The role of laparoscopy and laparoscopic ultrasound in the preoperative staging of pancreatico-biliary cancers--A meta-analysis.

BACKGROUND

Staging laparoscopy (SL) may prevent non-therapeutic laparotomy in patients with otherwise resectable pancreatico-biliary cancers, but evidence is inconclusive. This meta-analysis aims to ascertain the true benefit of SL.

METHODS

All studies undertaking SL as a diagnostic sieve were included and data homogenised. Standard meta-analytical tools with emphasis on sensitivity testing and meta-regression to detect the cause for heterogeneity between studies were used.

RESULTS

29 studies satisfied the criteria. 3305 patients underwent SL of which 12 were incomplete. Morbidity (n = 15) and mortality (n = 1) was low. True yield of SL for pancreatic/perpancreatic cancers (PPC) was 25% (95% CI 24-27) with a Diagnostic Odds Ratio (DOR) of 104 (95% CI 48-227). Resection rate improved from 61% to 80%. For proximal biliary cancers (PBC), SL increased the curative resection rate from 27% to 50%, with true yield of 47% (95% CI 42-52) and a DOR 61 (95% CI 19-189). Sub-group analysis for detection of liver and peritoneal lesions demonstrated a sensitivity of 88% (95% CI 83-92) and 92% (95% CI 84-96) for PPC; 83% (95% CI 69-92) and 93% (95% CI 81-99) for PBC, respectively. There was no between-study heterogeneity for peritoneal lesions. However for detection of local invasion, sensitivity was low: 58% (95% CI 51-65) for PPC and only 34% (95% CI 22-47) for PBC. Meta-regression did not reveal any cause for the observed heterogeneity between studies.

CONCLUSION

SL offers significant benefit to patients with resectable pancreatico-biliary cancers in avoiding non-therapeutic laparotomy and should be adopted in routine clinical practice in a judicious algorithm.