|
.gif)
|
|
Commercial LCR and
PCR assays were significantly more sensitive than a nonamplified DNA
probe test for the detection of C. trachomatis infection in
women.
Head-to-head
multicenter comparison of DNA probe and nucleic acid amplification
tests for Chlamydia trachomatis infection in women performed
with an improved reference standard.
Black CM, Marrazzo J, Johnson RE, Hook
EW III, Jones RB, Green TA, Schachter J, Stamm WE, Bolan G, St Louis
ME, Martin DH.
Journal of Clinical
Microbiology 2002;40:3757-3763.
Summary:
Question
What are the performances of commercially available PCR, LCR, and DNA
probe tests for the detection of C. trachomatis in cervical swabs
and urine samples (PCR and LCR only) compared to detection by culture of
cervical and urethral swabs?
Design
This study describes a head-to-head multicenter evaluation of commercial
cervical swab and urine PCR and LCR tests and a cervical swab DNA probe
test in a large and geographically diverse population of symptomatic and
asymptomatic women attending sexually transmitted disease and family
planning clinics.
Participants
Participants consisted of female patients (n = 3551) who presented to an
STD clinic in each of five participating centers (Birmingham, AL,
Indianapolis, IN, New Orleans, LA, San Francisco, CA, and Seattle, WA) and
a family planning clinic in New Orleans, with or without symptoms of an
STD and with an indication for a pelvic examination. Pregnant women and
women who had taken antibiotics in the previous 30 days were excluded.
Enrolled patients were 63% African-American, 25% Caucasian, 4%
Asian-Pacific Islander, and 7% other or unknown. The median age was 24
years.
Description of Tests and Diagnostic
Standard
Randomly ordered endocervical swabs were collected for C. trachomatis
detection by the Amplicor microwell plate PCR test (Roche Diagnostic
Systems, Branchburg, NJ), the LCx LCR test (Abbott Laboratories, Abbott
Park, IL), and the PACE 2 DNA probe test (Gen-Probe, Inc., San Diego, CA)
all performed according to the manufacturers' protocols. A first-catch
urine sample was also collected for analysis by PCR and LCR. A urethral
swab and an endocervical cytobrush specimen were collected and used for C.
trachomatis culture on McCoy cells. After incubation for up to 72 h,
cell cultures were stained with a major outer membrane protein-specific
fluorescein-conjugated antibody reagent.
Main Outcome Measures
The sensitivity and specificity for each test were calculated using the
following standards to classify patients as infected: 1) three different
cervical "specimen" reference standards (culture, LCR, or PCR
tests) and 2) three different two-specimen "patient" reference
standards (cervical or urethral culture, cervical or urine LCR, or
cervical or urine PCR tests).
Main Results
Overall, the percentage of subjects positive by cervical tests was 10.1%
by culture, 8.4% by PACE 2, 12.0% by LCR, and 10.8% by PCR. Positive tests
by LCR or PCR with cervical specimens exceeded those with urine specimens
by 0.6 and 0.2 percentage points, respectively. A comparison of the
sensitivities and specificities of the DNA probe, PCR, and LCR tests
determined by using the different, independent specimen and patient
reference standards are shown in the table. The sensitivities of the
cervical NAATs significantly exceeded those of PACE 2. The sensitivities
of the cervical and urine LCR and PCR tests were not significantly
different when a two specimen, patient reference standard was used.
| PACE 2, LCR, and PCR
sensitivities and specificities for detection of C. trachomatis
infection determined by comparison to different reference standards |
| Test
performance (%) |
Reference
standard by specimen type and test |
| Cervical
swab |
Cervical
and urethral swab |
Cervical
swab and urine |
| Culture |
LCR |
PCR |
Culture |
LCR |
PCR |
| Sensitivity
of: |
|
|
|
|
|
|
| PACE
2 |
78.1 |
68.4 |
74.9 |
71.6 |
60.8 |
61.9 |
| LCR
cervix |
96.9 |
|
95.8 |
91.4 |
|
85.5 |
| LCR
urine |
82.6 |
|
83.6 |
83.4 |
|
80.8 |
| PCR
cervix |
89.9 |
85.7 |
|
84.0 |
75.8 |
|
| PCR
urine |
79.2 |
75.7 |
|
79.5 |
74.9 |
|
| Specificity
of: |
|
|
|
|
|
|
| PACE
2 |
99.3 |
99.6 |
99.3 |
99.5 |
99.7 |
99.6 |
| LCR
cervix |
97.5 |
|
98.1 |
98.1 |
|
99.0 |
| LCR
urine |
96.6 |
|
97.2 |
97.8 |
|
98.9 |
| PCR
cervix |
98.2 |
99.5 |
|
98.7 |
99.7 |
|
| PCR
urine |
97.0 |
98.2 |
|
98.0 |
99.4 |
|
Authors' Conclusions
For test evaluation studies, including a
urine test in the reference standard will more nearly approximate a
patient standard and reduce overestimation of cervical test sensitivity.
For overall estimates of test performance, when comparing cervical and
urine tests, both cervical and urine tests should be included in the
reference standard. Culture tests are not required as a part of the
standard and can be replaced with any commercial NAAT.
Source of funding: none given
For correspondence: Carolyn M. Black,
National Center for Infectious Diseases, Mailstop C17, Centers for Disease
Control and Prevention, 1600 Clifton Rd. NE, Atlanta, GA, 30333. E-mail
address: cblack@cdc.gov.
|
