[Study of perioperative safety of laparoscopic pancreaticoduodenectomy in elderly patients]

To investigate the safety of laparoscopic pancreaticoduodenectomy (LPD) in elderly patients and the related risk factors admitted to the intensive care unit (ICU) after LPD.

The perioperative data of patients who underwent LPD in Tianjin Medical University General Hospital from February 2017 to June 2023 were retrospectively collected, including basic data, preoperative laboratory indicators, intraoperative and postoperative indicators, pathological results (tumor size, lymph node dissection and pathological type), postoperative complications, ICU postoperative management and prognosis. The patients were divided into the elderly group (≥ 65 years) and the non-elderly group (< 65 years) according to age. Perioperative data between two groups were compared. Kaplan-Meier survival curve was drawn to analyze the survival rate of the elderly group and the non-elderly group, and the pancreatic head carcinoma group and other type of tumors group after LPD. Logistic regression was used to analyze the risk factors of ICU stay (length of ICU stay > 1 day) after LPD in elderly patients. The receiver operator characteristic curve (ROC curve) was drawn to analyze the predictive value of this risk factor for ICU stay after LPD in elderly patients.

A total of 160 patients were enrolled, including 57 cases in the elderly group (17 cases of vascular reconstruction) and 103 cases in the non-elderly group (40 cases of vascular reconstruction). All patients underwent R0 resection and were transferred to the comprehensive ICU for treatment. The follow-up time of patients with malignant tumors was 43 (6, 72) months. The elderly group had significantly longer surgery time, postoperative hospital stay and oral feeding time than the non-elderly group, and the incidence of delayed gastric emptying (DGE) was significantly higher than that in the non-elderly group. There were no significant differences in intraoperative blood transfusion rate, intraoperative blood loss, pathological results, short-term and severe postoperative complications, reoperation rate and 90-day mortality between the two groups. In patients with vascular resection reconstruction, the intraoperative blood loss in the elderly group was significantly higher than that in the non-elderly group, and the operation time and postoperative hospital stay were significantly longer. During ICU, the acute physiology and chronic health evaluation II [APACHE II: 12 (9, 14) vs. 8 (7, 10)], sequential organ failure assessment [SOFA: 6 (4, 8) vs. 3 (2, 5)] within 24 hours after admission to ICU were significantly increased in the elderly group (both P < 0.05), the time of mechanical ventilation [hours: 12 (10, 15) vs. 9 (5, 13)] and the length of ICU stay [days: 2 (1, 2) vs. 1 (1, 1)] were significantly increased in the elderly group (both P < 0.05), and the proportion of multi-disciplinary team (MDT) was also significantly increased in the elderly group (33.3% vs. 17.4%, P < 0.05), there were no significant differences in the levels of hemoglobin (Hb), albumin, and blood lactic acid between the two groups. Logistic regression analysis showed that the APACHE II score was an independent risk factor for ICU stay after LPD in elderly patients (β = 1.737, P = 0.028). ROC curve showed that the prediction performance was the best when the APACHE II score was 13, with the sensitivity of 72.41% and the specificity of 96.43%, and the area under the ROC curve (AUC) of 0.884. The Kaplan-Meier survival curve showed that there were no significant difference in median survival time (months: 24.1 vs. 24.7) and 5-year survival rate (19.01% vs. 19.02%) between the elderly group (52 cases) and the non-elderly group (92 cases) among the 144 patients with malignant tumors (both P > 0.05). The median survival time in the pancreatic head carcinoma group was significantly shorter than that in the other tumors group (63 cases; months: 20.2 vs. 40.1, P < 0.05), 5-year survival rate was significantly lower than that in the other tumors group (21.98% vs. 30.91%, P < 0.05).

LPD is a safe and feasible treatment for elderly patients. APACHE II score has a certain predictive value for ICU stay after LPD in elderly patients.