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.

The study investigated the effectiveness of two different laparoscopic techniques for gastrojejunal (GJ) anastomosis. The techniques studied were circular end-to-end (EEA) and linear cutting (GIA) staplers.

Treatment.

Cost-effectiveness analysis.

The study population comprised patients undergoing stapled GJ anastomosis for isolated Roux-en-Y gastric bypass (RYGBP). The criteria for RYGBP included a body mass index (BMI) greater than 40 and medical co-morbidities.

The setting was not stated, but it is assumed to have been secondary care. The economic study was carried out in the USA.

The effectiveness data were gathered between September 1999 and September 2001. The resource dates and price year were not stated.

The effectiveness data were derived from a single study.

It was unclear whether the costing and collection of resource data were carried out prospectively on the same patient groups. However, given the study design used, it is likely that the cost data were derived retrospectively.

There was no evidence of any sample size or power calculations. The sample was selected on the basis of a medical chart review of all patients who underwent stapled GJ anastomosis for isolated RYGBP over the specified 2-year period. There was no full justification for the choice of patient sample in terms of disease, treatment and generalisability of the findings. The sample comprised 61 patients overall, with 32 in the EEA group and 29 in the GIA group. Further demographic details (e.g. age, gender distribution) were not given.

This was a retrospective cohort study that was conducted in a single centre over 2 years. The length of follow-up was 608 months. No patients appear to have been lost to follow-up.

The primary health outcomes used were:

complications such as wound infection, pneumonia and anastomotic leak;

weight loss, defined as the percentage excess weight lost (%EWL);

operative time; and

the length of hospital stay.

Weight was measured using clinical scales. Anastomotic leak was measured with a closed-suction drain. One of the 61 patients was not included in the analysis because of late conversion from a laparoscopic to an open surgical approach. There were no significant differences in the BMI and co-morbidities of the patient groups at baseline, although it was unclear whether this was the case at analysis.

The operative time was shorter for the GIA group (145 +/- 27.9 minutes) than for the EEA group (180 +/- 56.1 minutes), (p=0.003).

The average hospital length of stay was shorter (non significantly) for the EEA group (4.2 +/- 3.4 days) than for the GIA group (5.1 +/- 4.1 days), (p=0.34).

Postoperative complications (including wound infection, pneumonia and anastomotic leak) were higher (non significantly) for the EEA group (48.3%) than for the GIA group (27.6%), (p=0.19).

Two early reoperations were performed in the GIA group for anastomotic leaks.

Wound infections were higher (non significantly) in the EAA group (21.9%) than in the GIA group (6.9%), (p=0.08).

There was no significant difference between treatment groups in terms of weight loss.

The authors concluded that, although operative time was significantly shorter in the GIA group, there were no differences between the groups for the other outcomes. The technique to use should be selected on the basis of surgeon preference, operative time and the potential for serious complications.

No summary measure of benefits was used in the economic analysis. Whilst the authors demonstrated therapeutic equivalence in general terms, the study may best be described as a cost-consequences analysis.

Hospital (average) costs were included in the analysis. These were for the operating room, room or board, pharmacy, radiology, laboratory, cardiology and other. It is assumed that medical charts were used to determine resource use, but this was unclear. It was also unclear how, or from what year, the estimation of prices or unit costs was obtained. The resource quantities and the costs were not reported separately. Discounting was not reported, but it was unlikely to have been relevant given the time period during which the costs were incurred. The price years were not stated.

The costs were reported as descriptive statistics (mean values plus standard deviations).

No indirect costs were reported.

US dollars ($).

There was no sensitivity analysis of the costs.

See the 'Effectiveness Results' section.

The hospital costs were divided into operating room, room or board, pharmacy, radiology, laboratory, cardiology and other cost categories:

The total costs were $9,761 (+/- 6,567) for the EEA group and $10,830 (+/- 6,588) for the GIA group, (p=0.53).

The operating room costs were $6,291 (+/- 2,533) for the EEA group and $6,317 (+/- 1,452) for the GIA group, (p=0.96).

The room or board costs were $2,150 (+/- 2,546) for the EEA group and $2,580 (+/- 2,538) for the GIA group, (p=0.51).

The pharmacy costs were $472 (+/- 807) for the EEA group and $817 (+/- 1,810) for the GIA group, (p=0.35).

The radiology costs were $162 (+/- 259) for the EEA group and $322 (+/- 405) for the GIA group, (p=0.13).

The laboratory costs were $280 (+/- 412) for the EEA group and $391 (+/- 665) for the GIA group, (p=0.44).

The cardiology costs were $38.3 (+/- 48.0) for the EEA group and $29.9 (+/- 12.4) for the GIA group, (p=0.37).

Other costs were $452 (+/- 730) for the EEA group and $461 (+/- 695) for the GIA group, (p=0.96).

The costs and benefits were not combined.

Although operative time was significantly shorter in the linear cutting (GIA) group, other outcomes (complications, length of stay, weight loss) and costs were not significantly different between the study groups. The authors suggested that future practice should be based upon surgeon preference, operative time and the likelihood of serious complications.

The selection of the comparators was justified on the basis that they represented the most frequently employed anastomotic techniques. You should decide if the represent widely used technologies in your own setting.

The analysis was based upon a retrospective cohort study design. This is associated with threats to internal validity in relation to the comparability of the study groups and the potential effects of confounding upon the results in the absence of randomisation. In particular, it was unclear whether the reported baseline comparability of BMI and co-morbidities remained at analysis. It was not possible to determine whether the study sample was representative of the study population as insufficient details of the patient characteristics were reported. This potentially limits the external generalisability of the results.

No summary benefit measure was used in the economic analysis. The effectiveness analysis generally demonstrated that the two treatments were equally effective in terms of the clinical outcomes.

It appears that relevant costs pertaining to the hospital perspective have been included. These were appropriately justified on the basis of observed differences in operative times between the technologies, and in relation to the potential variation in costs due to patient co-morbidity and postoperative complications. However, results based upon this narrowly defined perspective should be treated with caution in practice. The sources and dates for the resource and price data were unclear, and no separation or statistical analysis of the quantities and costs was performed. This limits the reliability and generalisability of the authors' findings.

The study findings appeared to compare favourably with other studies in terms of weight loss. However, it was not clear whether these other studies had addressed the same research question. The authors did not specifically address the issue of generalisability of the results.

The authors suggested that the findings of this study should assist surgeons in making decisions about stapled anastomosis, based upon relative difficulty, operative time and the likelihood of complications. However, the authors also added that, as part of several changes to their surgical procedure, they had more recently abandoned the EEA technique because of the wound infection rates and the longer operative time involved. Further analysis, including a prospective randomised controlled trial is recommended.

None stated.