta-analysis. Brazilian Journal of Anesthesiology. 2016; 66(5):513–28 [PubMed: 27591466]
87.: Sadique Z, Harrison DA, Grieve R, Rowan KM, Pearse RM, group Os. Cost-effectiveness of a cardiac output-guided haemodynamic therapy algorithm in high-risk patients undergoing major gastrointestinal surgery. Perioperative Medicine. 2015; 4:13 [PMC free article: PMC4678585] [PubMed: 26674494]
88.: Salzwedel C, Puig J, Carstens A, Bein B, Molnar Z, Kiss K et al. Perioperative goal-directed hemodynamic therapy based on radial arterial pulse pressure variation and continuous cardiac index trending reduces postoperative complications after major abdominal surgery: a multi-center, prospective, randomized study. Critical Care. 2013; 17:R191 [PMC free article: PMC4057030] [PubMed: 24010849]
89.: Scheeren TW, Wiesenack C, Gerlach H, Marx G. Goal-directed intraoperative fluid therapy guided by stroke volume and its variation in high-risk surgical patients: a prospective randomized multicentre study. Journal of Clinical Monitoring and Computing. 2013; 27(3):225–233 [PubMed: 23558909]
90.: Schultz RJ, Whitfield GF, LaMura JJ, Raciti A, Krishnamurthy S. The role of physiologic monitoring in patients with fractures of the hip. Journal of Trauma. 1985; 25(4):309–16 [PubMed: 3989888]
91.: Senagore AJ, Emery T, Luchtefeld M, Kim D, Dujovny N, Hoedema R. Fluid management for laparoscopic colectomy: a prospective, randomized assessment of goal-directed administration of balanced salt solution or hetastarch coupled with an enhanced recovery program. Diseases of the Colon and Rectum. 2009; 52(12):1935–40 [PubMed: 19934912]
92.: Shillcutt SK, Montzingo CR, Agrawal A, Khaleel MS, Therrien SL, Thomas WR et al. Echocardiography-based hemodynamic management of left ventricular diastolic dysfunction: a feasibility and safety study. Echocardiography. 2014; 31(10):1189–98 [PubMed: 24660834]
93.: Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ. 1997; 315(7113):909–12 [PMC free article: PMC2127619] [PubMed: 9361539]
94.: Skues M, Jackson I. Ten year performance of ambulatory surgery in England. Ambulatory Surgery. 2018; 24(1):15–19
95.: Slagt C, Malagon I, Groeneveld AB. Systematic review of uncalibrated arterial pressure waveform analysis to determine cardiac output and stroke volume variation. British Journal of Anaesthesia. 2014; 112(4):626–37 [PubMed: 24431387]
96.: Smetkin AA, Kirov MY, Kuzkov VV, Lenkin AI, Eremeev AV, Slastilin VY et al. Single transpulmonary thermodilution and continuous monitoring of central venous oxygen saturation during off-pump coronary surgery. Acta Anaesthesiologica Scandinavica. 2009; 53(4):505–514 [PubMed: 19183113]
97.: Srinivasa S, Taylor MH, Sammour T, Kahokehr AA, Hill AG. Oesophageal Doppler-guided fluid administration in colorectal surgery: critical appraisal of published clinical trials. Acta Anaesthesiologica Scandinavica. 2011; 55(1):4–13 [PubMed: 21126237]
98.: Srinivasa S, Taylor MH, Singh PP, Yu TC, Soop M, Hill AG. Randomized clinical trial of goal-directed fluid therapy within an enhanced recovery protocol for elective colectomy. British Journal of Surgery. 2013; 100(1):66–74 [PubMed: 23132508]
99.: Stens J, Hering JP, van der Hoeven CWP, Boom A, Traast HS, Garmers LE et al. The added value of cardiac index and pulse pressure variation monitoring to mean arterial pressure-guided volume therapy in moderate-risk abdominal surgery (COGUIDE): a pragmatic multicentre randomised controlled trial. Anaesthesia. 2017; 72(9):1078–1087 [PubMed: 28543041]
100.: Stewart RM, Park PK, Hunt JP, McIntyre Jr RC, McCarthy J, Zarzabal LA et al. Less is more: improved outcomes in surgical patients with conservative fluid administration and central venous catheter monitoring. Journal of the American College of Surgeons. 2009; 208(5):725–735 [PubMed: 19476825]
101.: Sundaram SC, Salins SR, Nandha Kumar A, Korula G. Intra-operative fluid management in adult neurosurgical patients undergoing intracranial tumour surgery: randomised control trial comparing pulse pressure variance (PPV) and central venous pressure (CVP). Journal of Clinical and Diagnostic Research. 2016; 10(5):UC01–UC05 [PMC free article: PMC4948505] [PubMed: 27437329]
102.: Szakmany T, Toth I, Kovacs Z, Leiner T, Mikor A, Koszegi T et al. Effects of volumetric vs. pressure-guided fluid therapy on postoperative inflammatory response: a prospective, randomized clinical trial. Intensive Care Medicine. 2005; 31(5):656–63 [PubMed: 15812629]
103.: Szturz P, Folwarczny P, Kula R, Neiser J, Sevcik P, Benes J. Multi-parametric functional hemodynamic optimization improves postsurgical outcome after intermediate risk open gastrointestinal surgery: a randomized controlled trial. Minerva Anestesiologica. 2019; 85(3):244–254 [PubMed: 29756693]
104.: Valentine RJ, Duke ML, Inman MH, Grayburn PA, Hagino RT, Kakish HB et al. Effectiveness of pulmonary artery catheters in aortic surgery: a randomized trial. Journal of Vascular Surgery. 1998; 27(2):203–11; discussion 211-2 [PubMed: 9510275]
105.: Van der Linden PJ, Dierick A, Wilmin S, Bellens B, Hert SG. A randomized controlled trial comparing an intraoperative goal-directed strategy with routine clinical practice in patients undergoing peripheral arterial surgery. European Journal of Anaesthesiology. 2010; 27(9):788–793 [PubMed: 20613538]
106.: Venn R, Steele A, Richardson P, Poloniecki J, Grounds M, Newman P. Randomized controlled trial to investigate influence of the fluid challenge on duration of hospital stay and perioperative morbidity in patients with hip fractures. British Journal of Anaesthesia. 2002; 88(1):65–71 [PubMed: 11881887]
107.: Wakeling HG, McFall MR, Jenkins CS, Woods WG, Miles WF, Barclay GR et al. Intraoperative oesophageal doppler guided fluid management shortens postoperative hospital stay after major bowel surgery. British Journal of Anaesthesia. 2005; 95(5):634–42 [PubMed: 16155038]
108.: Walsh SR, Tang T, Bass S, Gaunt ME. Doppler-guided intra-operative fluid management during major abdominal surgery: systematic review and meta-analysis. International Journal of Clinical Practice. 2008; 62(3):466–70 [PubMed: 18031528]
109.: Warnakulasuriya SR, Davies SJ, Wilson RJ, Yates DR. Comparison of esophageal Doppler and plethysmographic variability index to guide intraoperative fluid therapy for low-risk patients undergoing colorectal surgery. Journal of Clinical Anesthesia. 2016; 34:600–8 [PubMed: 27687456]
110.: Wetterslev M, Moller-Sorensen H, Johansen RR, Perner A. Systematic review of cardiac output measurements by echocardiography vs. thermodilution: the techniques are not interchangeable. Intensive Care Medicine. 2016; 42(8):1223–33 [PubMed: 26932349]
111.: Wiles MD, Whiteley WJ, Moran CG, Moppett IK. The use of LiDCO based fluid management in patients undergoing hip fracture surgery under spinal anaesthesia: neck of femur optimisation therapy - targeted stroke volume (NOTTS): study protocol for a randomized controlled trial. Trials. 2011; 12(1):213 [PMC free article: PMC3191355] [PubMed: 21955538]
112.: Xu H, Shu SH, Wang D, Chai XQ, Xie YH, Zhou WD. Goal-directed fluid restriction using stroke volume variation and cardiac index during one-lung ventilation: A randomized controlled trial. Journal of Thoracic Disease. 2017; 9(9):2992–3004 [PMC free article: PMC5708410] [PubMed: 29221272]
113.: Yu Y, Dong J, Xu Z, Shen H, Zheng J. Pleth variability index-directed fluid management in abdominal surgery under combined general and epidural anesthesia. Journal of Clinical Monitoring and Computing. 2015; 29(1):47–52 [PubMed: 24557584]
114.: Zakhaleva J, Tam J, Denoya PI, Bishawi M, Bergamaschi R. The impact of intravenous fluid administration on complication rates in bowel surgery within an enhanced recovery protocol: a randomized controlled trial. Colorectal Disease. 2013; 15(7):892–9 [PubMed: 23905554]
115.: Zeng K, Li Y, Liang M, Gao Y, Cai H, Lin C. The influence of goal-directed fluid therapy on the prognosis of elderly patients with hypertension and gastric cancer surgery. Drug Design, Development and Therapy. 2014; 8:2113–9 [PMC free article: PMC4218889] [PubMed: 25378913]
116.: Zhang J, Chen CQ, Lei XZ, Feng ZY, Zhu SM. Goal-directed fluid optimization based on stroke volume variation and cardiac index during one-lung ventilation in patients undergoing thoracoscopy lobectomy operations: a pilot study. Clinics. 2013; 68(7):1065–70 [PMC free article: PMC3715019] [PubMed: 23917675]
117.: Zhang J, Qiao H, He Z, Wang Y, Che X, Liang W. Intraoperative fluid management in open gastrointestinal surgery: goal-directed versus restrictive. Clinics. 2012; 67(10):1149–55 [PMC free article: PMC3460017] [PubMed: 23070341]
118.: Zheng H, Guo H, Ye JR, Chen L, Ma HP. Goal-directed fluid therapy in gastrointestinal surgery in older coronary heart disease patients: randomized trial. World Journal of Surgery. 2013; 37(12):2820–9 [PubMed: 24048581]
119.: Zollner C, Goetz AE, Weis M, Morstedt K, Pichler B, Lamm P et al. Continuous cardiac output measurements do not agree with conventional bolus thermodilution cardiac output determination. Canadian Journal of Anaesthesia. 2001; 48(11):1143–7 [PubMed: 11744592]

Appendices

Appendix A. Review protocols

Table 13. Review protocol: Cardiac output monitoring (PDF, 198K)

