Although total elbow arthroplasty is undertaken in far smaller numbers than total hip and knee arthroplasty a recent review of the world literature indicated that aseptic loosening radiologically occurred in 17.2% whilst clinical loosening was present in 6.4%. In addition, infections were noted in 8.1%

With both aseptic and septic loosening bone loss can be a major problem and must be addressed if revision surgery is contemplated Options for treating bone loss include:

Revision with standard implant

When considering revision surgery it is essential to ascertain whether or not implant loosening is aseptic or septic. To this end screening blood tests including white blood count, ESR and CRP should be performed. A bone scan may also be helpful. In addition, it is my practice to perform an aspiration biopsy prior to revision surgery. A sample of fluid from the elbow joint is looked at microscopically and cultured for organism sensitivity.

If infection is present surgery is undertaken as a two-stage procedure. The first stage involves removal of the implant and bone cement together with the insertion of antibiotic beads specially prepared with added antibiotics appropriate to the sensitivity of the infectiong organism.

If infection is not present then a one-stage revision is performed.

Cobb and Morrey (1997) reported the use of Total Elbow Replacement (TER) for patients with distal humerus fractures. In this paper, 48% of the patients had a previous history of inflammatory arthropathy. Our aim was to determine the role of TER as treatment for complex distal humeral fractures in elderly patients with no previous history of inflammatory arthropathy. These patients have greater functional demands.

The complexity of the original injury was graded according to the AO Mullers’ classification of supracondylar humeral fractures. All patients were reviewed clinically and radiographically. Their daily activities and general post-operative quality of life was estimated with the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire. The Mayo elbow performance score was used for functional evaluation. Implant survivorship was assessed.

Fourteen patients (11 female and three male) underwent a Coonrad-Morrey TER via a standard posterior approach for humeral fractures. Their mean age was 73 years (range 61–84 years) at the time of injury. Ten patients had suffered a C3 injury, two a B3 and two an A3 according to the AO classification. The mean time to surgery after their injury was 8 days (range: 1–21 days). Complications: one myocardial infarct and one superficial wound infection.

Mean time to follow-up was three years and two months (range: 9–66 months). Nine (64%) reported no pain, four (29%) had mild pain with activity and one had mild pain at rest. The mean arc of extension-flexion movement was 24°–125°. Supination: mean = 90° (range: 70°–100°). Pronation: mean = 70° (range: 50°–110°). No elbow was unstable. The mean DASH score was 22.6 (range: 0.92–63.3). Zero reflects no disability, 100 reflects most severe disability. The mean Mayo elbow performance score was 90 (range: 80–100). X-rays revealed that all implants were well fixed with no evidence of loosening. One patient had severe hypertrophic bone.

Three and four part fractures of the proximal humerus can prove difficult to treat and results are generally poor. We used a Polarus Nail technique to treat seven consecutive patients who had sustained an isolated fracture to the proximal humerus. According to Neers classification, four patients had sustained a three-part fracture and three patients a four-part fracture. One patient had a fracture dislocation.

At review, six of the seven patients were assessed using the Constant and Dash scoring systems. One patient had died, but at last review had been discharged with a satisfactory result.

The average age of the patients reviewed was 62 years (range 48–79). The dominant hand was affected in 2 patients.

All six patients were followed up to fracture union and were happy with the result of treatment. All patients had mild or no pain. The average Constant score was 83 (range 59–98) and average Dash score was 131 (range 8–300)

When comparing our results to other methods of treatment already described, we found that fixation using a Polarus nail provided a satisfactory alternative method. In fact, our patients appeared to have less pain and a higher score to all elements of the Constant score.

We conclude therefore that the use of the Polarus Nail should be considered as a treatment option in this group of patients.