Skip to main content
  • Predictors of Surgical Site Infection After Non-Emergent Craniotomy: A Nationwide Readmission Database Analysis

    Final Number:

    Ian Andre Buchanan MD; Daniel Donoho MD; Michelle Lin BA; Arati B Patel BS; Krista Lamorie-Foote BA; Timothy Wen MPH; Li Ding MD MPH; Steven L. Giannotta MD; William J. Mack MD; Frank Attenello MD

    Study Design:
    Clinical Trial

    Subject Category:

    Meeting: Congress of Neurological Surgeons 2018 Annual Meeting

    Introduction: Introduction: Surgical site infections after craniotomy harm patients and increase healthcare expenditures. Reported infection rates are heterogeneous and few predictors of infection have been identified. We sought to identify risk factors associated with readmission for surgical site infections after non-emergent craniotomy.

    Methods: Methods: The 2010-2014 cohorts of the Nationwide Readmission Database were analyzed. Index admissions were defined as adults who underwent non-urgent craniotomy with at least 30 days of follow-up. Surgical site infection was determined by ICD-9CM diagnosis code. The rate of readmission was determined using de-identified linkage numbers permitting the tracking of patient readmissions at any hospital within the state of their index admission. Patient, hospital-level, and admission factors associated with surgical site infections were explored in multivariable analysis.

    Results: Results: Of 93,920 index admissions, 2.2% were readmitted for surgical site infections within 30 days and 0.9% required an operation. The median time to readmission was 13 days. The total cost of these readmissions was $206,519,262. 3.3% of patients died during their readmission. Patients with diabetes mellitus had increased odds of readmission (OR 1.29, 1.14-1.46). Patients undergoing craniotomy for benign (OR 1.93, 1.66–2.24) and malignant (OR 1.46, 1.26-1.68) tumors were more likely to be readmitted than patients operated for cerebrovascular disease or epilepsy. Patients who initially received a ventriculostomy had greater odds of readmission (OR 1.33, 1.07-1.67). Patients with Medicaid had increased odds of readmission (OR 1.29, 1.13–1.47) compared to those privately insured.

    Conclusions: Conclusions: Although readmissions for infection after craniotomy are rare, readmissions generate over $41 million in annual costs and carry a substantial risk of in-hospital death. Patient factors, such as tumor surgery, the use of a ventricular drain, and diabetes raise the risk of infection. Patients with Medicaid and concomitant lower household income have increased infection rates. Identifying these factors may assist quality improvement efforts in neurosurgery.

    Patient Care: By determining factors associated with postoperative infection, we aim to guide treatment plans to better manage patients to prevent infection and decrease patient morbidity and mortality.

    Learning Objectives: Understand that adult patients undergoing non-emergent craniotomy are more likely to develop a surgical site infection in the presence of an external ventricular drain. Understand that patients who receive craniotomies for tumors are more likely to be readmitted for infection compared to those undergoing craniotomies for other purposes. Identify insurance type as a predictor of 30-day readmission for surgical site infection in adult patients undergoing non-emergent craniotomy.


We use cookies to improve the performance of our site, to analyze the traffic to our site, and to personalize your experience of the site. You can control cookies through your browser settings. Please find more information on the cookies used on our site. Privacy Policy