BACKGROUND: Data have been published suggesting a bidirectional interaction between cytomegalovirus (CMV) infection and acute graft-versus-host disease (aGvHD) in allogeneic hematopoietic stem cell transplant (allo-HSCT) recipients. Here, we hypothesized that prospective CMV DNA monitoring in stool specimens may be useful for predicting subsequent occurrence of intestinal aGvHD (IaGvHD). METHODS: This two-center study enrolled 121 consecutive adult patients undergoing any modality of allo-HSCT. A total of 1,009 stool specimens were collected (a median of 7 specimens/patient; range, 1-18). CMV DNA monitoring in stools and plasma was performed using real-time PCR assays. RESULTS: CMV DNA was detected in stools in 20 patients (cumulative incidence, 16.9%; 95% CI, 6.3%-31.8%). Median CMV DNA level in stool specimens was 1,258 IU/0.1g (range, 210-4,087 IU/0.1 g). All these patients and their donors were CMV seropositive, and 16 of the 20 patients also had CMV DNAemia, while 4 patients had CMV DNA detected in stools without CMV DNAemia. No correlation was found between CMV DNA loads in plasma and stools (P = .40). Prior CMV DNAemia, aGvHD, or IaGvHD were not associated with presence of CMV DNA in feces. IaGvHD was present in 30 patients, in 5 of whom CMV DNA was detected in stools. Neither detection of CMV DNA in feces nor in plasma was associated with subsequent IaGvHD (OR, 0.67; 95% CI, 0.18-2.52; P = .55 and OR, 0.86; 95% CI, 0.38-1.96; P = .71, respectively). No patient in this cohort had CMV end-organ disease within the study period. CONCLUSION: Our study failed to provide evidence pointing to a reciprocal interaction between GI CMV infection and IaGvHD. CMV DNA monitoring in stools seems of no value to anticipate occurrence of IaGvHD.
BACKGROUND: Data have been published suggesting a bidirectional interaction between cytomegalovirus (CMV) infection and acute graft-versus-host disease (aGvHD) in allogeneic hematopoietic stem cell transplant (allo-HSCT) recipients. Here, we hypothesized that prospective CMV DNA monitoring in stool specimens may be useful for predicting subsequent occurrence of intestinal aGvHD (IaGvHD). METHODS: This two-center study enrolled 121 consecutive adult patients undergoing any modality of allo-HSCT. A total of 1,009 stool specimens were collected (a median of 7 specimens/patient; range, 1-18). CMV DNA monitoring in stools and plasma was performed using real-time PCR assays. RESULTS: CMV DNA was detected in stools in 20 patients (cumulative incidence, 16.9%; 95% CI, 6.3%-31.8%). Median CMV DNA level in stool specimens was 1,258 IU/0.1g (range, 210-4,087 IU/0.1 g). All these patients and their donors were CMV seropositive, and 16 of the 20 patients also had CMV DNAemia, while 4 patients had CMV DNA detected in stools without CMV DNAemia. No correlation was found between CMV DNA loads in plasma and stools (P = .40). Prior CMV DNAemia, aGvHD, or IaGvHD were not associated with presence of CMV DNA in feces. IaGvHD was present in 30 patients, in 5 of whom CMV DNA was detected in stools. Neither detection of CMV DNA in feces nor in plasma was associated with subsequent IaGvHD (OR, 0.67; 95% CI, 0.18-2.52; P = .55 and OR, 0.86; 95% CI, 0.38-1.96; P = .71, respectively). No patient in this cohort had CMV end-organ disease within the study period. CONCLUSION: Our study failed to provide evidence pointing to a reciprocal interaction between GI CMV infection and IaGvHD. CMV DNA monitoring in stools seems of no value to anticipate occurrence of IaGvHD.