TITLE: Esophageal Doppler Ultrasound-Based Cardiac Output Monitoring for Adults undergoing Surgery: A Review of Clinical and Cost-Effectiveness

Transcription

1 TITLE: Esophageal Doppler Ultrasound-Based Cardiac Output Monitoring for Adults undergoing Surgery: A Review of Clinical and Cost-Effectiveness DATE: 18 March 2013 CONTEXT AND POLICY ISSUES Optimal management of cardiac output and fluid balance is considered one of the key elements in improving outcomes in high-risk surgical patients and in critically ill patients. 1-3 Cardiac output refers to the amount of blood that is pumped by the heart in unit time and is calculated by multiplying the stroke volume by the heart rate. 2 Stroke volume refers to the amount of blood pumped by the left ventricle in one contraction. 2 Sufficient blood flow is required to provide adequate oxygen and nutrients to cells and tissues and to assist in the clearance of waste products. 2,4 For patients undergoing surgery or in intensive care, optimization of intravenous fluid replacement is essential for maintaining adequate organ perfusion. If patients do not receive adequate additional fluids, there is possibility of hypovolemia (abnormally low levels of blood plasma) followed by hypotension and organ failure. On the other hand, excessive fluid addition may trigger heart failure. 2 Several procedures for monitoring and optimizing intravenous fluid are available. These include thermodilution by pulmonary artery catheterization, the dye dilution method involving dye injection into the pulmonary artery and measurement of dye concentration at the femoral or radial artery, the lithium dilution method using an arterial catheter with an attached lithium sensor, pulse contour analysis involving measurement of arterial pulse pressure waveform using an arterial catheter, methods using the Fick principle, thoracic electrical bioimpedance, transesophageal echocardiography, ultrasonic cardiac output monitoring, central venous pressure monitoring (CVP), esophageal Doppler monitoring, and conventional clinical assessment (CAA) which generally refers to non-invasive measurement of clinical markers. 2,4,5 Esophageal Doppler monitoring (EDM) is the focus of this report. It is a minimally invasive procedure used to measure cardiac output. A small probe, which emits an ultrasound beam, is placed via the oral or nasal route in the esophagus of a patient, usually under anesthesia. Blood flow velocity in the descending aorta is measured by the change in frequency (Doppler shift) of this beam as it reflects off a moving object. Blood flow velocity together with an estimate of cross-sectional area of the aorta is used to determine the cardiac output. 2,5 Disclaimer: The Rapid Response Service is an information service for those involved in planning and providing health care in Canada. Rapid responses are based on a limited literature search and are not comprehensive, systematic reviews. The intent is to provide a list of sources of the best evidence on the topic that CADTH could identify using all reasonable efforts within the time allowed. Rapid responses should be considered along with other types of information and health care considerations. The information included in this response is not intended to replace professional medical advice, nor should it be construed as a recommendation for or against the use of a particular health technology. Readers are also cautioned that a lack of good quality evidence does not necessarily mean a lack of effectiveness particularly in the case of new and emerging health technologies, for which little information can be found, but which may in future prove to be effective. While CADTH has taken care in the preparation of the report to ensure that its contents are accurate, complete and up to date, CADTH does not make any guarantee to that effect. CADTH is not liable for any loss or damages resulting from use of the information in the report. Copyright: This report contains CADTH copyright material and may contain material in which a third party owns copyright. This report may be used for the purposes of research or private study only. It may not be copied, posted on a web site, redistributed by or stored on an electronic system without the prior written permission of CADTH or applicable copyright owner. Links: This report may contain links to other information available on the websites of third parties on the Internet. CADTH does not have control over the content of such sites. Use of third party sites is governed by the owners own terms and conditions.

2 The purpose of this report is to assist in the decision making with respect to EDM by providing evidence on the clinical effectiveness and cost-effectiveness of intraoperative fluid management guided by esophageal Doppler ultrasound-based cardiac output monitoring devices in adult patients undergoing surgery. RESEARCH QUESTIONS 1. What is the clinical-effectiveness of intraoperative use of esophageal Doppler ultrasoundbased cardiac output monitoring devices in adult patients undergoing surgery? 2. What is the cost-effectiveness of intraoperative use of esophageal Doppler ultrasoundbased cardiac output monitoring devices in adult patients undergoing surgery? KEY FINDINGS No studies were identified which compared esophageal Doppler monitoring (EDM) with methods specifically using an arterial line or arterial catheter. In most instances, available evidence suggested a trend towards reduction in length of hospital stay, complication rates and mortality for intraoperative fluid management strategies using EDM compared to those using central venous pressure monitoring (CVP) or conventional clinical assessment (CAA), in adults undergoing surgery. However results need to be interpreted with caution as the differences were not statistically significant in several instances. Strategies with EDM combined with CVP or CAA appeared to be more cost-effective than strategies which involved CVP or CAA alone or in combination in the absence of EDM. METHODS Literature Search Strategy A limited literature search was conducted on key resources including PubMed, The Cochrane Library (2013, Issue 1), University of York Centre for Reviews and Dissemination (CRD) databases, Canadian and major international health technology agencies, as well as a focused Internet search. No filters were applied to limit the retrieval by study type. Where possible, retrieval was limited to the human population. The search was also limited to English language documents published between January 1, 2008 and February 14, Selection Criteria and Methods One reviewer screened the titles and abstracts of the retrieved publications and selected potentially relevant articles for retrieval of full-text publications for further investigation. A second reviewer evaluated the full-text publications for final selection, according to the criteria listed in Table 1. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 2

