Abstract
INTRODUCTION:Reducing length of stay (LOS) and readmission improve healthcare value and outcomes. Peri-operative protocols have been shown to facilitate earlier discharge without compromising safety in nonfunctioning adenomas. No multicenter studies have investigated their effectiveness in Cushing's Disease (CD).METHODS:A retrospective analysis was conducted from the Registry of Adenomas of the Pituitary and Related Disorders (RAPID) consortium with 18 participating centers. Institutions were surveyed regarding formal protocols for patients with CD. 747 patients were analyzed to assess the impact of protocols on LOS and 90-day readmission. Multivariable regression was performed to evaluate independent factors.RESULTS:13/18 (72%) of institutions reported using a protocol, 8/18 (44%) utilized a formal document and 4/18 (22%) implemented a hospital policy approved by their respective medical boards. LOS was significantly reduced in centers that had an established protocol compared to no protocol (3.76 vs 3.99 days, p = 0.04), and more so with a formal document (3.59 vs. 4.17 days, p< 0.001) or hospital policy (3.39 vs. 4.01 days, p < 0.001). Other factors associated with decreased LOS were presence of a separate CD pathway, intraoperative checklist specific to pituitary surgery, non-narcotic pain regimen, foley removal order, outpatient APP follow-up, and a target discharge date of = 2 days. Multivariable regression identified hospital policy (p = 0.03), intraoperative checklist (p = <0.001), non-narcotic pain regimen (p = 0.01), and target discharge = 2 days (p = 0.048) as independent effects on LOS. A formal document (OR 0.44, p = 0.007) and nasal packing (OR 0.48, p = 0.006) significantly reduced readmission. No protocol factors reduced readmission for hyponatremia, but a formal document reduced readmission for adrenal insufficiency (OR 0.13, p = 0.03).CONCLUSIONS:This is the first multi-center study to show that perioperative protocols are associated with decreased LOS and readmission in CD, particularly if protocols are formalized.