ITERATIVE CHANGES IN RISK-STRATIFIED PANCREATECTOMY CLINICAL PATHWAYS AND FACTORS ASSOCIATED WITH ACCELERATED DISCHARGE AFTER PANCREATICODUODENECTOMY
Andrew D. Newton*, Timothy E. Newhook, Morgan L. Bruno, Laura R. Prakash, Yi-Ju Chiang, Natalia Paez Arango, Whitney L. Dewhurst, Elsa M. Arvide, Naruhiko Ikoma, Jessica E. Maxwell, Michael P. Kim, Jeffrey E. Lee, Matthew Katz, Ching-Wei D. Tzeng
Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX
Background: After implementing postoperative pancreatic fistula (POPF) risk-stratified pancreatectomy clinical pathways (RSPCPs) in 2016, the pancreaticoduodenectomy (PD) median length of stay (LOS) decreased from 10 to 6 days. We studied the association of iterative changes in pathways on accelerated discharge and early postoperative outcomes.
Methods: Original RSPCPs were revised in 1/2019, with recommended earlier nasogastric tube (NGT) removal (postoperative day (POD) 1 vs. POD2 for low-risk and POD2 vs. POD4 for high-risk) and new surgical drain amylase cutoffs for POD 1 or 3 removal. All patients had intraoperative NGT. We defined POD1 NGT removal as "early." LOS, readmission, POPF, and delayed gastric emptying (DGE) were compared between "original" (10/2016-1/2019) and "revised" (2/2019-9/2020). Logistic regression identified predictors of accelerated discharge (median LOS in revised RSPCPs or sooner; '‰¤POD5 low-risk; '‰¤POD6 high-risk).
Results: There were 233 (84 high-risk) patients in original and 131 (42 high-risk) patients in revised RSPCPs. After revision, early NGT removal was more common for low-risk (65.2 vs. 30.2%) and high-risk (45.2 vs. 15.5%, P<0.001). Drain removal by POD3 was similar for low-risk (69.7 vs. 75.2, P=0.368) and high-risk (21.4 vs. 16.7%, P=0.625). Median LOS decreased since 2/2019, for both low-risk (5 vs. 6 days, P=0.011) and high-risk (6 vs. 9 days, P=0.005). Between revised and original RSPCPs, there were no differences for low-risk or high-risk patients in readmissions (29.2 vs. 23.5%, P=0.359; 38.1 vs 29.2%, P=0.421), POPF (4.5 vs. 3.4%, P=0.731; 23.8 vs. 33.4%, P=0.309), or DGE (19.1 vs. 10.7%, P=0.082; 21.4 vs. 21.4%, P=1.00). The first full day of regular diet tolerance was earlier (POD4 vs. POD5, P<0.001) with no difference in NGT reinsertions between patients with and without early NGT removal (8.9 vs. 6.5%, P=0.416). In low-risk patients, younger age (OR 1.04, 95% CI 1.01-1.07, P=0.011), early NGT removal (OR 2.38, 95% CI 1.02-3.33, P=0.002), and drain removal by POD3 (OR 2.85, 95% CI 1.39-4.17, P<0.001) were independently associated with accelerated discharge ('‰¤POD5). In high-risk patients, early NGT removal (OR 4.76, 95% CI 1.89-11.1, P<0.001) and drain removal by POD3 (OR 5.26, 95% CI 1.89-14.3, P<0.001) were independently associated with accelerated discharge ('‰¤POD6).
Conclusions: Following iterative changes in RSPCPs in 1/2019, with median LOS targets of 6 and 8 days for low-/high-risk patients, LOS after PD decreased further with no increase in readmissions, POPF, or DGE. NGT removal on POD1 and drain removal by POD3 were associated with accelerated discharge ('‰¤5 and '‰¤6 days for low-/high-risk patients). With one-third of low-risk and half of high-risk patients keeping their NGT beyond POD1, further iterative strategies could include standardized POD1 NGT removal, independent of POPF risk.
Back to 2021 Abstracts