Impact of surgery for chronic pancreatitis on the risk of pancreatic cancer: Untying the Gordian knot

Table 1 Systematic reviews and meta-analyses evaluating the association between chronic pancreatitis and pancreatic cancer

Ref.	Number of studies	Number of patients	Risk estimate	Remarks
Raimondi et al[6]	22 (12 case-control, 10 cohort)	16124	RR, 11.8 (95%CI: for 1-yr lag period; RR, 5.8 (95%CI: 2.1-15.9) for 2-yr lag period; RR, 69.0 (95%CI: 56.4-84.4) for hereditary pancreatitis; RR, 100 (95%CI: 37.0-218.0) for tropical pancreatitis	Details of seven studies included for risk estimate calculation not mentioned; Details of number of patients with pancreatic cancer in included studies not mentioned; High relative risk reported for tropical pancreatitis based on a single study
Tong et al[7]	17 (14 case-control, 3 cohort)	14667 pancreatic cancer, 17587 pancreatitis	Pooled OR, 10.35 (95%CI: 9.13-11.75) for chronic pancreatitis; Pooled OR, 6.41 (95%CI: 4.93-8.34) for unspecified type of pancreatitis -	Inclusion of studies with both acute and chronic pancreatitis limits generalizability; Inclusion of acute, chronic and unspecified pancreatitis to calculate the temporality of the association might underestimate the relative risk
Kirkegård et al[8]	13 (4 case-control, 9 cohort)	674 pancreatic cancer, 25,329 chronic pancreatitis	EE, 16.16 (95%CI:12.59-20.73) for 2-yr lag period; EE, 7.90 (95%CI: 4.26-14.66) for 5-yr lag period; EE, 3.53 (95%CI: 1.69-7.38) for 9-yr lag period	Some of the included studies had a smaller number of patients with pancreatic cancer or chronic pancreatitis which might inflate the estimated risk; In some of the included studies more than one-third of patients were lost to follow-up

CI: Confidence interval; EE: Effect estimate; OR: Odds ratio; RR: Relative risk.