3 Table 1: Selection Criteria Population Adults undergoing surgery Intervention Comparator Outcomes Study Designs Esophageal Doppler Ultrasound-based cardiac output monitoring during surgery (aka intra-operatively) Arterial line or arterial catheter Question 1: length of stay, adverse events, resource utilization Question 2: cost-effectiveness and resource utilization Health technology assessment (HTA), systematic review (SR) and meta-analysis (MA), randomized controlled trial (RCT) and nonrandomized studies Economic study (cost-effectiveness study) Exclusion Criteria Studies were excluded if they did not satisfy the selection criteria in Table 1, if they were published prior to 2008, or duplicate publications of the same study and did not provide additional relevant information. Individual studies which were included in at least one of the included systematic reviews were excluded. Systematic reviews in which all included studies were included in more recent or comprehensive systematic reviews or health technology assessments were excluded. Non-comparative studies were excluded unless no comparative studies were available. Critical Appraisal of Individual Studies Critical appraisal of a study was conducted based on an assessment tool appropriate for the particular study design. The AMSTAR checklist 6 was used for systematic reviews; the Downs and Black checklist 7 for RCTs and non-randomized studies; the checklist of Drummond et al. 8 for economic studies. For the critical appraisal, a numeric score was not calculated. Instead, the strength and limitations of the study were described. SUMMARY OF EVIDENCE Quantity of Research Available The literature search yielded 153 citations. Upon screening titles and abstracts, 135 articles were excluded and 18 potentially relevant articles were selected for full-text review. Five potentially relevant articles were identified from the grey literature. Of these 23 articles, 15 did not satisfy the inclusion criteria and were excluded. One health technology assessment, two systematic reviews, two RCTs, two non-randomized studies, and one cost-effectiveness study were relevant and selected for inclusion. The health technology assessment included a systematic review of RCTs and an economic evaluation. Details of the study selection process are outlined in Appendix 1. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 3

4 Summary of Study Characteristics Characteristics of the included health technology assessment, systematic reviews, RCTs, nonrandomized studies and economic studies are summarized below and details are provided in Appendix 2. Health technology assessment The included health technology assessment 1 was published in 2009 from the United Kingdom (UK). It included a systematic review of 10 RCTs and an economic evaluation. This health technology assessment was based on a previously published AHRQ systematic review 2 which included eight RCTs and was supplemented with two additional RCTs. The 10 RCTs included 959 patients undergoing surgery or in critical care. Strategies with Esophageal Doppler monitoring (EDM) for optimizing fluid management in these patients were compared with control strategies including procedures such as central venous pressure monitoring (CVP) and conventional clinical assessment (CCA). The EDM device used was CardioQ in eight RCTs, Hemosonic 100 in one RCT and TECO in one RCT. The age of patients in the EDM group ranged between 33 and 82 years and in the control group between 40 and 85 years. Outcomes reported were length of hospital stay, complications and mortality. The authors of this health technology assessment intended to include also a systematic review of economic evaluations but no relevant studies could be identified. They conducted an economic evaluation using partial economic modeling involving pairwise comparisons between strategies with and without EDM. The EDM devices considered were CardioQ and CardioQP. CardioQ supports adult probes and CardioQP supports both adult and pediatric probes. The EDM device was assumed to have a lifetime of five years. Results were expressed as additional cost ( ) per additional QALY and average extra cost ( ) per additional survivor that would need to be incurred before EDM would no longer be considered cost-effective. Cost data used were for the period 2006 to Systematic reviews Two relevant systematic reviews 9,10 comparing strategies with and without EDM were identified. Both were from New Zealand with one 9 published in 2008 and one 10 in Both systematic reviews included five RCTs, of which four overlapped. In addition, these four RCTs were also included in the health technology assessment report 1 mentioned above. Of the four overlapping RCTs, three were specific to colorectal surgery and one included patients undergoing urological, gynaecological and general surgical procedures. The unique RCT in one systematic review 9 was specific to colorectal surgery and the unique RCT in the second systematic review 10 was specific to upper gastro-intestinal surgery. The total number of patients was not specified in one systematic review 9 and was 428 in the second systematic review. 10 The age of patients, the proportion of females and males or the EDM devices used were not specified. Both systematic reviews reported on length of hospital stay, complications and some physiological parameters. One systematic review 9 also reported on mortality. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 4

