|
Hepatic tumour imaging using iron oxide MRI: comparison with computed tomography, clinical impact, and cost analysis |
Schultz J F, Bell J D, Goldstein R M, Kuhn J A, McCarty T M |
|
|
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 Hepatic tumour imaging using iron oxide magnetic resonance imaging.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised patients who underwent both Fe-MRI and CECT, who had solid hepatic lesions detected by CECT.
Setting The setting was a hospital; the economic analysis was carried out in the USA.
Dates to which data relate Effectiveness, resource use, and cost data were collected between November 1996 and July 1998. The price year was not reported.
Source of effectiveness data Effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was carried out prospectively on the same patient sample as that used in the effectiveness analysis.
Study sample More than 1,000 patients underwent contrast-enhanced MRI of the abdomen. 57 were identified who underwent Fe-MRI. This included 23 men and 34 women, with a mean age of 58 years. Fifty patients received a concomitant CECT scan. Of these 50 patients, 35 had solid focal hepatic lesions and underwent comparison of CECT and Fe-MRI findings. No power calculations were reported. Baseline characteristics in terms of indication for imaging were given for some, but not all, patients and not according to those selected for inclusion.
Study design This was a retrospective, diagnostic evaluation by chart review, carried out at a single centre. Patient records were followed-up until after diagnosis or record of any change in clinical management. No patient was lost to follow-up. The extent to which recorded data were available was not reported.
Analysis of effectiveness Given that no indication was given of the lack of chart data it is impossible to determine whether the analysis was based on intention to treat or on treatment completers only. The primary health outcomes used in the analysis were the detection of lesions, size of lesions detected, sensitivity, and changes in clinical management. The reference standard for diagnosis was a combination of intraoperative ultrasonography (IOUS) and pathological specimens. It was stated that 22 of the 35 patients were selected for abdominal exploration. Of these 19 had IOUS, 2 had resection without IOUS and 1 had neither. Of the 19 who had IOUS, 11 also had resection and the correlation in terms of number of lesions between IOUS and resection was stated to be 100%.
Effectiveness results Fe-MRI identified 157 lesions in 57 patients, (2.8 lesions per patient, range: 0 - 14).
Of the 35 patients who underwent both CECT and Fe-MRI, Fe-MRI detected more lesions (3.9 per patient) than CECT (2.2 per patient; p=0.016).
The "average" size of lesions detected by Fe-MRI was smaller than those detected by CECT (2.5 versus 3.4cm; p=0.017).
From the 21 patients with IOUS and/or pathology data the sensitivity of CECT (58%) was found to be lower than that of Fe-MRI (86%; p<0.001) and IOUS (100%; p<0.001). The data presented in a table were similar, but different: sensitivity was 85% for Fe-MRI versus 55% for CT. There was 1 false positive for Fe-MRI and 2 for CT. The positive predictive value (PPV) was 98% for Fe-MRI and 95% for CT. 14 lesions were detected by pathology, 71 by IOUS, 61 by Fe-MRI and 41 by CT. Fewer lesions were detected by CECT than by IOUS and pathology (41 versus 71 lesions; p=0.003).
Clinical management was altered in 67% of the 57 patients who underwent Fe-MRI based on Fe-MRI findings.
Clinical conclusions In those patients with lesions shown to be solid by CT, Fe-MRI detected statistically significantly more lesions. In those patients with IOUS or pathology data, the sensitivity of CT was statistically significantly lower than that of Fe-MRI. Fe-MRI also detected fewer false positives, giving a higher PPV. Of those patients who underwent Fe-MRI for the liver, clinical management was altered in 67%.
Measure of benefits used in the economic analysis The authors did not report a summary health benefit and left clinical outcomes disaggregated. Hence, a cost-consequences analysis was conducted.
Direct costs Direct costs were not discounted due to the short time horizon of the study (less than one year). Quantities and unit costs were reported separately by procedure. The direct costs included were those that related to the cost of the test (as billed to the patient and including professional fees) and the savings from avoiding other investigations. The quantity/cost boundary adopted was that of the hospital. The source of cost data was the authors' hospital. The price year was not reported.
Statistical analysis of costs No statistical analysis of costs was reported.
Indirect Costs Indirect costs were not included.
Sensitivity analysis No sensitivity analyses were reported.
Estimated benefits used in the economic analysis The reader is referred to the effectiveness results reported earlier.
Cost results The costs of tests were $1,625 for CECT, $1,740 for MRI, and $3,746 with image-guided biopsy. Savings around changes in clinical management with Fe-MRI were $207,548. The costs of Fe-MRI (for 57 patients) were $99,180. This led to overall savings of $1,901 per patient.
