1. ¹Immunology Laboratory, Vancouver Hospital, Vancouver, Canada
2. ²Department of Statistics, Simon Fraser University, Vancouver, Canada
3. ³Division of Nephrology, Department of Medicine, University of British Columbia, Vancouver, Canada

Correspondence: PA Keown, Immunology Laboratory, Vancouver General Hospital, 855, West 12th Avenue, Vancouver BC, V5Z 1M9, British Columbia, Canada. E-mail: keown@interchange.ubc.ca

⁴Current address: Division of Nephrology, the Second Hospital, Shandong University, Shandong, PR China

Received 19 January 2006; Revised 1 July 2006; Accepted 11 July 2006; Published online 30 August 2006.

Abstract

This prospective, blinded observational study was conducted to measure the predictive value the of flow cytometric crossmatch for biopsy-proven acute rejection, graft loss, or death following kidney transplantation. Patients were selected for renal transplantation on the basis of a conventional antihuman globulin cytotoxic T-cell crossmatch. Flow crossmatch was performed simultaneously, but the results were not disclosed to the transplant team. A total of 257 kidney transplant recipients were enrolled in the study; 78 patients experienced biopsy-proven rejection in the first post-transplant year, and 41 patients lost their graft or died during the period of follow-up (mean: 2046 days). Kaplan–Meier estimates of rejection, graft loss, or patient death did not differ between subjects with a positive or negative flow crossmatch. Cox analyses showed no influence of the flow crossmatch on the risk of biopsy-proven acute rejection (P=0.987). The sensitivity and specificity of the flow crossmatch for prediction of biopsy-proven rejection were 0.128 and 0.883, and the positive and negative post-test probabilities were 0.323 and 0.301, respectively. The magnitude of the channel shift did not influence the multivariate Cox regression model. The area under the receiver operating characteristic curve of the flow crossmatch was 0.483 (P=0.71) and 0.572 (P=0.38), respectively for the living and cadaver transplant recipients, indicating no discriminative value in this study population. Flow crossmatch appears to have no significant incremental value in predicting biopsy-proven acute rejection, graft loss, or death following kidney transplantation in patients who have a negative antihuman globulin cytotoxic T-cell crossmatch against their donor.