Cytomegalovirus (CMV) infection is among the most common important viral infections in solid organ transplant (SOT) recipients. Diagnostic tests for detecting CMV replication are widely used for this group of patients, however there is no clear agreement on the cut-off levels for interpretation of clinical decisions especially when the low level of viral load is detected. In this study, CMV pp65 antigenemia test results were compared with plasma CMV-DNA levels detected by quantitative real-time polymerase chain reaction (qPCR) in samples of kidney and liver transplant recipients in the Central Laboratory of Dokuz Eylul University Hospital between 2011 and 2013, and the correlation between these two tests and viral load equivalent to antigenemia positivity were determined. In the study, pp65 antigenemia and CMV-DNA qPCR results were evaluated retrospectively. The samples from the same patients were included if the time between antigenemia and CMV-DNA qPCR tests were less than 48 hours. SPSS v15.0 was used for correlation, regression and ROC curve analysis. The results of the 217 samples collected from 100 patients (59 male, 41 female; age range: 16-71, mean age: 46 +/- 13 years), 36 liver and 64 kidney recipients were evaluated in the study. Of the patients 80% were CMV IgM negative, IgG positive; 1% was CMV IgG and IgM positive; 2% were CMV IgM and IgG negative, while for 17 patients serological results could not be reached. CMV pp65 antigenemia and CMV-DNA were both negative in 102 (47%) samples, while both were positive in 37 (17%) samples. The single sample from a case with CMV IgM and IgG positivity yielded negative results for both antigenemia and CMV-DNA tests. In 78 samples antigenemia were negative and CMV-DNA qPCR were positive, while there were no samples with antigenemia positive and qPCR negative. Mean values of antigenemia and qPCR tests were 23 positive cells/200.000 leukocytes (range: 1 to 230 positive cells) and 12.595 copies/ml (range: 180 to 106.311 copies/ml), respectively. There was a significant correlation between antigenemia and qPCR results among the samples that were positive by both assays (r=0.785). ROC curve analysis showed that CMV viral load of 205 copies/ml in plasma corresponds to >= 1 pp65 antigen positive cells per 200.000 leukocytes (sensitivity: 91.7%, specificity: 90.3%). Higher analytical sensitivity of qPCR test can be explained by the results of CMV-DNA PCR positive and antigenemia negative samples. Nonexistence of samples with antigen positive and PCR negative results supported this finding. ROC analysis showed that any sample with CMV-DNA qPCR result less than 205 copies/ml, could be accepted as pp65 antigenemia negative. This viral load value is valid only for the studied patient group and assays, therefore could be changed according to study population and tests.