5 Randomized controlled trials (RCTs) Two relevant RCTs 11,12 were identified. Both RCTs were published in 2012, one 11 from Denmark and one 12 from UK. Both RCTs involved patients undergoing colorectal surgery, one 11 included 150 patients and one 12 included 179 patients. The mean age was between 66 and 68 years. Fluid management with the goal of achieving near-maximal stroke volume guided by EDM was compared with the goal of zero balance and normal body weight (zero balance approach, formerly known as restricted approach ) in one RCT. 11 In the second RCT 12 goal directed fluid management with EDM was compared with standard care. The EDM device used was CardioQ in one RCT 12 and CardioQ-ODM in the second RCT. 11 Both RCTs reported on length of hospital stay, complications, mortality and physiological changes (such as bowel movement, gastrointestinal function, and diet tolerance). Non-randomized studies Two relevant non-randomized studies 13,14 were identified. Both were published from the UK, one 13 published in 2013 and one 14 in One study 13 was a prospective study comparing EDM with CVP monitoring. It included 104 patients undergoing free perforated flap surgery. The mean age in the EDM and CVP groups were 39 and 44 years respectively and both groups contained a higher proportion of female patients than males. The study reported on length of hospital stay, complications, and physiological changes. The second study 14 was a before and after EDM implementation study. It included 1,307 patients undergoing various types of surgery and of age 60 years. The study reported length of hospital stay, reoperation, readmission to hospital and mortality Economic evaluation Two economic evaluations 1,15 were identified of which one 1 was part of a health technology assessment report. Characteristics of this economic evaluation have been discussed with the characteristics of the health technology assessment. One economic study 15 was published in 2011 from a group in Spain. It examined the costeffectiveness of cardiac output monitoring and hemodynamic function optimization using various strategies including one or more options such as conventional clinical assessment (CCA), central venous pressure monitoring (CVP), and esophageal Doppler monitoring (EDM). The authors constructed an analytical decision model to compare four strategies ([CCA+ CVP+ EDM] versus [CCA+ CVP] versus [CCA+ EDM] versus CCA) as these were some of the most commonly used in clinical settings and RCTs. The time horizon considered was until discharge from hospital, assuming that cardiac output control systems and fluid administration during surgery would not impact outcomes after discharge. The perspective was that of a hospital which is considered to be a good proxy for an overall health care system perspective. Estimates of effectiveness were obtained from meta-analyses or individual RCTs. For patients undergoing CCA+ CVP or CCA alone it was assumed that the length of hospital stay was the same. For the base case analysis, it was assumed that the equipment would last for 5 years and usage would be 125 times per year. Sensitivity analyses were conducted with other utilization rates. To adjust for consecutive years of usage an inflation rate of 3% was applied. Cost data were obtained from the Madrid Health System, the La Paz University Hospital finance department and from Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 5

6 device manufactures. The cost data was compared with published cost data and no significant differences were observed. Costs were presented in for the year Quality adjusted life year (QALY) and incremental cost-effectiveness ratio (ICER) were calculated. Strategies with lower effectiveness and higher cost were considered as dominated and for these cases ICERs were not calculated. Summary of Critical Appraisal Critical appraisal of the included health technology assessment, systematic reviews, RCTs, nonrandomized studies and economic studies are summarized below and details are provided in Appendix 3. Health technology assessment The authors of the health technology assessment report 1 based their report on the AHRQ report 2 which they considered to be good quality. They supplemented the information with two additional studies. Objectives, inclusion and exclusion criteria were stated and a comprehensive literature was undertaken. Lists of included and excluded studies were provided. Characteristics of the individual RCTs were available. Quality assessments of the included studies were conducted. Article selection and data extraction was conducted by one reviewer and a second reviewer was consulted in case of uncertainty. Conflict of interest was declared and there was none. Publication bias was not explored. For the economic section, clinical effectiveness data were taken from the systematic review. The form of economic evaluation, time horizon and perspective were stated. Sensitivity analyses were conducted. Cost data were provided. Systematic reviews For the two included systematic reviews 9,10 the objectives, inclusion and exclusion criteria were stated, a comprehensive literature was undertaken, lists of included studies were provided and quality assessment was conducted. However, in these two systematic reviews the list of excluded studies, or description of the study selection process were not provided and it was unclear if article selection and data extraction were done in duplicate. In one systematic review 9 publication bias was explored and there appeared to be none. Conflict of interest statements were provided and no conflicts were declared. In this systematic review characteristics of the individual studies were not provided and the results were presented qualitatively. In the second systematic review 10 characteristics of the individual studies were provided but with few details, and pooled estimates were provided. However, publication bias was not explored and conflict of interest was not stated. Randomized controlled trials (RCTs) In the two included RCTs, 11,12 objectives were stated, patient characteristics, interventions, outcomes and sample size calculations were described. Intent-to treat analyses were conducted and P-values were provided. Both studies were double-blind. As with RCTs, generalizability was limited due to restrictive inclusion criteria. In one RCT 11 there was some potential for selection bias as the presence of the investigating anesthetist and surgeon were mandatory for screening of patients. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 6

7 Non-randomized studies Two relevant non-randomized studies 13,14 were included. In one prospective non-randomized study 13 objectives were stated and inclusion and exclusion criteria were provided. Patient characteristics, interventions and outcomes were described. There was no apparent significant difference between the two groups studied with respect to patient demographics and operative indications. All patients were included in the analyses and P-values were provided. Sample size calculation was not described and the authors mentioned the possibility of the study being underpowered. The second non-randomized study 14 was a before and after EDM implementation study. In this study, objectives were stated and patient characteristics, interventions and outcomes were described but few details were provided. All patients were included in the analyses and P-values were provided. Prospective data from consecutive patients following implementation of EDM were compared with retrospective data from controls matched by specialty and severity of surgery. The control group had differences in age and physical status scores but perioperative risk indicator was similar. This study was generalizable to some extent as data was collected from three hospitals in England with different size, geographical location and case mix. Economic evaluation Two economic evaluations 1,15 were identified, of which one 1 was part of a health technology assessment report. Quality assessment of this economic evaluation has been discussed in the health technology assessment section above. In the second economic evaluation the objective, form of economic evaluation, time horizon and perspective were stated. Clinical effectiveness data were obtained from meta-analyses or individual RCTs. Sensitivity analyses were conducted. Sources of cost data were stated. Summary of Findings The overall findings are summarized below and details of the findings of the included health technology assessment, systematic reviews, RCTs, non-randomized studies and economic evaluations are provided in Appendix 4. What is the clinical-effectiveness of intraoperative use of esophageal Doppler ultrasound-based cardiac output monitoring devices in adult patients undergoing surgery? Length of hospital stay was reported in one HTA, 1 two systematic reviews, 9,10 two RCTs, 11,12 and two non-randomized studies 13,14 and is shown in Table 2. Hospital stay was less for strategies with EDM compared to those without EDM but the difference was not statistically significant in the majority of cases. One RCT included in a systematic review reported an increase in hospital stay with EDM compared with no EDM, however it was not mentioned if the difference was statistically significant. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 7

