Safety evaluation of early drain removal following pancreaticoduodenectomy: A single-center retrospective cohort study

Predictive value of postoperative serum lipase level for postoperative pancreatic fistula after pancreaticoduodenectomy

BACKGROUND

The prediction of postoperative pancreatic fistula (POPF) is important. This study aimed to investigate the role of postoperative serum lipase level in predicting POPF.

METHODS

Data from 234 consecutive patients who underwent pancreaticoduodenectomy (PD) were collected. The predictive values of serum amylase and serum lipase during postoperative days (PODs) 1 to 3 for POPF were compared. Subgroup analyses were performed to determine the prognostic value of different levels and durations of elevated serum lipase.

RESULTS

Fifty-six patients developed POPF. The POPF group exhibited increased levels of serum amylase and lipase from PODs 1 to 3 (all P < 0.001). Compared with serum amylase, serum lipase has greater predictive value for POPF. Specifically, serum lipase had the highest area under the receiver operating characteristic curve (AUC) at POD 1 (0.791). Body mass index > 24 kg/m2 [odds ratio (OR) = 2.431, 95% confidence interval (CI): 1.094-5.404, P = 0.029], soft pancreatic texture (OR = 3.189, 95% CI: 1.263-8.056, P = 0.014), serum lipase > 60 U/L at POD 1 (OR = 5.135, 95% CI: 1.257-20.982, P = 0.023), and C-reactive protein > 167 mg/dL at POD 3 (OR = 3.607, 95% CI: 1.431-9.090, P = 0.007) were identified as independent risk factors for POPF. Patients with serum lipase ≤ 60 U/L at POD 1 (n = 104) exhibited lower rates of POPF (3.8% vs. 40.0%, P < 0.001) and severe complications (Clavien-Dindo ≥ IIIa) (4.8% vs. 25.4%, P < 0.001) than those with serum lipase > 60 U/L at POD 1. Moreover, no additional elevation or duration of serum lipase offered any further prognostic value.

CONCLUSIONS

Postoperative serum lipase outperformed serum amylase in the prediction of POPF, and patients with normal serum lipase level at POD 1 had favorable outcomes. A sustained increase in the serum lipase level offers no additional prognostic value.

Early Drain Removal Versus Routine Drain Removal After Pancreaticoduodenectomy and/or Distal Pancreatectomy: A Meta-Analysis and Systematic Review

BACKGROUND

Early drain removal (EDR) has been widely accepted, but not been routinely used in patients after pancreaticoduodenectomy (PD) and distal pancreatectomy (DP). This study aimed to evaluate the safety and benefits of EDR versus routine drain removal (RDR) after PD or DP.

METHODS

A systematic search was conducted on medical search engines from January 1, 2008 to November 1, 2023, for articles that compared EDR versus RDR after PD or DP. The primary outcome was clinically relevant postoperative pancreatic fistula (CR-POPF). Further analysis of studies including patients with low-drain fluid amylase (low-DFA) on postoperative day 1 and defining EDR timing as within 3 days was also performed.

RESULTS

Four randomized controlled trials (RCTs) and eleven non-RCTs with a total of 9465 patients were included in this analysis. For the primary outcome, the EDR group had a significantly lower rate of CR-POPF (OR 0.23; p < 0.001). For the secondary outcomes, a lower incidence was observed in delayed gastric emptying (OR 0.63, p = 0.02), Clavien-Dindo III-V complications (OR 0.48, p < 0.001), postoperative hemorrhage (OR 0.55, p = 0.02), reoperation (OR 0.57, p < 0.001), readmission (OR 0.70, p = 0.003) and length of stay (MD -2.04, p < 0.001) in EDR. Consistent outcomes were observed in the subgroup analysis of low-DFA patients and definite EDR timing, except for postoperative hemorrhage in EDR.

CONCLUSION

EDR after PD or DP is beneficial and safe, reducing the incidence of CR-POPF and other postoperative complications. Further prospective studies and RCTs are required to validate this finding.

Comparing the distal pancreatectomy fistula risk score (D-FRS) and DISPAIR-FRS for predicting pancreatic fistula after distal pancreatectomy

BACKGROUNDS

Distal pancreatectomy fistula risk score (D-FRS) and DISPAIR-FRS has not been widely validated for predicting postoperative pancreatic fistula (POPF) after distal pancreatectomy (DP).

METHODS

We retrospectively analysed 104 patients undergoing DP. The predictive value of the D-FRS and DISPAIR-FRS were compared. Risk factors associated with POPF were investigated by multivariate analysis.

RESULTS

Of the 104 patients, 23 (22.1%) were categorized into the POPF group (all grade B). The areas under the ROC (AUCs) of the D-FRS (preoperative), D-FRS (intraoperative), and DISPAIR-FRS were 0.737, 0.809, and 0.688, respectively. Stratified by the D-FRS (preoperative), the POPF rates in low-risk, intermediate-risk, and high-risk groups were 5%, 22.6%, and 36.4%, respectively. By the D-FRS (intraoperative), the POPF rates in low-risk, intermediate-risk, and high-risk groups were 8.8%, 47.1%, and 47.4%, respectively. By the DISPAIR-FRS, the POPF rates in low-risk, intermediate-risk, and extreme-high-risk groups were 14.8%, 23.8% and 62.5%, respectively. Body mass index and main pancreatic duct diameter were independent risk factors of POPF both in preoperative (P = 0.014 and P = 0.033, respectively) and intraoperative (P = 0.015 and P = 0.039) multivariate analyses.

CONCLUSION

Both the D-FRS (preoperative), D-FRS (intraoperative), and DISPAIR-FRS has good performance in POPF prediction after DP. The risk stratification was not satisfactory in current Asian cohort.