Intracapsular neck of femur fractures may be treated with fixation or arthroplasty, depending on fracture characteristics and patient factors. Two common methods of fixation are the sliding hip screw, with or without a de-rotation screw, and cannulated screws. Each has its merits, and to date there is controversy around which method is superior, with either method thought to risk avascular necrosis of the femoral head (AVN) rates in the region of 10–20%.

Fixation with cannulated screws may be performed in various ways, with current paucity of evidence to show an optimum technique. There are a multitude of factors which are likely to affect patient outcomes: technique, screw configuration, fracture characteristics and patient factors. We present a retrospective case series analysis of 65 patients who underwent cannulated screw fixation of a hip fracture.

Electronic operative records were searched from July 2014 until July 2019 for all patients with a neck of femur fracture fixed with cannulated screws: 68 were found. Three patients were excluded on the basis of them having a pathological fracture secondary to malignancy, cases were followed up for 2 years post-operatively. Electronic patient records and X-rays were reviewed for all included patients. All X-rays were examined by each team member twice, with a time interval of two weeks to improve inter-observer reliability.

65 patients were included with 2:1 female to male ratio and average age of 72 years. 36 patients sustained displaced fractures and 29 undisplaced. Ten patients sustained a high-energy injury, none of which developed AVN. Average time to surgery was 40 hours and 57 patients mobilised on day one post-operatively. All cases used either 7 or 7.3mm partially threaded screws in the following configurations: 2 in triangle apex superior, 39 triangle apex inferior, 22 rhomboid and 2 other, with 9 cases using washers. All reductions were performed closed.

Five (8%) of our patients were lost to follow-up as they moved out of area, 48 (74%) had no surgical complications, seven (11%) had mild complications, three (5%) moderate and two (3%) developed AVN. Both of these sustained displaced fractures with low mechanism of injury, were female, ASA 2 and both ex-smokers. One received three screws in apex inferior configuration and one rhomboid, neither fixed with washers.

Our AVN rate following intracapsular hip fracture fixation with cannulated screws is much lower than widely accepted. This study is under-powered to comment on factors which may contribute to the development of AVN. However, we can confidently say that our practice has led to low rates of AVN. This may be due to our method of fixation; we use three screws in an apex inferior triangle or four screws in a rhomboid, our consultant-led operations, closed reduction of all fractures, or our operative technique. We pass a short thread cannulated screw across the least comminuted aspect of the fracture first in order to achieve compression, followed by two or three more screws (depending on individual anatomy) to form a stable construct. Our series shows that fixation of intracapsular hip fractures with cannulated screws as we have outlined remains an excellent option. Patients retain their native hip, have a low rate of AVN, and avoid the risks of open reduction.

Infected non-unions of proximal femoral fractures are difficult to treat. If debridement and revision fixation is unsuccessful, staged revision arthroplasty may be required. Non-viable tissue must be resected, coupled with the introduction of an antibiotic-eluting temporary spacer prior to definitive reconstruction. Definitive tissue microbiological diagnosis and targeted antibiotic therapy are required. In cases of significant proximal femoral bone loss, spacing options are limited.

We present a case of a bisphosphonate-induced subtrochanteric fracture that progressed to infected non-union. Despite multiple washouts and two revision fixations, the infection remained active with an unfavourable antibiogram. The patient required staged revision arthroplasty including a proximal femoral resection. To enable better function by maintaining leg length and offset, a custom-made antibiotic-eluting articulating temporary spacer, the Cement-a-TAN, was fabricated. Using a trochanteric entry cephalocondylar nail as a scaffold, bone cement was moulded in order to fashion an anatomical, patient-specific, proximal femoral spacer. Following resolution of the infection, the Cement-a-TAN was removed and a proximal femoral arthroplasty was successfully performed.

Cement-a-TAN is an excellent temporary spacing technique in staged proximal femoral replacement for infected non-union of the proximal femur where there has been significant bone loss. It preserves mobility and maintains leg length, offset and periarticular soft-tissue tension.