Ten-year follow-up on Dutch orthopaedic blood management (DATA III survey).

INTRODUCTION: Hip and knee arthroplasties are frequently complicated by the need for allogeneic blood transfusions. This survey was conducted to assess the current use of perioperative blood-saving measures and to compare it with prior results. MATERIALS AND METHODS: All departments of orthopaedic surgery at Dutch hospitals were sent a follow-up survey on perioperative blood-saving measures, and data were compared to the results of two surveys conducted 5 and 10 years earlier. RESULTS: The response rate was 94 out of 108 departments (87%). Most departments used erythropoietin prior to hip and knee replacements at the expense of preoperative autologous blood donation. The use of intraoperative autologous retransfusion in revision hip (56 vs. 54%) as well as revision knee arthroplasty (26 vs. 24%), was virtually unchanged. Postoperative autologous retransfusion is still used by the majority of departments after both primary arthroplasty and revision of hip (58/53%) and knee (65/61%). CONCLUSIONS: Currently, just as in 2007, the majority of Dutch orthopaedic departments uses erythropoietin, normothermia and postoperative autologous retransfusion with hip and knee arthroplasty. Intraoperative retransfusion is used mainly with hip revision arthroplasty. Other effective blood management modalities such as tranexamic acid have not been widely implemented.

Reduction of blood loss with the use of a new combined intra-operative and post-operative autologous blood transfusion system compared with no drainage in primary total hip replacement.

Autologous retransfusion and no-drainage are both blood-saving measures in total hip replacement (THR). A new combined intra- and post-operative autotransfusion filter system has been developed especially for primary THR, and we conducted a randomised controlled blinded study comparing this with no-drainage. A total of 204 THR patients were randomised to autologous blood transfusion (ABT) (n = 102) or no-drainage (n = 102). In the ABT group, a mean of 488 ml (sd 252) of blood was retransfused. The mean lowest post-operative haemoglobin level during the hospital stay was higher in the autotransfusion group (10.6 g/dl (7.8 to 13.9) vs 10.2 g/dl (7.5 to 13.3); p = 0.01). The mean haemoglobin levels for the ABT and no-drainage groups were not significantly different on the first day (11.3 g/dl (7.8 to 13.9) vs 11.0 g/dl (8.1 to 13.4); p = 0.07), the second day (11.1 g/dl (8.2 to 13.8) vs 10.8 g/dl (7.5 to 13.3); p = 0.09) or the third day (10.8 g/dl (8.0 to 13.0) vs 10.6 g/dl (7.5 to 14.1); p = 0.15). The mean total peri-operative net blood loss was 1464 ml (sd 505) in the ABT group and 1654 ml (sd 553) in the no-drainage group (p = 0.01). Homologous blood transfusions were needed in four patients (3.9%) in the ABT group and nine (8.8%) in the no-drainage group (p = 0.15). No statistically significant difference in adverse events was found between the groups. The use of a new intra- and post-operative autologous blood transfusion filter system results in less total blood loss and a smaller maximum decrease in haemoglobin levels than no-drainage following primary THR.

Poor accuracy of freehand cup positioning during total hip arthroplasty.

Several studies have demonstrated a correlation between the acetabular cup position and the risk of dislocation, wear and range of motion after total hip arthroplasty. The present study was designed to evaluate the accuracy of the surgeon's estimated position of the cup after freehand placement in total hip replacement. Peroperative estimated abduction and anteversion of 200 acetabular components (placed by three orthopaedic surgeons and nine residents) were compared with measured outcomes (according to Pradhan) on postoperative radiographs. Cups were placed in 49.7 degrees (SD 6.7) of abduction and 16.0 degrees (SD 8.1) of anteversion. Estimation of placement was 46.3 degrees (SD 4.3) of abduction and 14.6 degrees (SD 5.9) of anteversion. Of more interest is the fact that for the orthopaedic surgeons the mean inaccuracy of estimation was 4.1 degrees (SD 3.9) for abduction and 5.2 degrees (SD 4.5) for anteversion and for their residents this was respectively, 6.3 degrees (SD 4.6) and 5.7 degrees (SD 5.0). Significant differences were found between orthopaedic surgeons and residents for inaccuracy of estimation for abduction, not for anteversion. Body mass index, sex, (un)cemented fixation and surgical approach (anterolateral or posterolateral) were not significant factors. Based upon the inaccuracy of estimation, the group's chance on future cup placement within Lewinnek's safe zone (5-25 degrees anteversion and 30-50 degrees abduction) is 82.7 and 85.2% for anteversion and abduction separately. When both parameters are combined, the chance of accurate placement is only 70.5%. The chance of placement of the acetabular component within 5 degrees of an intended position, for both abduction and anteversion is 21.5% this percentage decreases to just 2.9% when the tolerated error is 1 degrees . There is a tendency to underestimate both abduction and anteversion. Orthopaedic surgeons are superior to their residents in estimating abduction of the acetabular component. The results of this study indicate that freehand placement of the acetabular component is not a reliable method.

