Rates of Infection and Pathogen Detection for Patients Undergoing Hematopoietic Stem Cell Transplantation (HSCT)

With changing immunomodulation, prophylactic and empiric therapy in HSCT patients, continual reassessment of infecting pathogens and rates of recovery are required as infections remain a leading cause of morbidity and mortality.

We performed a retrospective chart review of 86 BMT patients who underwent allogeneic HSCT (AlloSCT) or autologous HSCT (ASCT) from 7/11 to 4/14 and with 100 days of post-transplant follow-up.  Evaluation included pathogen detection and recovery from D0-30 and D31-100 (defined as one set of testing per 24h period).

There were 86 patients, 30 AlloSCT (median age 35, 20-66) and 56 ASCT (median age 55, 20-72).  Age and gender had no statistically significant effect on overall mortality.  100% of patients underwent evaluations for infection.  Of 406 total samples obtained, only 22 (5%) revealed a pathogen. 158 (119 D0-30, 39 D31-100) blood culture evaluations were obtained in 66 patients (50 D0-30, 16 D31-100) with only 10 (5 D0-30, 5 D31-100) positive evaluations in 8 patients.  Overall, 11% of ASCT and 14% of AlloSCT patients had clinically significant bacteremia.  96 (71 D0-30, 25 D31-100) urine cultures were obtained in 55 patients with 3 (0 D0-30, 3 D31-100) positive evaluations in 2 patients resulting in 0% of ASCT and 7% of AlloSCT with clinically significant bacteriuria. 152 (130 D0-30, 22 D31-100) C. diff tests were obtained in 77 patients with 9 positive evaluations in 8 patients (6 D0-30, 2 D31-100) resulting in a C. diff detection rate of 12% and 11% in ASCT and AlloSCT patients, respectively.  All-cause mortality was 14% (7% ASCT, 27% AlloSCT; p = 0.021).  Of patients with positive blood, urine, and C.difftesting, mortality was 50% (p = 0.020), 100% (p = 0.042), and 25% (0.003), respectively.  No patients died during D0-D100.  After D100, infection-related mortality was 25%.

Actionable infections require rapid detection in BMT patients.  Clinical parameters often trigger extensive evaluations but our diagnostic yield was < 15%.  Infections during D0-D100 were associated with greater all-cause mortality and 25% of post-transplant death was attributed to infection.   While current empiric therapy has reduced infection rates, improved clinical algorithms and methods of pathogen detection are needed to improve HSCT care.