BACKGROUND: Complications of the biliary anastomosis are the principal cause of clinically serious bacterial sepsis in liver transplant recipients. Reported series suggest an association of bacterial and fungal infection with cytomegalovirus (CMV) infection, although the mechanism of this association is unclear. METHODS: We examined the association of serious bacterial sepsis with CMV replication in a cohort of 106 consecutive liver transplant recipients. Sequentially collected buffy coats were examined with a polymerase chain reaction (PCR) assay that has been shown to have good predictive value for the development of CMV infection. For selected patients, CMV-specific IgM response and serum tumor necrosis factor-alpha (TNF-alpha) were also measured. RESULTS: Ten of 13 patients with serious bacterial sepsis developed buffy coat PCR positivity, compared with 26 of 93 patients without bacterial sepsis (chi-square, P<0.001). Ten of 10 septic recipients with a seropositive liver donor developed PCR positivity. For 9 of 10 recipients, bacterial sepsis developed before PCR positivity. Bacterial sepsis was associated with high serum levels of TNF-alpha. Immune response to CMV (reflected by the appearance CMV-specific IgM) was apparently affected by bacterial sepsis, and IgM response was not observed for the three septic patients who died during the study period. CONCLUSIONS: We conclude that CMV replication is encouraged by serious bacterial sepsis. Replication may be promoted by high antecedent levels of TNF-alpha, and/or by poor immune response to CMV in the context of serious bacterial infection.
BACKGROUND: Complications of the biliary anastomosis are the principal cause of clinically serious bacterial sepsis in liver transplant recipients. Reported series suggest an association of bacterial and fungal infection with cytomegalovirus (CMV) infection, although the mechanism of this association is unclear. METHODS: We examined the association of serious bacterial sepsis with CMV replication in a cohort of 106 consecutive liver transplant recipients. Sequentially collected buffy coats were examined with a polymerase chain reaction (PCR) assay that has been shown to have good predictive value for the development of CMV infection. For selected patients, CMV-specific IgM response and serum tumor necrosis factor-alpha (TNF-alpha) were also measured. RESULTS: Ten of 13 patients with serious bacterial sepsis developed buffy coat PCR positivity, compared with 26 of 93 patients without bacterial sepsis (chi-square, P<0.001). Ten of 10 septic recipients with a seropositive liver donor developed PCR positivity. For 9 of 10 recipients, bacterial sepsis developed before PCR positivity. Bacterial sepsis was associated with high serum levels of TNF-alpha. Immune response to CMV (reflected by the appearance CMV-specific IgM) was apparently affected by bacterial sepsis, and IgM response was not observed for the three septicpatients who died during the study period. CONCLUSIONS: We conclude that CMV replication is encouraged by serious bacterial sepsis. Replication may be promoted by high antecedent levels of TNF-alpha, and/or by poor immune response to CMV in the context of serious bacterial infection.