Thirty-Day Hospital Readmission following Transsphenoidal Surgery: Can Cause Decide the Timing of Readmission?

Introduction: Thirty-day hospital readmission can significantly affect healthcare costs and is considered as a yardstick for health care quality.

The purpose of this study was to analyze the frequency, causes and risk factors for 30-day readmission following trans-sphenoidal surgery for pituitary adenoma. Additionally, it was analyzed whether timing of readmission depended on the cause.

Materials and Methods: Patients who underwent endoscopic trans-sphenoidal resection (2013–2017) for pituitary adenomas at our tertiary referral center were included. Readmission was defined as any hospital admission after the patient was discharged following index surgery. Date of discharge was considered as day zero. Outpatient visits to emergency department were excluded. Among 429 patients evaluated, 346 met the selection criteria.

Results: Thirty-seven (11%) patients had at least one 30-day readmission. Total number of readmissions was 41 (mean: 1.1 ± 0.39). Majority of these were unplanned (90%) and seen within day 10 (63%). Mean day of admission was 9.73 ± 7.77 days following discharge. Mean duration of hospital stay during readmission was 4.56 ± 3.87 days. Approximately 24% patients needed surgical intervention.

Preoperative BMI (OR: 1.05; CI: 1.004–1.098; p = 0.03; 33.86 ± 13.19 vs. 31.04 ± 6.66) and thromboembolic history (OR: 47.59; CI: 9.91–228.55; p = 0.0001; 24% vs. 0.7%) were significantly higher in the cohort with 30-day readmissions.

Intraoperative CSF leaks, type of sellar reconstruction, cavernous sinus involvement, extent of resection (partial vs. complete), tumor type (functioning vs. nonfunctioning) and tumor volume were not associated with readmission.

Causes of readmission were classified into three groups – directly related to surgery (Group R—49%), accentuated by surgery (Group A—17%) and unrelated to surgery (Group U—34%). SIADH (15%) and deep vein thrombosis (12%) were the most common causes.

Causes in Group R included SIADH (n = 6), reoperation (n = 3), CSF leak (n = 2), diabetes insipidus (n = 2), epistaxis (n = 2), vision problems/third cranial nerve palsy (n = 2), steroid withdrawal (n = 1), debridement of extensive sellar crusts (n = 1) and septal hematoma (n = 1). Group A causes included deep vein thrombosis (n = 5), pulmonary embolus (n = 1) and inferior venacaval thrombus (n = 1). Causes unrelated to surgery (n = 13) were cardiac, pulmonary, renal issues etc.

Although, association between preoperative thromboembolic event and postoperative thromboembolic event (leading to 30-day readmission) trended toward significance (p = 0.052), most patients (70%) with preoperative thromboembolic event were admitted for non–thromboembolic-related causes.

Timing of readmission was significantly different depending on cause (Group R: 6.95 ± 6.15; Group A: 15.85 ± 8.11; Group U: 10.64 ± 8.19; p = 0.02) while duration of hospital stay was comparable.

Conclusion: Preoperative thromboembolic history is a strong predictor of 30-day readmission. The timing of readmission following trans-sphenoidal surgery for pituitary adenoma depends on cause.