8 Table 2: Length of hospital stay (days) Study Comparison Finding HTA 1 (2 RCTs; N= 157 surgical patients) HTA 1 (1 RCT; N= 61 surgical patients) HTA 1 (3 RCTs; N= 170 surgical patients) HTA 1 (2 RCTs; N= 236 critically ill patients) (EDM+ CVP+ CCA) vs (CVP+ CAA) (EDM+ CCA) vs (CVP+ CAA) (EDM+ CCA) vs (CAA) (EDM+ CVP+ CCA) vs (CVP+ CAA) Effect measure Effect WMD (95% CI) (-2.98, -0.65) Mean 13.5 vs 13.3 (p= 0.96) Mean difference Decrease: Significant (1 RCT), non-significant (1 RCT) and significance NR (1 RCT) Mean difference Decrease : Significant (1 RCT), and significance NR (1 RCT) SR 9 (4 RCTs, N= NR) EDM vs no EDM Mean difference Decrease: (2 RCT) Increase: (1 RCT) SR 10 (4 RCTs; N= 368 surgical patients) EDM vs conventional WMD (95% CI) (-2.58, -0.62) RCT 11 (N= 150 surgical patients) RCT 12 (N= 179 surgical patients)* Non randomized prospective study 13 (N= 104 surgical patients) Non randomized implementation study 14 (N= methods EDM vs zero fluid balance approach (Z) EDM vs standard care Mean ± SD Median (interquartile range)* 8.45 ± 7.5 vs 7.66 ± 8.2 (p= 0.54) 8.8 ( ) vs 6.7 ( ); (p= 0.09) EDM vs CVP Mean ± SD 8.9 ± 3.6 vs 10.8 ± 6.2; (p= 0.147) Before and after implementation of EDM Mean ± SD 15.1 ± 16.7 vs 18.7 ± 24.4 (p= 0.002) 1,307 surgical patients) CCA= conventional clinical assessment, CI= confidence interval, CVP= central venous pressure monitoring, EDM= esophageal Doppler monitoring, HTA= health technology assessment, N= number of patients, RCT= randomized controlled trial, SD= standard deviation, SR= systematic review, WMD= weighted mean difference *mentioned as post-operative days not specified as length of hospital stay Complication rates were reported in one HTA, 1 two systematic reviews, 9,10 two RCTs, 11,12 and two non-randomized studies 13,14 and are shown in Table 3. Complications appeared to be fewer for strategies with EDM compared to those without EDM. However, in three instances the differences were statistically significant, in five instances the differences were not statistically significant and in two instances statistical significance was not mentioned. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 8

9 Table 3: All complications Study Comparison Finding HTA 1 (3 RCTs; N= 288 surgical patients) HTA 1 (1 RCT; N= 61 surgical patients) HTA 1 (1 RCT; N= 59 surgical patients) HTA 1 (2 RCTs; N= 236 critically ill patients) (EDM+ CVP+ CCA) vs (CVP+ CAA) (EDM+ CCA) vs (CVP+ CAA) (EDM+ CCA) vs (CAA) (EDM+ CVP+ CCA) vs (CVP+ CAA) Effect measure Effect OR (95% CI) 0.43 (0.26, 0.71) OR (95% CI) 0.61 (0.21, 1.72) OR (95% CI) 0.41 (0.14, 1.16) OR (95% CI) 0.49 (0.30, 0.81) SR 9 (4 RCTs, N= NR) EDM vs no EDM Rate Decrease: (3 RCT) Increase: (1 RCT) SR 10 (4 RCTs; N= 368 surgical patients) EDM vs conventional OR (95% CI) 0.28 (0.17, 0.46) RCT 11 (N= 150 surgical patients) Non randomized prospective study 13 (N= 104 surgical patients) Non randomized implementation study 14 (N= 1,307 surgical patients)* methods EDM vs zero fluid balance approach (Z) Rate 23 (32%) vs 24 (30%); (p= 0.791) EDM vs CVP Incidence EDM: None CVP: 1 episode of Horner syndrome and 5 incidences of symptomatic hematomas Before and after implementation of EDM Rate 5.9% vs 8.4%; (p=0.08) CCA= conventional clinical assessment, CI= confidence interval, CVP= central venous pressure monitoring, EDM= esophageal Doppler monitoring, HTA= health technology assessment, N= number of patients, OR= odds ratio, RCT= randomized controlled trial, SR= systematic review * complications were not mentioned but reoperation was mentioned Major or serious complications were reported in one HTA, 1 and two RCTs 11,12 and are shown in Table 4. Major or serious complications appeared to be fewer for strategies with EDM compared to those without EDM. However, in one instance the difference was statistically significant, and in two instances the differences were not statistically significant. Table 4: Major or serious complications Study Comparison Finding HTA 1 (3 RCTs; N= 220 surgical patients) RCT 11 (N= 150 surgical patients) (EDM+ CVP+ CCA) vs (CVP+ CAA) EDM vs zero fluid balance approach (Z) EDM vs standard Effect measure Peto OR (95% CI) Effect 0.12 (0.04, 0.31) Rate 14% vs 10% (p= 0.616) RCT 12 (N= 179 surgical Rate 10/89 vs 13/90 (p= 0.47) patients) care CCA= conventional clinical assessment, CI= confidence interval, CVP= central venous pressure monitoring, EDM= esophageal Doppler monitoring, HTA= health technology assessment, N= number of patients, OR= odds ratio, RCT= randomized controlled trial, Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 9

