|
Clinical value of acute rest technetium-99m tetrofosmin tomographic myocardial perfusion imaging in patients with acute chest pain and non-diagnostic electrocardiograms |
Heller G V, Stowers S A, Hendel R C, Herman S D, Daher E, Ahlberg A W, Baron J M, de Leon C F, Rizzo J A, Wackers F J |
|
|
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 Acute rest technetium-99m (Tc-99m) tetrofosmin single-photon emission computed tomographic (SPECT) myocardial perfusion imaging in patients with a suspected myocardial infarction (MI) but a normal or non-diagnostic electrocardiogram (ECG). Patients were given an injection of 20 to 30 MCI of Tc-99 tetrofosmin at rest and SPECT imaging data were acquired between 15 minutes and 3 hours afterwards.
Economic study type Cost-effectiveness analysis.
Study population Patients presenting to emergency departments with chest pain, possibly of cardiac origin, but with a non-diagnostic ECG.
Setting Hospital emergency department. The economic study was carried out in the USA.
Dates to which data relate Source of effectiveness data Effectiveness data were based on a single study.
Link between effectiveness and cost data Costing was undertaken on the same patient sample as that used in the effectiveness study. It was unclear whether costing was performed prospectively or retrospectively.
Study sample Power calculations did not determine the sample size. Entry criteria for enrolment in the study were: acute chest pain suggestive of myocardial ischemia, a normal or non-diagnostic ECG, hospital admission and informed consent. Exclusion criteria were: acute MI at the time of evaluation, a history of MI or diagnostic Q waves, ECG changes in 2 or more leads, a history consistent with unstable angina or pregnancy. 385 non-consecutive patients were enrolled in the study at 6 centres. 28 patients (7%) were retrospectively excluded for various reasons including 5 whose SPECT images were considered uninterpretable leaving 357 patients who were included in the analysis (44% were aged over 55 years). Study criteria were applied to 4,704 consecutive patients during a 5-6 month period at 4 of the US centres to evaluate if the patients enrolled in the study were representative of eligible patients; 615 met the entry criteria. Of the eligible patients not enrolled, 35 (6.2%) had an acute MI. SPECT images were interpreted by consensus of 4 investigators blinded to patient identification or clinical information. Consensus of three investigators blinded to patient identification or image results reinterpreted all entry ECGs.
Study design This was a prospective multi-centred cohort study. Five sites were in the USA and one in Israel. Follow-up was for 30 days after discharge. 31 patients (9%) were lost to follow-up. The lost to follow-up group was reported to be not statistically different from the follow-up group in terms of admission demographics. The investigators evaluating SPECT images were blinded to patient identity and clinical information.
Analysis of effectiveness It was not explicitly stated whether the analysis was based on intention to treat or treatment completers only. The primary health outcome measures were primary and secondary cardiac events in hospital and at 30-day follow-up. Primary cardiac events consisted of ischemic cardiac death and nonfatal MI, while secondary cardiac events constituted cardiac catheterisation, coronary revascularisation and readmission for unstable angina. Sensitivity, specificity, and positive and negative predictive values were among other outcomes reported. A receiver operating characteristic (ROC) curve analysis was performed to evaluate potential differences in diagnostic approaches. A multiple logistic regression model was produced to analyse univariate predictors of acute myocardial infarction.
Effectiveness results The primary cardiac events in hospital and at 30-day follow-up, respectively, were as follows:
ischemic death, 0 and 0;
nonfatal MI, 20 (6%) and 0.
The secondary cardiac events in hospital and at 30-day follow-up, respectively, were:
cardiac catheterisation, 88 (25%) and 6 (2%);
coronary revascularisation, 34 (10%) and 4 (1%).
204 (57%) images were interpreted as normal by consensus and 153 (43%) were considered abnormal. Of the 20 (6%) patients with an acute myocardial infarction during their hospital stay, 18 had abnormal images. This represented a sensitivity of 90% (95% CI: 86% - 93%). Specificity was 59.5% (95% CI: 55% - 65%). Positive and negative predictive values were 12% and 99%, respectively. Predictors of acute MI in the multiple logistic regression analysis were male gender, diabetes, non-diagnostic ECG, and an abnormal SPECT.
Clinical conclusions SPECT imaging is a good predictor of MI and can safely be used to reduce unnecessary hospital admissions.