Table 14Health economic review protocol

Review question	All questions – health economic evidence
Objectives	To identify health economic studies relevant to any of the review questions.
Search criteria	Populations, interventions and comparators must be as specified in the clinical review protocol above. Studies must be of a relevant health economic study design (cost–utility analysis, cost-effectiveness analysis, cost–benefit analysis, cost–consequences analysis, comparative cost analysis). Studies must not be a letter, editorial or commentary, or a review of health economic evaluations. (Recent reviews will be ordered although not reviewed. The bibliographies will be checked for relevant studies, which will then be ordered.) Unpublished reports will not be considered unless submitted as part of a call for evidence. Studies must be in English.
Search strategy	A health economic study search will be undertaken using population-specific terms and a health economic study filter – see appendix B below.
Review strategy	Studies not meeting any of the search criteria above will be excluded. Studies published before 2003, abstract-only studies and studies from non-OECD countries or the USA will also be excluded. Each remaining study will be assessed for applicability and methodological limitations using the NICE economic evaluation checklist which can be found in appendix H of Developing NICE guidelines: the manual (2014).⁶⁵ Inclusion and exclusion criteria If a study is rated as both ‘Directly applicable’ and with ‘Minor limitations’ then it will be included in the guideline. A health economic evidence table will be completed and it will be included in the health economic evidence profile. If a study is rated as either ‘Not applicable’ or with ‘Very serious limitations’ then it will usually be excluded from the guideline. If it is excluded then a health economic evidence table will not be completed and it will not be included in the health economic evidence profile. If a study is rated as ‘Partially applicable’, with ‘Potentially serious limitations’ or both then there is discretion over whether it should be included. Where there is discretion The health economist will make a decision based on the relative applicability and quality of the available evidence for that question, in discussion with the guideline committee if required. The ultimate aim is to include health economic studies that are helpful for decision-making in the context of the guideline and the current NHS setting. If several studies are considered of sufficiently high applicability and methodological quality that they could all be included, then the health economist, in discussion with the committee if required, may decide to include only the most applicable studies and to selectively exclude the remaining studies. All studies excluded on the basis of applicability or methodological limitations will be listed with explanation in the excluded health economic studies appendix below. The health economist will be guided by the following hierarchies. Setting: UK NHS (most applicable). OECD countries with predominantly public health insurance systems (for example, France, Germany, Sweden). OECD countries with predominantly private health insurance systems (for example, Switzerland). Studies set in non-OECD countries or in the USA will be excluded before being assessed for applicability and methodological limitations. Health economic study type: Cost–utility analysis (most applicable). Other type of full economic evaluation (cost–benefit analysis, cost-effectiveness analysis, cost–consequences analysis). Comparative cost analysis. Non-comparative cost analyses including cost-of-illness studies will be excluded before being assessed for applicability and methodological limitations. Year of analysis: The more recent the study, the more applicable it will be. Studies published in 2003 or later but that depend on unit costs and resource data entirely or predominantly from before 2003 will be rated as ‘Not applicable’. Studies published before 2003 will be excluded before being assessed for applicability and methodological limitations. Quality and relevance of effectiveness data used in the health economic analysis: The more closely the clinical effectiveness data used in the health economic analysis match with the outcomes of the studies included in the clinical review the more useful the analysis will be for decision-making in the guideline. For example, economic evaluations based on observational studies will be excluded, when the clinical review is only looking for RCTs,

Appendix B. Literature search strategies

The literature searches for this review are detailed below and complied with the methodology outlined in Developing NICE guidelines: the manual 2014, updated 2018.⁶⁵

For more detailed information, please see the Methodology Review.

B.1. Clinical search literature search strategy

Searches were constructed using a PICO framework where population (P) terms were combined with Intervention (I) and in some cases Comparison (C) terms. Outcomes (O) are rarely used in search strategies for interventions as these concepts may not be well described in title, abstract or indexes and therefore difficult to retrieve. Search filters were applied to the search where appropriate.

Table 15Database date parameters and filters used

Database

Dates searched

Search filter used

Medline (OVID)

1946 – 30 May 2019

Exclusions

Randomised controlled trials

Systematic review studies

Embase (OVID)

1974 – 30 May 2019

Exclusions

Randomised controlled trials

Systematic review studies

The Cochrane Library (Wiley)

Cochrane Reviews to 2019 Issue 5 of 12

CENTRAL to 2019 Issue 5 of 12

DARE, and NHSEED to 2015 Issue 2 of 4

HTA to 2016 Issue 4 of 4

None

Medline (Ovid) search terms

1.	Intraoperative Care/ or exp Intraoperative Period/ or exp Perioperative Nursing/ or exp Monitoring, Intraoperative/
2.	((intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
3.	((care* or caring or treat* or nurs* or recover* or monitor) adj3 during adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
4.	or/1-3
5.	limit 4 to English language
6.	(exp child/ or exp pediatrics/ or exp infant/) not (exp adolescent/ or exp adult/ or exp middle age/ or exp aged/)
7.	5 not 6
8.	letter/
9.	editorial/
10.	news/
11.	exp historical article/
12.	Anecdotes as Topic/
13.	comment/
14.	case report/
15.	Intraoperative Care/ or exp Intraoperative Period/ or exp Perioperative Nursing/ or exp Monitoring, Intraoperative/
16.	((intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
17.	((care* or caring or treat* or nurs* or recover* or monitor) adj3 during adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
18.	or/1-3
19.	limit 4 to English language
20.	(exp child/ or exp pediatrics/ or exp infant/) not (exp adolescent/ or exp adult/ or exp middle age/ or exp aged/)
21.	5 not 6
22.	letter/
23.	editorial/
24.	news/
25.	exp historical article/
26.	Anecdotes as Topic/
27.	comment/
28.	case report/
29.	(letter or comment*).ti.
30.	or/8-15
31.	randomized controlled trial/ or random*.ti,ab.
32.	16 not 17
33.	animals/ not humans/
34.	exp Animals, Laboratory/
35.	exp Animal Experimentation/
36.	exp Models, Animal/
37.	exp Rodentia/
38.	(rat or rats or mouse or mice).ti.
39.	or/18-24
40.	7 not 25
41.	exp Echocardiography/
42.	Cardiography, Impedance/
43.	((oesophageal or esophageal or intra?esophageal or trans?esophageal) adj5 (echocardiogra* or doppler)).ti,ab.
44.	TEE.ti,ab.
45.	Plethysmography, Impedance/
46.	((bioimpedance* or impedance) adj (cardiograp or plethysmogra* or phlebogra*)).ti,ab.
47.	((thoracic or transthoracic) adj electric* bioimpedance*).ti,ab.
48.	Pulse Wave Analysis/
49.	((pulse* or arterial) adj3 (contour or power or wave*)).ti,ab.
50.	Dye Dilution Technique/
51.	((dye or indicator or lithium) adj3 dilut*).ti,ab.
52.	(electric* adj (cardiometry or velocimetry)).ti,ab.
53.	Pressure recording analy*.ti,ab.
54.	bioreactance*.ti,ab.
55.	Hemodynamic Monitoring/
56.	(h?emodynamic adj3 (output* or index or monitor* or measur* or record* or reading* or track* or assess*)).ti,ab.
57.	exp Cardiac Output/
58.	(cardiac adj3 (output* or index or monitor* or measur* or record* or reading* or track* or assess*)).ti,ab.
59.	or/27-44
60.	26 and 45
61.	randomized controlled trial.pt.
62.	controlled clinical trial.pt.
63.	randomi#ed.ab.
64.	placebo.ab.
65.	randomly.ab.
66.	clinical trials as topic.sh.
67.	trial.ti.
68.	or/47-53
69.	Meta-Analysis/
70.	Meta-Analysis as Topic/
71.	(meta analy* or metanaly* or metaanaly* or meta regression).ti,ab.
72.	((systematic* or evidence) adj2 (review or overview*)).ti,ab.
73.	(reference list* or bibliograph* or hand search* or manual search* or relevant journals).ab.
74.	(search strategy or search criteria or systematic search or study selection or data extraction).ab.
75.	(search* adj4 literature).ab.
76.	(medline or pubmed or cochrane or embase or psychlit or psyclit or psychinfo or psycinfo or cinahl or science citation index or bids or cancerlit).ab.
77.	cochrane.jw.
78.	((multiple treatment* or indirect or mixed) adj2 comparison*).ti,ab.
79.	or/55-64
80.	46 and (54 or 65)

Embase (Ovid) search terms

1.	peroperative care/ or intraoperative period/ or perioperative nursing/ or surgical patient/ or *intraoperative monitoring/
2.	((intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
3.	((care* or caring or treat* or nurs* or recover* or monitor) adj3 during adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
4.	or/1-3
5.	limit 4 to English language
6.	(exp child/ or exp pediatrics/) not (exp adult/ or exp adolescent/)
7.	5 not 6
8.	letter.pt. or letter/
9.	note.pt.
10.	editorial.pt.
11.	case report/ or case study/
12.	(letter or comment*).ti.
13.	or/8-12
14.	randomized controlled trial/ or random*.ti,ab.
15.	13 not 14
16.	animal/ not human/
17.	nonhuman/
18.	exp Animal Experiment/
19.	exp Experimental Animal/
20.	animal model/
21.	exp Rodent/
22.	(rat or rats or mouse or mice).ti.
23.	or/15-22
24.	7 not 23
25.	exp echocardiography/
26.	impedance cardiography/
27.	((oesophageal or esophageal or intra?esophageal or trans?esophageal) adj5 (echocardiogra* or doppler)).ti,ab.
28.	TEE.ti,ab.
29.	impedance plethysmography/
30.	((bioimpedance* or impedance) adj (cardiograp or plethysmogra* or phlebogra*)).ti,ab.
31.	((thoracic or transthoracic) adj electric* bioimpedance*).ti,ab.
32.	pulse wave/
33.	((pulse* or arterial) adj3 (contour or power or wave*)).ti,ab.
34.	dye dilution curve/
35.	((dye or indicator or lithium) adj3 dilut*).ti,ab.
36.	(electric* adj (cardiometry or velocimetry)).ti,ab.
37.	Pressure recording analy*.ti,ab.
38.	bioreactance*.ti,ab.
39.	hemodynamic monitoring/
40.	(h?emodynamic adj3 (output* or index or monitor* or measur* or record* or reading* or track* or assess*)).ti,ab.
41.	heart output/
42.	(cardiac adj3 (output* or index or monitor* or measur* or record* or reading* or track* or assess*)).ti,ab.
43.	or/25-42
44.	24 and 43
45.	random*.ti,ab.
46.	factorial*.ti,ab.
47.	(crossover* or cross over*).ti,ab.
48.	((doubl* or singl) adj blind).ti,ab.
49.	(assign* or allocat* or volunteer* or placebo*).ti,ab.
50.	crossover procedure/
51.	single blind procedure/
52.	randomized controlled trial/
53.	double blind procedure/
54.	or/45-53
55.	systematic review/
56.	Meta-Analysis/
57.	(meta analy* or metanaly* or metaanaly* or meta regression).ti,ab.
58.	((systematic* or evidence) adj2 (review or overview*)).ti,ab.
59.	(reference list* or bibliograph* or hand search* or manual search* or relevant journals).ab.
60.	(search strategy or search criteria or systematic search or study selection or data extraction).ab.
61.	(search* adj4 literature).ab.
62.	(medline or pubmed or cochrane or embase or psychlit or psyclit or psychinfo or psycinfo or cinahl or science citation index or bids or cancerlit).ab.
63.	cochrane.jw.
64.	((multiple treatment* or indirect or mixed) adj2 comparison*).ti,ab.
65.	or/55-64
66.	44 and (54 or 65)