10 Mortality was reported in one HTA, 1 one systematic review, 10 two RCTs, 11,12 and one nonrandomized study 14 and is shown in Table 5. Mortality appeared to be less for strategies with EDM compared to those without EDM. However, the difference was not statistically significant in seven instances and statistically significant in only one instance. Table 5: Mortality Study Comparison Finding HTA 1 (4 RCTs; N= 338 surgical patients) HTA 1 (1 RCT; N= 61 surgical patients) HTA 1 (3 RCTs; N= 139 surgical patients) HTA 1 (2 RCTs; N= 236 critically ill patients) (EDM+ CVP+ CCA) vs (CVP+ CAA) (EDM+ CCA) vs (CVP+ CAA) (EDM+ CCA) vs (CAA) (EDM+ CVP+ CCA) vs (CVP+ CAA) Effect measure Effect OR (95% CI) 0.13 (0.02, 0.96) Rate 3/30 versus 6/31; (p= 0.3) OR (95% CI) 0.81 (0.23, 2.77) OR (95% CI) 0.84 (0.41, 1.70) SR 9 (5 RCTs, N= NR) EDM vs no EDM - NR SR 10 (4RCTs; N= 300 EDM vs OR (95% CI) 0.62 (0.16, 2.45) surgical patients) conventional RCT 11 (N= 150 surgical patients) methods EDM vs zero fluid balance approach (Z) Rate 1% vs 1%; (p= 1.0) RCT 12 (N= 179 surgical EDM vs standard Rate 2/89 vs 2/90; (p= 1.0) patients) care Non randomized EDM vs CVP - NR prospective study 13 (N= 104 surgical patients) Non randomized Before and after Rate 2.8% vs 3.5%; (p=0.37) implementation study 14 (N= 1,307 surgical patients implementation of EDM CCA= conventional clinical assessment, CI= confidence interval, CVP= central venous pressure monitoring, EDM= esophageal Doppler monitoring, HTA= health technology assessment, N= number of patients, OR= odds ratio, RCT= randomized controlled trial, SR= systematic review. What is the cost-effectiveness of intraoperative use of esophageal Doppler ultrasound-based cardiac output monitoring devices in adult patients undergoing surgery? Two relevant economic studies 1,15 were identified. One economic study 15 evaluated four strategies and conducted a cost-effectiveness analysis. These strategies were (EDM+ CVP+ CCA), (CVP+ CCA), (EDM+ CCA) and CCA alone. Considering that there were 1.1 fewer days of hospital stay for patients managed with (EDM+ CVP+ CCA) compared with those managed with (CVP+ CCA), the cost savings per patient was 911. As length of hospital stay is the main driver of cost, if there was no difference in length of hospital stay for these strategies, the cost savings per patient with (EDM+ CVP+ CCA) would be reduced to 369. Cost values were those for year As the difference in mortality rates of (EDM+ CVP+ CCA) compared with (CVP+ CCA) was not statistically significant, (EDM+ CVP+ CCA) did not always appear to be the best option. The ICER for (EDM+ CVP+ CCA) compared with (EDM+ CCA) was per QALY. Strategies such as that did not include esophageal Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 10

