Fracture healing, functional outcome and health related quality of life in younger patients with a femoral neck fracture

Abstract

Few prospective studies have investigated the patient-reported outcome measures (PROM) in patients less than 70 years that sustained a femoral neck fracture (FNF) treated with closed reduction and internal fixation (CRIF). The health-related quality of life (HRQoL) using the self-administered questionnaires EuroQol 5 Dimension (EQ-5D) and Short Form-36 Health Survey (SF-36) and with a long term follow-up has not previously been examined in this group of patients. Factors associated with healing complications leading to a major re-operation with a total hip replacement as well as mortality has not been investigated for this group. The ability of the questionnaires EQ-5D and SF-36 to reveal clinical important changes in health over time, the so called responsiveness, has not been studied for patients younger than 70 years with a FNF. The comprehensive goal of this thesis was to investigate the HRQoL, functional outcome, factors associated with a major re-operation as well as mortality with a 4, 12, 24-month and 10-year follow-up in patients 20-69 years that sustained a FNF treated with CRIF. Another aim was to determine the responsiveness of EQ-5D and SF-36 in this specific group of patients.

Methods: Study I. 182 patients with displaced and non-displaced FNF were included in a prospective multicenter study. At each follow-up, clinical and radiographic examinations were performed at 4, 12 and 24 months. Gathered data included HRQoL (EQ-5D and SF-36), hip function according to Harris Hip Score (HHS), fracture healing complications such as nonunion (NU) and avascular necrosis (AVN), co-morbidities and major re-operations with a hip arthroplasty. Study II. 128 patients from the same cohort as Study I with a displaced FNF were included. At inclusion, bone mineral density (BMD) was measured by DXA. A logistic regression was performed to find associated variables with a re-operation due to NU or AVN detected at 4, 12 and 24-months follow-up. Study III. A 10-year follow-up of the cohort in Study I. The hip function was assessed using the Hip Disability Outcome Score (HOOS) and HRQoL was evaluated using the EQ-5D questionnaire. Deceased patients had the date of death recorded and associated factors with mortality were analyzed by regression logistic calculation of baseline data. Study IV. The responsiveness of the HRQoL instruments SF-36 and EQ-5D was evaluated by estimating the standardized effect size (SES), standardized response mean (SRM), receiver operating characteristic curve (ROC) and area under the curve (AUC).

Results: Study I. At 24 months, the hip function according to HHS in patients with a displaced fracture showed a good or excellent score in 73% and in those with a non-displaced fracture 85% (p = 0.15). EQ-5D and SF-36 were the lowest at 4 months and improved at 12 and 24 months but did not regain pre-fracture level (p<0.01) regardless of fracture type. Patients with a displaced FNF had a NU in 23% of the cases, AVN in 15% and a total of 28% had a major re-operation. No patients with a non-displaced FNF developed NU, 12% developed an AVN and a total of 8% had a major re-operation with a total hip replacement (THR). Study II. The rate of a major re-operation was 6%, 16% and 28% at 4, 12 and 24 months. Patients having a low BMD (OR 5.5, CI 1.1-27) and harmful alcohol consumption (OR 3.2, CI 1.2-8.8) were more likely to undergo a major re-operation due to NU or AVN. Study III. From initial 182 included patients, a total of 88 patients participated. The EQ-5D improved compared to 24-month follow-up (p=0.006). However, the EQ-5D did not recover to the pre-fracture level (p<0.001). The score, however was equivalent to population data of Sweden. Factors associated with mortality at 10 years were higher age, co-morbidity, osteoporosis and smoking. Study IV. SES at 4 months was large for EQ-5D and SF-36 and moderate at 12- and 24- month follow-up. The correlation between the changes in HHS (4–24 months) and HRQoL were moderate to weak but the correlation between total scores were strong. The ability to predict and follow the external standard as well as the effect sizes implies that the internal and external responsiveness of SF-36 and EQ-5D were good.

Conclusion: The functional outcome was good or excellent in more than two thirds and the HRQoL continues to improve up to 10 years after sustaining a FNF in patients younger than 70 years treated with CRIF. However, the HRQoL did not reach the level before fracture, probably because all included subjects were 10 years older. The EQ-5D at 10-year followup were equivalent to sex- and age-matched reference population in Sweden. The majority of the patients with a displaced FNF healed and less than 10% of patients with a nondisplaced FNF underwent a major re-operation. A low BMD and harmful alcohol consumption according to AUDIT increased the risk for undergoing a major re-operation with a total hip replacement. One third of the patients were deceased 10 years later and they were more compromised with illnesses and smoking comparing to surviving subjects. The HRQoL questionnaires EQ-5D and SF-36 were both responsive for changes in health over time. EQ-5D was easier to administer and can be used alone compared to the more complex and time-consuming SF-36.