Cochrane Library (Wiley) search terms

#1.	MeSH descriptor: [Intraoperative Care] this term only
#2.	MeSH descriptor: [Intraoperative Period] this term only
#3.	MeSH descriptor: [Perioperative Nursing] this term only
#4.	(or #1-#3)
#5.	((perioperative* or peri-operative* or intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) near/3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)):ti,ab
#6.	((care* or caring or treat* or nurs* or recover* or monitor) near/3 (during) near/3 (surg or operat* or anaesthes* or anesthes*)):ti,ab
#7.	(or #4-#6)
#8.	MeSH descriptor: [Echocardiography] explode all trees
#9.	MeSH descriptor: [Cardiography, Impedance] explode all trees
#10.	((oesophageal or esophageal or intraesophageal or transesophageal) near/5 (echocardiogra* or doppler)):ti,ab
#11.	TEE:ti,ab
#12.	MeSH descriptor: [Plethysmography, Impedance] explode all trees
#13.	((bioimpedance* or impedance) near/1 (cardiograp or plethysmogra* or phlebogra*)):ti,ab
#14.	((thoracic or transthoracic) near/1 electric* bioimpedance*):ti,ab
#15.	MeSH descriptor: [Pulse Wave Analysis] explode all trees
#16.	((pulse* or arterial) near/3 (contour or power or wave*)):ti,ab
#17.	MeSH descriptor: [Dye Dilution Technique] explode all trees
#18.	((dye or indicator or lithium) near/3 dilut*):ti,ab
#19.	(electric* near/1 (cardiometry or velocimetry)):ti,ab
#20.	(Pressure recording analy*):ti,ab
#21.	bioreactance*:ti,ab
#22.	MeSH descriptor: [Hemodynamic Monitoring] explode all trees
#23.	(hemodynamic near/3 (output or index or monitor* or measur* or record* or reading* or track* or assess*)):ti,ab
#24.	MeSH descriptor: [Cardiac Output] explode all trees
#25.	(cardiac near/3 (output* or index or monitor* or measur* or record* or reading* or track* or assess*)):ti,ab
#26.	(or #8-#25)
#27.	#7 and #26

B.2. Health Economics literature search strategy

Health economic evidence was identified by conducting a broad search relating to the perioperative care population in NHS Economic Evaluation Database (NHS EED – this ceased to be updated after March 2015) and the Health Technology Assessment database (HTA) with no date restrictions. NHS EED and HTA databases are hosted by the Centre for Research and Dissemination (CRD). Additional health economics searches were run on Medline and Embase.

Table 16Database date parameters and filters used

Database

Dates searched

Search filter used

Medline

2014 – 30 May 2019

Exclusions

Health economics studies

Embase

2014 – 30 May 2019

Exclusions

Health economics studies

Centre for Research and Dissemination (CRD)

HTA - Inception – 02 May 2019

NHSEED - Inception to 02 May 2019

None

Medline (Ovid) search terms

1.	exp Preoperative Care/ or exp Perioperative Care/ or exp Perioperative Period/ or exp Perioperative Nursing/
2.	((pre-operative* or preoperative* or preop* or pre-op* or pre-surg* or presurg) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
3.	((perioperative* or peri-operative* or intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
4.	((postoperative* or postop* or post-op* or post-surg* or postsurg) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
5.	((care* or caring or treat* or nurs* or recover* or monitor) adj3 (before or prior or advance or during or after) adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
6.	1 or 2 or 3 or 4 or 5
7.	(intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat* or perioperat* or peri-operat*).ti,ab.
8.	((during or duration) adj3 (surg* or operat* or anaesthes* or anesthes*)).ti,ab.
9.	7 or 8
10.	postoperative care/ or exp Postoperative Period/ or exp Perioperative nursing/
11.	(postop* or post-op* or post-surg* or postsurg* or perioperat* or peri-operat*).ti,ab.
12.	(after adj3 (surg* or operat* or anaesthes* or anesthes*)).ti,ab.
13.	(post adj3 (operat* or anaesthes* or anesthes*)).ti,ab.
14.	10 or 11 or 12 or 13
15.	exp Preoperative Care/ or Preoperative Period/
16.	(pre-operat* or preoperat* or pre-surg* or presurg*).ti,ab.
17.	((before or prior or advance or pre or prepar) adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
18.	15 or 16 or 17
19.	6 or 9 or 14 or 18
20.	letter/
21.	editorial/
22.	news/
23.	exp historical article/
24.	Anecdotes as Topic/
25.	comment/
26.	case report/
27.	(letter or comment*).ti.
28.	or/20-27
29.	randomized controlled trial/ or random*.ti,ab.
30.	28 not 29
31.	animals/ not humans/
32.	exp Animals, Laboratory/
33.	exp Animal Experimentation/
34.	exp Models, Animal/
35.	exp Rodentia/
36.	(rat or rats or mouse or mice).ti.
37.	or/30-36
38.	19 not 37
39.	limit 38 to English language
40.	(exp child/ or exp pediatrics/ or exp infant/) not (exp adolescent/ or exp adult/ or exp middle age/ or exp aged/)
41.	39 not 40
42.	economics/
43.	value of life/
44.	exp “costs and cost analysis”/
45.	exp Economics, Hospital/
46.	exp Economics, medical/
47.	Economics, nursing/
48.	economics, pharmaceutical/
49.	exp “Fees and Charges”/
50.	exp budgets/
51.	budget*.ti,ab.
52.	cost*.ti.
53.	(economic* or pharmaco?economic*).ti.
54.	(price* or pricing*).ti,ab.
55.	(cost* adj2 (effectiv* or utilit* or benefit* or minimi* or unit* or estimat* or variable*)).ab.
56.	(financ* or fee or fees).ti,ab.
57.	(value adj2 (money or monetary)).ti,ab.
58.	or/42-57
59.	and 58

Embase (Ovid) search terms

1.	preoperative period/ or intraoperative period/ or postoperative period/ or perioperative nursing/ or *surgical patient/
2.	((pre-operative* or preoperative* or preop* or pre-op* or pre-surg* or presurg) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
3.	((perioperative* or peri-operative* or intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)).ti,ab.
4.	((care* or caring or treat* or nurs* or recover* or monitor) adj3 (before or prior or advance or during or after) adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
5.	1 or 2 or 3 or 4
6.	peroperative care/ or exp peroperative care/ or exp perioperative nursing/
7.	(intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat* or perioperat* or peri-operat*).ti,ab.
8.	((during or duration) adj3 (surg* or operat* or anaesthes* or anesthes*)).ti,ab.
9.	6 or 7 or 8
10.	postoperative care/ or exp postoperative period/ or perioperative nursing/
11.	(postop* or post-op* or post-surg* or postsurg* or perioperat* or peri-operat*).ti,ab.
12.	(after adj3 (surg* or operat* or anaesthes* or anesthes*)).ti,ab.
13.	(post adj3 (operat* or anaesthes* or anesthes*)).ti,ab.
14.	10 or 11 or 12 or 13
15.	exp preoperative care/ or preoperative period/
16.	(pre-operat* or preoperat* or pre-surg* or presurg*).ti,ab.
17.	((before or prior or advance or pre or prepar) adj3 (surg or operat* or anaesthes* or anesthes*)).ti,ab.
18.	15 or 16 or 17
19.	5 or 9 or 14 or 18
20.	letter.pt. or letter/
21.	note.pt.
22.	editorial.pt.
23.	case report/ or case study/
24.	(letter or comment*).ti.
25.	or/20-24
26.	randomized controlled trial/ or random*.ti,ab.
27.	25 not 26
28.	animal/ not human/
29.	nonhuman/
30.	exp Animal Experiment/
31.	exp Experimental Animal/
32.	animal model/
33.	exp Rodent/
34.	(rat or rats or mouse or mice).ti.
35.	or/27-34
36.	19 not 35
37.	limit 36 to English language
38.	(exp child/ or exp pediatrics/) not (exp adult/ or exp adolescent/)
39.	37 not 38
40.	health economics/
41.	exp economic evaluation/
42.	exp health care cost/
43.	exp fee/
44.	budget/
45.	funding/
46.	budget*.ti,ab.
47.	cost*.ti.
48.	(economic* or pharmaco?economic*).ti.
49.	(price* or pricing*).ti,ab.
50.	(cost* adj2 (effectiv* or utilit* or benefit* or minimi* or unit* or estimat* or variable*)).ab.
51.	(financ* or fee or fees).ti,ab.
52.	(value adj2 (money or monetary)).ti,ab.
53.	or/40-52
54.	39 and 53

