Non-operative predictors for subsequent interventions after intracapsular femoral neck fractures in elderly: A comprehensive analysis of the Hungarian nationwide health insurance database.

OBJECTIVES

To investigate the correlation between non-operative prognostic factors and non-prosthetic fracture-related treatments following internal fixation of intracapsular femoral neck fractures in elderly patients.

DESIGN AND SETTING

Retrospective observational cohort study. Comprehensive analysis of the Hungarian nationwide health insurance database.

PARTICIPANTS

Data of in-patient records with S7200 ICD-10 codes were collected from the Hungarian National Health Insurance Fund Administration (HNHIFA) and from the health care provider institutes. The patients with femoral neck fractures in the year of 2000, following reduction and internal fixation aged 60 years or older were evaluated. The secondary, non-prosthetic fracture related treatments during the 8 year follow-up period were registered.

MAIN OUTCOME MEASURES

Of the prognostic factors, age, gender, type of fracture, season and day of the primary surgery, length of waiting time to the operation and the accompanied diseases were evaluated as risk factors for all type of fracture-related further interventions, with the exception of arthroplasties.

RESULTS

A total of 2895 patients with intracapsular femoral neck fractures met the study criteria. The mean age was 77.96 years (SD: 8.54). The cohort of the patients was observed for a total of 10,077.8 person-years. The non-prosthetic fracture related treatment was performed in 265 patients (9,2%); the median of the time elapsed to the secondary definitive treatment was 3.5 months. With Cox regression analysis, significant correlation was revealed between the incidence of non-prosthetic treatment and younger age (year, HR = 0.977, p = 0.002), surgical delay (12-24 h vs 0-6 h, HR = 1.518, p = 0.023; 24h+ vs 0-6 h, HR = 1.372, p = 0.050), season of primary osteosynthesis (fall vs summer, HR = 0.636, p = 0.012), and type of femoral neck fracture (intracapsular displaced vs intracapsular undisplaced, HR = 1.340, p = 0,047). There was no significant effect of the day of primary surgery, the gender and the presence of co-morbidities on the incidence of further surgical interventions.

CONCLUSION

The summertime primary surgical intervention, delay of surgery longer than 12 h and type of femoral neck fracture are independent predictors of non-prosthetic further treatment of femoral neck fractures in elderly patients.

LEVEL OF EVIDENCE

Level IV, evidence from cohort studies.