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.

Off-pump coronary artery bypass grafting (OP-CAB) was compared with conventional coronary artery bypass grafting (CABG).

Treatment.

Cost-effectiveness analysis.

The study population comprised patients eligible for isolated CABG surgery, which was performed by one surgeon (the lead author of this paper).

The setting was secondary care. The economic study was conducted in Cleveland, Ohio, USA.

The effectiveness and resource use data for OP-CAB related to operations conducted between August 1998 and October 1999. The effectiveness and resource use data for conventional CABG surgery related to operations conducted between August 1998 and February 1999. The price year was not reported.

The effectiveness data were derived from a single study.

The costing was undertaken retrospectively on the same patient sample as the effectiveness study.

Calculations to assess the power of the study sample to detect statistically significant differences in the primary outcome measure were not reported. The sample was selected retrospectively from consecutive patients who had undergone surgery with one surgeon. A total of 100 patients had undergone OP-CAB procedures and 100 patients had undergone conventional CABG surgery. Thus, the sample seems to have been appropriate to the study question.

Initially (until May 1999), the patients for OP-CAB were restricted to those with isolated single- or double-vessel coronary artery disease, and those with significant co-morbidities (renal failure or calcified aortas) or significant peripheral cerebrovascular disease. After May 1999, all patients were generally considered eligible for OP-CAB procedures. The exceptions were patients with very unstable preoperative haemodynamics and those who underwent re-operative procedures with the presence of patent but atheromatous grafts. The number of patients excluded from the study was not reported.

This was a case-control study conducted at a single centre. The patients were followed for 30 days or until the end of the hospital stay. No patients were lost to follow-up. The authors did not report any procedures to mask the investigators to the treatment method for the analysis of the outcome.

The analysis was conducted on the basis of treatment completers only. The primary health outcomes used in the analysis were:

mortality within 30 days of the operation or during the same hospital stay;

readmission within 30 days;

perioperative myocardial infarction;

postoperative stroke, defined as any new persistent postoperative neurologic deficit that was also confirmed with a positive head computed tomography scan;

mediastinitis, defined as sternal wound infection that required a second surgical procedure for closure;

bleeding, defined as the need to perform re-exploratory procedures in the patient;

postoperative dialysis; and

postoperative atrial fibrillation.

The patient groups were not shown to be comparable at baseline. There was a higher prevalence of females in the OP-CAB group (45%) compared with the CABG group (35%), (p<0.001). In addition, there were three demographic differences between the OP-CAB and CABG group. First, the incidence of 3-vessel coronary artery disease was more prevalent in the CABG group. Second, all patients with dialysis-dependant renal failure were relegated to the OP-CAB group. Third, the CABG group had a higher proportion of patients who had undergone reoperation, 10% compared with 2% in the OP-CAB group, (p<0.03).

Mortality within 30 days of the operation or during the same hospital stay was 2% in the CABG group and 2% in the OP-CAB group, (non significant).

The percentage of patients being readmitted within 30 days was 14% in the CABG group and 5% in the OP-CAB group, (p=0.024).

Perioperative myocardial infarction was 4% in the CABG group and 2% in the OP-CAB group, (non significant).

The percentage of patients experiencing postoperative stroke was 1% in the CABG group and 0% in the OP-CAB group, (p=0.05).

No patients in either the CABG or OP-CAB group experienced mediastinitis.

The percentage of patients bleeding was 7% in the CABG group and 1% in the OP-CAB group, (p<0.009).

The percentage of patients experiencing postoperative atrial fibrillation was 24% in the CABG group and 24% in the OP-CAB group, (non significant).

The percentage of patients requiring postoperative dialysis was 4% in the CABG group and 0% in the OP-CAB group, (p<0.004).

Compared with CABG procedures, OP-CAB procedures were equally efficacious and safe for most patients who required complete revascularisation. The OP-CAB procedure may lead to accelerated recovery and decreased morbidity.

No summary measure of benefit was used in this study. The study was therefore categorised as a cost-consequences analysis.

With the exception of the length of hospital stay, the resource quantities and the costs were not reported separately. The costs to the hospital were reported in the analysis. The variable direct costs included the length of hospital stay, ancillary costs, pharmacy costs and nursing costs. The authors did not report explicitly the source of the direct cost data. They did refer briefly to the use of an integrated hospital cost management and decision system software to estimate the cost. However, it is unclear whether this system produces cost or charge data. The timeframe was the length of hospital stay. The authors reported that the costs were estimated using actual resource use data, but did not detail how the costs were generated. Discounting was not performed as it was irrelevant (the timeframe of the study was short). The authors reported average costs per case. The date to which the prices refer was not reported.