NHS EED and HTA (CRD) search terms

#1.	MeSH DESCRIPTOR Preoperative Care EXPLODE ALL TREES
#2.	MeSH DESCRIPTOR Perioperative Care EXPLODE ALL TREES
#3.	MeSH DESCRIPTOR Perioperative Period EXPLODE ALL TREES
#4.	MeSH DESCRIPTOR Perioperative Nursing EXPLODE ALL TREES
#5.	(((perioperative* or peri-operative* or intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)))
#6.	(((care* or caring or treat* or nurs* or recover* or monitor) adj3 (before or prior or advance or during or after) adj3 (surg or operat* or anaesthes* or anesthes*)))
#7.	(((pre-operative* or preoperative* or preop* or pre-op* or pre-surg* or presurg) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)))
#8.	(((postoperative* or postop* or post-op* or post-surg* or postsurg) adj3 (care or caring or treat* or nurs* or monitor* or recover* or medicine)))
#9.	#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8
#10.	(* IN HTA)
#11.	(* IN NHSEED)
#12.	#9 AND #10
#13.	#9 AND #11
#14.	MeSH DESCRIPTOR Intraoperative Care EXPLODE ALL TREES
#15.	#1 OR #2 OR #3 OR #4 OR #14
#16.	((intraoperative* or intra-operative* or intrasurg* or intra-surg* or peroperat* or per-operat* or perioperat* or peri-operat*))
#17.	(((during or duration) adj3 (surg* or operat* or anaesthes* or anesthes*)))
#18.	((postop* or post-op* or post-surg* or postsurg* or perioperat* or peri-operat*))
#19.	((after adj3 (surg* or operat* or anaesthes* or anesthes*)))
#20.	((post adj3 (operat* or anaesthes* or anesthes*)))
#21.	((pre-operat* or preoperat* or pre-surg* or presurg*))
#22.	(((before or prior or advance or pre or prepar) adj3 (surg or operat* or anaesthes* or anesthes*)))
#23.	#15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22
#24.	#10 AND #23
#25.	#11 AND #23
#26.	#12 OR #13 OR #24 OR #25

Appendix C. Clinical evidence selection

Figure 1Flow chart of clinical study selection for the review of Cardiac Output Monitoring

Appendix D. Clinical evidence tables

Download PDF (590K)

Appendix E. Forest plots

E.1. Oesophageal Doppler compared to pulse contour analysis

Figure 2Patients with complications

E.2. Cardiac output monitoring compared to conventional assessment

Figure 3Mortality

Figure 4Patients with complications

Figure 5Patients with complications (POMS ≥1 (3-days))

Figure 6Patients with complications (POMS ≥1 (5-days))

Figure 7Patients with complications (POMS ≥1 (8-days))

Figure 8Length of hospital stay

Figure 9Length of stay in ICU

Figure 10Hospital readmission

Appendix F. GRADE tables

Table 17Clinical evidence profile: Oesophageal Doppler monitoring versus pulse contour analysis

Quality assessment

No of patients

Effect

Quality

Importance

No of studies

Design

Risk of bias

Inconsistency

Indirectness

Imprecision

Other considerations

Oesophageal Doppler

Pulse contour analysis

Relative (95% CI)

Absolute

Patients with complications (follow-up 8 days)

randomised trials

no serious risk of bias

no serious inconsistency

no serious indirectness

serious¹

none

6/11

(54.5%)

90%

RR 0.61 (0.34 to 1.08)

351 fewer per 1000 (from 594 fewer to 72 more)

⨁⨁⨁◯

MODERATE

CRITICAL

1: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs.

Table 18Clinical evidence profile: Cardiac output monitoring versus conventional clinical assessment

Quality assessment							No of patients		Effect		Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Non-invasive cardiac output monitoring	Conventional clinical assessment	Relative (95% CI)	Absolute	Quality	Importance
Mortality (follow-up <90 days)
12	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	very serious¹	none	26/982 (2.6%)	3.3%	RR 0.87 (0.53 to 1.42)	4 fewer per 1000 (from 16 fewer to 14 more)	⨁⨁◯◯ LOW	CRITICAL
Patients with complications (follow-up <45 days)
13	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	serious¹	none	315/1019 (30.9%)	41.6%	RR 0.75 (0.67 to 0.84)	104 fewer per 1000 (from 67 fewer to 137 fewer)	⨁⨁⨁◯ MODERATE	CRITICAL
Complications: POMS ≥1 (3-days)
1	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	serious¹	none	89/109 (81.7%)	73%	RR 1.12 (0.97 to 1.29)	88 more per 1000 (from 22 fewer to 212 more)	⨁⨁⨁◯ MODERATE	CRITICAL
Complications: POMS ≥1 (5-days)
1	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	serious¹	none	55/109 (50.5%)	48.7%	RR 1.04 (0.79 to 1.35)	19 more per 1000 (from 102 fewer to 170 more)	⨁⨁⨁◯ MODERATE	CRITICAL
Complications: POMS ≥1 (8-days)
1	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	serious¹	none	40/109 (36.7%)	28.8%	RR 1.27 (0.87 to 1.87)	78 more per 1000 (from 37 fewer to 251 more)	⨁⨁⨁◯ MODERATE	CRITICAL
Length of hospital stay (Better indicated by lower values)
8	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	no serious imprecision	none	446	495	-	MD 0.57 lower (1.12 lower to 0.03 higher)	⨁⨁⨁⨁ HIGH	IMPORTANT
Length of stay in ICU (Better indicated by lower values)
2	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	no serious imprecision	none	107	107	-	MD 0.36 lower (0.59 to 0.12 lower)	⨁⨁⨁⨁ HIGH	IMPORTANT
Readmission rate (follow-up 30-60 days)
5	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	serious¹	none	41/347 (11.8%)	9.5%	RR 1.23 (0.81 to 1.87)	22 more per 1000 (from 18 fewer to 83 more)	⨁⨁⨁◯ MODERATE	IMPORTANT

1: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs.
2: Downgraded by 1 or 2 increments because of heterogeneity, I2>50%, p=0.05, unexplained by subgroup analysis.

Appendix G. Health economic evidence selection

Figure 11Flow chart of health economic study selection for the guideline

Appendix H. Health economic evidence tables

Download PDF (261K)

Appendix I. Excluded studies

I.1. Excluded clinical studies

Table 19Studies excluded from the clinical review

Study	Exclusion reason
Abbas 2008¹	Systematic review is not relevant to review question or unclear PICO. References screened
Bahlmann 2016²	Systematic review is not relevant to review question or unclear PICO
Bartha 2016⁴	Post-hoc analysis of Bartha 2013
Benes 2010⁶	Interventions including starch bolus
Benes 2014⁷	No relevant outcomes
Bisgaard 2013⁸	Interventions including starch bolus
Bisgaard 2013⁹	Interventions including starch bolus
Bock 2007¹⁰	Not in English
Bonazzi 2002¹¹	Inappropriate comparison
Brandstrup 2012¹²	Interventions including starch bolus
Buhre 1999¹³	Inappropriate study design
Bundgaard-Nielsen 2013¹⁴	Interventions including starch bolus
Calvo-Velcino 2018¹⁵	Interventions including starch bolus
Cecconi 2011¹⁶	Interventions including starch bolus
Challand 2012¹⁷	Interventions including starch bolus
Chytra 2007¹⁸	Interventions including starch bolus
Colantonio 2015¹⁹	Interventions including starch bolus
Colbert 1998²⁰	No relevant outcomes
Concha 2011²¹	Not in English
Conway 2002²²	Interventions including starch bolus
Donati 2007²⁶	Inappropriate comparison
El Sharkawy 2013²⁷	Interventions including starch bolus
Elgendy 2017²⁸	Interventions including starch bolus
Forget 2010³⁰	Interventions including starch bolus
Funk 2015³¹	Interventions including starch bolus
Gan 2002³²	Interventions including starch bolus
Giglio 2012³³	Systematic review is not relevant to review question or unclear PICO. References screened
Gomez-Izquierdo 2015³⁴	Systematic review is not relevant to review question or unclear PICO. References screened
Gómez-Izquierdo 2017³⁵	Interventions including starch bolus
Gurgel 2011³⁶	Systematic review is not relevant to review question or unclear PICO. References screened
Harten 2008³⁸	Interventions including starch bolus
Joyce 1990⁴⁰	Inappropriate comparison
Kapoor 2008⁴¹	Incorrect interventions
Kawahito 1999⁴⁴	Inappropriate study design
Krishnamurthy 1997⁴⁵	Inappropriate study design
Laupland 2002⁴⁸	Systematic review is not relevant to review question or unclear PICO
Lee 2015⁴⁹	Interventions including starch bolus
Legrand 2015⁵⁰	Systematic review is not relevant to review question or unclear PICO. References screened
Lewis 2016⁵¹	Systematic review is not relevant to review question or unclear PICO. References screened
Linden 2010¹⁰⁵	Interventions including starch bolus
Lopes 2007⁵³	Interventions including starch bolus
Maheshwari 2018⁵⁵	Inappropriate comparison
Mayer 2009⁵⁷	Systematic review is not relevant to review question or unclear PICO
McKendry 2004⁵⁸	Inappropriate intervention
McKenny 2013⁵⁹	Interventions including starch bolus
Michard 2017⁶⁰	Systematic review is not relevant to review question or unclear PICO. References screened
Mowatt 2009⁶³	Systematic review is not relevant to review question or unclear PICO. References screened
Mythen 1995⁶⁴	Interventions including starch bolus
NCT 2010⁶⁷	Clinical trial website with no published results
NCT 2013⁶⁸	Clinical trial website with no published results
NCT 2017⁶⁹	Clinical trial website with no published results
Owall 1992⁷³	No relevant outcomes
Pavlovic 2016⁷⁴	Interventions including starch bolus
Peng 2014⁷⁶	Interventions including starch bolus
Pestana 2014⁷⁷	Interventions including starch bolus
Phan 2014⁷⁸	Interventions including starch bolus
Picard 2016⁷⁹	Interventions including starch bolus
Polonen 2000⁸¹	Inappropriate comparison
Poso 2014⁸²	Interventions including starch bolus
Ramsingh 2016⁸³	Clinical trial website with no published results
Ripolles 2016⁸⁶	Systematic review is not relevant to review question or unclear PICO. References screened
Scheeren 2013⁸⁹	Interventions including starch bolus
Schultz 1985⁹⁰	Inappropriate comparison
Senagore 2009⁹¹	Interventions including starch bolus
Sinclair 1997⁹³	Interventions including starch bolus
Slagt 2014⁹⁵	Systematic review is not relevant to review question or unclear PICO
Smetkin 2009⁹⁶	Interventions including starch bolus
Srinivasa 2011⁹⁷	Systematic review is not relevant to review question or unclear PICO. References screened
Stewart 2009¹⁰⁰	Inappropriate comparison
Sundaram 2016¹⁰¹	No relevant outcomes
Szakmany 2005¹⁰²	Interventions including starch bolus
Valentine 1998¹⁰⁴	Inappropriate comparison
Walsh 2008¹⁰⁸	Systematic review is not relevant to review question or unclear PICO. References screened
Warnakulasuriya 2016¹⁰⁹	Interventions including starch bolus
Wetterslev 2016¹¹⁰	Systematic review is not relevant to review question or unclear PICO. References screened
Wiles 2011¹¹¹	Study protocol
Xu 2017¹¹²	Interventions including starch bolus
Yu 2015¹¹³	No relevant outcomes
Zeng 2014¹¹⁵	Study has since been retracted
Zhang 2012¹¹⁷	Interventions including starch bolus
Zhang 2013¹¹⁶	Interventions including starch bolus
Zheng 2013¹¹⁸	Interventions including starch bolus
Zollner 2001¹¹⁹	Inappropriate comparison

I.2. Excluded health economic studies

Published health economic studies that met the inclusion criteria (relevant population, comparators, economic study design, published 2003 or later and not from non-OECD country or USA) but that were excluded following appraisal of applicability and methodological quality are listed below. See the health economic protocol for more details.

