Is Laparoscopic Hartmann′s Reversal a Safe Option? A Systematic Review and Meta-Analysis

Abstract

Background: Hartmann's reversal is a major operation to restore colorectal continuity. Traditionally, an open Hartmann's reversal (OHR) has been performed but there is a trend toward performing laparoscopic Hartmann's reversal (LHR). With the increasing number of publications comparing these two, it is important to ascertain whether the benefits of LHR outweigh the risks. Objective: To compare LHR and OHR with respect to morbidity and mortality rates. Design: A systematic review and meta-analysis. Setting: The study was conducted at the Peter MacCallum Cancer Centre in Melbourne, Australia. Patients and Methods: A detailed systematic search was performed through PubMed, SCOPUS, TRIP, EMBASE, and ClinicalKey from 1990 to October 26, 2016. A review was undertaken in accordance with PRISMA guidelines. Main Outcome Measures: The primary outcome measure was 30-day morbidity. Secondary outcome measures included estimated intraoperative blood loss, conversion from laparoscopic to open approach, length of hospital stay, and 30-day mortality. Sample Size: Eighteen eligible studies were identified, comprising a total of 7824 patients: 1586 in the laparoscopic group and 6238 in the open group. Results: There was no statistical difference in mean operative time between the two groups. Overall morbidity was lower in the LHR group (16.8% vs 23.7%, P < 0.0001). Subgroup-analysis showed a higher risk of sepsis (6.5% vs 3.2%; P < 0.0001), wound infection (22.5% vs 12.6%; P < 0.0001), and ileus (13.4% vs 5.5%; P = 0.001) in the OHR group. Conclusion: LHR was associated with a lower morbidity rate and shorter hospital stay with an equivalent operative time. There is a moderate rate of conversion and appropriate case selection is important. Limitations: An absence of prospective or randomized trials comparing the two approaches for Hartmann's reversal, contributing to selection bias in our study. It was difficult to combine patient characteristics data due to the heterogeneity in the reported parameters. Conflict of Interest: None.