Content » Vol 87, Issue 2

Investigative Report

Pooling Samples: the Key to Sensitive, Specific and Cost-effective Genetic Diagnosis of Chlamydia trachomatis in Low-Resource Countries

Elena Shipitsyna1, Kira Shalepo1, Alevtina Savicheva1,2, Magnus Unemo2,3 and Marius Domeika2 

1Laboratory of Microbiology, DO Ott Research Institute of Obstetrics and Gynecology RAMS, St Petersburg, Russia, 2Department of Medical Sciences, Uppsala University, Uppsala, and 3Department of Clinical Microbiology, Örebro University Hospital, Örebro, Sweden

The aims of this study were to compare the performance characteristics and cost-effectiveness of pooling endocervical samples for screening and diagnosis of Chlamydia trachomatis, and to investigate the prevalence of C. trachomatis infection in women in Leningrad Oblast, Russia. A total of 1500 endocervical samples were tested individually and when pooled in groups of 5 and 10 samples, respectively. A previously evaluated in-house diagnostic polymerase chain reaction (PCR) assay was utilized. The sensitivity and specificity of the PCR were not affected by either pooling strategy. The estimated prevalence of genital C. trachomatis infection was 6.6%, 6.1% and 6.0% based on individually tested samples, and pools of 5 and 10, respectively. For diagnosis of individual samples, the pooling strategies resulted in cost savings of 53.3% (5 samples per pool) and 44.0% (10 samples per pool). Pooling samples for PCR detection of C. trachomatis is an accurate and cost-saving approach for diagnosis and large-scale prevalence studies in St Petersburg, Russia. Key words: Chlamydia trachomatis; diagnostics; PCR; NAAT; pooling; Russia.

(Accepted September 4, 2006.)

Acta Derm Venereol 2007; 87: 140–143.

Marius Domeika, Department of Medical Sciences, Uppsala University, Box 552, SE-751 22 Uppsala, Sweden. E-mail:

In 1999, approximately six million people in Eastern Europe were infected with genital Chlamydia trachomatis (1). In Russia, the reported incidences of chlamydial infections vary significantly between different regions. However, due to suboptimal laboratory diagnostics and incomplete case reporting and epidemiological surveillance, the reported incidences are presumably not reliable (2).

More than half of chlamydial infections in women may be asymptomatic and these individuals may not be identified without specially designed screening programs. Furthermore, improved control of genital chlamydial infections can also reduce the transmission of other sexually transmitted infections (STIs) including human immunodeficiency virus (HIV) (3).

In Russia, there are no local, regional or national screening programs for genital chlamydial infections. It is recommended to test pregnant women, infants, children, and socially unsupported persons attending outpatient gynaecological, paediatric, or skin and venereal disease departments for STIs. However, because of the lack of resources, the implementation of this decision is limited. The cheapest test systems, such as Russian produced, sub-optimally evaluated and quality assured antigen and/or nucleic acid detection assays, may be funded by the state. Consequently, the laboratory assays used for diagnosis of genital chlamydial infection in Russia have to be optimized, standardized and quality assured (2) to be utilized effectively in comprehensive and broad screening programs.

Pooling can be a cost beneficial approach for screening and diagnosis of C. trachomatis in both male and female samples (4–6). It allows estimation of population-based prevalence, which is needed to design screening programs for specific core or target groups and for individual-based prevalence, which form the basis for implementing effective intervention strategies. Pooling strategies could be highly applicable in resource-poor countries, such as the post-Soviet countries, e.g. Russia.

The objectives of the present study were: (i) to evaluate the sensitivity and specificity of an in-house polymerase chain reaction (PCR) using pooled cervical samples in comparison with testing the samples individually; (ii) to investigate the prevalence of C. trachomatis infections in the general female population of Leningrad Oblast, Russia; and (iii) to estimate the cost-effectiveness of pooling strategies.

Materials and Methods

Study population

Leningrad Oblast is a region comprising a territory of 83,900 km2 that surrounds the city of St Petersburg, Russia. Leningrad Oblast has approximately 1.6 million inhabitants, out of which approximately 387,000 are females of reproductive age.

