|
A urine-based PCR
method may be used for detection of trichomoniasis in men in
settings in which collecting urethral swabs and performing culture
are not feasible.
Validation of a
urine-based PCR-enzyme-linked immunosorbent assay for use in
clinical research settings to detect Trichomonas vaginalis in
men.
Kaydos-Daniels SC, Miller WC, Hoffman
I, Banda T, Dzinyemba W, Martinson F, Cohen MS, Hobbs MM.
Journal of Clinical
Microbiology 2003;41:318-323.
Summary:
Question
How does a PCR assay performed on urine samples compare to culture of
urine and urethral swabs for the detection of T. vaginalis
infection in men?
Design
This article describes a cross-sectional study to validate detection of T.
vaginalis in men by a urine-based PCR-enzyme-linked immunosorbent
assay (ELISA) with urine and urethral swab culture as the reference
standard.
Participants
One thousand three hundred sixty-one men attending outpatient STD and
dermatology clinics at Lilongwe Central Hospital, Lilongwe, Malawi, were
evaluated. Men enrolled through the STD clinic (n=929) presented with
urethritis (4 or more polymorphonuclear cells per high power field),
urethral discharge, dysuria, or genital ulcers; men enrolled through the
dermatology clinic (n=432) were asymptomatic for STD infection. The ages
of the participants ranged from 18 to 61 years (mean=26.1 years), 43.1%
were married, and 78% reported fewer than 2 sexual partners in the
previous 4 weeks (range=0 to 30). HIV was diagnosed in 46.7% and 27.6% of
STD and dermatology clinic subjects, respectively. One thousand two
hundred twenty-five urine specimens were tested with the T. vaginalis
PCR assay.
Description of Tests and Diagnostic
Standard
Urethral swabs were collected and immediately used to inoculate TV InPouch
culture systems (Biomed, San Jose, CA) according to the manufacturer's
protocol. After the physical examination, 20 to 30 ml of first-void urine
was collected. Ten ml was centrifuged within 4 hr of collection and the
sediment was used to inoculate a culture. Both cultures were examined
microscopically on days 2 and 5 for the presence of motile trichomonads.
Subjects with any positive culture were considered positive for T.
vaginalis infection.
For PCR analysis, 1028 urine samples were
processed with the Amplicor CT Urine Specimen Prep Kit (Roche Applied
Science, Indianapolis, IN) and 197 with the Amplicor CT/NG Urine Specimen
Prep Kit (Roche) in accordance with the manufacturer's instructions, then
frozen, and shipped in liquid nitrogen to the testing laboratory in North
Carolina. Specimens prepared from the CT Urine Specimen Prep Kit were
extracted with phenol chloroform and the DNA was precipitated. The PCR
assay was performed with primers TVK3 and TVK7 (digoxigenin labeled).
Products were detected in duplicate with the PCR DIG ELISA detection kit
(Roche Diagnostic Systems) using biotinylated probe TVK. The absorbance
cutoff value for a positive test was determined by constructing a ROC
curve.
Main Outcome Measures
The sensitivity and specificity of the PCR assay compared to culture were
calculated. Although culture is considered 100% specific, its imperfect
sensitivity would underestimate the specificity of the PCR. To compensate,
the specificity was adjusted by the method of Staquet et al. using a high
estimate of 70% and a low of 50% for the sensitivity of T. vaginalis
culture for men.
Main Results
T. vaginalis was detected by culture of urethral swab or urine
sediment in 9.0% and 3.2% of subjects enrolled from the STD and
dermatology clinics, respectively. The sensitivity and specificity of the
urine PCR compared to culture were 88.9% (95% CI, 79.2-94.4) and 88.6%
(95% CI, 86.8-90.4), respectively. The adjusted specificities were 91.0%
for the high and 94.5% for the low culture estimates. Sixty-four percent
of the specimens with ODs greater than the ELISA cutoff were negative by
culture. The performance of the T. vaginalis PCR with samples from
subjects with different clinical characteristics is shown in the table. No
significant differences were observed.
| Performance
of T. vaginalis PCR assay with specimens from subjects with
different characteristics |
| Characteristics |
No. of
subjects |
%
Senstitivity (95% CI) |
%
Specificity (95% CI) |
Adjusted
% specificty |
|
Urethritis |
| Present |
521 |
87.2 (73.6-94.7) |
89.0 (85.8-91.6) |
97.5 |
| Absent |
701 |
90.9 (74.5-97.6) |
88.5 (85.7-90.7) |
92.7 |
|
HIV |
| Positive |
496 |
84.6 (68.8-93.6) |
86.9 (83.3-89.8) |
93.5 |
| Negative |
711 |
92.7 (79.0-98.1) |
89.7 (87.1-91.8) |
95.1 |
| N.
gonorrhoeae |
| Positive |
364 |
85.0 (61.1-96.0) |
89.5 (85.7-92.5) |
94.1 |
| Negative |
858 |
90.0 (78.8-95.9) |
88.3 (85.9-90.4) |
94.7 |
| Dysuria |
| Present |
554 |
86.8 (71.1-95.1) |
89.7 (86.7-92.1) |
95.8 |
| Absent |
671 |
90.5 (76.5-96.9) |
87.8 (84.9-90.2) |
93.4 |
| Urethral
discharge |
| Present |
464 |
87.9 (70.9-96.0) |
90.3 (87.0-92.8) |
96.7 |
| Absent |
761 |
89.4 (76.1-96.0) |
87.7 (85.0-90.0) |
93.1 |
| Urine
volume |
| >30 ml |
646 |
84.2 (68.1-93.4) |
89.0 (86.2-91.3) |
93.7 |
| <30 ml |
559 |
92.7 (79.0-98.1) |
88.2 (85.1-90.8) |
95.2 |
Authors' Conclusions
The performance of the T. vaginalis
PCR-ELISA assay performed on urine compared favorably to urethral swab and
urine sediment culture for the detection of T. vaginalis in men.
The addition of a urine-based screening test for T. vaginalis would
complement routinely used similar tests for gonorrhoea and chlamydial
infections.
Source of funding:
National Institutes of Health, NIH Sexually Transmitted Disease
Cooperative Research Centers, NIH National Study of Adolescent Health:
Survey 2000, and NIH training grants; NIH Sexually Transmitted Disease
Clinical Trials Unit contract; and the Clinical Associate Physician
Program of the General Clinical Research Center, Division of Research
Health.
For correspondence:
Marcia M. Hobbs, Department of Medicine, Campus Box 7030, University of
North Carolina at Chapel Hill, Chapel Hill, NC 27599. E-mail address:
mmhobbs@med.und.edu.
|
.gif){kind=small}