Table 20Studies excluded from the health economic review

Reference	Reason for exclusion
None.

Tables

Table 1PICO characteristics of review question

Population	Adults 18 years and over having major or complex or high risk surgery (based on NICE preoperative tests for elective surgery guideline categorisation) and high risk patients (based on American Society of Anesthesiologists physical status grade) undergoing any surgery.
Interventions	non-invasive cardiac output monitoring oesophageal doppler monitor trans-oesophageal echocardiography thoracic electrical bioimpedance pulse pressure waveform analysis systems based on pulse contour analysis and dye dilution
Comparisons	pulmonary artery catheter conventional clinical assessment
Outcomes	Critical outcomes: health-related quality of life mortality perioperative complications Important outcomes: length of hospital stay length of stay in intensive care unit hospital readmission
Study design	Randomised controlled trials (RCTs), systematic reviews of RCTs. Prospective cohort studies if no RCT evidence is identified.

Table 2Summary of studies included in the evidence review

Study	Intervention and comparison	Population	Outcomes	Comments
Bartha 2013³	Pulse contour analysis: Fluid challenge (3 ml kg21) with colloid administered and repeated if a 10% increase in stroke volume achieved. If no increase occurred, and if oxygen delivery was, 600 ml min21m22, then a dobutamine infusion was started at 0.2–10 mgkg21 min21. The infusion was stopped if tachycardia occurred. Further fluid challenges were given if the SV decreased by 10%. The research team administered GDT, which was discontinued at the end of the operation. N=70 Conventional clinical assessment: The attending anaesthesia team managed the routine fluid treatment. Colloids were administered before spinal anaesthesia followed by the background infusion of buffered glucose and Ringer’s acetate according to the treatment algorithm. Other fluids or vasopressor treatment (e.g. phenylephrine and ephedrine) for correction of decreasing arterial pressure were administered at the attending anaesthetist’s discretion. N=75	Patients aged ≥70 years and weight ≥40 kg who were undergoing proximal femoral fracture surgery. Median age (range): 85 years (71-101) Sweden	Length of hospital stay Complications	All patients were monitored with a lithium dilution cardiac output monitor (LiDCO). The LiDCO monitor was covered for the attending anaesthesia team.
Correa-Gallego 2015²³	Pulse contour analysis: At the completion of transection goal directed fluid management was initiated following an algorithm using a FloTrac monitor; 1:1 blood loss replacement with colloid, and albumin bolus infusions to restore SVV to a value ≤ 2 standard deviations from their baseline after induction. Crystalloid infusion was continued at 1ml/kg/hr. N=69 Conventional clinical assessment: Standard fluid management; 1:1 blood loss replacement with colloid, and crystalloid infusion at 6 ml/kg/hr of total operative time to restore the calculated insensible losses and maintenance requirements. N=66	All adult patients scheduled to undergo an open, elective liver resection. Mean age (SD): 56.5 years (13.5) USA	Mortality Length of hospital stay Readmission Complications	All patients had continuous arterial waveform monitoring from the beginning of the operation and their SVV after induction was recorded using the FloTrac sensor and EV1000 clinical platform.
Dhawan 2018²⁵	Transesophageal echocardiography: Patients had TEE used throughout the intraoperative period to assist with fluid and hemodynamic management. N=38 Conventional clinical assessment: Standardized general anaesthetic with radial artery blood pressure monitoring. N=39	Patients undergoing elective radical cystectomy for invasive bladder cancer. Mean age (SD): 67 (10) USA	Mortality Length of hospital stay Complications	While TEE was not routinely used in this group, it was allowed if requested by the general anaesthesiologist during the intraoperative period in a “rescue” role to evaluate life-threatening hemodynamic instability.
Feldheiser 2015²⁹	Oesophageal Doppler monitoring: CardioQ-ODM shown to the treating personnel and the goal-directed algorithm was performed according to the values measured by the ODM. N=11 Pulse contour analysis: LiDCOrapid was shown to the treating personnel and the algorithm was performed according to the values measured by the LiDCO. N=10 The goal-directed algorithm guides the administration of intravenous colloid solution to maintain preload, the titration of norepinephrine to maintain arterial blood pressure and if necessary the titration of enoximone or nitroglycerine to lower central venous pressure.	Patients aged at least 18 years and undergoing elective liver resection (hemihepatectomy or extended liver resection). Median age (IQR): ODM: 69 years (56-75) PPA: 52 years (41-65) Germany	Length of hospital stay Complications	Conventional care arm implemented in study; due to an unbalance in the extension of the surgical procedures with a high rate of only minor procedures the conventional group was dropped from the analysis. In each of the three allocation groups both the ODM and PPA were established. ASA ≥IV excluded
Hand 2016³⁷	Pulse contour analysis: Goal-directed haemodynamic therapy using Vigileo and EV-1000. Measures BP, SV variation, cardiac index, and systemic vascular resistance via central line. Hypotension defined as man arterial pressure <75mm Hg or >10% below baseline. Treatment algorithm dictates no action or treatment with IVF bolus, dobutamine, epinephrine or phenylephrine. N=47 Conventional clinical assessment: Standard management of hypotension, utilising only IV fluids (crystalloid and colloid). Goal BP was set as mean arterial pressure >70 or within 20% of baseline. N=47	All patients scheduled for primary free tissue transfer reconstruction with head and neck oncologic surgeons were enrolled. Mean age (SD): 58.4 years (13) USA	Length of hospital stay Length of stay in ICU Complications
Kapoor 2016⁴²	Pulse contour analysis: Goal-directed haemodynamic therapy. Received standard haemodynamic monitoring. In addition, the cardiac index and the continuous central venous oxygen saturation were monitored. A cardiac output monitoring sensor was connected to the radial arterial cannula. If the CI was <2.5 L/min/m2, CVP <6 mmHg, or SVV >10%, fluids were given intravenously until the target CVP and SVV levels were achieved. N=60 Conventional clinical assessment: Received standard haemodynamic monitoring including Electrocardiogram oxygen saturation, invasive blood pressure, central venous pressure and arterial blood gas, urine output, and EtCO₂ monitoring were common to both the groups. All patients received fluids to maintain the CVP between 6 and 8 mmHg and MAP was maintained between 90 and 105 mmHg using inotropic agents and vasodilators. N=60	Patients with a European system for cardiac operative risk evaluation ≥3 undergoing coronary artery bypass grafting. Mean age (SD): 61.2 years (5.4) India	Mortality Length of hospital stay Length of stay in ICU
Kaufmann 2017⁴³	Oesophageal Doppler monitoring: Managed according to an algorithm utilising ODM variables (i.e., mean arterial pressure (MAP), stroke volume and cardiac index). Fluid responsiveness and haemodynamic variables were reassessed at least every 15min. N=50 Conventional clinical assessment: Subjects allocated to the control group were haemodynamically managed as follows. Isolated hypotension (defined as 20% decrease in MAP below baseline or<60mmHg) was treated by a continuous infusion of norepinephrine up to a maximum rate of 0.25 mg.kg⁻¹ min⁻¹. N=50	Patients with indication for elective lung parenchyma resection. Mean age (range): 65 years (55-74) Germany	Complications Length of hospital stay
Lai 2015⁴⁶	Pulse contour analysis: Goal-directed haemodynamic therapy. A medically qualified investigator monitored patients throughout surgery with a LiDCOrapid. The concealed investigator administered warmed colloid fluid challenges with Gelofusine directed by an algorithm to achieve an SVV goal of less than 10% throughout surgery. N= 110 Conventional clinical assessment: All patients received mechanical ventilation whilst under general anaesthesia; tidal volume was not protocolised. The anaesthetist administered intraoperative crystalloid, colloid, blood products, and inotropes or vasopressors based on estimated patient requirements, losses, and standard haemodynamic variables. All participants had arterial line monitoring. Central venous pressure monitoring was permitted. Standard fluid therapy was not defined, but a general recommendation was made that perioperative fluid excess should be avoided. N= 111	Patients having major elective rectal resection or cystectomy with ileal conduit. Mean age (SD): 63 years (15) UK	Mortality Complications Length of hospital stay Readmission
Mayer 2010⁵⁶	Pulse contour analysis: Standard monitoring plus enhanced hemodynamic monitoring with the FloTrac/Vigileo device and an attempted cardiac index of at least 2.5 L·min-1·m-2. The arterial line was connected to the Vigileo monitor via the FloTrac pressure transducer. The shape of the arterial curve was checked visually for damping throughout the study period. CI, stroke volume index (SVI), as an indicator for fluid status, and stroke volume variation, (SVV) as an indicator for fluid responsiveness during mechanical ventilation and sinus rhythm, were continuously measured. Blood loss was substituted with crystalloid/colloid fluids according to an algorithm and a haemoglobin value below 8 mg dL-1 was considered to be a trigger for transfusion of packed red blood cells. N= 30 Conventional clinical assessment: Standard monitoring included electrocardiogram, invasive arterial blood pressure via right or left radial artery, CVP, pulse oximetry, temperature, inspiratory and expiratory gas concentrations. Crystalloid/colloid delivered to ensure MAP was kept between 65 and 90 mmHg, CVP between 8 and 12 mmHg and urinary output more than 0.5 mL kg-1 h-1. N= 30	Patients with an ASA status of III with two or more risk factors undergoing open major abdominal surgery (intestine resection, gastric resection, liver resection, esophageal resection, Whipple). Mean age (range): 72.5 years (68-78) USA	Mortality Complications Length of hospital stay
Moppett 2015⁶²	Pulse contour analysis: A LiDCO monitor was attached and calibrated, the use of vasoactive agents was at the discretion of the attending anaesthetist, as was target arterial pressure during surgery. Also received targeted i.v. colloid boluses using invasive pulse contour analysis continuous cardiac output monitoring to optimize SV following a predetermined algorithm. The attending anaesthetist was aware of the fluids being given. N=68 Conventional clinical assessment: A LiDCO monitor was attached and calibrated, the use of vasoactive agents was at the discretion of the attending anaesthetist, as was target arterial pressure during surgery. Operative anaesthetists were not allowed to view the LiDCO monitor for patients in the control group unless they believed that there was a strong clinical need to do so. N=62	Patients admitted through the emergency department with primary fragility hip fracture, aged over 60 who were listed for surgical repair under spinal anaesthesia. Median age (range): 85 years (63-95) UK	Mortality Complication Length of hospital stay
