|
Cardioversion in patients with atrial fibrillation and left atrial thrombi on initial transesophageal echocardiography: should transesophageal echocardiography be repeated before elective cardioversion? A cost-effectiveness analysis |
Seto T B, Taira D A, Manning W T |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The study evaluated a follow-up transesophageal echocardiography (TEE) after four weeks of warfarin treatment following left atrial thrombus on transesophageal echocardiography.
Study population The study population comprised patients aged over 70 who presented with atrial fibrillation of more than 2 days' duration as determined by TEE, and for whom elective cardioversion was desired. Patients were those with a left atrial thrombus identified on initial TEE who were discharged from hospital to receive 3-4 weeks of warfarin therapy.
Setting The setting was the community and secondary care.
Dates to which data relate Effectiveness estimates were based on studies published between 1995 and 1998. Prices were from 1996. It was not clear how resource use estimates were derived.
Source of effectiveness data Effectiveness estimates were based on a review of previously completed studies and on assumptions.
Modelling A computerised decision-analytic model was used to synthesise costs and effectiveness estimates.
Outcomes assessed in the review The authors assessed the following outcomes of the follow-up TEE strategy:
probability of a residual thrombus after 1 month of warfarin therapy;
haemorrhage for those patients with residual thrombus;
mortality following residual thrombus;
stroke following no residual thrombus; and
mortality following stroke.
The authors assessed mortality from stroke for the "no follow-up" TEE strategy. In addition, the authors estimated risks associated with TEE, and the risk of haemorrhage.
Study designs and other criteria for inclusion in the review Sources searched to identify primary studies The authors searched MEDLINE and reviewed reference lists manually.
Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included About 35 studies were used as background for the clinical assumptions in the model.
Methods of combining primary studies The results from studies were combined using both weighted averages and narrative methods. For example, the estimate of the probability of residual thrombus after 1 month of warfarin therapy was based on a weighted average of estimates from four studies of follow-up TEE. In contrast, the authors used a narrative estimate of the probability of stroke following cardioversion of 0.5% when the study they quote estimated 0% - 0.4%. This was done in order to be conservative and to bias the estimate against the follow-up TEE.
Investigation of differences between primary studies The authors discussed the patients included in the studies underlying the estimates of effectiveness in the model and reported that they used conservative estimates.
Results of the review FOLLOW-UP TEE STRATEGY.
The authors estimated that 24% of patients would experience residual thrombus as detected in follow-up TEE.
The risk of haemorrhage in these patients was 0.12% per month, and mortality from haemorrhage was 41%.
Stroke following no residual thrombus was 0.5%, and mortality following stroke was 18%.
NO FOLLOW-UP TEE STRATEGY
The mortality from stroke was 18%.
The likelihood of esophageal perforation following TEE was estimated at 0.02%, and the mortality from such perforation was estimated at 8%.
The probability of major haemorrhage during warfarin therapy was estimated at 1.7% per year.
Methods used to derive estimates of effectiveness The authors based some assumptions on clinical evidence from the literature, but tried to be conservative in their estimates. For example, the probability of death following perforation during TEE was estimated at 8% whereas the literature presented estimates ranging from 0% to 4%.
Estimates of effectiveness and key assumptions Key assumptions of the model were that the only morbidity from cardioversion is cerebral embolism and that the only complication of TEE is esophageal perforation.
Measure of benefits used in the economic analysis The measure of benefits used in the economic analysis was quality-adjusted life years (QALYs). Utilities for the different health states of the model were derived from published studies. The quality of life of stroke patients with long-term morbidity was 0.5, major haemorrhage was 0.6, minor stroke was 0.7 and haemorrhage without long-term morbidity was 0.9. It was not clear on whose values these estimates were based.
Direct costs Direct costs of TTE, cardioversion, stroke, haemorrhage, complications and anticoagulation therapy were included. Costs were discounted at 5% per year. The authors stated that the cost of TEE and cardioversion was obtained by converting the current procedural terminology codes into resource-based relative value units. Unit costs were then obtained from published studies and from Medicare hospital accounting information. All costs were from 1996, and costs dating from earlier were reflated to 1996 costs.
Statistical analysis of costs Statistical tests were not carried out.
Indirect Costs No indirect costs were included.
Sensitivity analysis Sensitivity analyses were carried out by varying the estimates of costs, clinical probabilities, life expectancies and utilities. One- and two-way sensitivity analyses, as well as threshold analyses, were used. Threshold analyses were based on an incremental cost-effectiveness ratio (ICER) of $50,000 and dominance.
