Utility of Outpatient Surveillance Blood Cultures in Hematopoietic Allograft Recipients on High-Dose Glucocorticoids for Treatment of Graft-Versus-Host-Disease

Background:  Glucocorticoids used to treat acute graft-versus-host-disease (GVHD) are thought to blunt clinical symptoms of infection.  We set out to address the value of weekly surveillance blood cultures drawn from patients receiving high-dose glucocorticoids in the outpatient department (OPD), and to characterize isolated pathogens, antibiotic use and outcomes in patients undergoing such surveillance.  Methods:  We conducted a retrospective review of blood cultures collected from a cohort of hematopoietic cell transplantation (HCT) patients enrolled in a clinical trial of GVHD (≥ grade II) therapy who were treated with high-dose glucocorticoids (defined as a prednisone dose ≥ 0.5 mg/kg/day). Surveillance blood cultures were defined as those collected weekly in the OPD while patients were receiving high-dose glucocorticoids. Cultures obtained as part of a symptom work-up (e.g. fevers, chills or rigors) or as follow-up for documented bacteremia were considered non-surveillance cultures. Clinical data were collected using center databases supplemented by medical record review.  Results:  A total of 138 adult HCT recipients were eligible for inclusion in the study.  Patients were excluded from analyses if they never had surveillance blood cultures (n=7) or were inpatient during the study period (n=4).  Of the remaining 127 subjects, 1082 outpatient cultures were obtained, for a median of 8 cultures (interquartile range [IQR]; 5-10) per patient; the majority of which were surveillance cultures (1020 [94%]).  Bacteria were isolated from 48 of 1082 cultures (4.4%), or 1 positive culture per 24.5 blood cultures drawn. Significantly fewer surveillance compared to non-surveillance cultures isolated organisms (41/1020 [4.0%] vs. 7/82 [8.5%], p=0.007). The most frequently detected pathogen was coagulase-negative Staphylococcus (28/1082 [2.7%]). Gram negative rod (GNR) pathogens were rare (10/1082 [0.9%], 3 Serratia, 2 Enterobacter and 1 Pseudomonas, E. coli, Bacteroides, Stenotrophomonas and Klebsiella), and a similar number GNRs were detected from surveillance and non-surveillance cultures (p=0.56). Antibiotics were administered to nearly all patients with positive surveillance cultures (38/41 [93%]), and one in 4 was admitted to the hospital for treatment; none needed ICU care or died from their infection.  Conclusions: Weekly outpatient surveillance blood cultures obtained from hematopoietic allograft recipients on high-dose glucocorticoids were infrequently positive, and the majority of isolates were low-risk pathogens. Although this screening approach appeared to have limited value and may lead to excess antibiotic exposure, future prospective studies are needed to confirm our findings in this high-risk population.