In the present study, endocervical samples obtained from 1500 asymptomatic females attending 16 gynaecological departments in Leningrad Oblast, Russia for ordinary gynaecological check-ups were analysed at the Laboratory of Microbiology, DO Ott Research Institute of Obstetrics and Gynecology RAMS, St Petersburg, Russia. The inclusion criteria for enrolment in the study were: (i) not being pregnant; (ii) not having used antibiotics during the last 6 months; and (iii) being of reproductive age. Informed consent was obtained from all patients.

The study was approved by the Local Institutional Review Board at DO Ott Institute of Obstetrics and Gynecology and the Russian Academy of Medical Sciences (RAMS), St Petersburg, Russia.

Specimen collection and storage

Endocervical samples were collected using Dacron swabs, which were placed in empty 5-ml plastic tubes. Specimens were kept at 4ºC for up to 4 days before they were shipped to the laboratory. At the laboratory, samples were stored at 4ºC and tested within 1–3 days.

Isolation of genomic DNA

Processing of individual samples. Isolation of genomic DNA was performed according to the manufacturer′s instructions (DNA-express, Lytech, Moscow, Russia) with minor modifications. Briefly, 1 ml of 0.9% saline was added to the specimen, the specimen was vortexed and 100 µl aliquot was centrifuged at 13,000 × g for 10 min. Subsequently, the supernatant was discarded and 200 µl of lysis buffer was added. After being vortexed for 15 s, the samples were incubated at 95°C for 10 min, centrifuged at 13,000 × g for 15 s, and finally the supernatant was used as template in the PCR.

Processing of pooled samples. The pooling of samples was carried out by combining 100 µl aliquots from 5 and 10 consecutive samples to yield a total volume of 500 and 1000 µl, respectively. After centrifugation of the pooled samples at 13,000 ×g for 10 min, the subsequent steps of the procedure were performed as described above for the individual samples.

Polymerase chain reaction

A previously evaluated in-house PCR (2), which amplifies a 273-bp fragment of the C. trachomatis cryptic plasmid, was used in accordance with the manufacturer’s instructions (Lytech). In brief, 5 µl DNA template was amplified in a total reaction volume of 25 µl using a thermocycler Thercyc (DNA-technology, Moscow, Russia) and the following parameters: 94°C for 5 min, followed by 35 cycles of 94°C for 10 s, 60°C for 10 s, and 72°C for 10 s. The amplification was ended with an extension step at 72°C for 5 min. To visualize amplified products, 1.5% agarose gel electrophoresis with ethidium bromide staining was performed. DNA extracted from a C. trachomatis reference strain D/UW-3/Cx (ATCC VR-885) was included in each PCR run as a positive control, and distilled water as a negative control. To determine the size of the amplicons, Gene Ruler DNA Molecular Weight Marker (Fermentas, Kaunas, Lithuania) was used.

In order to ascertain the validity of negative results, i.e. identify the presence of PCR inhibitors, an internal control (IC) comprising a 227-bp sequence with annealing sites for the C. trachomatis primers was included in the PCR master mix. In case of inhibition, dilution of the sample was performed according to the manufacturer's instructions (Lytech). In brief, 100 µl of the processed sample was mixed with 200 µl of lysis buffer; the sample was vortexed for 15 s, incubated at 95°C for 10 min, centrifuged at 13,000 × g for 15 s, and finally the supernatant was used as template in a new PCR. For validation of positive results, all positive samples and pools were re-tested.

Estimation of population prevalence based on pooled samples

The estimation of the population prevalence was performed as described previously (7). Consequently, the following equations were used: (i) p = {1 – [1 – (s/n)]1/c} × 100%; (ii) SD = {{s/n[1 – (s/n)] (2/c)–1} /c2n}0,5 × 100%; (iii) 95% CI = p ± (2SD), where p is the estimated prevalence, SD is standard deviation, 95% CI is 95% confidence interval, s is the total number of positive pools, n is the total number of pools, and c is the number of specimens in each pool.

Cost analysis

At the Laboratory of Microbiology, DO Ott Institute of Obstetrics and Gynecology, RAMS, the cost of diagnostic C. trachomatis PCR of one sample including test kits, consumables, technician’s salary and other related costs is US$2.6. In the present study, the potential cost saving percentages (CS) associated with a reduction of the number of tests due to the pooling strategies were calculated as previously described (5), i.e. CS = n – (P + pp)/n × 100%, where P is the number of tests required to screen all the pools, pp is the number of tests required to identify the positive samples in each of the positive pools, and n is the total number of tests.


