Session VI - Geriatrics

Outcomes after Hip Fracture: The Results of a Prospective Multicenter Database

Kenneth J. Koval, MD; Andrew L. Chen, MD; Ethan A. Halm, MD; Sean R. Morrison, MD; Mary Ann McLaughlin, MD; Gretchen Orosz, MD; Jay Magaziner, PhD; Albert Siu, MD, Hospital for Joint Diseases-New York University, New York, New York, USA

Purpose: Hip fracture in the elderly population is associated with disability and increased age-adjusted mortality. We sought to 1) identify the clinical impact of potentially modifiable risk factors that may affect the timing of hip fracture surgery in the elderly; 2) assess the impact of active clinical issues or functional impairments on discharge on clinical and functional outcomes in patients hospitalized with hip fractures; 3) identify and compare the significance of pre-fracture predictors of functional status and mortality at 6 months after hospitalization for hip fracture; and 4) compare risk-adjusted outcomes for hospitals providing initial care.

Methods: This prospective, multi-institutional observational study included 571 consecutive patients more than 50 years of age admitted to four institutions between August 1997 and August 1998 with a hip (femoral neck or intertrochanteric) fracture. Structured interviews with patients and proxies were performed, with identification of pre-injury ambulatory capacity and cognitive status and comorbid conditions. The presence of delirium and laboratory and physical examination findings were recorded on admission and prior to surgery. Physical and laboratory abnormalities were classified as moderate (level 1) or severe (level 2). Complications were recorded as they occurred, supplemented by medical review. At 2 and 6 months after the initial injury, ambulatory capacity and functional status were re-assessed. Multivariate logistic regression analysis was used to determine the significance of independent variables.

Results: Patients averaged 82 years of age with 2.3 comorbidities. Multivariate logistic regression analysis showed that dependency in ambulation, a history of chronic obstructive pulmonary disease, and the presence of one (OR 2.2; 95% CI ,1.1 to 4.3) or more (OR 5.9; CI 2.5 to 14.3) level 2 criteria on admission were associated with the development of a major perioperative complication. On discharge, 16.3% of patients had one or more active clinical issues, most commonly low diastolic blood pressure, fever, or tachycardia; 41% of patients had one or more new impairments, most commonly urinary tract infections, decubitus ulcers, and being bedbound. Active clinical issues or new impairments on discharge were associated with increased risk of death, readmission, and major events (P < 0.05);18.4% of patients with no active clinical issues on discharge died or were readmitted within 60 days compared with 28.0% of those with one and 44.4% of those with more than two (P < 0.001). New impairments were also associated with lower functional independent mobility scores at 60 days. These relationships were maintained in multivariate analyses that controlled for a previously validated hip fracture-specific risk-adjustment indicator.

The in-hospital mortality rate was 1.6%; this increased to 13.5% at 6 months; 26.3% of patients experienced an "adverse outcome" (death or need of total assistance for ambulation) after hip fracture. Medical comorbidities and abnormal laboratory indices on admission were strong predictors of mortality, but were not significantly associated with impaired ambulatory capacity at 6 months. Increased age and pre-fracture residence at a nursing home were not associated with increased mortality, but were predictive of diminished ambulatory capacity. Adjustment for baseline characteristics either substantially augmented or diminished inter-hospital differences in outcomes: two hospitals had functional status or mortality that was significantly worse than the overall mean, although the other outcome was better, but not significantly, than average.

Discussion and Conclusions: Potentially reversible abnormalities in laboratory and physical examination occurred frequently on admission and, when severe, significantly increased the risk of peri-operative complications. We also found that level 1 criteria, while warranting correction, did not increase risk and may not be an indication for delaying surgery. Hip fracture in the elderly is associated with increased mortality and diminished ambulatory capacity 6 months after injury. Patients with multiple medical problems are more likely to die within 6 months of injury, but increasing age and pre-fracture residence in a nursing home are more predictive of inability to ambulate independently. Active clinical issues and new impairments on discharge adversely affect clinical outcomes even after controlling for other important prognostic factors. Individual hospitals may have worse mortality rates, but this is not necessarily associated with worse functional outcome.