Noblett 2006⁷¹	Oesophageal Doppler monitoring: Routine perioperative monitoring. Crystalloid, colloid or blood products administered by the anaesthetist based on intraoperative losses and standard parameters. Received an additional colloid (Volpex) bolus to maintain a descending aortic corrected flow time of more than 0.35 seconds and further bolus given to optimise the stroke volume. Further bolus only given if SV fell by >10% or FTc fell below 0.35 seconds. N=54 Conventional clinical assessment: Routine perioperative monitoring included ECG, pulse oximetry, end-tidal carbon dioxide monitoring, and non-invasive BP monitoring. Crystalloid, colloid or blood products administered by the anaesthetist based on intraoperative losses and standard parameters. N=54	Consecutive patients undergoing colorectal resection. Mean age (SD): 64.9 years (14.6) UK	Mortality Complication Length of hospital stay Readmission	All patients had oesophageal Doppler monitors, but fluid administration for the intervention group was based solely on the Doppler-assessed parameters, following a strict algorithm. Volpex administered by separate medically qualified reviewer to maintain blinding.
Pearse 2014⁷⁵	Pulse contour analysis: Received intravenous fluid and inotropes according to a cardiac output–guided hemodynamic therapy algorithm using a cardiac output monitor (LiDCOrapid). Intravenous colloid solution was administered in 250mL boluses to achieve and maintain a maximal value of stroke volume. N=368 Conventional Clinical assessment: All patients received standard measures to maintain oxygenation, haemoglobin, core temperature, and HR. Additional fluid was administered at the discretion of the treating clinician guided by pulse rate, arterial pressure, urine output, core-peripheral temperature gradient, serum lactate, and base excess. N=366	Patients aged 50 years or older undergoing major gastrointestinal surgery. Mean age (SD): 71.8 years (8.5) UK	Mortality Complications Length of hospital stay
Pillai 2011⁸⁰	Oesophageal Doppler monitoring: Standard respiratory and cardiovascular monitoring including BP. Standard intraoperative fluids at the discretion of the consultant anaesthetists, and additional fluid from a researcher via esophageal Doppler determined protocol. Fluid given if SV increase by >10% and FTc <0.35 seconds N=32 Conventional Clinical assessment: Standard respiratory and cardiovascular monitoring including BP. Standard intraoperative fluids at the discretion of the consultant anaesthetists. N=32	Patients undergoing radical cystectomy as curative treatment for muscle invasive transitional cell carcinoma of the bladder. Mean age: 67.5 years UK	Length of hospital stay Complications	All patients had oesophageal Doppler inserted as to allow for double blinding.
Ramsingh 2013⁸⁴	Pulse contour analysis: FloTrac/Vigileo system was used. GDT patients were managed by an SVV guided protocol to maintain SVV\12 %. N=18 Conventional Clinical assessment: Control patients had fluid management guided by routine cardiovascular monitoring at the discretion of their Staff Anesthesiologist, who was blinded to SVV data. N=20	Patients scheduled for major abdominal, non-vascular surgery. Mean age (SD): 59.2 years (16.9) USA	Length of hospital stay
Ratti 2016⁸⁵	Pulse contour analysis: ECG and MAP were obtained using a radial or humeral catheterisation, pulse oxymetry and diuresis were monitored. Arterial access was connected to the FloTrac sensor of the Vigileo monitor system to measure SVV. In this group SV, CO and CI were monitored; VO2 and DO2 were calculated on the basis of blood gas analysis. The goal was to maintain SVV over 12% during resection. Fluid therapy with crystalloids was guided by SVV values to achieve a hypovolaemic state. N=45 Conventional Clinical assessment: ECG and MAP were obtained using a radial or humeral catheterization, pulse oxymetry and diuresis were monitored. CVP was measured through a CVC inserted in the internal jugular vein after the induction of general anaesthesia. In this group SvO2 was monitored as well. The goal was to maintain CVP under or equal to 5 cm H20. Fluid therapy with crystalloids was guided by CVP values to achieve a hypovolaemic state. N=45	Patients scheduled for laparoscopic liver resection (LLR) for primary or secondary liver tumours. Mean age (SD): 59.5 years (10) Italy	Morbidity Length of hospital stay Complications
Salzwedel 2013⁸⁸	Pulse contour analysis: Conventional Clinical assessment: Received basic anaesthetic monitoring. Additionally, hemodynamic therapy was guided by a predetermined algorithm accounting for pulse pressure variation, cardiac index trending and mean arterial pressure as measured by a cardiac index trending monitor (ProAQT). N=79 Conventional Clinical assessment: Received basic anaesthetic monitoring by five-lead-electrocardiogram, pulse oximetry and blood pressure cuff, at least one peripheral i.v., a central venous catheter and invasive radial arterial blood pressure monitoring. Treatment was entirely performed at the discretion of the care-giving anaesthesiologist. N=81	Patients undergoing elective abdominal surgery including general, gynecological and urological surgery with anticipated duration of surgery of more than 120 minutes or an estimated blood loss of more than 20% of blood volume, ASA classification 2 or 3, and an indication for an arterial line and central venous catheter. Mean age (SD): 64 years (17.6) Germany	Complications Length of hospital stay
Shillcutt 2014⁹²	Trans-oesophageal echocardiography: Echocardiography-guided hemodynamic management. Received hemodynamic management of crystalloid or colloid fluid based on left ventricular filling patterns on transesophageal echocardiography, according to a predetermined algorithm. N=14 Conventional Clinical assessment: Standard hemodynamic management using non-invasive blood pressure monitoring or invasive monitoring (arterial/central venous line) if so indicated by the anaesthetist. Target of keeping intraoperative BP within 10-15% of patient baseline readings. Fluids given as deemed appropriate by anaesthetist who were blinded to the study. N=14	Patients aged >65 years or aged >19 years with at least one risk factor for left ventricular diastolic dysfunction undergoing major non-cardiac surgery.	Length of hospital stay	Type of surgery: Orthopaedic: 9 General: 12 Vascular: 4 Thoracic: 3
Srinivasa 2013⁹⁸	Oesophageal Doppler monitoring: Patients randomized to GDT were treated with a weight-based bolus of colloid, permitted based on cardiac function measured by means of an oesophageal Doppler monitor (CardioQ). An algorithm based on FTc and SV dictated fluid administration. N=37 Conventional Clinical assessment: Patients were allowed to receive up to 1500 ml crystalloid solution during surgery. They were also permitted to receive a total of 500 ml succinylated gelatine colloid solution titrated by heart rate, blood pressure, urine output and invasive measures (arterial lines) when used. N=37	Patients undergoing elective open or laparoscopic colectomy for any indication. Mean age (SD): 70.5 years (14) New Zealand	Complications Length of hospital stay Readmissions
Stens 2017⁹⁹	Pulse contour analysis: Non-invasive continuous arterial blood pressure monitor was used for PPV and CI measurements in all patients. The anaesthetist was required to keep MAP > 70 mmHg, CI > 2.5 l min -1.m-2 and PPV < 12% using a predefined protocol. N=122 Conventional Clinical assessment: The attending anaesthetist was blinded to the PPV/CI values and maintained target MAP values > 70 mmHg (as measured by the Nexfin device) with intravenous fluids of any type, vasopressors and/or inotropes, based on their clinical judgement. N=122	Patients with elective moderate-risk abdominal surgery planned. Mean age (SD): 63 (12.5) The Netherlands	Mortality Complications Length of hospital stay Length of ICU stay Readmissions
Szturz 2019¹⁰³	Oesophageal Doppler monitoring: The haemodynamic optimisation in the intervention group was based on papremters acquired from the CadioQ monitor according to a predefined algorithm, N=75 Conventional Clinical assessment: The intraoperative hemodynamic care was purely on the treating anaesthetist decision based on the knowledge of standard parameters (i.e heart rate, IBP, CVP, diuresis, clinical findings). N=75	All intermediate and high-risk patients scheduled for abdominal gastrointestinal surgery with presumed length of more than 120 minutes or with expected blood loss if more than 15% of blood volume. Mean age (SD): 65.5 (10.5) Czech republic	Mortality Complications Length of hospital stay Length of stay in ICU
Venn 2002¹⁰⁶	Oesophageal Doppler monitoring: Patients received additional 200 ml gelofusine fluid challenges guided by Doppler measurements of stroke volume and corrected flow time from the investigator, in addition to any fluid given by the clinician. N=30 Conventional Clinical assessment (CVP guided): Patients received additional 200 ml gelofusine fluid challenges guided by the response of the central venous pressure to a fluid challenge from the investigator, in addition to any fluid given by the clinician. N=31 Conventional Clinical assessment: Clinicians were able to give i.v. fluid as they thought appropriate. Although central venous pressure was monitored and recorded by the investigator, the clinician was unaware of these measurements and so was unable to use them to guide therapy. The investigator gave no additional fluids in this group. N=29 Intraoperatively, all patients received i.v. crystalloid (Hartmann’s solution), colloid in the form of gelofusine, or blood to replace estimated and measured fluid losses, in an attempt to maintain heart rate and arterial pressure within 20% of pre-induction baseline levels.	Patients aged >65 years with fractured hips. Mean age (SD): 83.8 years (8.3) UK	Mortality Complications Length of hospital stay	CVP guided conventional care and conventional care groups were pooled for comparison with oesophageal Doppler monitoring group.
Wakeling 2005¹⁰⁷	Oesophageal Doppler monitoring: In addition to routine fluid management, fluids adjusted according to ODM values following a predetermined algorithm. Patients received 250 ml boluses of colloid solution. If the stroke volume increased by 10% or more but the CVP did not rise by 3 mm Hg or more, the fluid challenge was repeated. The fluid challenges of 250 ml were repeated until the stroke volume failed to rise by 10% and/or the CVP rose by 3 mm Hg or more. No further colloid fluid boluses were given until a 10% decrease in stroke volume occurred. N=67 Conventional Clinical assessment: Patients were managed using routine cardiovascular monitoring and CVP measurements. The CVP was used to guide i.v. fluid administration and was kept between 12 and 15 mm Hg. N=67	Patients requiring elective or semi-elective large bowel surgery. Median age (IQR): ODM: 69.9 years (10.2) Conventional: 69.1 years (12.3) UK	Quality of life Complications Length of stay	All patients had OD but only intervention group had fluids adjusted according to ODM values.
Zakhaleva 2013¹¹⁴	Oesophageal Doppler monitoring: Naso-oesophageal Doppler placed by anaesthesiologist. Patients received intraoperative boluses of water and electrolyte according to a predetermined algorithm which incorporated the variables of cardiac output, SV and systemic vascular resistance. N=32 Conventional Clinical assessment: Preoperative crystalloid loading at 2 ml/kg/h of fasting, and given infusion of crystalloid in volume of three to four times the actual blood loss. Additional crystalloid was given at 4-8ml/kg/h based on estimated insensible loss. N=40	Patients over 18 years of age presenting for bowel resection, defined as open or laparoscopic with primary anastomosis. Mean age (range): 57 years (22-80) UK	Mortality Complication Length of hospital stay

Table 3Clinical evidence summary: Oesophageal Doppler monitoring versus pulse contour analysis

Outcomes

No of Participants (studies) Follow up

Quality of the evidence (GRADE)

Relative effect (95% CI)

Anticipated absolute effects

Risk with Pulse contour analysis

Risk difference with Oesophageal Doppler (95% CI)

Patients with complications

(1 study)

8 days

⊕⊕⊕⊝

MODERATE¹

due to imprecision

RR 0.61 (0.34 to 1.08)

Moderate

900 per 1000

351 fewer per 1000

(from 594 fewer to 72 more)

(a): Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs.