Sensitivity and specificity using pooled cervical samples

Out of the 1500 cervical samples, 99 were positive for C. trachomatis. Two individual samples were repeatedly negative in the IC; however, these samples were not inhibited after being diluted.

Out of the 300 pools (5 samples per pool), 80 were positive in the C. trachomatis PCR. Out of these 80 pools, 65 contained 1, 11 contained 2, and 4 contained 3 PCR-positive samples. Out of the 150 pools (10 samples per pool), 69 were positive in the PCRs. Out of these 69 pools, 46 contained 1, 18 contained 2, 3 contained 3, and 2 contained 4 PCR-positive samples. Furthermore, none of the pools were negative in the IC.

All positive individual samples and pools were confirmed in the re-testing. Consequently, PCR of pooled cervical samples (pools of 5 or 10) comprised 100% sensitivity in comparison with individual testing, i.e. all the pools containing positive samples were positive and 100% specific, and no false positive pools were detected.

Population prevalence

The prevalence of C. trachomatis in asymptomatic women in Leningrad Oblast, Russia determined by testing individual samples was 6.6% (99 out of 1500). The estimated prevalence when specimens were tested pooled in groups of 5 and 10 was highly similar, i.e. 6.1% (95% CI 4.5–7.7) and 6.0% (95% CI 4.3–7.7), respectively (Table I).

Cost analysis

For the estimation of the population prevalence based on pooled samples, without further testing of the positive pools, pooling endocervical samples by 5 reduced the costs by 80% (300 vs. 1500), and pooling by 10 decreased the costs by 90% (150 vs. 1500). For the diagnosis of individual cases, the pooling strategy resulted in cost savings of 53.3% and 44.0% when samples were pooled by 5 and by 10, respectively (Table I).

Table I. Characteristics of pooling endocervical samples for detection of Chlamydia trachomatis using an in-house PCR in St Petersburg, Russia


Individual samples


by 5


by 10

Individual cervical samples or pools tested (n)




Positive pools (n)



Samples or pools inhibited (n)




Samples to be tested totally (n)




Positive samples revealed (n)




Cost to detect one positivec (US$)




Estimated prevalence (95% CI)






Cost savings for population-based screening (%)



Cost savings for diagnosis of individual samples (%)



Sensitivity (%)



Specificity (%)



aSum of 300 pools and individual samples from positive pools (80 × 5).

bSum of 150 pools and individual samples from positive pools (69 × 10).

cNumber of samples tested × price per test/number of positive detected.

CI: confidence interval


Molecular diagnostic methods for genital chlamydial infections are increasingly utilized in the countries of Eastern Europe (2, 8, 9). The costs for such analyses and performance quality of the different assays used significantly vary in these countries, mainly depending on the origin of the test kits, i.e. whether they are national diagnostic products or imported kits. Genital chlamydial infections are, to a great extent, asymptomatic, especially in women, and these infections can be detected mainly by actively looking for the infected individuals in the entire population or in defined target groups. However, screening programs do not exist in most East European countries, mainly because effective test kits are too expensive. Most importantly, the chlamydial control programs that have been developed and implemented in some other countries are considered to have a significant effect on the prevalence of C. trachomatis infections (10–13).

Recently, highly sensitive and specific molecular methods for diagnosis of C. trachomatis have opened the door for large-scale low-cost screening. In the present study, pooling of cervical samples proved to have identical sensitivity and specificity compared with individual testing using an in-house PCR assay for detection of C. trachomatis, which is in concordance with several previous studies (6, 14–16).

Interestingly, two samples seemed to contain PCR inhibitory substances, i.e. they were repeatedly negative in the IC when individually tested. However, none of the pools of samples was inhibited. This has earlier been observed using urine samples (17). As previously described (18–21), pre-treatment procedures, such as dilution of the samples, storage in refrigerator/freezer, resuspending dried endocervical swabs in saline, delayed testing, etc., can decrease the number of PCR-inhibited samples.