11 Doppler ultrasound appeared to have lower effectiveness and higher cost and were considered as dominated. For these cases ICERs were not calculated. The economic study included in the health technology assessment report 1 was an economic evaluation using partial economic modeling involving pairwise comparisons. The strategies investigated were (EDM+ CVP+ CCA), (CVP+ CCA), (EDM+ CCA) and CCA alone. From a Monte Carlo analysis with 1000 iterations and considering a threshold of 30,000/QALY it appeared that compared with (CVP+ CCA), (EDM+ CVP+ CCA) was more effective and less costly for both best case and worst case scenarios. Best and worst case scenarios differed in terms of the costs associated with EDM, length of hospital stay, facility used (general ward or intensive care unit [ICU]) and length of survival per additional survivor. Similar analyses suggested that (EDM+ CCA) was more effective and less costly than CCA. The additional cost per additional survivor that would need to be incurred before EDM would no longer be considered cost-effective is shown in Table 6. The results were statistically significant for (EDM+ CVP+ CCA) compared with (CVP+ CAA) but not statistically significant for (EDM+ CCA) compared with CCA. Table 6: Average additional cost per additional survivor Comparison Patient type Scenario Average extra cost per additional survivor that would need to be incurred before EDM would no longer [EDM+ CVP+ CCA] versus [CVP+ CCA] [EDM+ CCA] versus [CCA] High risk surgical patients High risk surgical patients be considered cost-effective (95% CI) Best case 4,441 ( 2,151 to 6,732) Worst case 642 ( 225 to 1060) Best case 11,588 (- 2,529 to 25,705) Worst case 1,879 (- 920 to 4678) CCA= conventional clinical assessment, CI= confidence interval, CVP= central venous pressure monitoring, EDM= esophageal Doppler monitoring Limitations Though evidence on EDM was available, none of the studies compared EDM with methods specifically using an arterial line or arterial catheter. There was overlap of RCTs included in the health technology assessment report and the two systematic reviews; four RCTs were included in all three reports. Hence, it should be noted that the results of the three reports are not completely exclusive and effects may be overemphasized. Not all outcomes were reported in all studies. Complication rates were reported in most studies but data on major or serious complications were sparse. It was not always clear what constituted major or severe complications. Though in most of the studies, colorectal surgery was mainly considered there was some heterogeneity in the surgical procedures considered. Laparoscopic and open surgeries, and colon and rectal surgeries were considered together. While this could increase generalizability, it is difficult to know specifically which types of surgeries would benefit most from using EDM. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 11

12 Standard of care could be different in different institutions and this could affect the outcomes achieved and confound the comparison of outcomes achieved by different methods. Length of the surgical procedure could have an impact on the outcome investigated. There is potential for subjectivity in the determination of length of hospital stay as the surgeon s perception of readiness of a patient to be discharged could vary. Comparisons of EDM strategies with all available strategies for intraoperative fluid management in adults undergoing surgery were not available. The results presented here pertain to some specific strategies. None of the included studies were conducted in Canada hence results may not be generalizable to the Canadian setting. CONCLUSIONS AND IMPLICATIONS FOR DECISION OR POLICY MAKING Evidence for EDM was available from one health technology assessment report comprising of a systematic review and economic evaluation, two systematic reviews, four clinical studies comprising of two RCTs and two non-randomized studies, and one economic study. No studies comparing EDM with methods specifically using an arterial line or arterial catheter were identified. In most instances, available evidence suggests a trend towards reduction in length of hospital stay, complication rates and mortality for intraoperative fluid management strategies using EDM compared to those using CVP or CAA in adults undergoing surgery. However, results need to be interpreted with caution as the differences were not statistically significant in several instances. Strategies with EDM appeared to be more cost-effective than CVP or CAA strategies without EDM. It should be noted that recent advances in peri-operative care could off- set some of the clinical benefits observed in previously published studies. Standard of care could vary at different institutions and this could impact the outcomes achieved. Such implications could impact the decision making. PREPARED BY: Canadian Agency for Drugs and Technologies in Health Tel: Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 12

13 REFERENCES 1. Mowatt G, Houston G, Hernandez R, de VR, 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 Technol Assess [Internet] Jan [cited 2013 Feb 20];13(7):iii-xii, 1. Available from: 2. Agency for Healthcare Research and Quality. Esophageal Doppler ultrasound-based cardiac output monitoring for real-time therapeutic management of hospitalized patients: a review [Internet]. Rockville (MD): AHRQ; p. [cited 2013 Mar 1]. (Technology Assessment Program). Available from: Based_Cardiac_Output_Monitoring.pdf 3. Walsh SR, Tang T, Bass S, Gaunt ME. Doppler-guided intra-operative fluid management during major abdominal surgery: systematic review and meta-analysis. Int J Clin Pract Mar;62(3): Straight from the heart [Internet]. Plymouth Meeting (PA): ECRI; 2009 Dec. (Health Devices). [cited 2013 Mar 7]. Available from: 5. Alhashemi JA, Cecconi M, Hofer CK. Cardiac output monitoring: an integrative perspective. Crit Care [Internet] [cited 2013 Jul 3];15(2):214. Available from: 6. Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, Hamel C, et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Methodol [Internet] Feb 15 [cited 2013 Feb 25];7:10. Available from: 7. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health [Internet] Jun [cited 2013 Jan 10];52(6): Available from: 8. Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. BMJ [Internet] Aug 3 [cited 2013 Mar 4];313(7052): Available from: 9. Srinivasa S, Taylor MH, Sammour T, Kahokehr AA, Hill AG. Oesophageal Dopplerguided fluid administration in colorectal surgery: critical appraisal of published clinical trials. Acta Anaesthesiol Scand Jan;55(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 Jan;63(1): Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 13

14 11. 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? Br J Anaesth Aug;109(2): 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. Br J Anaesth Jan;108(1): Figus A, Wade RG, Oakey S, Ramakrishnan VV. Intraoperative esophageal Doppler hemodynamic monitoring in free perforator flap surgery. Ann Plast Surg Dec Kuper M, Gold SJ, Callow C, Quraishi T, King S, Mulreany A, et al. Intraoperative fluid management guided by oesophageal Doppler monitoring. BMJ. 2011;342:d Maeso S, Callejo D, Hernandez R, Blasco JA, Andradas E. Esophageal Doppler monitoring during colorectal resection offers cost-effective improvement of hemodynamic control. Value Health Sep;14(6): Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 14

