The impact of surveillance and rapid reduction in immunosuppression to control BK virus-related graft injury in kidney transplantation.

We prospectively screened 609 consecutive kidney (538) and kidney-pancreas (71) transplant recipients for BK viremia over a 4-year interval using polymerase chain reaction viral load detection and protocol kidney biopsies. We found that BK viremia is common at our center: total cases 26.7%, cases during first year 21.3% (mean 4 months), and recipients with ≥ 10 000 copies/ml 12.3%. We found few predictive clinical or demographic risk factors for any BK viremia or viral loads ≥ 10,000 copies/ml, other than prior treatment of biopsy confirmed acute rejection and/or higher immunosuppressive blood levels of tacrolimus (P = 0.001) or mycophenolate mofetil (P = 0.007). Viral loads at diagnosis (<10 000 copies/ml) demonstrated little impact on graft function or survival. However, rising copy numbers demand early reductions in immunosuppressive drug doses of at least 30-50%. Viral loads >185 000 copies/ml at diagnosis were predictive of BK virus-associated nephropathy (BKVAN; OR: 113.25, 95% CI: 17.22-744.6, P < 0.001). Surveillance for BK viremia and rapid reduction of immunosuppression limited the incidence of BKVAN to 1.3%. The addition of leflunomide or ciprofloxacin to immunosuppressive dose reduction did not result in greater rates of viral clearance. These data support the role of early surveillance for BK viremia to limit the impact on transplant outcome, although the most effective schedule for screening awaits further investigation.