Table 4Clinical evidence summary: Cardiac output monitoring versus conventional clinical assessment

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with Conventional clinical assessment	Risk difference with Non-invasive cardiac output monitoring (95% CI)
Mortality	2012 (12 studies) <90 days	⊕⊕⊝⊝ LOW¹ due to imprecision	RR 0.87 (0.53 to 1.42)	Moderate
Mortality	2012 (12 studies) <90 days	⊕⊕⊝⊝ LOW¹ due to imprecision	RR 0.87 (0.53 to 1.42)	33 per 1000	4 fewer per 1000 (from 16 fewer to 14 more)
Patients with complications	2049 (13 studies) <45 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 0.75 (0.67 to 0.84)	Moderate
Patients with complications	2049 (13 studies) <45 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 0.75 (0.67 to 0.84)	416 per 1000	104 fewer per 1000 (from 67 fewer to 137 fewer)
Complications: POMS ≥1 (3-days)	220 (1 study) 3 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 1.12 (0.97 to 1.29)	Moderate
Complications: POMS ≥1 (3-days)	220 (1 study) 3 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 1.12 (0.97 to 1.29)	730 per 1000	88 more per 1000 (from 22 fewer to 212 more)
Complications: POMS ≥1 (5-days)	220 (1 study) 5 days	⊕⊕⊝⊝ LOW¹ due to imprecision	RR 1.04 (0.79 to 1.35)	Moderate
Complications: POMS ≥1 (5-days)	220 (1 study) 5 days	⊕⊕⊝⊝ LOW¹ due to imprecision	RR 1.04 (0.79 to 1.35)	487 per 1000	19 more per 1000 (from 102 fewer to 170 more)
Complications: POMS ≥1 (8-days)	220 (1 study) 8 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 1.27 (0.87 to 1.87)	Moderate
Complications: POMS ≥1 (8-days)	220 (1 study) 8 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 1.27 (0.87 to 1.87)	288 per 1000	78 more per 1000 (from 37 fewer to 251 more)
Length of hospital stay	941 (8 studies)	⊕⊕⊕⊕ HIGH		The mean length of hospital stay in the control groups was 12.3 days	The mean length of hospital stay in the intervention groups was 0.57 lower (1.12 to 0.03 lower)
Length of stay in ICU	214 (2 studies)	⊕⊕⊕⊕ HIGH		The mean length of stay in ICU in the control groups was 3.26 days	The mean length of stay in ICU in the intervention groups was 0.36 lower (0.59 to 0.12 lower)
Readmission rate	707 (5 studies) 30-60 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 1.23 (0.81 to 1.87)	Moderate
Readmission rate	707 (5 studies) 30-60 days	⊕⊕⊕⊝ MODERATE¹ due to imprecision	RR 1.23 (0.81 to 1.87)	94 per 1000	22 more per 1000 (from 18 fewer to 82 more)

(a): Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs.
(b): Downgraded by 1 or 2 increments because of heterogeneity, I2>50%, p=0.05, unexplained by subgroup analysis.

Table 5Evidence not suitable for GRADE analysis: Oesophageal Doppler monitoring versus pulse contour analysis

Outcome

Study (no. of participants)

Risk of bias

Comparison results (pulse contour analysis)

Intervention results (Oesophageal Doppler)

P value

Length of hospital stay (days)

Feldheiser 2015

(21)

Low

Median (IQR):

13 (9.75-22.5)

Median (IQR):

13 (12-19)

0.91

Table 6Evidence not suitable for GRADE analysis: Cardiac output monitoring versus conventional clinical assessment

Outcome	Study (no. of participants)	Risk of bias	Comparison results (conventional clinical assessment)	Intervention results (cardiac output monitoring)	P value
Mortality	Moppett 2015 (114)	Low	There was no significant difference (P=0.148) with outcomes adjusted for NHFS or age. Values reported on Kaplan Meier curve
Quality of life	Wakeling 2005 (128)	High	The EORTC QLQ C-30 and QLQ CR38 quality of life questionnaires completed 4–6 weeks after surgery showed no differences between the groups.
Total number of complications	Pillai 2011 (66)	High	35/34	16/32	n/a
	Stens 2017 (188)	High	42/94	38/94	0.78
	Venn 2002 (79)	Low	21/52	11/27	0.98
	Zakhaleva 2013 (72)	Low	19/40	7/31	0.03
Length of hospital stay (days)	Bartha 2013 (142)	High	Median (range): 9 (3-20)	Median (range) 10 (1-38)	>0.05
	Kaufmann 2017 (96)	High	Median (IQR): 11 (9-12)	Median (IQR): 9 (5-11)	0.005
	Noblett 2006 (103)	High	Median (range): 9 (4-45)	Median (range): 7 (3-35)	n/a
	Srinivasa 2013 (85)	Low	Median (range): 5 (2-29)	Median (range): 6 (3-31)	n/a
	Stens 2017 (175)	High	Median (IQR): 6 (4-11)	Median (IQR): 6 (4-9)	n/a
	Szturz 2019 (140)	Low	Median: 11	Median: 9	0.301
	Wakeling 2005 (128)	High	Median (IQR): 11.5 (4.75)	Median (IQR): 10 (5.75)	0.031
	Zakhaleva 2013 (72)	Low	Median (range): 5 (3-16)	Median (range): 6 (3-30)	n/a
	Correa-Gallego 2015 (135)	Low	Median (range): 6 (5-8)	Median (range): 7 (6-8)	n/a
	Mayer 2010 (60)	Low	Median (IQR): 19 (14-23.5)	Median (IQR): 15 (12-17.75)	0.006
	Pearse 2014 (730)	Low	Median (IQR): 11 (7-17)	Median (IQR): 10 (7-14)	0.05
	Ramsingh 2013 (38)	Low	Median (IQR): 7.5 (5.25-10.75)	Median (IQR): 5 (3.75-8.25)	n/a
	Ratti 2016 (90)	Low	Median (range): 5 (3-13)	Median (range): 4 (2-10)	n/a
	Shillcutt 2014 (29)	High	Median (range): 5 (1-36)	Median (range): 3 (1-10)	0.058
Length of ICU stay (days)	Stens 2017 (175)	High	Median (IQR): 0 (0-0)	Median (IQR): 0 (0-0)	n/a
Length of ICU stay (days)	Szturz 2019 (140)	Low	Median: 4	Median: 4	0.339

Table 7Health economic evidence profile: Cardiac output monitoring (Cardio-Q ODM) versus pulse contour analysis (PCA) versus central venous pressure (CVP) versus conventional clinical assessment (CCA)

Study

Applicability

Limitations

Other comments

Incremental cost

Incremental effects

Cost effectiveness

Uncertainty

NICE 2011⁶⁶ (UK)

Partially applicable ^(a)

Potentially serious limitations ^(b)

Population: Adults undergoing moderate and major risk surgery and high risk adults undergoing any surgery
Comparators:
- 1: CCA
- 2: CVP & CCA
- 3: PCA^(d) & CCA
- 4: CVP & ODM & CCA
- 5: CVP & PCA^(d) & CCA
- 6: ODM & CCA
Cost comparison
Time horizon: until discharged from hospital

6−1: −£966^(c)

6−2: −£1,088^(c)

6−3: −£1,150^(c)

6−4: −£55^(c)

6−5: −£1,091^(c)

None

ODM & CCA was cost-saving

Increasing the effectiveness of general ward length of stay for CVP & CCA and keeping the effectiveness of ODM constant resulted in ODM no longer being cost-saving.

PSA demonstrated that ODM was cost-saving in comparison to CVP & CCA with a saving of £1,378.

: Abbreviations: CCA= conventional clinical assessment; CVP= central venous pressure; ODM= oesophageal Doppler monitor; PCA= pulse contour analysis; PSA= probabilistic sensitivity analysis
(a): UK NHS perspective, costs from 2008/09 and changes in practice mean that it may not be as relevant to current practice. Measure of effect is not in line with NICE reference case methods as the analysis does not measure QALYs.
(b): Time horizon is too short and may not fully capture differences in costs and health outcomes. Some of the health benefits have not been captured and some of the treatment effects were based on assumptions. The treatment effects used in the analysis do not reflect the full body of available evidence for this area (23 RCTs included in the clinical review). Five out of eleven of the RCTs included in the meta-analysis used starch boluses and were excluded from the NGC clinical review. Funded by Deltex Medical.
(c): 2008/09 UK Pounds. Cost components included: Length of hospital stay (ICU, HDU and general ward), device costs, maintenance and consumables, fluids and staffing.
(d): Note: Pulse contour analysis was used as the name of the intervention throughout the review instead of pulse pressure waveform analysis to be in line with the clinical review.

Table 8Health economic evidence profile: Cardiac output monitoring (LiDCO plus) versus usual care

Study

Applicability

Limitations

Other comments

Incremental cost

Incremental effects

Cost effectiveness

Uncertainty

Bartha 2012 (Sweden)⁵

Partially applicable^(a)

Potentially serious limitations^(b)

Population: Adults 80 years and over undergoing surgery for hip fractures
Comparators:
- Standard care
- Cardiac output-monitoring (LiDCO rapid)
Cost utility analysis
Probabilistic decision analytic model
Time horizon: 5 years

−£1,436^(c)

QALYs:

0.344

Cardiac output monitoring was dominant^(d) (cheaper and more effective)

96.4% of simulations resulted in cardiac output monitoring being dominant.

