Since the landmark studies of Patel and Terasaki in the late 1960s, pretransplant cross-matching has been performed by HLA laboratories on a 24-hr/7-day basis. In fact, regulating agencies such as the American Society for Histocompatibility and Immunogenetics and the United Network for Organ Sharing have mandated prospective crossmatching for selected solid organ transplants. However, two recent publications (Transplantation 1998; 66: 1833; and Transplantation 1998; 66: 1835) have suggested a change to this approach. Specifically, those authors advocate the transplantation of non-sensitized individuals without a final prospective cross-match as a means to reduce cold ischemia time and the incidence of delayed graft function. Such considerations were predicated upon results generated by cytotoxicity-based antibody screening. We and others, however, have reported that a flow cytometric-based assay is a more sensitive method to detect alloantibodies than cytotoxicity. Furthermore, an increasing number of reports document that graft survival is improved among patients whose final flow cytometric crossmatches were negative compared to patients with positive flow cytometric crossmatches. Although we agree that it is reasonable to transplant truly non-sensitized patients without a prospective final crossmatch, our data demonstrate that a large number of patients deemed non-sensitized by cytotoxicity-based antibody assessment are, in fact, sensitized.Panel-reactive antibody (PRA) testing was performed with 703 sera from 527 patients. The patient population consisted of individuals awaiting either renal or cardiac transplantation. PRA evaluations were performed using lymphocyte cytotoxicity (antiglobulin-enhanced, complement-dependent cytotoxicity [AHG-CDC]) or assays (enzyme-linked immunosorbent assay [ELISA]; flow cytometry) in which solubilized HLA molecules were affixed to solid phase matrices.PRA activity in 264 sera from 88 patients was evaluated by AHG-CDC, ELISA, and flow cytometry. Results among the three methods were concordant for 83% of these sera. Discordant results occurred with 32 samples and demonstrated a distinct hierarchy in the sensitivity of the three techniques to detect alloantibodies. None of the 32 sera were positive by AHG-CDC, 20/32 were positive by ELISA, and 32/32 were positive by flow cytometry. Subsequent studies revealed that, among 527 patients, 302 (57%) exhibited 0% PRA by AHG-CDC. Of these 302 AHG-CDC-negative patients, 76 (25%) had class I or class II antibodies detectable using a flow cytometric approach. Within the AHG-CDC-negative/flow cytometric-positive patients, PRA values exhibited a wide range (6-99%) for both class I and class II antibodies. The average PRA was 27% and 38% for class I and II, respectively. Retrospective flow cytometric crossmatches performed for 30 recipients of cardiac allografts whose AHG-CDC PRA were 0% revealed that 11/30 crossmatches were positive.The concept of transplanting non-sensitized patients without a prospective final crossmatch is appealing and, if bona fide, clearly makes sense. However, our data demonstrate that how a patient is deemed non-sensitized is critical. The difference between AHG- and flow cytometric-based PRA testing is significant and can result in transplantation of alloimmunized patients considered to be non-sensitized. Therefore, we recommend that, if a transplant center chooses to forego a prospective final crossmatch, the decision to do so should be based on methods more sensitive than AHG-CDC.