Comparison of quantitative PCR after automatic DNA extraction and pp65 antigen detection for diagnosis of CMV infection in stem cell transplant recipients

Abstract number: P1145

Navarro J., Cardeñoso L., Azcona J., Arenal N., Lomas E., Sanchez P., Camara R.

Objectives: Cytomegalovirus (CMV) infection has been recognised as one of the major causes of morbidity in stem cell and solid organ transplant recipients. Efficacy of treatment is closely related to accurate early diagnostic laboratory tests. Quantitation of CMV DNA by conventional polymerase chain reaction (PCR) and pp65 antigen detection in peripheral blood specimens are performed in our department. The aim of this study is the comparison and correlation between the two methods in the diagnosis of CMV infection in our stem cell transplant recipients.

Methods: We retrospectively reviewed peripheral blood samples from stem cell transplant recipients obtained between January 1st 2005 and September 10th 2007. Quantitative PCR was performed using COBAS Ampliprep TNAI kit® (Roche) after DNA automatic extraction, and pp65 antigen was detected by CINApool Anti HCMH ppUL83® (Argene) kit. A positive sample was defined by a pp65 antigenaemia geqslant R: gt-or-equal, slanted2 cel/4×105 cel and/or PCR geqslant R: gt-or-equal, slanted 600 copies/mL. Episodes were defined as serial positive samples until continued negativity for at least 15 days. Samples analysis and episodes analysis were performed independently. Qualitative variables are expressed as proportions. Concordance between both methods was determined by testing Kappa index and Spearman coefficient. P-values lower than 0.05 were considered statistically significant.

Results: We studied 439 samples, identifying 84 episodes from 53 different patients. Samples analysis: 10.3% of pp65 determinations were non-evaluable and 1.1% of PCR were invalidated. 212 (48.3%) positive samples were detected (18.4% were detected only by pp65, 33.5% only by PCR and 48.1% by both methods). Correlation between pp65 and PCR was medium (Spearman coefficient=0.37, p < 0.001). Assuming qualitative interpretation of pp65 and PCR, medium concordance was also found (Kappa index=0.42, p < 0.001). Episodes analysis: 19.0% were detected only by pp65, 16.7% only by PCR and 64.3% by both methods. 36.9% episodes were detected at the same time by PCR and pp65. 8.3% were early detected by pp65 and 19.0% by PCR, and in both cases with a median of 7 days.

Conclusions: Medium correlation has been found between pp65 antigen detection and conventional PCR. An important proportion of samples and episodes have been detected only by one test, so both methods are complementary, and PCR cannot replace antigenaemia.

Session Details

Date: 19/04/2008
Time: 00:00-00:00
Session name: 18th European Congress of Clinical Microbiology and Infectious Diseases
Location: Barcelona, Spain
Presentation type:
Back to top