We reviewed 183 patients who had undergone reconstruction of the anterior cruciate ligament. The incidence of meniscal tears and degenerative change was assessed and related to the timing from injury to surgery. Degenerative change was scored using the French Society of Arthroscopy system. The patients were divided into an early (surgery within 12 months of injury) and a late group (surgery more than 12 months from injury). The late group was also subdivided into four groups of 12-month periods ranging from one year to more than four years after injury.

There was a significantly higher incidence of meniscal tears in patients undergoing reconstruction after 12 months compared with those in the early group (71.2% vs 41.7%; p < 0.001). This was due to a large increase in medial meniscal tears in the late group. An increased incidence of degenerative change was also found in the late group (31.3% vs 10.7%; p < 0.001). Analysis of the subgroups showed that the incidence of meniscal tears and degenerative change did not differ significantly when surgery was performed after 12 months from injury. We conclude that reconstruction of the anterior cruciate ligament should be carried out within 12 months of injury to minimise the risk of meniscal tears and degenerative change.

We assessed the functional outcome following fracture of the tibial plateau in 63 consecutive patients. Fifty-one patients were treated by internal fixation, five by combined internal and external fixation and seven non-operatively. Measurements of joint movement and muscle function were made using a muscle dynamometer at three, six and 12 months following injury. Thirteen patients (21%) had a residual flexion contracture at one year. Only nine (14%) patients achieved normal quadriceps muscle strength at 12 months, while 19 (30%) achieved normal hamstring muscle strength. Recovery was significantly slower in patients older than 40 years of age. We conclude that there is significant impairment of movement and muscle function after fracture of the tibial plateau and that the majority of patients have not fully recovered one year after injury.

We prospectively analysed the epidemiology of acetabular fractures over a period of 16 years in order to identify changes in their incidence or other demographic features. Our study cohort comprised a consecutive series of 351 patients with acetabular fractures admitted to a single institution between January 1988 and December 2003.

There was no significant change in the overall incidence of acetabular fractures, which remained at 3 patients/100 000/year. There was, however, a significant reduction in the number of men sustaining an acetabular fracture over the period (p < 0.02). The number of fractures resulting from falls from a height < 10 feet showed a significant increase (p < 0.002), but there was no change in those caused by motor-vehicle accidents. There was a significant reduction in the median Injury Severity score over the period which was associated with a significant decrease in mortality (p < 0.04) and a reduction in the length of hospital stay. The incidence of osteoarthritis noted during follow-up of operatively-treated fractures declined from 31% to 14%, reflecting improved results with increasing subspecialisation. Our findings suggest that there will be a continuing need for some orthopaedic surgeons to specialise in the management of these fractures. In addition, the reductions in the Injury Severity score and mortality may be associated with improved road and vehicle safety.

Injury to the common peroneal nerve was present in 14 of 55 patients (25%) with dislocation of the knee. All underwent ligament reconstruction. The most common presenting direction of the dislocation was anterior or anteromedial with associated disruption of both cruciate ligaments and the posterolateral structures of the knee. Palsy of the common peroneal nerve was present in 14 of 34 (41%) of these patients. Complete rupture of the nerve was seen in four patients and a lesion in continuity in ten.

Three patients with lesions in continuity, but with less than 7 cm of the nerve involved, had complete recovery within six to 18 months. In the remaining seven with more extensive lesions, two regained no motor function, and one had only MRC grade-2 function. Four patients regained some weak dorsiflexion or eversion (MRC grade 3 or 4). Some sensory recovery occurred in all seven of these patients, but was incomplete. In summary, complete recovery occurred in three (21%) and partial recovery of useful motor function in four (29%). In the other seven (50%) no useful motor or sensory function returned.

The identification of high-risk factors in patients with fractures of the pelvis at the time of presentation would facilitate investigation and management. In a series of 174 consecutive patients with unstable fractures of the pelvic ring, clinical data were used to calculate the injury severity score (ISS), the triage-revised trauma score (T-RTS), and the Glasgow coma scale (GCS). The morphology of the fracture was classified according to the AO system and that of Burgess et al. The data were analysed using univariate and multivariate methods in order to determine which presenting features were identified with high risk.

Univariate analysis showed an association between mortality and an ISS over 25, a T-RTS below eight, age over 65 years, systolic blood pressure under 100 mmHg, a GCS of less than 8, blood transfusion of more than ten units in the first 24 hours and colloid infusion of more than six litres in the first 24 hours. Multivariate analysis showed that age, T-RTS and ISS were independent determinants of mortality. A T-RTS of eight or less identified the cohort of patients at greatest risk (65%). The morphology of the fracture was not predictive of mortality. We recommend the use of the T-RTS in the acute situation in order to identify patients at high risk.

We treated 21 patients with 22 dislocations of the knee by repair or reconstruction of all injured ligaments. Eight knees were treated in the acute phase (less than two weeks after injury); the remainder were treated more than six months after injury (6 to 72). Reconstructions were carried out with a combination of autograft and allograft tendons and by direct ligament repair where possible. At a mean follow-up of 32 months (11 to 77) the mean Lysholm score was 87 (81 to 91) in the acute group and 75 (53 to 100) in the delayed group. The mean Tegner activity rating was 5 in the acute group and 4.4 in the delayed group.