Population-based C. trachomatis prevalence is considered to be a prerequisite for making a cost analysis to determine whether a screening program should be initiated in a certain population. Pooling urine specimens has been shown to give an accurate estimate of the population prevalence (22, 23). In the present study using endocervical specimens, the C. trachomatis prevalence determined by individual testing (6.6%) was very similar to, and clearly within 95% CI of, the estimated prevalence, i.e. 4.5–7.7 and 4.3–7.7, when pooling samples by 5 and by 10, respectively. In addition, the cost savings associated with using the pooling strategies for the estimation of the population prevalence were 73.4% and 54.0% when samples were pooled by 5 and by 10, respectively. Furthermore, pooling endocervical samples by 5 and by 10 in combination with subsequent individual re-testing for diagnosis of individual samples reduced the test costs by 53.3% and 44.0%, respectively. Given that the cost of the diagnostic in-house PCR used for C. trachomatis at the microbiology laboratory at Ott Institute, St Petersburg, Russia is US$2.6 per individual sample, pooling samples by 5 reduces the cost of the PCR assay to US$1.2 per individual sample.

Consequently, the present pooling strategy would be most useful and cost saving for diagnostics, especially at laboratories that process a large number of specimens and for large-scale studies aiming to estimate population-based prevalence of C. trachomatis infection. However, the application of this approach depends on: (i) the acceptance of the delay of the test result, i.e. the additional time needed to test individual samples from the positive pool and (ii) the prevalence of the population (4, 22, 23). Thus, the cost-effectiveness of utilizing pooling strategies is highly dependent on the prevalence of infection and the pool size used. As shown earlier (4), pooling is not cost-effective when the prevalence of disease is 20% or higher. However, when the prevalence is approximately 5%, a cost saving of more than 50% is possible. Overall, the introduction of pooling strategies into laboratory practice may require a revision of the laboratory logistics (22, 23) in order to avoid technical mistakes (22). Another possible drawback might be that a positive specimen is pooled with specimens containing inhibitors, which might require additional time to resolve.

In conclusion, the introduction of effective programs for control of genital chlamydial infections, especially in resource-poor countries, relies on the introduction of targeted screening of high-risk core groups or populations. Pooling of genital samples could help to overcome budgetary constrains in such countries, e.g. Russia, and make control programs for genital C. trachomatis infection feasible.


We are grateful for the collaboration of the staff of all the 16 gynaecological departments in Leningrad Oblast, Russia. The present study was supported by a grant from the East Europe Committee of the Swedish Health Care Community, SIDA, Sweden.


1. Global prevalence and incidence of selected curable sexually transmitted infections. Overview and estimates. Geneva: WHO, 2001: 123

2. Shalepo K, Savicheva A, Shipitsyna E, Unemo M, Domeika M. Diagnosis of Chlamydia trachomatis in Russia – in-house PCR assays may be effective but overall optimisations and quality assurance are urgent. APMIS, 2006; 114: 500–507.

3. Plummer FA, Simonsen JN, Cameron DV, Ndinya-Achola JO, Kreiss JK, Gakinya MN, et al. Cofactors in male-female sexual transmission of human immunodeficiency virus type 1. J Infect Dis 1991; 163: 233–239.

4. Peeling R, Toye B, Jessamine P, Gemmill I. Pooling of urine specimens for PCR testing: a cost saving strategy for Chlamydia trachomatis control programmes. Sex Transm Inf 1998; 74: 66–70.

5. Krepel J, Patel J, Sproston A, Hopkins F, Jang D, Mahony J, et al. The impact on accuracy and cost of ligase chain reaction testing by pooling urine specimens for the diagnosis of Chlamydia trachomatis infections. Sex Transm Dis 1999; 26: 504–507.

6. Bang D, Angelso L, Schirakov B, Westh H. Comparison of the Becton Dickinson strand displacement amplification and Cobas Amplicor Roche PCR for the detection of Chlamydia trachomatis: pooling versus individual tests. Clin Microbiol Infect 2003; 9: 1020–1023.

7. Kline RL, Brothers TA, Brookmeyer R, Zeger S, Quinn TC. Evaluation of human immunodeficiency virus seroprevalence surveys using pooled sera. J Clin Microbiol 1989, 27: 1449–1452.

