Screening for Gonorrhea: Update of the Evidence: Evidence Table 3

You Are Here: AHRQ Home > Clinical Information > U.S. Preventive Services Task Force > Topic Index > Screening: Gonorrhea > Update of the Evidence

Evidence Table 3. Studies of Gonorrhea Screening Tests

Author, Year	N	Study Test/ Specimen	Gold Standard/ Expanded Gold Standard	Discrepant/ Confirmatory Test	Prevalence in Study Population	Population/ Setting	Results	Limitations	Quality Rating
Bassiri, 1997⁶⁹	3,340 women	PCR (Amplicor® CT/NG) Urine	None	Positive samples retested by same PCR assay and 16S RNA-based PCR.	0.3%	Asymptomatic women aged 15-44 years attending 24 family planning centers or gynecology clinics in 14 European countries for contraceptive advice (n=2,987) or pregnancy termination (n=353).	Of 3,340 samples tested, 9 (0.3%) were positive for N. gonorrhoeae; all were confirmed positive by repeat testing using the same assay; all 9 were negative using 16S RNA-based PCR.	Prevalence was too low to adequately determine test performance; no gold standard.	Poor
Beltrami, 1998⁷¹	640 men	DNA probe (PACE® 2) Urethral swabs	Culture Plated to Thayer-Martin agar; typical appearing colonies were identified presumptively by gram stain and the oxidase reaction; final identification was performed by carbohydrate utilization.	None	3%	Asymptomatic arrestees in New Orleans, LA (1993-1994).	DNA probe was positive in 9 samples (2%), culture in 13 (3%). Sensitivity=54%, specificity=99.5%, PPV=78%, NPV=99%. Sensitivity and PPV for detection of gonorrhea were lower than those found in other studies, but this study included only asymptomatic men.	Arrestees do not represent primary care screening population; no resolution of discrepant results.	Poor
Chan, 2000⁷²	825 men 399 women	1. SDA (BDProbeTec™ ET) 2. PCR (Amplicor) Urine	None	In-house PCR assays	3% men 1.5% women	Symptom status not known; consecutive patients attending 3 STD clinics and other family physicians' offices in Saskatchewan, Canada.	Women PCR: sensitivity=100%, specificity=98.4%, PPV=50%, NPV=100% SDA: sensitivity=100%, specificity=99.2%, PPV=66.7%, NPV=100% Men PCR: sensitivity=96.2%, specificity=99.1%, PPV=78.1%, NPV=99.9% SDA: sensitivity=100%, specificity=99.9%, PPV=96.3%, NPV=100% Throughput: 46 specimens/8 hours for PCR and 150/8 hours with SDA.	Predominantly STD clinic population; unknown symptom status.	Fair
Cosentino, 2003⁶⁵	455 women	SDA (BDProbeTec™ ET) Vaginal and cervical swabs; study compares specimen sites	Culture plated to modified Thayer Martin medium and chocolate agar; identification was based on gram staining, oxidase test, and Gonochek II.	Samples with discrepant results (SDA/culture) were evaluated by LCR. Samples were considered true positive if they were positive by culture or by two molecular tests (SDA and LCR).	3.1%	Symptomatic women attending reproductive health clinics in Pittsburgh, PA (1997-2000).	Culture (cervix): sensitivity=76.9%, specificity=100%, PPV=100%, NPV=97.9% SDA (vagina): sensitivity=100%, specificity=99.8%, PPV=97.5%, NPV=100% SDA (cervix): sensitivity=100%, specificity=100%, PPV=100%, NPV=100%	Women were symptomatic.	Fair
Crotchfelt, 1997⁷⁰	344 men 192 women	PCR (Amplicor® CT/NG) Women: endocervical swabs and urine Men: urethral swabs and urine	Culture plated to modified Thayer Martin medium; colonies containing oxidase-positive and gram-negative diplococci were presumptively identified as N. gonorrhoeae, confirmed by using fluorescent-antibody staining.	For specimens that were PCR positive and culture negative, PCR was repeated using a confirmatory 16S rRNA PCR assay.	17.2% men 7.8% women	Symptom status not known; men and women attending two Baltimore, MD (city) STD clinics (1995).	Men: PCR (urine): sensitivity=94.4%, specificity=98.5% PCR (urethra): sensitivity=97.3%, specificity=97.0% Culture: sensitivity=76.6%, specificity=100% Women: PCR (urine): sensitivity=90.0%, specificity=95.9% PCR (cervix): sensitivity=100%, specificity=99.4% Culture: sensitivity=65.2%, specificity=100%	STD clinic population; unknown symptom status.	Fair