Synthesis of costs and benefits Authors' conclusions The authors argued that Fe-MRI is a powerful imaging technique, with greater hepatic tumour detection sensitivity than CECT. Moreover, it is an economically feasible imaging method that will alter the clinical management in most patients imaged.
CRD COMMENTARY - Selection of comparators The explicit comparator was a traditional imaging technique. However, there was a contradiction between the technologies studied in terms of effectiveness and cost. In terms of cost, the total was given for the use of Fe-MRI in hepatic pathology. However, this was compared with the comparator of no Fe-MRI, i.e. the net savings resulting from changing management due to the use of Fe-MRI. In terms of effectiveness, only 35 of the patients having Fe-MRI for liver pathology were selected for comparison with CT. Therefore, the costing and effectiveness differences do not refer to a comparison of the same technologies.
Validity of estimate of measure of effectiveness The analysis was based on a diagnostic test evaluation, which was appropriate for the study question. Data were collected from a retrospective chart review. However, it was not clear how patients were selected since no selection criteria were given for the 57 hepatic Fe-MRI patients or for the 35 patients stated to have a primary or metastatic cancer by CT. Baseline characteristics were not given for these samples. Therefore, we do not know to which population these results are relevant. The authors noted that the accuracy of Fe-MRI is still unknown. The effectiveness measures should have been appropriate. However, in terms of the size of lesion there was no reference test. In terms of number of lesions, a reference was available, i.e. IOUS and/or pathology. However, this was only available for 21 patients, all of whom had lesions. Sensitivity seems to have been calculated in terms of the probability of detecting a lesion given the presence of lesions, but, since there were no reference data on patients testing negative for lesions, it must have been assumed that there were no false negatives. Therefore, this will probably over-estimate sensitivity. A similar problem exists for the PPV in that those Fe-MRI which did not detect any lesions were not included. Some of these might have been false negatives, which would reduce the PPV. The evidence for a change in management by Fe-MRI, as stated, applies for the entire 57 Fe-MRIs and it was not stated which Fe-MRI result resulted in which change. Therefore, we do not know the role of any diagnostic procedure, including CT and Fe-MRI, in the decision-making process.
Validity of estimate of measure of benefit The authors did not derive a measure of health benefit. The analysis was therefore categorised as a cost-consequences study.
Validity of estimate of costs It is useful that costs were presented in such a way that total costs can be broken down to number of procedures and their unit costs. However, it would be difficult to generalise to other settings without resource quantities and unit costs. As stated above, the cost comparator was different from the effectiveness comparator. In fact, we do not know the extent of bias by patient record selection or confounding due to other technologies such as other tests, interventions or the role of judgement, since the management strategies used were not reported and it was not clear how record information was selected.
Other issues The authors made appropriate comparisons of their findings with those from other studies, but did not address the issue of generalisability to other settings. The variability in comparators and their lack of precise definition and the absence of sample characteristics means that the authors' conclusions are highly questionable. The authors acknowledged that the study did not consider the costs and burden of complications for patients undergoing further invasive diagnostic evaluations or invasive pre-operative evaluation with CTAP. Thus, Fe-MRI is likely to have additional clinical and economic impact on patient management that is yet to be measured.
Implications of the study The authors argued that Fe-MRI is a cost-effective, non-invasive method for pre-operative evaluation. "It should be used as an integrated method of pre-operative evaluation along with sonography and CT, but may replace the need for CTAP". They also stated that "By selecting more appropriate surgical candidates and integrating other treatment methods, the morbidity and mortality associated with hepatic resection can be potentially lowered with improved patient selection". Given the methodological problems outlined above, these implications seem difficult to support. The study can serve as a source of ideas for further research.
Bibliographic details Schultz J F, Bell J D, Goldstein R M, Kuhn J A, McCarty T M. Hepatic tumour imaging using iron oxide MRI: comparison with computed tomography, clinical impact, and cost analysis. Annals of Surgical Oncology 1999; 6(7): 691-698 Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Aged, 80 and over; Cost Savings; Cost-Benefit Analysis; Female; Ferric Compounds; Health Care Costs; Humans; Liver Neoplasms /diagnosis /radiography; Magnetic Resonance Imaging /economics /standards; Male; Middle Aged; Prognosis; Retrospective Studies; Tomography, X-Ray Computed /economics /standards; Ultrasonography, Interventional /economics /standards AccessionNumber 21999002217 Date bibliographic record published 30/06/2002 Date abstract record published 30/06/2002 |
|
|
|