The authors of this review concluded that mini-open donor nephrectomy appeared to provide advantages for the kidney donor in comparison to standard open donor nephrectomy. They also stated that the technique had a shorter operative time when compared with laparoscopic living donor nephrectomy. This review had several methodological limitations and the authors' conclusions may not be wholly reliable.

To compare mini-open donor nephrectomy (MODN) with standard open (SODN) and laparoscopic (LDN) living donor nephrectomy

The authors searched MEDLINE and Cochrane (database unspecified) for studies between 1980 and September 2007. Search terms were detailed in the report. References of the articles acquired were scanned for further studies.

To be eligible for the review, studies needed to compare MODN to SODN or LDN in a sample of living kidney donors. All techniques were defined in the report. Relevant outcomes were operative and postoperative parameters and adverse events for the donor and recipient outcomes, including graft function. Two reviewers were involved in selecting studies for the review and discrepancies were resolved by consensus.

The quality of the non-randomised studies was assessed using a modified Newcastle-Ottawa Scale (NOS). This evaluated method of patient selection, comparability of the study groups and number of outcomes reported. Each study could receive a star rating of zero to nine. A high-quality study was defined as one with five or more stars. Randomised controlled trials (RCTs) were assumed to be high-quality studies unless there was any doubt (unspecified). The authors did not state how many reviewers performed the validity assessment.

The authors extracted donor operative parameters (operative time, warm ischaemic time and estimated blood loss) and donor postoperative parameters (total in-patient analgesia requirement, hospital stay and time to return to work). They divided donor adverse events into overall number of complications, intraoperative complications and postoperative complications (further divided into in-patient and long-term post-hospital discharge). A complication was defined as any event that caused patient morbidity. Data were extracted on one-year recipient and graft survival rates, delayed graft function and ureteric leaks, strictures and total complications. For studies that presented continuous data as means and range values, standard deviations (SD) were calculated using published statistical algorithms. In papers with case overlap, the ones with the best methodology and larger number of patients were used.

Two reviewers extracted data independently and discrepancies were resolved by consensus.

Analysis of dichotomous variables was carried out using the odds ratio (OR) and continuous variables were analysed using the weighted mean difference (WMD), both with 95% confidence intervals (CI). A random-effects model of meta-analysis was used. Yates' correction was used for studies that contained a zero in one cell for the number of events of interest in one of the two groups. Publication bias was assessed using funnel plots. Heterogeneity was assessed using a Χ² test. Sensitivity analysis was undertaken on high-quality studies and in those studies with more than 50 patients in the MODN treatment group.

Nine studies were included in the review (n=1,038): one RCT (n=100); four prospective comparisons (n=489); and four studies with retrospective control groups (n=449). Non-randomised studies received between two and seven stars. There was a suggestion of publication bias for hospital stay, total donor complications and postoperative in-hospital complications for MODN compared to SODN and for operative time with MODN compared to LDN.

MODN versus SODN: Operative time, warm ischaemia time and blood loss were not significantly different between treatment groups. Hospital stay was shorter by 1.67 (95% CI -2.35 to -0.99, p=0.001) days for patients treated with MODN (based on seven studies with statistically significant heterogeneity, p = 0.001)). Time to return to work was shorter by five weeks (95% CI -9.38 to -0.62, p=0.03), but this was based on one study only. Analgesia requirement (also based on one study) was 96mEq less in the MODN group (95% CI -149.81 to -42.19mEq, p<0.001). There were fewer overall complications for the MODN group (OR 0.47, 95% CI 0.24 to 0.93, p =0.03; five trials). Data on recipient parameters was very limited, but suggested no differences between treatments.

MODN versus LDN: Operative time was statistically significantly shorter by 55 minutes for MODN (95% CI - 93.46 to -16.36 min, p = 0.005; three studies). Warm ischaemia times were also significantly shorter by 147 seconds (95% CI -205.10 to -88.23 seconds, p<0.001; two studies). Other intra-operative parameters were not statistically significant. Total quantity of analgesia was significantly less by 9.6mEq of morphine (95% CI 0.36 to 18.87mEq, p=0.04; two studies) in the LDN group. There were no other statistically significant differences in any of the postoperative parameters and donor complication rates. No statistically significant differences were noted for recipient parameters.

Details of sensitivity analyses and their effect on donor and recipient parameters were provided in the report.

MODN appeared to provide advantages for the donor in comparison to SODN and had a shorter operative time when compared with LDN.

The review had detailed inclusion criteria for eligible interventions. Study design, patients and eligible outcomes were more broadly defined. Searching used a range of methods, although it was unclear whether unpublished research was eligible for inclusion in the review. Attempts were made to reduce bias and error by the involvement of two reviewers in the study selection and data extraction processes. The quality of non-randomised studies was assessed, but it was unclear how many reviewers were involved in the assessment. The randomised controlled trial was assumed to be of high quality, but such an assumption cannot normally be made. There were problems with the meta-analyses: different types of study design are not normally combined in meta-analysis due to their varying levels of bias; and there was evidence of statistical (and possibly clinical) heterogeneity. It was also possible that chance findings arose from the multiple testing for significance between treatments. Given the above limitations, the authors' conclusions may not be wholly reliable.

Practice: The authors did not state any implications for practice.

Research: The authors stated that there was a need for more detailed studies of MODN, especially in the form of RCTs and studies that examined recipient outcomes. Further studies with larger series of patients who underwent MODN were needed to confirm the results of the review.

None.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.