Which CMV viral load threshold should be defined as CMV infection in kidney transplant patients?

Cytomegalovirus infection (CMV) is prevalent in kidney transplant patients. Which level of CMV viral load should be accepted as the gold standard for CMV infection diagnosis is a relatively unsettled issue.

Seventy-three kidney transplant patients (mean age = 35.97 ± 14.07 years, 39 male and 34 female) entered this retrospective study. The area under the curve (AUC) of the receiver operative curve (ROC) characteristics was used to define which level of CMV viral load results in the most sensitivity and specificity for different clinical and para clinical parameters differing infected and non-infected patients. Quantitative real-time polymerase chain reaction (TaqMan method) was used for measuring CMV viral load. Written consent was obtained from all patients.

Platelets, compared with the other clinical and para-clinical parameters, had the strongest correlation with CMV viral load in kidney transplant patients (r = -.314, P = .007). There was no correlation between CMV viral load and other laboratory parameters including clinical manifestation. Choosing a threshold of more than 10,000 copies/mL of CMV viral load for defining CMV infection resulted in significance for differing in both white blood cell and platelet count between infected and non-infected patients (AUC = .68, P = .023; AUC = .70, P = .014, respectively).

Accepting a CMV viral load threshold of more than 10,000 copies/mL as CMV infection has the most sensitivity and specificity for predicting both white blood cell and platelet counts in kidney transplant patients. No CMV viral load threshold as the gold standard for CMV infection diagnosis has the discriminatory power for differing clinical and para-clinical parameters other than platelet and white blood cell count between presumably infected kidney transplant patients and those not infected.