Failure of the uncoated titanium ProxiLock femoral hip prosthesis.

UNLABELLED: New prostheses should be evaluated for stability and clinical performance. In a prospective randomized clinical trial, we implanted 22 titanium (Ti) and 20 hydroxyapatite-coated (HA) ProxiLock femoral hip prostheses during total hip arthroplasty in 42 patients. The patients were followed for 24 months with clinical, radiographic and radiostereometric analysis. Full weightbearing was allowed immediately postoperatively. One patient with a titanium stem was lost to followup. During the first two months, 34 of the 41 stems subsided and/or rotated towards retroversion, regardless of stem type. At the 24-month followup 35 of the 41 prostheses were either fully stabilized (16 HA and 11 Ti stems) or had clinical irrelevant migration (four HA and four Ti stems). Six Ti prostheses showed continuous migrations with maximums of 4.7 mm translation and 12.2 degrees retroversion; four of these were revised, the other two had no clinical complaints. CLINICAL RELEVANCE: The migration pattern we found indicates insufficient primary fixation of the ProxiLock stem in an immediate full weightbearing protocol. The HA coating improves the secondary stability of the prosthesis compared to the uncoated stem. Early migration is associated with an increased risk of possible future loosening and revision, and therefore we discontinued the use of this prosthesis. LEVEL OF EVIDENCE: Therapeutic Level I. See Guidelines for Authors for a complete description of levels of evidence.

An injectable calcium phosphate cement as a bone-graft substitute in the treatment of displaced lateral tibial plateau fractures.

In the treatment of displaced tibial plateau fractures, filling the resultant metaphyseal defect is essential for maintaining reduction. This study reports on 14 patients with an average age of 53 (range 34-63) years who sustained such a fracture and whose tibial defect was augmented with an injectable calcium phosphate cement under arthroscopic and fluoroscopic control. Eight fractures were treated with subcortical percutaneous screw fixation, plates were used in four cases and two fractures were treated with bone cement only. Full weight-bearing was allowed after 6-12 weeks. At the last follow-up which averaged 28 (range 18-47) months, the reduction achieved at the index operation was not altered in any of the patients except one. Most patients recorded little or no pain with an average VAS score of 1.2 (0-3.4). Ten patients had no limitation of their walking distance. Patient satisfaction was good or excellent in 12 cases. Flexion averaged 140 degrees (range 130-150 degrees ) and extension was unlimited in 13 patients. The Lysholm knee score averaged 80 and the Knee Society score 180 (range 127-195). In our opinion the injectable calcium phosphate cement used here is a promising alternative for filling metaphyseal defects in the treatment of displaced tibial plateau fractures.

Lack of adhesion molecules in testicular diffuse centroblastic and immunoblastic B cell lymphomas as a contributory factor in malignant behaviour.

Diffuse large B-cell lymphoma of the testis is a rare tumour, often with disseminated disease. According to the Kiel Classification, these lymphomas are of centroblastic or immunoblastic type, corresponding in the Working Formulation to malignant lymphoma, large cell non-cleaved and large cell immunoblastic, respectively. Adhesive cell-cell and cell-matrix interactions are generally assumed to play an important part in the metastatic process, and to find clues to the highly malignant biological behaviour of this tumour we examined expression of integrins and other adhesion molecules on the tumour cells and the presence of matrix proteins. Few adhesion molecules appeared to be expressed. CD44 was expressed in 10/12 lymphomas, CD49f/VLA-6 was positive in 5/12 cases, and CD49d/VLA-4, CD54 and CD62L were detectable in a small number (2-3) of lymphomas. All other adhesion molecules were lacking. This expression pattern is suggestive of a high metastatic potential: the tumour cells seem to be poorly attached to the extracellular matrix, to each other and to other cells (CD54-, CD11a-, CD58-). The adhesion molecules expressed, CD44, CD49f/VLA-6 and CD49d/VLA-4, have been reported to play a part in dissemination, mediating intravasation (CD49f/VLA-6) and extravasation (CD44, CD49d/VLA-4). This profile of adhesion molecules may explain, at least in part, the specific biological behaviour of these lymphomas with early and rapid dissemination.