15 ABBREVIATIONS ASA CCA CI CVP DB EDM GDT HTA ICER ICU N NR OR QALY RCT SD SR TECO UK WMD Z American Society of Anesthesiologists Score conventional clinical assessment confidence interval central venous pressure double blind esophageal Doppler monitoring goal-directed therapy health technology assessment incremental cost effectiveness ratio intensive care unit number of patients not reported odd s ratio quality adjusted life year randomized controlled trial standard deviation systematic review transesophageal cardiac output United Kingdom weighted mean difference zero fluid balance approach Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 15

16 APPENDIX 1: Selection of Included Studies 153 citations identified from electronic literature search and screened 135 citations excluded 18 potentially relevant articles retrieved for scrutiny (full text, if available) 5 potentially relevant reports retrieved from other sources (grey literature, hand search) 23 potentially relevant reports 15 reports excluded: - intervention or comparison not of interest (2) - outcomes not of interest (2) -study design not of interest (2) -studies in systematic review already included in a more recent or comprehensive HTA or systematic review (3) - duplicate (1) -article unavailable (1) -other (review articles, editorials, guidance document)(4) 8 reports included in review Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 16

17 APPENDIX 2: Characteristics of Included Studies First Author, Publication Year, Country Study Design, Duration Health Technology Assessment (HTA) Mowatt, , UK HTA (SR included10 RCTs, 1 economic modeling); The SR was based on the AHRQ report 2 plus 2 additional RCTs) During surgery or hospitalizatio n for critical care Patient Characteristic s, Sample Size (n) SR Adults undergoing surgery (hip fracture repair, cardiac, bowel, colorectal, general, urological or gynaecological) or adults managed in critical care facility who require cardiac output monitoring. Age (years): EDM: 33-82, Control: Female: 273 (39%) Male: 421 (61%), NR: 265 Intervention SR EDM (CardioQ [8 RCTs], HemoSonic 100 [1 RCT], TECO [1 RCT]) Comparator s/ comparison s SR During surgery: (EDM+ CVP+ CCA) vs (CVP+ CCA), 5 RCT, 453 patients. (EDM+ CCA) vs (CVP+ CCA), 1 RCT, 61 patients. (EDM+ CCA) vs CCA, 3 RCTs, 139 patients. In critically ill patients: (EDM+ CVP+ CCA) vs (CVP+ CCA), 2 RCT, 336 patients. Outcomes Measured SR Mortality, length of hospital stay, complications HTA (economic modeling and SR); Costeffectiveness analysis (partial economic modeling using pairwise comparisons) Systematic review and meta-analysis N= 959 Economic evaluation High risk surgical patients. Critically ill hospitalized patients. Economic evaluation EDM+ CVP+ CCA] versus [CVP+ CCA] and [EDM+ CCA] versus [CCA] for high risk surgical patients EDM+ CVP+ CCA] versus [CVP+ CCA] for critically ill patients. (EDM device: CardioQ and CardioQP. CardioQ supports only adult probes and CardioQP supports both adult and pediatric probes) Economic evaluation Additional cost per additional QALY (presented as Incremental cost effectiveness plane) and average extra cost per additional survivor (presented as histograms) Srinivasa, , New Zealand SR (included 5 RCTs) Adults undergoing EDM (Device Control (details not Length of hospital stay, Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 17

18 First Author, Publication Year, Country Study Design, Duration Duration not specified Patient Characteristic s, Sample Size (n) surgery (4 RCTs specifically on colorectal surgery and 1 RCT included patients undergoing urological, gynecological and general surgical procedures) Intervention names not reported) Comparator s/ comparison s provided) Outcomes Measured complications, and physiological parameters Age (years): NR Female/Male: NR Abbas, , New Zealand SR (included 5 RCTs) N= NR Adults undergoing surgery (colorectal, upper GI, major abdominal) Age (years): NR Female/Male: NR N= 428 Randomized controlled trial (RCT) Brandstrup, , Denmark Multicenter, DB RCT FU= 30 days Adults undergoing elective colorectal surgery Age (years) (mean± SD)- EDM: 66.9±14.9, Z group: 68.1± 14.9 Male (number [%]): EDM: 39 (55%) Z group: 47 (59%) N= 150 EDM (Device names not reported) EDM (CardioQ- ODM) Intravenous fluid therapy according to conventional measures of CVP, heart rate and arterial blood pressure. Zero fluid balance approach (formerly known as restricted approach ). In this approach, all measured fluid losses are replaced with a goal of zero fluid balance without the replacement of the loss-to- Mortality, hospital stay, admission to ICU, return of gut function, use of inotropes, complications, and other outcomes (such as colloid/crystalloi d use, urine output, oxygen delivery) Mortality, complications, length of hospital stay, readiness for discharge, need for antiemetic or diuretic treatment and physiological changes Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 18

