The volume effect in liver surgery: a systematic review and meta-analysis

Arthur J. Richardson, Tony C. Y. Pang, Emma Johnston, Michael J. Hollands, Vincent W. T. Lam, Henry C. C. Pleass

Research output: Contribution to journalArticlepeer-review

32 Citations (Scopus)


Background: There is an inverse relationship between hospital and surgeon volume and mortality in many types of complex surgery. The aim of this paper is to investigate the volume effect on outcomes of liver surgery. Methods: A systematic review and meta-analysis was performed. A literature search was conducted using Medline and EMBASE from 1995 to 2012. A random effects model was used. Results: Seventeen studies were selected for detailed analysis. Definition of a high-volume institution varied from 2 to more than 33 procedures per year. The pooled odds ratio of mortality rate in low- vs high-volume centres was 2.0 [95 % confidence interval (CI), 1.6-2.4; P < 0.001]. Some studies divided centres into more than two groups and compared the highest and lowest volume groups. The pooled odds ratio of mortality rate for this comparison type was 3.2 (95 % CI, 1.7-5.8; P < 0.001). Funnel plots suggest possible publication bias. There was inadequate data to compare morbidity. Only two of seven studies demonstrated a shorter length of stay in the high-volume centres. There was no convincing volume effect on long-term survival. Conclusions: This study suggests a strong relationship between volume and perioperative mortality. No difference in morbidity, length of stay or survival was demonstrated.

Original languageEnglish
Pages (from-to)1984-1996
Number of pages13
JournalJournal of Gastrointestinal Surgery
Issue number11
Publication statusPublished - 1 Nov 2013
Externally publishedYes


  • Cancer surgery
  • Hepatectomy
  • High volume
  • Liver surgery
  • Morbidity
  • Mortality
  • Regionalisation


Dive into the research topics of 'The volume effect in liver surgery: a systematic review and meta-analysis'. Together they form a unique fingerprint.

Cite this