Fewer than 10% of pancreaticoduodenectomy patients required postoperative nasogastric drainage, the current postoperative standard of care for pancreaticoduodenectomy, the results of a longitudinal observational cohort study have shown.
Dr. John W. Kunstman and his colleagues in the surgery department of Yale University, New Haven, Conn., observed two consecutive cohorts of 125 pancreaticoduodenectomy (PD) patients. The first cohort had nasogastric tubes (NGTs) maintained postoperatively until clinically indicated, while the second cohort had NGTs maintained postoperatively "only in rare circumstances, such as inability to extubate the patient postoperatively," the authors said. All patients were treated by the same surgeon between July 2003 and February 2012, most commonly for pancreatic neoplasm.
There were no statistically significant differences between the two groups in patient demographics, including indications for surgery. The routine NGT group was 51% male with an average age of 63 years, while the selective NGT group was 46% male with an average age of 67 years. Bias was reduced in that both cohorts were analyzed in an intent-to-treat manner.
Pylorus-preserving pancreaticoduodenectomy (PPPD) was performed in the absence of oncologic or other disease-specific considerations; otherwise, a classic (Kausch-Whipple) PD was performed. The most common comorbidity overall was hypertension, although the routine NGT group had a higher incidence of coronary artery disease than the selective NGT group (20.8% vs. 8.8%) and a higher mean creatinine level (0.97 vs. 0.88 mg/dL) (J. Am. Coll. Surgeons 2013;217:481-8).
Primary outcomes included postoperative NGT insertion and reinsertion, delayed gastric emptying (DGE) incidence, time to dietary tolerance, and length of stay.
In the selective NGT cohort, only 9 patients required continued NGT, 5 of them for postoperative endotracheal intubation and 4 for surgical considerations.
Neither the incidence of NGT insertion and reinsertion, nor the duration of NGT replacement, differed significantly between groups. Overall complication rates were also similar, although patients in the selective NGT group were less likely to have DGE. Compared with the routine NGT cohort, the selective cohort had a shorter mean length of stay (10 vs. 7 days) and mean time to dietary tolerance.
Additionally, multivariate analysis of all outcome variables indicated that DGE, the most common adverse event associated with PD, was independently correlated with the routine use of NGT.
"This is of particular interest given the role that gastric decompression plays in treatment of DGE. However, this finding must be interpreted with care, as incidence of DGE varies from 10% to 50% of PD cases, depending on the reporting group," the authors wrote.
In terms of the study’s limitations, the higher number of PPPD and retrocolic gastroenteric anastomoses performed in the routine cohort may have been a factor in the differing DGE incidence rates, since pylorus preservation and anastomotic positioning are considered historically correlated to DGE, the authors said.
Among individually examined adverse events, routine NGT patients were more likely than selective patients to be reintubated (11.2% vs. 3.2%), require a postoperative blood transfusion (16.0% vs. 6.4%), and be diagnosed with DGE (18.4% vs. 8.0%).
There were a total of four deaths in the 30-day postoperative period, all occurring in patients over age 80. There were eight octogenarians in each cohort; three of the deaths occurred in the routine group.
Improvements in perioperative care and the surgeon’s increased skill over time may also have affected the results. However, the authors concluded, "based on these data, persistent concerns, such as fear of increased anastomotic leak or DGE, which have mandated customary nasogastric decompression after PD, can be safely reevaluated."
Dr. Kunstman and his coauthors reported no relevant disclosures.