19 First Author, Publication Year, Country Challand, , UK Study Design, Duration Single center, DB RCT Patient Characteristic s, Sample Size (n) (EDM=71, Z group= 79) Adults undergoing major colorectal surgery (open or laparoscopic) Age (years) (mean± SD)- EDM/GDT: 66± 15.36, Control: 65.9± 14.1 Intervention EDM guided intraoperative goal-directed fluid therapy (GDT) (CardioQ) Comparator s/ comparison s third space. Control (Standard care) Outcomes Measured Mortality, complications, length of hospital stay, readiness for discharge, physiological changes Male/Female- EDM/GDT: 54/35, Control: 48/42 Non-randomized study (NRS) Figus, , Single center, UK nonrandomized study, prospective N= 179 (EDM/GDT: 89, control: 90) Adults undergoing free perforated flap surgery Age (years) (mean± SD)- EDM: 38.9± 18.0 CVP: 44.4± 17.1 EDM (DP12 CardioQ) CVP monitoring (CVP and/or arterial monitoring considered in this group) Length of hospital stay, complications, return to theater, flap survival, fluid input, output and balance Male/Female- EDM:11/39=21.6 %/88.4% CVP:16/36= 30.2%/69.8% N= 104 (EDM: 51, CVP: 53) Kuper, , UK Multi-center nonrandomized implementati on study, (before and after Patients undergoing surgery (colorectal surgery [1 center], major After EDM implementatio n Before EDM implementatio n (Control) Length of hospital stay, inhospital mortality, readmission to hospital, readmission to Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 19

20 First Author, Publication Year, Country Study Design, Duration implementati on of EDM) Patient Characteristic s, Sample Size (n) elective and emergency surgery [1 center], colorectal and orthopaedic surgery [1 center]) Intervention Comparator s/ comparison s Outcomes Measured critical care, reoperation, Number (%) of patients in various age ranges 60 years EDM: 237 (36.5%) Control: 196 (29.8%) years EDM: 167 (25.7%) Control: 175 (26.6%) 71 years EDM: 245 (37.8%) Control: 287 (43.6%) N= 1307 (After EDM: 649, Control: 658) Economic evaluation Maeso, , Costeffectiveness Spain analysis. Time horizon: until discharge Patients undergoing colorectal surgery Four strategies compared: [CCA+ CVP+ EDM] versus [CCA+ CVP] versus [CCA+ EDM] versus CCA (EDM device used not specified) ICER Hospital perspective (a good proxy for overall health care system perspective) CCA= conventional clinical assessment, CVP= central venous pressure, DB= double blind, EDM= esophageal Doppler monitoring, GDT= goal directed fluid therapy, ICER= incremental cost-effectiveness ratio, ICU= intensive care unit, N= number of patients, NR= not reported, SD= standard deviation, Z= zero fluid balance approach Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 20

21 APPENDIX 3: Summary of Study Strengths and Limitations First Author, Publication Year, Country Strengths Health Technology Assessment (HTA) Mowatt, , SR UK The objective was stated. The inclusion and exclusion criteria were stated. Comprehensive literature search (multiple databases) Study selection described and flow chart presented List of included and excluded studies provided Characteristics of individual studies were provided or were available in the AHRQ report on which this HTA was based Methods used to combine the findings of studies were appropriate Quality assessment was conducted of the AHRQ report and the additional individual studies. Conflict of interest was stated and there was none Limitations SR Article selection and data extraction were done by one reviewer and a second reviewer was consulted in case uncertainty Publication bias was not explored. Economic Objectives were stated. The strategies compared were stated The form of economic evaluation, the rationale for the choice of alternative strategies were stated Clinical effectiveness data were obtained from meta-analyses or individual RCTs. Time horizon and perspective were stated Cost data were provided Sensitivity analyses were conducted Economic Several assumptions were made which may not be applicable in all cases Systematic review and meta-analysis (SR/MA) Srinivasa, 9 The objective was stated. 2011, New The inclusion and exclusion criteria Zealand were stated. Comprehensive literature search (two databases: Medline and Study selection process was not described List of excluded studies was not provided Characteristics of the individual studies Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 21

22 First Author, Publication Year, Country Strengths EMBASE and reference lists of relevant articles) was conducted List of included studies was provided Quality assessment of studies was conducted Publication bias (constructing Funnel plot) was explored and the authors mentioned there was no evidence of publication bias Conflict of interest was stated and there was none Limitations were not described in detail Unclear if article selection or data extraction were done in duplicate Results were presented qualitatively Abbas, , New Zealand The objective was stated. The inclusion and exclusion criteria were stated. Comprehensive literature search (two databases: Medline and EMBASE) was conducted List of included studies was provided Methods used to combine the findings of studies were appropriate Quality assessment of studies was conducted Randomized controlled trial (RCT) Brandstrup, 11 Objectives were stated. 2012, Denmark Inclusion/ exclusion criteria were stated but few details. Patient characteristics, interventions, and outcomes were described. Randomized; double blind - patients and surgeons were blinded. Sample size calculation was described Intent-to-treat analysis P-values provided Study selection was not described List of excluded studies was not provided Characteristics of the individual studies were not described in detail. (However, it was mentioned that the patients in the included studies were comparable with respect to age, type of surgery, pre-operative hemoglobin levels and physiological scores and that there were no difference between theses parameters between the experimental and control groups.) Unclear if article selection or data extraction were done in duplicate Publication bias was not explored No mention of conflict of interest Potential for selection bias as the presence of both the investigating anaesthetist and surgeon was mandatory for screening patients for inclusion Generalizability limited; uncertain as to whether study patients were representative of all patients. Challand, , UK Objectives were stated. Inclusion/ exclusion criteria were stated but few details. Patient characteristics, interventions, and outcomes were Generalizability limited; uncertain as to whether study patients were representative of all patients. Esophageal Doppler Ultrasound-based Cardiac Output Monitoring for Adults 22