Estimated benefits used in the economic analysis The follow-up TEE generated 8.0 QALYs, whereas the "no follow-up" strategy generated 7.8 QALYs. The discount rate was 5%.
Cost results Costs for the two strategies were $2,380 (follow-up TEE) and $2,644 (no follow-up TEE). The discount rate was 5%.
Synthesis of costs and benefits The follow-up TEE strategy was dominant in the base-case analysis (i.e. more effective and less costly), and the estimated costs and benefits were not combined. The authors commented that the cost-effectiveness ratio increased to near $50,000 per QALY when the risk of thromboembolism after cardioversion in patients with residual left atrial thrombi was reduced from the baseline estimate of 50% to 35%. As long as the risk of thromboembolism in patients with residual thrombi was more than 42% the follow-up TEE strategy is more effective and less costly. If the risk is as low as 35% one must be willing to pay $50,000 per QALY to choose follow-up TEE strategy. If the risk is less than 34% the follow-up TEE strategy is dominated by the "no follow-up" strategy. The risk of postcardioversion stroke (estimated at 0.5% in the baseline analysis) could be as much as 6.1% before the cost per QALY exceeds $50,000. The results did not change significantly when probabilities and costs of complications after TEE were varied across plausible ranges, nor when quality of life estimates were varied.
Authors' conclusions The authors concluded that a follow-up TEE with cardioversion, only after thrombus resolution, may be less costly and slightly more effective than a strategy in which patients undergo cardioversion without a follow-up TEE. They commented that the decision is dependent on the risk of postcardioversion stroke in patients with undetected residual left atrial thrombi.
CRD COMMENTARY - Selection of comparators The authors compared follow-up TEE with no follow-up TEE, which is a relevant alternative in clinical decision-making and therefore appropriate as a comparator in this analysis.
Validity of estimate of measure of effectiveness The measure of effectiveness was based on probabilities of developing a range of cardiovascular conditions following atrial fibrillation and left atrial thrombi. The authors recognised that many of these probabilities were based on assumptions and on uncertain estimates. The authors therefore made all assumptions transparent and explicit and justified their estimates with relevant clinical studies. Indeed, the authors were conservative in their estimates for the base-case analysis, and estimated thresholds for values at which follow-up TEE would no longer be dominant or cost-effective at $50,000/QALY. The cost-effectiveness estimate was sensitive to the estimate of risk of thromboembolism following cardioversion of patients with non-resolved thrombus.
Validity of estimate of measure of benefit The utilities were based on published data but were assigned by the authors. It was difficult to assess the validity of the utilities that were used in the model, although a reference was provided.
Validity of estimate of costs It was difficult to assess the validity of the cost estimates. The authors claimed not to have used charges in their study, however costs and unit prices were not estimated separately, and sources for unit prices were not explicitly reported in the paper.
Other issues The authors did not discuss their results in the light of other economic studies of the follow-up TEE strategy. The issue of generalisability of the results was not explicitly discussed but was addressed to some extent in the sensitivity analysis of the effect estimates. It is implicit that the results will depend on baseline disease risks in populations that are considered for follow-up TEE, and also that results cannot be extrapolated beyond a population of patients with nonvalvular atrial fibrillation with left atrial thrombi on initial TEE. The authors recognised a number of limitations to their study, primarily related to the clinical assumptions in the model.
Implications of the study The authors commented that this study was conducted in the absence of data from randomised controlled trials. The study is very relevant for the systematic consideration of available clinical data and for the identification of critical values that influence the decision-making process.
Source of funding Dr Manning is part supported by an Established Investigator Grant (9740003N) of the American Heart Association, Dallas, Tex.
Bibliographic details Seto T B, Taira D A, Manning W T. Cardioversion in patients with atrial fibrillation and left atrial thrombi on initial transesophageal echocardiography: should transesophageal echocardiography be repeated before elective cardioversion? A cost-effectiveness analysis. Journal of the American Society of Echocardiography 1999; 12(6): 508-516 Other publications of related interest Comment in: Journal of the American Society of Echocardiography 1999;12(12):1125-6.
Indexing Status Subject indexing assigned by NLM MeSH Anticoagulants /therapeutic use; Atrial Fibrillation /economics /therapy; Cost-Benefit Analysis; Decision Trees; Echocardiography, Transesophageal /economics; Electric Countershock /economics; Humans; Probability; Quality-Adjusted Life Years; Risk Factors; Thrombosis /prevention & control /ultrasonography AccessionNumber 21999001209 Date bibliographic record published 31/03/2002 Date abstract record published 31/03/2002 |
|
|
|