Results were sensitive to relative risks for mortality and morbidity. When clinical effect was reduced by increasing the relative risk by 90% the ICER was £292 per QALY gained.

Sadique 2015⁸⁷ (UK)

Partially applicable^(e)

Potentially serious limitations^(f)

Population: Adults 50 years and over undergoing major gastrointestinal surgery
Comparators:
- Standard care
- Cardiac output monitoring (LiDCO rapid)
Cost utility analysis
Within trial analysis (RCT) with modelled post-trial extrapolation
Time horizon: 10 years

Lifetime:

−£404^(g)

Six months:

−£404^(g)

Lifetime QALYs:

0.19

Six month QALYs:

0.01

Cardiac output monitoring was dominant^(d) (cheaper and more effective)

Different scenario analyses were conducted, which did not affect the results.

: Abbreviations: QALY= quality-adjusted life years; RCT= randomised controlled trial
(a): Swedish healthcare perspective and 2012 Swedish kroners may not be relevant to current UK practice. Study focuses on one type of surgery instead of all major surgery. Unclear what tariff and population was used for quality of life weights, cost year is not reported and discount rate used is not in line with NICE reference case.
(b): Time horizon may be too short to fully capture costs and outcomes. Baseline probabilities and treatment effects for complications were based on a single RCT therefore the treatment effects used in the analysis do not reflect the full body of available evidence for this area (23 RCTs included in the clinical review).
(c): 2012 Swedish Kroner covered to UK pounds. ⁷² Cost components included: monitor costs (LiDCO rapid), hospital costs, costs of various complications in hospital, costs of long-term medical care costs after stroke and cardiovascular complications and death.
(d): Interventions are dominant when they are both less costly and more effective.
(e): UK NHS perspective and costs from 2012/13 may not reflect current practice. Study is based on one type of surgery and not the whole surgical population. Unclear if costs are discounted.
(f): Baseline and treatment effects are based on a single RCT therefore the treatment effects used in the analysis do not reflect the full body of available evidence for this area (23 RCTs included in the clinical review). The analysis did not include complications as a health outcome. Cost source slightly unclear and costing methods to avoid double counting could impact results.
(g): 2012/13 UK Pounds. Cost components included: surgical costs, length of stay in critical care and surgical ward, blood products and device costs.

Table 9Health economic evidence profile: ODM & CCA versus PCA & CCA versus CCA

Study

Applicability

Limitations

Other comments

Incremental cost

Incremental effects

Cost effectiveness

Uncertainty

Legrand 2015⁵⁰ (France)

Partially applicable ^(a)

Potentially serious limitations ^(b)

Population: Adults undergoing intermediate and high risk abdominal surgery.
Comparators:
- 1: CCA alone
- 2: PCA^(d) & CCA
- 3: ODM & CCA
Cost-effectiveness analysis
Time horizon: until discharged from hospital

2−1: −£334^(c)

3−1: −£134^(c)

3−2: £200^(c)

Major complication avoided:

2−1: 0.129

3−1: 0.072

3−2: −0.057

Death avoided:

2−1: 0.018

3−1: 0.021

3−2: 0.003

Both PCA and ODM were dominant^(e) when they were compared to CCA.

One-way sensitivity analysis was conducted varying each of the parameters.

Probabilistic sensitivity analysis was conducted by performing 1000 iterations. For mortality avoided PPWA and ODM were dominant compared with CCA in 92.9% and 69.5% of cases, respectively.

For major complications avoided PPWA and ODM were dominant compared with CCA in 97.3% and 76.1% of cases, respectively.

: Abbreviations: CCA= conventional clinical assessment; ODM= oesophageal Doppler monitor, PCA= pulse contour analysis
(a): French healthcare perspective and 2011 euros may not be relevant to current UK practice. Study focuses on one type of surgery and does not include all major surgery. Measure of effect is not in line with NICE reference case methods as the analysis does not measure QALYs.
(b): Time horizon may be too short to fully capture costs and outcomes. The treatment effects used in the analysis do not reflect the full body of available evidence for this area (23 RCTs included in the clinical review). Five out of thirteen of the RCTs included in the meta-analysis used starch boluses and were excluded from the NGC clinical review.
(c): 2011 French Euros covered to UK pounds.⁷². Cost components included: Medical devices (CardioQ-ODM and Vigileo/FloTrac), hospital costs such as procedures performed, length of stay and complications.
(d): Note: Pulse contour analysis was used as the name of the intervention throughout the review instead of pulse pressure waveform analysis to be in line with the clinical review.
(e): Interventions are dominant when they are both less costly and more effective.

Table 10Health economic evidence profile: CCA & CVP & ODM versus CCA & CVP versus ODM & CCA versus CCA

Study

Applicability

Limitations

Other comments

Incremental cost

Incremental effects

Cost effectiveness

Uncertainty

Maeso 2011⁵⁴ (Spain)

Partially applicable^(a)

Potentially serious limitations ^(b)

Population: Adults undergoing colorectal resection.
Comparators:
- 1: CCA
- 2: ODM & CCA
- 3: CCA & CVP
- 4: CCA & CVP & ODM
Cost-effectiveness analysis and cost-utility analysis
Time horizon: for the CEA until discharged, and for the CUA a lifetime horizon

Costs until discharged (mean per patient):

4−1: −£931^(c)

4−2: −£364^(c)

4−3: −£882^(c)

Costs for lifetime horizon (mean per patient):

4−1: −£402^(c)

4−2: £154^(c)

4−3: −£803^(c)

Survival rate (mean per patient):

4−1: 0.093

4−2: 0.091

4−3: 0.014

Free of major complication rate (mean per patient):

4−1: 0.232

4−2: 0.152

4−3: 0.115

QALYs (mean per patient):

4−1: 1.37

4−2: 1.34

4−3: 0.21

CEA: CCA & CVP & ODM was dominant^(d) for survival and major complication avoided compared to other interventions.

CUA: CCA & CVP & ODM dominated CCA and CVP and CCA alone. Compared against CCA & ODM it resulted in an ICER of £114.93 per QALY.

Deterministic sensitivity analyses were performed by varying uncertain parameter values.

Probabilistic sensitivity analysis was conducted by performing 10,000 iterations.

The probability of CCA & CVP & ODM being cost-effective ranged from 40% to 60% at €50,000 per death avoided.

: Abbreviations: CCA= conventional clinical assessment; CVP= central venous pressure; ICER= incremental cost-effectiveness ratio; ODM= oesophageal Doppler monitor; QALY= quality-adjusted life years; RCT= randomised controlled trial
(a): Spanish healthcare perspective and 2007 euros may not be relevant to current UK practice. Study focuses on one type of surgery instead of all major surgery. QALYs were only included in a sensitivity analysis.
(b): Time horizon of until discharge was too short to fully capture outcomes and costs. Did conduct a sensitivity analysis with long-term horizon but assumed that people alive would incur the same costs and QALYs. The treatment effects used in the analysis do not reflect the full body of available evidence for this area (23 RCTs included in the clinical review) and some of the treatment effects were obtained from other high risk surgeries where there was missing data for certain comparisons. One out of four of the RCTs included in the meta-analysis used starch boluses and was excluded from the NGC clinical review.
(c): 2007 Spanish Euros covered to UK pounds.⁷². Cost components included: Device costs, surgery time, hospital stay and high dependency unit stay. Staff costs were assumed to be included in the surgery time cost.
(d): Interventions are dominant when they are both less costly and more effective.

Table 11Health economic evidence profile: ODM & CCA versus CCA and CCA & CVP & ODM versus CVP & CCA

Study

Applicability

Limitations

Other comments

Incremental cost

Incremental effects

Cost effectiveness

Uncertainty

Mowatt 2009⁶³ (UK)

Partially applicable^(a)

Potentially serious limitations ^(b)

Population: Adults undergoing high risk surgery
Comparisons:
- ODM & CCA versus CCA
- CCA & CVP & ODM versus CVP & CCA
Cost-utility analysis
Time horizon: 5 years

Study concluded that both ODM strategies are cost-effective at a threshold of £30,000 per QALY.

Probabilistic sensitivity analysis was conducted by performing 1000 iterations.

The probability of ODM being cost-effective was not reported however most of the iterations fell in the south-east quadrant for both of the ODM & CVP & CCA versus CVP & CCA and ODM & CCA versus CCA comparisons, meaning ODM was more effective and less costly.

: Abbreviations: CCA= conventional clinical assessment; CVP= central venous pressure; ODM= oesophageal Doppler monitor; QALY= quality-adjusted life years;
(a): UK NHS perspective, costs from 2006/07 and changes in practice mean that it may not be relevant to current practice. Did not state whether discounting was used in 5 year analysis. Utilities were not from the relevant population as it was obtained from ICU survivors instead of surgery survivors.
(b): Does not give a breakdown of the costs for each interventions and a breakdown of the QALYs for each intervention. Shows the probability that ODM would be considered cost-effective at a £30,000 per QALY threshold, not £20,000. Assumes that people survive on average for 5 years after surgery. The treatment effects used in the analysis do not reflect the full body of available evidence for this area (23 RCTs included in the clinical review). Five out of nine of the RCTs included in the meta-analysis used starch boluses and were excluded from the NGC clinical review.

Table 12Probabilistic base case results (per person)

	Total cost	Total QALYs	ICER	Probability COM CE at £20k
30 day results
CCA	£1,309	0.050
COM	£1,108	0.051
Incremental (COM vs CCA)	−£201	0.001	COM dominant^(a)	n/a
Lifetime results (discounted)
CCA	£27,570	6.99
COM	£27,572	7.07
Incremental (COM vs CCA)	£2.00	0.08	£25 per QALY gained	94%

: Abbreviations: CE = cost-effective; ICER = incremental cost-effectiveness ratio; n/a = not applicable; QALYs = quality-adjusted life years
(a): Interventions are dominant when they are both less costly and more effective.

Final

Evidence reviews underpinning recommendation 1.4.5 in the NICE guideline

This evidence review was developed by the National Guideline Centre

Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and, where appropriate, their carer or guardian.

Local commissioners and providers have a responsibility to enable the guideline to be applied when individual health professionals and their patients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.

Bookshelf ID: NBK561961PMID: 32931181