Elsevier

Surgery

Volume 134, Issue 4, October 2003, Pages 647-653
Surgery

Central surgical association
Perioperative outcomes of laparoscopic versus open splenectomy: A meta-analysis with an emphasis on complications

https://doi.org/10.1016/S0039-6060(03)00312-XGet rights and content

Abstract

Background

The purpose of this study was to analyze the published perioperative results of laparoscopic splenectomy (LS) compared to open splenectomy (OS), and to determine the impact of LS on the incidence and type of splenectomy-related complications.

Methods

Perioperative results and complications were tabulated from all English-language reports of LS from 1991 through 2002, and complications were analyzed further by type. Data were taken from 26 series that compared OS to LS within an institution (paired analysis) and from an additional 25 series of only LS (unpaired analysis), and a meta-analysis was performed.

Results

A total of 2940 patients from 51 published series were included (LS, 2119 patients; OS, 821 patients). Age, gender, and American Society of Anesthesiologists class were similar. In the analysis of paired OS and LS studies, the mean operative time for LS was significantly longer (LS, 180 minutes; OS, 114 minutes; P<.0001,) but the postoperative hospital stay was shorter (LS, 3.6 days; OS, 7.2 days; P<.001). Accessory spleens were identified in 11% of cases in both groups. The total complication rate for LS was 15.5%, compared with 26.6% for OS (P<.0001). LS was associated with significantly fewer pulmonary, wound, and infectious complications (P<.001 for all) but with more hemorrhagic complications, when conversions for bleeding were included. Mortality rates for LS and OS were similar (OS, 1.1%; LS, 0.6%; P = not significant). Comparable results were obtained when the unpaired LS series were added to the analysis.

Conclusions

Although operative times are longer for LS than OS, LS is associated with a significant reduction in splenectomy-related morbidity, primarily as a function of fewer pulmonary, wound, and infectious complications.

Section snippets

Study selection

A Medline search was done for all English-language reports that used the terms splenectomy or LS from January of 1991 through December of 2002. Surgical series that described outcomes were reviewed in detail, and the references were examined for additional articles that were not identified by the Medline search. Studies were excluded if any of the following parameters were present: fewer than 20 cases, splenectomy for trauma or incidentally, data not reported on an intention-to-treat basis, or

Patient population

A total of 51 reports met the criteria for inclusion in this analysis. Twenty-six studies compared OS to LS (paired), and 25 studies reported only on LS (unpaired). Of the 2940 patients in those studies, 2119 patients (72.1%) underwent LS, and 821 patients (27.9%) underwent OS. Of the patients who underwent LS, 41.3% were in studies that reported paired groups, with the remaining 58.7% of LS patients in series that described only the laparoscopic approach. Most studies were retrospective, with

Discussion

For LS to be regarded as the preferred approach to splenectomy, it should be associated not only with less postoperative pain and a faster recovery but also should have a lower complication rate. Data from some institutional series have suggested that LS is associated with fewer complications,16., 17., 18. but other series have failed to show this.19., 20. Three separate descriptive reviews attempted to address the issue of complications after LS,21., 22., 23. and 2 reviews concluded that LS is

Acknowledgements

We thank Paul Thompson, PhD, from the Division of Biostatistics at Washington University, for assistance with the statistical analysis.

References (25)

  • R.L. Friedman et al.

    Laparoscopic splenectomy for ITP: the gold standard

    Surg Endosc

    (1996)
  • J.L. Flowers et al.

    Laparoscopic splenectomy in patients with hematologic diseases

    Ann Surg

    (1996)
  • Cited by (254)

    • Surgery of the spleen

      2019, Surgery (United Kingdom)
      Citation Excerpt :

      The overall incidence of PSVT is reported to be approximately 3%, although alternative series using Doppler ultrasound 1 week post-operatively identified PSVT in around 20% of patients after laparoscopic splenectomy. Nevertheless, the overall incidence appears to be similar between open and laparoscopic procedures in most studies, typically occurring around 2 weeks following surgery and is best identified using CT where other complications can be excluded.1–3,19 While certain haematological disorders associated with a pro-thrombotic tendency are thought to contribute to a greater incidence of PSVT, correlation with increasing splenic size has also been reported.1,2

    View all citing articles on Scopus

    Presented at the 60th Annual Meeting of the Central Surgical Association, Toronto, Ontario, Canada, March 20-22, 2003.

    View full text