Performance Characteristics of the Becton Dickinson ProbeTec System for Direct Detection of Chlamydia trachomatis and Neisseria gonorrhoeae in Male and Female Urine Specimens in Comparison With the Roche Cobas System Open Access

Chlamydia trachomatis and Neisseria gonorrhoeae are the 2 leading bacterial sexually transmitted infections worldwide. Genital infection with Chlamydia is the most common bacterial sexually transmitted disease in the United States1 and is responsible for about 40% of nongonococcal urethritis in men.2 In women, genital chlamydial infections are a major cause of pelvic inflammatory disease, which is an important cause of chronic pelvis pain, ectopic pregnancy, and infertility. Perinatal transmission of C trachomatis to infants can cause neonatal conjunctivitis and pneumonia. In 1994, the estimated cost of untreated chlamydial infections and their complications exceeded $2 billion in the United States.3 Screening and treatment of C trachomatis in asymptomatic women reduced the incidence of pelvic inflammatory disease by more than 50%,4 thus routine screening of women at moderate risk proved cost-effective.5 Gonorrhea is an important cause of urethritis in men and cervicitis in women. Approximately 20% to 40% of pelvic inflammatory disease and 14% of tubal infertility can be attributed to gonococcal infections. The estimated cost associated with female gonococcal infections to the Canadian health care system is in excess of $43 million (Canadian) annually; 82% of this cost relates to the diagnosis and treatment of pelvic inflammatory disease, ectopic pregnancy, and tubal infertility.6 Despite the overall decline in the incidence of gonorrhea in the United States, the prevalence of male rectal gonorrhea is increasing among men who have sex with men.7 

The Becton Dickinson BDProbTec ET system is a new semiautomated system using strand displacement amplification technology for the simultaneous amplification and detection of Chlamydia trachomatis and Neisseria gonorrhoeae.8 Being an amplification assay, it has the potential to increase the sensitivity and specificity of detection, especially in the case of C trachomatis. The automated system has been designed for higher throughput testing, reduced “hands-on” time, and more rapid time to results. The Provincial Laboratory (Regina, Saskatchewan, Canada) is a central public health laboratory that performs about 3000 chlamydia detection tests a month. In this study, we compared the performance characteristics of the BDProbTec ET system with those of the Roche Amplicor Cobas system for the detection of C trachomatis and N gonorrhoeae. Separately, we assessed the throughput and ease of use of the BDProbTec ET system based on our daily workload of chlamydia specimens.

Eight hundred twenty-five male and 399 female consecutive first-void urine specimens were collected from 3 sexually transmitted disease clinics and other family physicians' offices located in Saskatchewan. All urine specimens were collected and transported to the Provincial Laboratory overnight and were processed within 24 to 48 hours after collection.

The urine specimens were divided into 2 portions; 1 portion was processed for DNA extraction according to Roche Amplicor package insert instructions. The other portion, tested by the BDProbTec ET system, was left at room temperature overnight for next-day processing after the BDProbTec ET urine processing pouch was added to the urine. The remainder of the Roche Amplicor sample was saved to be used for in-house polymerase chain reaction (PCR) studies when discrepant results occurred.

The BDProbeTec ET C trachomatis and N gonorrhoeae amplified DNA assay8 was used with the BDProbTec ET system (BD Biosciences, Sparks, Md). The manufacturer's protocol was strictly followed. Briefly, a urine processing pouch was added to at least 20 mL of first-voided urine and left at room temperature for a minimum of 2 hours before proceeding. In this study, we left the specimens overnight for next-day processing. A 4-mL aliquot of the treated urine was then transferred to a sample tube and centrifuged at 2000g for 30 minutes at room temperature. The supernatant was then decanted, and the pellet was resuspended in 20 mL of diluent and then vortexed for 60 seconds. Positive and negative controls were processed according to the manufacturer's instructions. The tubes were then incubated at 114°C for 30 minutes in a lysing heater and allowed to cool at least 15 minutes at room temperature. Priming and amplification microwells for chlamydia, gonorrhea, and the amplification control were placed in appropriate plates. One hundred fifty microliters of lysed specimen and controls was then transferred into each of the corresponding columns of the priming microwells and incubated at room temperature for a minimum of 20 minutes. After incubation, the priming microwell plate was placed into the priming heater (72.5°C). The amplification microwell plate was placed into the warming heater (54°C), and both were incubated for 10 minutes. At the end of this incubation period, 100 μL from each column of the priming microwell plate was transferred to the corresponding column of the amplification microwell plate. The amplification microwell plate was then sealed and immediately transferred to the instrument for testing.

