Predictors of mortality in hip fracture patients

Systematic review and meta-analysis of preoperative predictors for early mortality following hip fracture surgery

Hip fractures are a global health problem with a high postoperative mortality rate. Preoperative predictors for early mortality could be used to optimise and personalise healthcare strategies. This study aimed to identify predictors for early mortality following hip fracture surgery. Cohort studies examining independent preoperative predictors for mortality following hip fracture surgery were identified through a systematic search on Scopus and PubMed. Predictors for 30-day mortality were the primary outcome, and predictors for mortality within 1 year were secondary outcomes. Primary outcomes were analysed with random-effects meta-analyses. Confidence in the cumulative evidence was assessed using the GRADE criteria. Secondary outcomes were synthesised narratively. Thirty-three cohort studies involving 462,699 patients were meta-analysed. Five high-quality evidence predictors for 30-day mortality were identified: age per year (OR: 1.06, 95% CI: 1.04-1.07), ASA score ≥ 3 (OR: 2.69, 95% CI: 2.12-3.42), male gender (OR: 2.00, 95% CI: 1.85-2.18), institutional residence (OR: 1.81, 95% CI: 1.31-2.49), and metastatic cancer (OR: 2.83, 95% CI: 2.58-3.10). Additionally, six moderate-quality evidence predictors were identified: chronic renal failure, dementia, diabetes, low haemoglobin, heart failures, and a history of any malignancy. Weak evidence was found for non-metastatic cancer. This review found relevant preoperative predictors which could be used to identify patients who are at high risk of 30-day mortality following hip fracture surgery. For some predictors, the prognostic value could be increased by further subcategorising the conditions by severity.

Epidemiology and mortality in older patients treated by reverse shoulder arthroplasty for displaced proximal humerus fractures

INTRODUCTION

Reverse shoulder arthroplasty (RSA) is rapidly becoming the preferred treatment for displaced proximal humerus fractures in older patients. However, few studies have analyzed the target population and the effect of RSA on survival, although the socioeconomic impact of this type of surgery is considerable.

PATIENTS AND METHODS

This was a retrospective epidemiological study of all patients with a displaced proximal humeral fracture treated by RSA in 14 public and private hospitals throughout France between 1995 and 2016. The French hospital discharge database (PMSI) was analyzed to isolate an 898-patient cohort who underwent RSA within 6 weeks of the fracture event. In 87% of cases, this was a 3- or 4-fragment fracture. We analyzed the epidemiological characteristics of the patients at the time of fracture, their survival (Kaplan-Meier estimate) and factors that may impact survival.

RESULTS

The mean age at the time of fracture and surgery was 79 years (46-98 years). Eighty percent of the cohort was female (sex ratio: 0.18 [p=0.0042], with 21% obesity rate [BMI>30]) and 60% of patients were ASA 1-2. The most common comorbidities were cardiovascular and neurological. The survival rate after RSA was 94% at 1 year and 73% at 5years. At the latest follow-up of 19 years, 42% of patients were still alive. In 18% of cases, the patient died within the first 15 days. The presence of comorbidities (ASA score>3-4) (p<0.004) and/or cognitive disorders (p<0.0001) were risk factors for early mortality. The time to surgery, type of fracture, associated fractures and discharge destination (return home, transfer to nursing home) had no effect on postoperative mortality in our cohort.

CONCLUSION

Despite being older (79 years) at the time of proximal humerus fracture, patients who underwent RSA treatment had a high survival rate (94% at 1 year, 73% at 5years), which is better than the survivorship reported after surgical treatment of femoral neck fractures (81-87% at 1 year, 38% at 5years). The presence of comorbidities (ASA>3-4) and/or cognitive disorders are risk factors for early mortality and should be taken into account to prevent early death.

LEVEL OF EVIDENCE

IV, Retrospective study.

Patient and Perioperative Factors Influencing the Functional Outcomes and Mortality in Elderly Hip Fractures

This study aimed to evaluate the functional outcomes and mortality following hip fracture surgery in elderly patients, and to identify the associated risk factors. Between January 2016 and December 2017, 480 consecutive patients were finally included for data analyses. The Harris score and Barthel index were used to evaluate the hip function and ability to perform activities of daily living (ADL). Univariate and multivariate logistics regression analyses were performed to determine the independent risk factors for mortality, poor hip function or poor ability to perform ADL. The mortality rate was 15.6% (75/480). In the survivors, poor outcome developed in 133 (32.8%) patients and poor ADL was in 72 (17.8%) patients. The independent factors that influenced mortality were advanced age (p = 0.033), male gender (0.031), living in rural area (p < 0.001), self-reported diabetes (p = 0.005), tumor (p = 0.024), preoperative delay >7 days (p = 0.020), postoperative drainage use (p = 0.034), WBC > 10 × 10⁹/L (p = 0.005), reduced RBC (p = 0.011), PLT < 100 × 10⁹/L (p < 0.001), ALB < 35 g/L (p < 0.001) and CK > 200 U/L (p = 0.003). The independent factors that influenced the hip function were male gender (p = 0.009), WBC > 10 × 10⁹/L (p < 0.001), lower HBG (p = 0.005), and ALB < 35 g/L (p < 0.001). The independent factors that influenced the ability to perform ADL were diagnosis of trochanteric fracture (p = 0.048), preoperative delay > 7 days (p = 0.027), postoperative drainage use (p = 0.010), elevated WBC (p = 0.020), lower HGB (p < 0.001), PLT < 100 × 10⁹/L (p = 0.002), and ALB < 35 g/L (p < 0.001). Although most of risk factors were not modifiable, they aid in patient individual risk evaluation, risk stratification, and counseling patients or relatives.