Darwin, 2002⁷⁵	669 women	DNA probe (Hybrid Capture® 2) Endocervical swab (culture) Endocervical brush (DNA probe)	Culture plated to modified Thayer Martin medium; oxidase-positive, gram-negative culture colonies were confirmed using fluorescein-conjugated antibodies specific for N. gonorrhoeae.	Discrepant results between DNA probe and culture were adjudicated by direct fluorescent-antibody (DFA).	13.5%	High-risk women attending public STD clinics in Birmingham, AL and Baltimore, MD.	DNA probe vs. culture: sensitivity=90.3% (81.0-96.0), specificity=96.8%, PPV=77.4%, NPV=98.8% DNA probe vs. adjudicated: sensitivity=92.2% (84.6-96.8), specificity=99.8%, PPV=98.8%, NPV=98.8% Culture vs. adjudicated: sensitivity=80.0% (70.3-87.7), specificity=100%, PPV=100%, NPV=97.0%	STD clinic population; unknown symptom status.	Fair
Darwin, 2002⁵⁸	1,202 men	All subjects were tested with DNA probe (PACE® 2C), then 140 positives were tested with 2 additional DNA probes (Hybrid Capture® 2 and PACE® 2). Urethral swab	None	PCR (SHARP assay)	14.6%	Symptomatic and asymptomatic men from a number of STD clinics.	PACE 2 and Hybrid Capture 2 tests had positive and negative agreements of 98.3% and 99.8% respectively. Hybrid Capture 2: sensitivity=98.9%, specificity=99.9%, PPV=99.4%, NPV=99.8% PACE 2: sensitivity=99.4%, specificity=99.9%, PPV=99.4, NPV=99.9%	Results reflect second step in testing process; STD clinic population; unknown symptom status.	Poor
Diemert, 2002⁵⁹	9,834 men and women	PCR (Cobas Amplicor® CT/NG) Women: endocervical swabs and urine Men: urethral swabs and urine	737 also tested by culture plated to modified Thayer Martin and chocolate agar. Identities of presumptive isolates were confirmed by carbohydrate utilization profiles.	Specimens positive by PCR underwent testing by 16S rRNA-based PCR.	0.4%	Symptomatic and asymptomatic consecutive patients undergoing evaluation for N. gonorrhoeae infection at outpatient clinics affiliated with a large Montreal tertiary-care center and 2 health centers in northern Quebec serving a primarily native Canadian population (1999-2000).	PCR vs. culture: sensitivity=100%, specificity=99.9% (99.2-100).	Prevalence was too low to adequately determine test performance; unknown symptom status; data not provided by sex.	Poor
Farrell, 2001⁶¹	260 men and women	1. PCR (Cobas Amplicor® CT/NG) 2. LCR (LCx NG) 3. PCR (In-house cppB gene) Urine	Infected=at least 2 positive tests. Not infected=at least 2 negative tests.	Retested at least twice by same 3 methods.	16.9%	Asymptomatic patients from widespread geographical sites in Queensland, Australia included populations known to have a high prevalence of asymptomatic N. gonorrhoeae and high PCR false-positivity.	PCR: sensitivity=97.9%, specificity=93.9%, PPV=78.0%, NPV=99.5% LCR: sensitivity=95.7%, specificity=100%, PPV=100%, NPV=99.1% In-house PCR: sensitivity=97.9%, specificity= 100%, PPV=100%, NPV=99.5%	Data not provided by sex.	Fair
Gaydos, 2003⁷⁸	1,484 women	TMA (Combo 2 assay) Urine Endocervical swab	Swabs: LCR, culture Urine: LCR Infected=culture positive or LCR swab positive + urine specimen positive. Not infected=both culture and LCR on both specimens types negative.	TMA positive specimens from patients classified as not infected underwent masked testing together with a subset of negative specimens by confirmatory TMA assays.	8.6%	Symptomatic and asymptomatic 18-35 year old women from 7 geographically diverse STD, family planning, and OB/GYN clinics with wide prevalence rates of N. gonorrhoeae.	Asymptomatic: Swab: sensitivity=96.9% (83.8-99.9), specificity=99.6% (98.7-100), PPV=93.9%, NPV=99.8% Urine: sensitivity=87.5% (71.0-96.5), specificity=99.5% (98.5-99.1), PPV=90.3%, NPV=99.3%. Symptomatic: Swab: sensitivity=100% (96.2-100), specificity=98.1% (96.9-98.9), PPV=86.2%, NPV=100% Urine: sensitivity=92.6% (85.3-97.0), specificity=99.1% (98.2-99.6), PPV=92.6%, NPV=99.1%. For 34 pregnant women, TMA results were concordant with patient infected status in all specimens (100% sensitivity and 100% specificity).	Good
