The hip fracture assessment tool: A scoring system to assess high risk geriatric hip fracture patients for post-operative critical care monitoring

Abstract

IntroductionIntensive care unit risk stratification models have been utilized in elective joint arthroplasty; however, hip fracture patients are fundamentally different in their clinical course. Having a critical care risk calculator utilizing pre-operative risk factors can improve resourcing for hip fracture patients in the peri‑operative period. MethodsA cohort of geriatric hip fracture patients at a single institution were reviewed over a three-year period. Non-operative patients, peri‑implant fractures, additional procedures performed under the same anesthesia period, and patients admitted to the intensive care unit (ICU) prior to surgery were excluded. Pre-operative laboratory values, Revised Cardiac Risk Index (RCRI), and American Society of Anesthesiologists (ASA) scores were calculated. Pre-operative ambulatory status was determined. The primary outcome measure was ICU admission in the post-operative period. Outcomes were assessed with Fisher's exact test, Kruskal-Wallis test, logistic regression, and ROC curve. Results315 patient charts were analyzed with 262 patients meeting inclusion criteria. Age ≥ 80 years, ASA ≥ 4, pre-operative hemoglobin < 10 g/dL, and a history of CVA/TIA were found to be significant factors and utilized within a “training” data set to create a 4-point scoring system after reverse stepwise elimination. The 4-point scoring system was then assessed within a separate “validation” data set to yield an ROC area under the curve (AUC) of 0.747. Score ≥ 3 was associated with 96.8 % specificity and 14.2 % sensitivity for predicting post-op ICU admission. Score ≥ 3 was associated with a 50 % positive predictive value and 83 % negative predictive value. ConclusionA hip fracture risk stratification scoring system utilizing pre-operative patient specific values to stratify geriatric hip patients to the ICU post-operatively can improve the pre-operative decision-making of surgical and critical care teams. This has important implications for triaging vital hospital resources. Level of evidenceIII (retrospective study)