8. Naaber P, Uuskula A, Naaber J, Poder A, Hjelm E, Hallén A, et al. Laboratory diagnosis of sexually transmitted infections in Estonia in 2001–2002: shortcomings with impact on diagnostic quality and surveillance. Sex Transm Dis 2005; 32: 759–764.

9. Vagoras A, Butylkina R, Jusevičiūtė V, Hallén A, Unemo M, Domeika M. Diagnosis of non-viral sexually transmitted infections in Lithuania and international recommendations. Euro Surveill 2006; 20: 11.

10. Walckiers D, Piot P, Stroobant A, van der Veken J, Declercq E. Declining trends in some sexually transmitted diseases in Belgium between 1983–1989. Genitourine Med 1991; 67: 374–377.

11. Addis DG, Vaughan ML, Ludka D, Pfister J, Davis JP. Decreased prevalence of Chlamydia trachomatis infection associated with a selective screening program in family planning clinics in Wisconsin. Sex Transm Dis 1993; 20: 28–35.

12. Crowley T, Horner PJ, Nelki J, Caul FO. Screening associated with reduced infection rates. BMJ 1994; 308: 716–717.

13. Aavitsland P. Use of laboratory testing for genital chlamydial infection in Norway. Quality Health Care 1993; 2: 91–95

14. Currie MJ, McNiven M, Yee T, Sciemer U, Bowden FJ. Pooling of clinical specimens prior to testing for Chlamydia trachomatis by PCR is accurate and cost saving. J Clin Microbiol 2004; 42: 4866–4867.

15. Gomes JP, Viegas S, Pailino A, Catry MA. Sensitivity evaluation of the Gen-Probe AMP-CT assay by pooling urine samples for the screening of Chlamydia trachomatis urogenital infection. Int J STD AIDS 2002; 13: 540–542.

16. Diamant J, Benis R, Schachter J, Moncada J, Pang F, Jha HC, et al. Pooling of Chlamydia laboratory tests to determine the prevalence of ocular Chlamydia trachomatis infection. Opthalmic Epidemiol 2001; 8: 109–117.

17. Lindan C, Mathur M, Kumta S, Jerajani H, Gogate A, Schachter J, Moncada J. Utility of pooled urine specimens for detection of Chlamydia trachomatis and Neisseria gonorrhoeae in men attending public sexually transmitted infection clinics in Mumbai, India, by PCR. J Clin Microbiol 2005; 43: 1674–1677.

18. Mahony J, Chong S, Jang D, Luinstra K, Faught M, Dalby D, et al. Urine specimens from pregnant and nonpregnant women inhibitory to amplification of Chlamydia trachomatis nucleic acid by PCR, ligase chain reaction, and transcription-mediated amplification: identification of urinary substances associated with inhibition and removal of inhibitory activity. J Clin Microbiol 1998; 36: 3122–3126.

19. Kellogg JA, Seiple JW, Keinedinst JL, Stroll ES, Cavanaugh SH. Improved PCR detection of Chlamydia trachomatis by using an altered method of specimen transport and high-quality endocervical specimens. J Clin Microbiol 1995; 33: 2765–2767.

20. Mahony JB, Luinstra KE, Sellors JW, Pickard L, Chong S, Jang D, et al. Role of confirmatory PCRs in determining performance of Chlamydia Amplicor PCR with endocervical specimens from women with a low prevalence of infection. J Clin Microbiol 1994; 32: 2490–2493.

21. Verkooyen RP, Luijendijk A, Huisman WM, Goessens WH, Kluytmans JA, van Rijsoort-Vos JH, et al. Detection of PCR inhibitors in cervical specimens by using the Amplicor Chlamydia trachomatis assay. J Clin Microbiol 1996; 34: 3072–3074.

22. Kacena KA, Quinn SB, Howell MR, Madico GE, Quinn TC, Gaydos CA. Pooling urine samples for ligase chain reaction screening for genital Chlamydia trachomatis infection in asymptomatic women. J Clin Microbiol 1998; 36: 481–485.

23. Morre S, Meijer M, Munk Ch, Kruger-Kjaer S, Winther J, Jorgensens H, Van Den Brule A. Pooling of urine specimens for detection of asymptomatic Chlamydia trachomatis infections by PCR in a low prevalence population: cost-saving strategy for epidemiological studies and screening programs. J Clin Microbiol 2000; 38: 1679–1680.