|
.gif)
|
|
Self-obtained low
vaginal swabs performed as well as conventional clinician-obtained
samples for identifying C. trachomatis, N. gonorrhoeae,
and T. vaginalis infection by PCR in a remote population.
The diagnosis of chlamydia, gonorrhoea,
and trichomonas infections by self obtained low vaginal swabs, in remote
northern Australian clinical practice.
Garrow SC, Smith DW, Harnett GB.
Sex Transm Infect.
2002;78:278-281
Summary:
Question
What is the performance and utility of self-obtained low vaginal swabs (SOLVS)
for the diagnosis of C. trachomatis, N. gonorrhoeae, and T.
vaginalis infection by PCR in a variety of clinical practice settings
in remote northwestern Australia?
Design
This paper describes a cross-sectional study of specimen collection
techniques for diagnosis by PCR of sexually transmitted infections in
women undergoing gynecological examinations in remote settings, performed
by a variety of practitioner types.
Participants
Three hundred forty-nine women from remote towns and communities in
Western Australia having a vaginal examination for any reason, including
well women, antenatal screening, and sexual health examinations were
included. The average age of participants was 28.8 years; 74% were
Australian Aboriginal women. Nurse practitioners attended 61% of the
patients in remote areas, while medical officers provided 63% of the town
patients' specimens.
Description of Tests and Diagnostic
Standard
Each woman provided a first-void urine (FVU) and a SOLVS, in that order,
followed immediately by a clinical examination in which the clinician
collected an endocervical swab (ECS), and a low vaginal swab (LVS). All
samples were transported at ambient temperature to one of three regional
laboratories and then by air transport to a central laboratory for
processing the next day. An in-house, nested, multiplex PCR was performed
for detection of N. gonorrhoeae and C. trachomatis using
primers to the N. gonorrhoeae cryptic plasmid and the C.
trachomatis plasmid. Positive samples were confirmed for N.
gonorrhoeae using two PCR assays with primers for two other target
sequences and for C. trachomatis using a PCR assay with primers to
different sequences in the plasmid. A nested PCR directed at repetitive
DNA sequences was used for T. vaginalis detection. Samples were
reported positive only if they were positive by all the PCR assays for
that organism. A positive result from any collection site was taken to
indicate that the patient was infected.
Main Outcome Measures
The sensitivity, specificity, positive predictive value (PPV), and
negative predictive value (NPV) of each collection technique for the
detection of C. trachomatis or N. gonorrhoeae, as determined
by a positive PCR test in the FVU or ECS, and for T. vaginalis, as
determined by a positive PCR test in the FVU, ECS or LVS were calculated.
Main Results
The overall prevalence of infection in the study population was 9.2%,
7.6%, and 16.1% for C. trachomatis, N. gonorrhoeae, and T.
vaginalis, respectively. Complete sample sets including SOLVS, ECS,
and FVU for C. trachomatis and N. gonorrhoeae detection and
SOLVS, ECS, FVU, and LVS for T. vaginalis detection were collected
simultaneously from 303 and 286 patients, respectively. The performance of
each collection technique by organism is shown in the table. A combination
of SOLV and FVU detected 96% of the C. trachomatis cases, 100% of the N.
gonorrhoeae cases, and 96% of the T. vaginalis cases.
| Diagnostic
performance by organism of each collection technique |
| Organism |
Specimen
collection |
Parameter
(%) |
| Sensitivity |
Specificity |
PPV |
NPV |
| C.
trachomatis |
SOLVS |
89 |
100 |
100 |
99 |
| FVU |
79 |
100 |
100 |
98 |
| ECS |
79 |
100 |
100 |
98 |
| N.
gonorrhoeae |
SOLVS |
96 |
99 |
92 |
100 |
| FVU |
83 |
100 |
100 |
99 |
| ECS |
91 |
100 |
100 |
99 |
| T.
vaginalis |
SOLVS |
93 |
99 |
93 |
99 |
| FVU |
89 |
100 |
100 |
98 |
| ECS |
87 |
100 |
100 |
98 |
| LVS |
93 |
100 |
100 |
99 |
Authors' Conclusions
A self-obtained low vaginal swab is at least
as good as a conventional clinician-obtained sample for identifying C.
trachomatis, N. gonorrhoeae, and T. vaginalis infections
by PCR in a remote population.
Source of funding: None given
For correspondence: Stuart Garrow,
North Peterborough Primary Care Trust, St. John's, Thorpe Road,
Peterborough PE3 6JG, UK. E-mail address: stuartgarrow@yahoo.co.uk
|