The amplification control microwell was included for each patient sample. The amplification control microwells contained nucleic acid target that is amplified in the presence of the sample matrix. The microwells are designed to identify samples that contain amplification inhibitors. A predetermined cutoff value is used to indicate presence or absence of amplification inhibitor.

Roche Amplicor C trachomatis and N gonorrhoeae PCR (Roche Diagnostic Systems, Mississauga, Ontario, Canada) was performed according to the manufacturer's instructions. An internal DNA control provided with the Roche assay was added to the master mix to monitor inhibition of the PCR assay and to ensure the validity of negative results in each run.

Aliquots of urine identified to be positive by either commercial method were tested by an in-house PCR method. Aliquots were typically extracted the same day as the commercial assay was performed. Urine specimens positive for either C trachomatis or N gonorrhoeae were processed for in-house PCR as follows: specimens were vortexed briefly and 1-mL aliquots were centrifuged at 20 000 rcf for 10 minutes at 4°C. The supernatant was discarded, and the pellet was resuspended in 300 μL of 5% chelex 100 (Bio-Rad Laboratories, Mississauga, Ontario). After a quick vortex, the sample was incubated at 56°C with occasional mixing for 15 minutes, followed by a heat shock at 100°C for 10 minutes. Vortexed and centrifuged samples were stored at −20°C until PCR was performed. Any samples determined to be inhibitory with the in-house PCR internal control were silica purified, based on the purification procedure of Boom et al.9 

In-house PCR was performed on the ABI 7700 (Perkin-Elmer, PE Biosystems, Foster City, Calif), which effected both thermal cycling and detection simultaneously. Three primer pairs and 3 probes were included in 2 multiplex PCRs, one for detecting the presence of C trachomatis and the other for N gonorrhoeae. Both PCRs were performed on a single extract and included an internal control to detect the presence of amplification inhibitors. Primers and probe for C trachomatis were designed from the genetically conserved cryptic plasmid, pCHL1.10 ,Neisseria gonorrhoeae primers and probe were based on the detection of cppB gene.11 

One hundred fifty urine specimens saved during the 3-week period were used for the throughput study. The specimens were stored at room temperature and tested to evaluate if a single BDProbTec ET instrument was capable of processing 150 specimens for the detection of both chlamydia and gonorrhea in an 8-hour shift using the amplification control. This throughput study was repeated once to ensure its reproducibility.

Sensitivity, specificity, positive predicative value, negative predicative value, and the inhibition rate were calculated after resolution of discrepant results for both assays for male and female, as well as for each organism. Specimens with inhibitors were not included in the calculation.

The actual number breakdown for chlamydia and gonorrheae for both assays and each sex is shown in Table 1. Of a total of 825 male and 399 female urine samples (after resolution), the BDProbTec ET system was able to detect 162 chlamydia-positive specimens, while the Roche Amplicor system was able to detect 163. The BDProbTec ET system detected 32 N gonorrhoeae–positive specimens, while the Roche Amplicor system detected 31. Although the sensitivity of both systems was essentially equivalent, the Roche Amplicor system was shown to be somewhat less specific (30 false positives for Amplicor vs 11 for the BDProbTec ET system). The Amplicor false-positive results were particularly pronounced for N gonorrhoeae, for which the positive predictive value was show to be only 69%. Of the total 1224 specimens tested, 40 results were shown to be inhibitory in the BDProbTec ET assay, as compared to 25 for the Roche Amplicor assay. All of the specimens shown to be initially inhibitory to either system were demonstrated to be negative after further testing.

Table 2 shows the sensitivity, specificity, and positive and negative predicative values for male and female specimens and for both the Roche Amplicor and the BDProbTec ET system.

One hundred fifty urine specimens saved during the 3-week study period were used to evaluate the throughput of the BDProbTec ET system to handle our daily specimen load. These specimens included 9 chlamydia-positive and 3 gonorrhea-positive specimens and 1 specimen showing evidence of both chlamydia and gonorrhea coinfection in the original runs. These specimens were stored at room temperature during the 3-week interval. Using a single BDProbTec ET instrument, our laboratory was able to test all 150 specimens for chlamydia and gonorrhea, including the amplification control, within a normal 8-hour shift on 2 consecutive days. Specimens that were positive in the original run were still positive 3 weeks later, even when stored at room temperature. A few of the negative specimens became indeterminate, but retested negative when run the next day.

