J Neurol Surg B Skull Base 2016; 77 - A126
DOI: 10.1055/s-0036-1579913

Patient and Hospital Factors Are Associated with Postoperative Outcomes and Hospital Charges in Patients Undergoing Transsphenoidal Surgery for Pituitary Adenomas: an Update from the Nationwide Inpatient Sample

Daniel A. Donoho 1, Andrew Platt 2, Diana Jiang 1, John Carmichael 1, Steven Cen 1, William J. Mack 1, Gabriel Zada 1
  • 1University of Southern California, Los Angeles, California, United States
  • 2University of Chicago, Chicago, Illinois, United States

Background: Race and socioeconomic status are associated with disparities in outcomes across a variety of medical conditions in the United States. However, prior analysis of nationwide data showed no association between patient factors and outcomes in transsphenoidal surgery (TSS) for pituitary adenomas. Similarly, although hospital factors have been shown to be associated with outcomes in TSS, these factors were not associated with hospital charges. We hypothesized that both patient and hospital factors would be associated with outcomes and hospital charges in patients undergoing TSS for pituitary adenomas.

Methods: The 2002–2010 HCUP/NIS database is an all payer database containing 20% of all hospital discharges in the United States. We included admissions for patients undergoing TSS for pituitary adenomas. Patient factors including age, race and payer status, and hospital factors including location, size, teaching status and procedural volume were collected. Outcome variables included endocrine and non-endocrine complications, mortality, hospital charges, and length of stay. Both univariate and multivariate models were constructed to examine associations between patient and hospital factors and outcome.

Results: From 2002–2010, data for 71,892 TSS performed during 71,245 total admissions for pituitary adenomas were analyzed.

After controlling for other patient and hospital factors, patients with Medicaid had an increased risk of mortality (RR = 2.31;p = 0.03), endocrine complications (RR = 1.18;p = 0.02) and increased hospital charges (RR = 1.28;p<0.01) compared with privately insured patients. Medicare patients had increased hospital charges (RR = 1.20; p = 0.01) compared with privately insured patients.

After controlling for other patient and hospital factors, African-American patients had increased risk of endocrine complications (RR = 1.15; p = 0.02) and increased hospital charges (RR = 1.25; p < 0.01) compared with white patients. Hispanic patients had increased hospital charges (RR = 1.34; p < 0.01) compared with white patients.

After controlling for other patient and hospital factors, patients treated at teaching hospitals were less likely to have increased hospital charges (RR = 0.82, p < 0.01). Patients treated at hospitals with procedural volume in the upper tertile had decreased length of stay (RR = 0.73; p < 0.01), lower hospital charges (RR= 0.77; p = 0.01), and fewer total complications (RR= 0.85; p = 0.04), endocrine complications (RR= 0.75; p < 0.01), and electrolyte abnormalities (RR= 0.67;p<0.01).

Conclusion: The HCUP/NIS database allows large-scale analyses of patient and hospital factors and outcomes. We demonstrate associations between both patient and hospital factors and outcomes in patients undergoing TSS for pituitary adenomas. We describe significant racial and socioeconomic disparities in outcomes and charges in these patients. High volume and academic centers are associated with improved outcomes and lower charges in these patients. These patient and hospital factors should be considered in today’s climate of health care and health insurance reform and represent important areas for future study.