The cost data were treated as point estimates and analysed as deterministic data. The statistical analysis of the costs used t-tests.

The indirect costs were not included in the analysis.

US dollars ($). No currency conversions were reported.

A sensitivity analysis was not performed.

No measure of benefit was used in this cost-consequences analysis. See the 'Effectiveness Results' section.

The total intervention and comparator costs were not reported.

The variable direct cost per case was US$13,045 for conventional CABG and US$9,185 for OP-CAB, (p<0.001).

The benefits and the costs were not combined in this cost-consequences analysis.

Off-pump coronary artery bypass grafting (OP-CAB) was equally efficacious and safe as conventional coronary artery bypass grafting (CABG). However, OP-CAB led to a shorter length of hospital stay with a corresponding 29% reduction in the costs.

The comparator used was justified on the grounds that it was the standard procedure used in the authors' hospital. You should decide if the intervention (OP-CAB) and the comparator (CABG) represent current practice in your own setting.

The analysis used a retrospective case-control study, which means that the study was prone to patient selection bias. This was apparent in the lack of comparability between the two patient groups. For instance, there was a higher prevalence of females in the OP-CAB group, and the incidence of 3-vessel coronary artery disease was more prevalent in the CABG group. In addition, all patients with dialysis dependant renal failure were relegated to the OP-CAB group, while the CABG group had a higher proportion of patients who had undergone reoperation. The presence of selection bias limits the generalisability of the study's findings. A more appropriate study design would have involved randomisation of patients to either the intervention or comparator group. This would have allowed potential biases and confounding factors to be accounted for in the analysis.

The study sample was representative of the study population. The statistical analyses were conducted on the basis of treatment completers only. Power calculations were not reported, but there was still a statistical difference between the intervention and comparator groups for some of the clinical outcomes. The lack of a statistically significant difference for the remaining clinical outcomes may have been due to the sample being of an insufficient size to detect a difference, rather than the presence of clinical equivalence in these outcomes.

The authors did not derive a summary measure of health benefit. The study was therefore categorised as a cost-consequences analysis.

The perspective from which the study was conducted was not reported. In addition, there were no details of the methods used to estimate the resource use and the costs. Therefore, it was not possible to determine if all of the relevant variable costs were included in the analysis. The authors reported that they omitted the costs associated with the perfusionists hired by the hospital because these were a fixed rather than a variable cost. Therefore, the costs of the intervention and comparator group may be incomplete. With the exception of length of hospital stay, the costs and the quantities were not reported separately. The resource use was taken from a single study. No statistical analysis of the quantities was performed, except for length of hospital stay. The authors did not explicitly report the source of the unit costs used in the analysis. However, they did refer to an integrated hospital management system, which may mean that charges rather than unit prices were used. If charges were used in the analysis, this limits the generalisability of the study's findings to other settings. The price year was not reported. Discounting was not performed due to the short timeframe of the study. The authors did not conduct any sensitivity analyses.

The authors made appropriate comparisons of their findings with those from other studies. They concluded that the results could not be generalised to other settings because one surgeon had performed all operations. There are two further reasons why the results cannot be generalised to other settings. First, the lack of sensitivity analyses to explore the robustness of the study's findings. Second, the fact that the analysis may have used charges rather than unit prices to estimate the total cost. The authors did not present their results selectively. The study enrolled consecutive patients undergoing conventional CABG or OP-CAB and this was reflected in the authors' conclusions. The authors reported a number of limitations to their study. In particular, it was a retrospective non-randomised study and the number of end points studied was relatively small.

The authors suggest that OP-CAB represents the ideal revascularisation strategy for patients at high risk for undergoing cardiopulmonary bypass. The authors report that, in their hospital, they are pursuing a strategy of OP-CAB in all patients who must undergo isolated CABG procedures. Conditions that were thought to be contraindications were, in the opinion of the authors, only relative contraindications and techniques to deal with these situations could be mastered with time and patience. Future studies are needed to address whether specific sub-groups of patients derive the maximal benefit from OP-CAB procedures.

Supported in part by honoraria from Medtronics Inc for peer-to-peer training with the Octopus device.