We studied the epidemiology of 401 fractures of the shaft of the humerus in 397 patients aged 16 years or older. The incidence was 14.5 per 100 000 per year with a gradually increasing age-specific incidence from the fifth decade, reaching almost 60 per 100 000 per year in the ninth decade. Most were closed fractures in elderly patients which had been sustained as the result of a simple fall. The age distribution in women was characterised by a peak in the eighth decade while that in men was more even. Simple fractures were by far the most common and most were located in the middle or proximal shaft. The incidence of palsy of the radial nerve was 8% and fractures in the middle and distal shaft were most likely to be responsible. Only 2% of the fractures were open and 8% were pathological. These figures are representative of a population with a low incidence of high-energy and penetrating trauma, which probably reflects the situation in most European countries.

We studied 60 patients with an acute displaced fracture of the femoral neck and with a mean age of 84 years. They were randomly allocated to treatment by either internal fixation with cannulated screws or hemiarthroplasty using an uncemented Austin Moore prosthesis. All patients had severe cognitive impairment, but all were able to walk independently before the fracture. They were reviewed at four, 12 and 24 months after surgery. Outcome assessments included complications, revision surgery, the status of activities of daily living (ADL), hip function according to the Charnley score and the health-related quality of life (HRQOL) according to the Euroqol (EQ-5D) (proxy report).

General complications and the rate of mortality at two years (42%) did not differ between the groups. The rate of hip complications was 30% in the internal fixation group and 23% in the hemiarthroplasty group; this was not significant. There was a trend towards an increased number of re-operated patients in the internal fixation group compared with the hemiarthroplasty group, 33% and 13%, respectively (p = 0.067), but the total number of surgical procedures which were required did not differ between the groups. Of the survivors at two years, 54% were totally dependent in ADL functions and 60% were bedridden or wheelchair-bound regardless of the surgical procedure. There was a trend towards decreased mobility in the hemiarthroplasty group (p = 0.066). All patients had a very low HRQOL even before the fracture. The EQ-5D_index score was significantly worse in the hemiarthroplasty group compared with the internal fixation group at the final follow-up (p < 0.001).

In our opinion, there is little to recommend hemiarthroplasty with an uncemented Austin Moore prosthesis compared with internal fixation, in patients with severe cognitive dysfunction.

We studied 217 patients with an unstable trochanteric or subtrochanteric fracture who had been randomly allocated to treatment by either internal fixation with a standard Gamma nail (SGN) or a Medoff sliding plate (MSP, biaxial dynamisation mode). Their mean age was 84 years (65 to 99) and they were reviewed at four and 12 months after surgery. Assessments of outcome included general complications, technical failures, revision surgery, activities of daily living (ADL), hip function (Charnley score) and the health-related quality of life (HRQOL, EQ-5D).

The rate of technical failure in patients with unstable trochanteric fractures was 6.5% (6/93) (including intra-operative femoral fractures) in the SGN group and 5.2% (5/96) in the MSP group. In patients with subtrochanteric fractures, there were no failures in the SGN group (n = 16) and two in the MSP group (n = 12). In the SGN group, there were intra-operative femoral fractures in 2.8% (3/109) and no post-operative fractures. There was a reduced need for revision surgery in the SGN group compared with the MSP group (8.3%; 9/108; p = 0.072). The SGN group also showed a lower incidence of severe general complications (p < 0.05) and a trend towards a lower incidence of wound infections (p = 0.05). There were no differences between the groups regarding the outcome of ADL, hip function or the HRQOL. The reduction in the HRQOL (EQ-5D_indexscore) was significant in both groups compared with that before the fracture (p < 0.005).

Our findings indicate that the SGN showed good results in both trochanteric and subtrochanteric fractures. The limited number of intra-operative femoral fractures did not influence the outcome or the need for revision surgery. Moreover, the SGN group had a reduced number of serious general complications and wound infections compared with the MSP group. The MSP in the biaxial dynamisation mode had a low rate of failure in trochanteric fractures but an unacceptably high rate when used in the biaxial dynamisation mode in subtrochanteric fractures.

The negative influence of an unstable trochanteric or subtrochanteric fracture on the quality of life was significant regardless of the surgical method.

We randomised 40 elderly patients of mean age 74 years with displaced three- or four-part fractures of the humerus to either conservative treatment or tension-band osteosynthesis.

At one year and after three to five years, clinical follow-up showed no functional differences between the two groups of patients, with optimal function achieved within one year. There were major complications only in the surgically-treated group. Radiological review showed that surgery had improved the position of the fractured humeral head, but this was not reflected in improved function.

Semi-rigid fixation with tension-band wiring of displaced multifragment fractures of the proximal humerus in the elderly did not improve the functional outcome when compared with conservative treatment.