Factors predictive of response to plasma exchange (PE) in treatment of transplantation-associated thrombotic microangiopathy (TA-TMA) are generally poorly understood. To determine any clinical or laboratory factors predictive of response to PE in treatment of TA-TMA, we retrospectively reviewed all 11 cases of TA-TMA treated with PE at out institution between December 2001 and March 2008. Response to PE was correlated with associated clinical conditions, grade of TA-TMA, TA-TMA index and lactate dehydrogenase (LDH) level at diagnosis. Overall response to PE was 27%, with three complete responses (CRs) and eight non-responses (NRs) seen. Response to PE was significantly associated with the absence of acute GVHD at TA-TMA diagnosis, with three CR in four patients without active acute GVHD, and NR in seven patients with acute GVHD (P=0.024). There was no significant association seen between response to PE grade of TA-TMA (P=0.179), TA-TMA index (P=0.25) or LDH measurements (P=0.31). Our experience in use of therapeutic PE for management of TA-TMA suggests that PE may indeed be effective in the treatment of this disorder, depending on the clinical circumstance in which it develops. PE is potentially efficacious in the absence of active acute GVHD, and ineffective when acute GVHD is manifest.