Laboratory diagnoses of sexually transmitted diseases have evolved rapidly during the last 20 years. This is especially true in the area of C trachomatis detection. The evolution from technically demanding cell culture for C trachomatis to enzyme immunoassay or direct immunofluorescent assay was a big step in public health for the detection of this major sexually transmitted pathogen. Later, confirmation assay for enzyme immunoassay assay greatly improved the sensitivity and specificity of the assay, especially for those specimens falling below the cutoff or in the gray zone area,12 as recommended by Centers for Disease Control and Prevention (Atlanta, Ga).13 The development of nucleic acid amplification methods, such as PCR, ligase chain reaction, and transcription-mediated amplification, has improved the sensitivity and specificity of detection greatly.14,15 In a similar tract, an enzyme immunoassay for the detection of N gonorrhoeae has been marketed by Abbott Laboratories for more than 10 years and is now available in nucleic acid amplification assays together with C trachomatis as a panel.16 Both bacteria can be detected from a single swab or urine specimen with increased sensitivity.17 The sensitivity of most of these amplification methods is close to or exceeds 90%.15 Using these nucleic acid amplification assays to screen a low-prevalence population is still a cost-effective intervention strategy.18 

The disadvantages of nucleic acid amplification methods are that they are labor intensive and do not have high-throughput capacity. The Roche Cobas Amplicor assay is a semiautomated system, but can process only about 46 specimens in an 8-hour shift with 1 instrument. With the recent introduction of the BDProbTec ET system, this limitation has now been overcome. The BDProbTec ET system combines the sensitivity and specificity of nucleic acid amplification technology with a throughput capable of meeting the workflow demands of a high-volume laboratory.

In this study, we used urine specimens from both male and female patients for the following reasons: our feedback from family physicians and nurses from our sexually transmitted disease clinics is that most of the patients preferred to give a urine specimen if given a choice; secondly, urine collection is much less technically demanding in terms of specimen collection; and finally, a previous study had shown that urine specimens tested with PCR assay produced a much superior sensitivity rating than did enzyme immunoassay on combined cervical and urethral swabs.19 Urine probably will be the specimen of choice in the future, as it is easy to collect and can be used for detection of multiple pathogens.

The BDProbTec ET system achieved an overall sensitivity of 95.3% for chlamydia and 100% for gonorrhea, which is similar to the performance of the Roche Amplicor sensitivities of 95.9% and 96.7%, respectively. The overall specificity for the BDProbTec ET system was 99.3% for chlamydia and 99.7% for gonorrhea, whereas the Roche Amplicor specificities of 98.3% and 98.9%, respectively, were slightly lower but not statistically significant. The Roche gonorrhea test produced 13 false-positive results, yielding a positive predictive value of only 69%. The Roche test has been shown to cross-react with nongonococcal Neisseria species,20 resulting in a 15% false-positivity rate in the Australian population tested. Data presented on the BDProbTec ET system at the Fifteenth Annual Clinical Virology Symposium gave sensitivity and specificity values similar to those reported in this study.21,22 Given the performance of both assays in our study, either system would be suitable for the detection of these 2 pathogens in moderate- or high-risk populations. Both assays have an amplification control that will indicate inhibition to prevent reporting false negatives. In our study, the inhibition rates were 3.3% for the BDProbTec ET system and 2% for Amplicor. The BDProbTec ET system appears to have more inhibition with male urine specimens, whereas the Amplicor system appears to have higher inhibition with female urine specimens. This is similar to our previous findings.19,23 However, none of the inhibited urine specimens in this study tested positive for either pathogen.

While the clinical performance of the 2 systems was comparable, a marked difference in throughput performance was shown. In a single 8-hour shift, the maximum throughput of 1 Cobas Amplicor instrument for the detection of chlamydia and gonorrhea was shown to be 46 specimens. On the other hand, using 1 BDProbTec ET system, we were able to process 150 specimens for chlamydia and gonorrhea using the amplification control, within the same 8-hour shift. This has a great impact in a high-volume laboratory like ours that processes between 120 and 180 specimens a day. We were able to process a full day's work within an 8-hour shift with only 1 instrument and report the results at the end of the shift. The high throughput and performance of the BDProbTec ET instrument allow us to achieve increased sensitivity and specificity with greater efficiency than previously possible. We view this as a significant step forward in the automation of nucleic acid amplification assays in the clinical laboratory and as a positive step toward the control of sexually transmitted chlamydia and gonorrhea infections.

The authors thank Janette Romanuik for her excellent secretarial work to put the manuscript together and Becton Dickinson for providing the reagents for the study.