The International Knee Documentation Committee assessment revealed no differences between the two groups. Instrumented testing of knee stability indicated better results for anterior cruciate ligament reconstructions which had been undertaken in the acute phase, but no difference in the outcome of posterior cruciate ligament reconstructions. There was no difference in the loss of knee movement between the two groups. Although the differences were small, the outcome in terms of overall knee function, activity levels and anterior tibial translation were better in those knees which had been reconstructed within two weeks of injury.

We randomised 31 patients with acute tears of the anterior cruciate ligament (ACL) to receive either early (within 2 weeks) or delayed (8 to 12 weeks) reconstruction using a quadruple hamstring graft in order to determine if there was any functional advantage to early reconstruction. Outcome measures included return of knee movement, muscle dynamometry, International Knee Documentation Committee (IKDC) scores and Tegner activity levels.

Two cases of symptomatic deep-venous thrombosis occurred in the early group. The mean range of movement was significantly greater at two weeks in the delayed group. Quadriceps muscle power was significantly better in the delayed group at 12 weeks (p < 0.05). These trends were evident at later time points, but were not statistically significant. One patient in each group had clinically significant knee stiffness at 52 weeks. IKDC scores and Tegner activity levels were not significantly different at 52 weeks. We conclude that there is no functional advantage to be gained by early reconstruction of the ACL.

We used calcium-phosphate cement combined with minimal internal fixation to treat 49 fractures of the lateral tibial plateau. There were 25 split depression fractures, 22 pure depression fractures and two bicondylar fractures. Anatomical reduction was obtained in 38 fractures, satisfactory reduction in nine and imperfect reduction in two. Of 44 patients reviewed at one year, 33 were rated as having an excellent reduction. Functional outcome as measured by the Rasmussen score was good or excellent at six months in 92% (44/48) of patients and in 95% (42/44) at one year. Eight (16%) showed some loss of reduction of the plateau. In seven of these the loss of reduction was slight (< 3 mm) and no action was taken. One patient with a deep infection had gross loss of reduction and a poor functional outcome. Calcium-phosphate cement is a useful alternative to bone grafting for the treatment of fractures of the tibial plateau.

In a series of 150 consecutive patients with unstable fractures of the pelvis, angiography was performed in 23 (15%) who had uncontrolled hypotension. There were three anteroposterior compression (APC), eight lateral compression (LC) and 12 vertical shear (VS) injuries. Arterial sources of haemorrhage were identified in 18 (78%) patients and embolisation was performed. Angiography was required in 28% of VS injuries. The morphology of the fracture was not a reliable guide to the associated vascular injury. Ten (43%) patients died, of whom six had had angiography as the first therapeutic intervention. Five of these had a fracture which was associated with an increase in pelvic volume (APC or VS) which could have been stabilised by an external fixator. Based on our findings we recommend skeletal stabilisation and, if indicated, laparotomy to deal with sources of intraperitoneal blood loss before pelvic angiography. Embolisation of pelvic arterial bleeding is a worthwhile procedure in patients with hypotension which is unresponsive to these interventions.

We reviewed 286 consecutive patients with a fracture of a pubic ramus. The overall incidence was 6.9/100 000/year in the total population and 25.6/100 000/year in individuals aged over 60 years. The mean age of the patients was 74.7 years and 24.5% suffered from dementia. Women were affected 4.2 times more often than men. After injury, geriatric rehabilitation was frequently required and although most surviving patients returned to their original place of residence, their level of mobility was often worse. The overall survival rates at one and five years were 86.7% and 45.6%, respectively. Multiple logistic regression analysis showed that age and dementia were the only independent significant factors to be predictive of mortality (p < 0.05).

Patients with a fracture of a pubic ramus had a significantly worse survival than an age-matched cohort from the general population (log-rank test, p < 0.001), but this was better than patients with a fracture of the hip during the first year after injury, although their subsequent mortality was higher. Five years after the fracture there was no significant difference in survival between the two groups.

Reamed intramedullary nailing was carried out on 57 Gustilo grade-IIIB tibial fractures in 55 patients. After debridement, there was substantial bone loss in 28 fractures (49%). The mean time to union was 43 weeks (14 to 94). When there was no bone loss, the mean time to union was 32 weeks; it was 45 weeks if there was bone loss. Fractures complicated by infection took a mean of 53 weeks to heal. Revision nailing was necessary in 13 fractures (23%) and bone grafting in 15 (26%). In ten fractures (17.5%) infection developed, in four within six weeks of injury and in six more than four months later. Of these, nine were treated successfully, but one patient required an amputation because of osteomyelitis.

Our results indicate that reamed intramedullary nailing is a satisfactory treatment for Gustilo grade-III tibial fractures.

We made a prospective study of 208 patients with tibial fractures treated by reamed intramedullary nailing. Of these, 11 (5.3%) developed dysfunction of the peroneal nerve with no evidence of a compartment syndrome.

The patients with this complication were significantly younger (mean age 25.6 years) and most had closed fractures of the forced-varus type with relatively minor soft-tissue damage. The fibula was intact in three, fractured in the distal or middle third in seven, with only one fracture in the proximal third. Eight of the 11 patients showed a ‘dropped hallux’ syndrome, with weakness of extensor hallucis longus and numbness in the first web space, but no clinical involvement of extensor digitorum longus or tibialis anterior. This was confirmed by nerve-conduction studies in three of the eight patients.

There was good recovery of muscle function within three to four months in all cases, but after one year three patients still had some residual tightness of extensor hallucis longus, and two some numbness in the first web space. No patient required further treatment.