Aim: To evaluate the short to medium term outcome of the Souter-Strathclyde prosthesis when used as a primary elbow arthroplasty in rheumatoid arthritis.

Introduction: The Souter-Strathclyde prosthesis has been evaluated in several studies. In our hospital the operation is carried out using the same strict precautions as for lower limb arthroplasty. At the onset of surgery the ulnar nerve is handled minimally. This study looked at infection and complication rates, and also at outcome.

Methods: The operations were carried out in laminar flow theatres with exhaust suits. At the start of the procedure the ulnar nerve was decompressed, but not mobilised from its bed, and held loosely with tape while being kept moist.

Cases were identified from theatre and implant records. 61 implants in 53 patients were identified. Complication and revision rates were established from case sheets. Of this group 19 patients had died leaving 40 implants in 34 patients suitable for review. Case notes were analysed. Patients were contacted by post and were asked to fill out a DASH form. 30 patients (34 implants) responded. 26 patients (30 implants) attended a review clinic.

Results: From the 61 implants there were 4 complications (one ulnar nerve dysthesia – transient, one dislocated prosthesis – open reduction, two wound infections) within the first year. Overall, 4 implants from 62 required revision, a rate of 6.5% (range 11–92 months, mean 37 months). Reasons – One loose, two periprosthetic fractures and one infection.

Mean follow up was 74 months (range 36–120). Mean DASH score at follow up was 48.7 (range 4.5–81.8).

Conclusion: The low incidence of ulnar nerve palsy, compared to published studies, may be attributable to the surgical technique. The use of laminar flow theatres and exhaust suits may account for the low infection rate. The complication rate and revision rate is comparable to previous published studies.

Patients undergoing primary knee arthroplasty at our unit routinely have two units of red cell concentrate cross-matched preoperatively. We assessed whether postoperative blood salvage and auto-transfusion reduced the need for allogenic blood transfusion as compared with standard suction drainage.

Haemoglobin and haematocrit preoperatively and postoperatively were recorded. The amount of drainage and re-transfusion was noted as well as the amount of blood saved, used and wasted.

182 patients were audited. Patients were placed in one of two groups: Group A receiving auto-transfusion of blood salvaged postoperatively (128 patients); and Group B where suction wound drainage without salvage was utilized (54 patients). Pre- and postoperative haemoglobin and haematocrit estimations were similar in both groups. Total drainage averaged 1061mI (range 175–2230 ml) for group A and 760mI (range 100–2280 ml) for group B. Auto-transfusion volumes averaged 814ml (range 0–1700m1) in group A.

Allogenic blood transfusion was required in 23% of patients in group A in contrast to 50% in group B. This resulted in an average wastage of 1.6 units in group A as opposed to 1.1 units in group B.

Postoperative auto-transfusion of salvaged blood in patients undergoing primary total knee arthroplasty reduces the need for allogenic blood transfusion requirements. It would also serve to reduce the wastage in units of whole blood allowing such patients to be routinely “grouped and retained” rather than “crossmatched”.