Increased mortality after a hip fracture has been associated with age, sex, and comorbidity. In order to estimate the long-term mortality with reference to hip fracture type, we followed 499 patients older than 60 years who had been treated surgically for a unilateral hip fracture for 10 years. At admission, patients with femoral neck fractures (n = 172) were 2 years younger than intertrochanteric patients (77.6 +/- 7.7 [SD] vs. 79.9 +/- 7.4 [SD], P = 0.001) and had a greater prevalence of heart failure (57% vs. 40.3%, P = 0.03). Similar mortality rates were observed at 1 year in both types of fracture (17.9% vs. 11.3%, log rank test P = 0.112). Mortality rates were significantly higher for intertrochanteric fractures at 5 years (48.8% vs. 34.7%, P = 0.01) and 10 years (76% vs. 58%, P = 0.001). Patients 60-69 years old with intertrochanteric fractures had significantly higher 10-year mortality than patients of similar age with femoral neck fractures (P = 0.008), while there was no difference between the groups aged 70-79 (P> 0.3) and 80-89 (P = 0.07). Women were less likely to die in 5 years (relative risk [RR] = 0.57, 95% confidence interval [CI] 0.41-0.79, P = 0.0007) and 10 years (RR = 0.65, 95% CI 0.49-0.85, P = 0.002). Age, sex, the type of fracture, and the presence of heart failure were independent predictors of 10-year mortality (Cox regression model P < 0.0001). The intertrochanteric type was independently associated with 1.37 (95% CI 1.03-1.83) times higher probability of death at 10 years (P = 0.002). In conclusion, the type of fracture is an independent predictor of long-term mortality in patients with hip fractures, and the intertrochanteric type yields worse prognosis.

Among 1685 patients who sustained a hip fracture at the island of Crete (Greece) in a 4-year period we found 106 patients with bilateral noncontemporary hip fractures. Pathologic hip fractures and fractures that emerged from high energy trauma were excluded. To investigate the possible factors predisposing to the later fracture in the sound hip, we studied these 106 patients and compared them with the 1579 patients who sustained a single hip fracture (control group). There was no significant difference between the mean age of the bilateral group at the time of the first hip fracture (78.3 +/- 7.4 years, range 52-94 years) and the mean age of the control group (77.3 +/- 11.9 years, range 50-101 years). We found no significant difference in the bone status between the two groups, using both lumbar spine dual photon absorptiometry (DXA) and calcaneus broadband ultrasound attenuation (BUA). Falls, which were the main cause of all the hip fractures, were much more common in the bilateral group. The second hip fracture was of the same location (trochanteric or cervical) in 92% of the trochanteric and 68% of cervical fractures and a tendency to greater displacement or instability was observed. Of the second hip fractures 75% happened in the first 48 months after the first one. The mean interval time was much longer (160 months) when a neck fracture was followed by a trochanteric one.

We assessed the bone mineral density (BMD), the body mass index (BMI), and the hip axis length (HAL) in 78 postmenopausal women with 38 cervical and 40 trochanteric hip fractures. The results were compared with those of age-matched, control postmenopausal women. No statistically significant difference was found in the values of BMD, BMI, and HAL between the groups of patients with cervical and those with trochanteric fractures, but lower BMD and BMI were found in fracture patients compared with the corresponding values of the control subjects. Contrary to the existing data, HAL was found to be shorter in the fracture patients compared with the controls. Thus, the type of hip fracture was found to be independent of the value of BMD, BMI, and the length of the patient's hip axis. The fact that a shorter hip axis was found in the group of fracture patients compared with that found in the control subjects raises questions about the significance of this parameter as an independent risk factor for hip fracture.

No abstract

Aseptic loosening after total hip arthroplasty is related to bone loss of the operated leg. The aim of the present study was to investigate the effect of aseptic loosening on volumetric bone mineral density (vBMD) and bone geometry in the operated leg, in postmenopausal women with a loosened cemented femoral implant using peripheral quantitative computed tomography (pQCT). We matched 12 postmenopausal women with aseptic loosening of cemented femoral implant, with 12 women without aseptic loosening (control group) according to age, BMI, and years from operation. All patients underwent pQCT of both tibias, DXA of the lumbar spine, and determination of biochemical markers of bone turnover. pQCT values in the control group as well as the nonoperated legs between groups had no significant difference. In the aseptic loosening group, there was significant reduction of cortical vBMD (cort vBMD) at 14% and 38% sites (cortical site), cortical thickness at 38% site, and of polar stress strength index (SSIp) at 14% site (transition zone) in the operated compared with the nonoperated leg. Similarly, there was significant reduction of cort vBMD at 14% and 38% sites and total vBMD and trabecular vBMD (trab vBMD) at the 14% site in the operated legs between the two groups. The aseptic loosening group had increased osteocalcin and serum collagen cross-linked N- and C-telopeptides (sNTX and sCTX) levels compared with controls. Aseptic loosening is associated with significant decrease of cortical and trabecular vBMD, and impairment of bone geometry and strength only in the operated leg. Increased bone turnover probably represents a local phenomenon, and is not associated with systemic skeletal disease.