Fig. 1 Algorithm for THA patient selection.
Courtesy of James Pegrum, MBBS, BSc

AAOS Now

Published 10/1/2016
|
Peter Pollack

Can an Algorithm Reduce Mortality for Femoral Neck Fracture Patients?

Adoption of scoring system linked to fewer patient deaths at 1 and 3 years
Increased use of total hip arthroplasty (THA) in the treatment of intracapsular proximal femoral fractures has highlighted the need for an improved patient-selection strategy, according to researchers in the United Kingdom. They displayed their findings in a poster presentation at the 2016 AAOS Annual Meeting.

"Hip fracture is a significant public health issue," explained James Pegrum, MBBS, BSc, "with up to 75,000 cases per year at an annual cost of approximately £2 billion in the United Kingdom. Globally there are more than 1.6 million femoral neck fractures associated with 740,000 deaths, and the number of fractures is expected to increase to 6.3 million by 2050. The occurrence of a fall and hip fracture often signifies ill-health due to multiple underlying comorbidities, and a life expectancy of less than 1 year in a third of cases. Selecting the cohort of patients who will benefit from THA over hemiarthroplasty remains a challenge.

"Current guidelines from the U.K. National Institute for Clinical Excellence (NICE) recommend THA for patients who can walk independently with one stick or less, are not cognitively impaired, and are medically fit for anesthesia and the THA procedure," he continued. "However, this score does not take into account the multiple comorbidities found within this patient cohort and does not sufficiently predict patient survival. In addition, an unpublished questionnaire to 107 orthopaedic surgeons in regional hospitals in the United Kingdom showed that only 50 percent of respondents use the NICE guidelines."

The researchers developed a scoring algorithm based on a combination of NICE guidelines, Charlson Comorbidity Index, and Sernbo score (Fig. 1). The Sernbo score stratifies patients into high- and low-risk groups based on four factors: age (< 80 years/ ≥ 80 years); social situation (independent or not); mobility; and mental state (normal/history of dementia). The researchers also proposed a separate scoring system to decide between hemiarthroplasty and THA, but it remains unvalidated.

Improved survivorship
Mr. Pegrum and his colleagues used an institutional database to review information on 376 consecutive intracapsular proximal femoral fractures treated with either THA or cemented hip hemiarthroplasty at a single trauma center. Of those, 250 consecutive fractures were selected for treatment based on existing NICE guidelines (Group 1), and 126 consecutive fractures were selected for treatment based on the new algorithm (Group 2). The research team assessed 3-year survivorship using a Kaplan-Meier survival plot. In addition, they used a Cox proportional hazards model to assess the multivariate model of predictors during procedure selection. Data were analyzed to assess appropriateness of selection criteria for THA or hemiarthroplasty.

"A log-rank test for equality of survivor functions revealed no significant difference between hemiarthroplasty and THA survivorship at 3 years in Group 1," said Mr. Pegrum, "while in Group 2, a significant difference was seen at the same time point. Among other things, the use of this new algorithm helped reduce 1-year mortality in patients undergoing THA in our institution from 15.4 percent to 3.8 percent. A statistically significant difference in Kaplan-Meier survival was apparent from 6 months, increasing out toward 3 years."

In addition, Mr. Pegrum noted that overall risk of mortality increased with age in an essentially linear manner, so the research team could not recommend an absolute age cut-off point for appropriateness of treatment.

Future efforts
"This new scoring algorithm does not predict which patients are likely to do better, but rather those with improved survivorship who will likely tolerate the longer surgery and have the long-term benefit from a THA," Mr. Pegrum explained. "The long-term survival of this group of patients is based on comorbidities within the Charlson score, but there may be other factors involved with patient outcome that are not currently included within the Charlson score. Additional studies using this algorithm and future patient follow-up may be better able to define these parameters in the future.

"Further work is needed to determine if patients with a pertrochanteric or subtrochanteric fracture can be risk-stratified in a similar way," he continued, "especially if there could be a decision-making algorithm for the use of dynamic hip screw versus intramedullary nailing for the pertrochanteric group."

Mr. Pegrum's coauthors are Geraint E. Thomas, MA, MBBS, PhD; Reza Mayahi, MD; and Gregoris Kambouroglou.

Peter Pollack is the electronic content specialist for AAOS Now. He can be reached at ppollack@aaos.org

Bottom Line

  • Researchers reviewed 376 consecutive patients with intracapsular proximal femoral fractures treated with either THA or cemented hip hemiarthroplasty.
  • Introduction of an improved algorithm for patient selection was associated with a reduction in 1-year mortality from 15.4 percent to 3.8 percent.
  • The algorithm does not predict improved outcomes—only improved survivorship.
  • Further work is needed to see if patients with a pertrochanteric or subtrochanteric fracture can be risk stratified in a similar manner.