| Abstract: | Pneumocystis carinii pneumonia (PCP) poses a serious risk to allogeneic bone marrow transplant (BMT) patients, who are often intolerant of trimethoprim‐sulfamethoxazole (TMP‐SMX), the traditional first‐line prophylactic agents. There are limited published data supporting the use of aerosolized pentamidine (AP) prophylaxis in the BMT population. We assessed the effectiveness of AP in BMT recipients by reviewing the experience at our center. We divided our review into four time periods from January 1990 to March 2000, during which approximately 700 BMTs were performed. The first period includes patients receiving AP treatments from January 1990 to July 1997 (baseline), the second from August 1997 to July 1998 (pre‐outbreak), the third from August 1998 to October 1999 (outbreak), and the fourth from November 1999 to March 2000 (post‐outbreak). At our center, TMP‐SMX is the first‐line agent for PCP prophylaxis, which is routinely continued for at least one year, or for the duration of enhanced immunosuppression. During the baseline period, 505 BMTs were performed and 192 patients (38%) received AP for part of their time at risk. Six patients (3%) experienced toxicities requiring discontinuation of AP. Three cases of PCP were diagnosed over 1114 patient‐months of treatment in the baseline period. During the last 42 months of the baseline period, 2/154 patients receiving AP and 2 of an estimated 293 patients receiving exclusively oral prophylaxis developed breakthrough PCP (p = 0.61). During the outbreak period, 9 of 180 patients receiving AP developed PCP compared to none in the group receiving exclusively oral prophylaxis. Either changes in our AP protocol during the pre‐outbreak period or pentamidine resistance may have led to this failure of prophylaxis. There were no further cases during the 5‐month post‐outbreak period. Our observed overall breakthrough rate was 12 cases out of 439 patients (2.7%). Our study shows that AP is an effective and well‐tolerated second‐line agent in preventing PCP post BMT and we recommend its continued use in this regard. However, it should be administered using a well‐studied protocol, and only when TMP‐SMX is not tolerated. |