Measure of benefits used in the economic analysis The benefit measure was acute MI detected (sensitivity).
Direct costs No discounting was performed because of the short time frame of the study. Net cost analysis was performed. Charge data were obtained from the 5 US medical centres in the study and consisted of hospital charges for SPECT imaging, and costs for hospital stay. An average cost was calculated based on 2 alternative admission strategies:
(1) that all patients with abnormal SPECT images would be admitted and patients with normal images would not be admitted; and
(2) that in addition to the above an additional 20% of patients would be admitted on clinical grounds.
Two cost scenarios were considered:
(1) the mean cost for a hospital stay and for SPECT imaging were used in the calculation and
(2) the mean cost of SPECT imaging and the mean of the lower 10th percentile of hospital costs were used.
The perspective adopted in the cost analysis was not explicitly specified. The date of the price data was not explicitly specified.
Sensitivity analysis Although not described by the authors as sensitivity analyses, they did calculate costs based on 3 assumed levels of sensitivity: 65%, 80% and 90% (based on the strictness of criteria for consideration of a normal image).
Estimated benefits used in the economic analysis SPECT imaging would have a sensitivity of 90%, 80%, or 65% in predicting myocardial infarction depending on the strictness of criteria for consideration of a normal image (strict, intermediate, and less strict, respectively).
Cost results Assuming a sensitivity of 90%, 43% of patients would be admitted at a cost saving of $4,258 using the first cost scenario (mean costs for both hospital stay and SPECT imaging) and $789 using the lower cost scenario for hospital costs. Details were not given for the scenario that a further 20% of patients with normal images would be admitted, except that, in the least favourable cost scenario, it would result in a saving per patient of $543.
Synthesis of costs and benefits Mean cost per MI missed and mean low cost per MI missed were calculated as measures of cost-effectiveness, yielding values of:
$81,372 and $44,520 for the strict scenario,
$73,850 and $37,317 for the intermediate scenario,
and $73,850 and $37,317 for the less strict interpretative scenarios.
Authors' conclusions Imaging can accurately predict acute MI in patients with a non-diagnostic ECG and can safely reduce the number of unnecessary hospital admissions with consequent reduction in costs.
CRD COMMENTARY - Selection of comparators Although the no-SPECT imaging option was implicitly regarded as the comparator, the study suffered from the lack of a properly defined comparator.
Validity of estimate of measure of benefit The estimate of benefit is likely to be internally valid given the blinded method use in image interpretation.
Validity of estimate of costs It is not possible to assess the internal validity of the costing as insufficient details were provided in the paper. Only cost savings were given, as opposed to two transparent sets of total costs corresponding to the intervention and the comparator. No prices or dates were provided, and there appears to have been no extensive analysis of resource use. Charge data were used as opposed to true costs, which may have adversely affected the generalisability of the cost results to other settings or countries.
Other issues The authors' conclusion seems to be reasonably justified. The fact that this study was carried out in institutions of varying size and geographic location may have increased the generalisability of the results. Adequate comparisons were made with other studies.
Implications of the study In this study, and that of Tatum et al, 2% to 5% of patients with normal rest SPECT images underwent subsequent revascularisation. This indicates that selected patients with normal SPECT images warrant further (outpatient) evaluation.
Bibliographic details Heller G V, Stowers S A, Hendel R C, Herman S D, Daher E, Ahlberg A W, Baron J M, de Leon C F, Rizzo J A, Wackers F J. Clinical value of acute rest technetium-99m tetrofosmin tomographic myocardial perfusion imaging in patients with acute chest pain and non-diagnostic electrocardiograms. Journal of the American College of Cardiology 1998; 31(5): 1011-1017 Other publications of related interest Tatum J L, Jesse R L, Kontos M C et al. Comprehensive strategy for the evaluation and triage of the chest pain patient. Annals of Emergency Medicine 1997;29:116-23.
Indexing Status Subject indexing assigned by NLM MeSH Electrocardiography; Female; Heart /radionuclide imaging; Humans; Male; Middle Aged; Myocardial Infarction /radionuclide imaging; Organophosphorus Compounds; Organotechnetium Compounds; Predictive Value of Tests; ROC Curve; Radiopharmaceuticals; Tomography, Emission-Computed, Single-Photon AccessionNumber 21998000650 Date bibliographic record published 30/09/2000 Date abstract record published 30/09/2000 |
|
|
|