Leslie, 2003⁵⁵	4,324 men and women (PCR) 2,305 men and women (culture)	PCR (Amplicor® CT/NG) FDA-listed sites: penile/urethral swabs urine cervical/vaginal swabs Non FDA-listed sites: ano-rectal oropharyngeal other	Culture plated to New York City agar and chocolate or horse-blood agar (depending on site).	16S rRNA assay used through Aug. 2001; then cppB PCR assay following discontinuation of 16S rRNA.	5.4%	Tests performed in a tertiary referral public health laboratory; specimens referred from Victoria, Australia. Patients were predominantly men with high rates of STDs.	PCR overall: sensitivity=81.7%, specificity=99.5%, PPV=92.7%, NPV=98.5%. PCR for FDA-listed specimens (553 penile/urethral swabs, urine, and cervical/vaginal swabs): sensitivity=96.7%, specificity=99.8%, PPV=98.9%, NPV=99.4%. PCR for non FDA-listed specimens (827 ano-rectal, oropharyngeal and other specimen types): sensitivity=65.1%, specificity=99.4%, PPV=84.8%, NPV=98.1%. No significant difference between the 2 confirmatory assays (p=0.65).	STD clinic population; unknown symptom status; sensitivity/specificity data not provided by sex or specific specimen.	Fair
Livengood, 2001⁵⁷	618 women	PCR (Cobas Amplicor® CT/NG) Endocervical swabs	Culture plated to Jembec media; gram-negative, oxidase-positive, and Gonogen II-positive results were reported as a positive culture. Infected=culture positive or 2 or more other tests positive.	PCR assay was repeated once in another laboratory in addition to a LCR.	4.4%	Women undergoing testing for endocervical N. gonorrhoeae infection, as determined by their practitioners, in the emergency department, private and staff OB/GYN clinics, and inpatient units of Duke University Medical Center, Durham, NC.	Samples were negative for 591, positive for 24, and discrepant for 3. After discrepancy resolution, 591 true negatives and 27 true positives. PCR: sensitivity=96.3%, specificity=100%, PPV=100%, NPV=99.8% Culture: sensitivity=92.6%, specificity=100%, PPV=100%, NPV=99.7%	Unknown symptom status.	Fair
Martin, 2000⁷⁶	2,192 women 1,981 men	1. PCR (Amplicor® CT/NG) 2. PCR (Cobas Amplicor® CT/NG) Women: endocervical swabs and urine Men: urethral swabs and urine	Culture; gram-negative diplococci were confirmed as N. gonorrhoeae by glucose utilization profiles.	16S rRNA PCR	Asymptomatic women: 5.4% asymptomatic men: 3.6% symptomatic men: 29.4%	Symptomatic and asymptomatic consecutive patients seen in STD clinics and family-planning centers in multiple U.S. cities.	The 2 PCR assays were 98.8% concordant and exhibited identical performance characteristics. Results for PCR (Cobas Amplicor): All women (145 cases; results similar if symptoms or not) Endocervical: sensitivity=92.4%, specificity=99.5% Urine: sensitivity=64.8%, specificity=99.8% Asymptomatic men (26 cases) Urethral: sensitivity=73.1%, specificity=99.0% Urine: sensitivity=42.3%, specificity=99.9% Symptomatic men (372 cases) Urethral: sensitivity=98.1%, specificity=98.8% Urine: sensitivity=94.1%, specificity=99.9% Sensitivity of culture Women: asymptomatic=86.2%, symptomatic=83.9% Men: asymptomatic=46.2%; symptomatic=92.7	Some subjects from STD clinics.	Good
Modarress, 1999⁶³	1,727 women 19 men	1. DNA probe (Hybrid Capture® 2) 2. DNA probe (PACE® 2) Women: endocervical brush Men: urethral swab	Infected=2 of 3 tests positive.	PCR	Site 1: 2.8% Site 2: 0.8%	Symptom status unknown for patients attending STD, family planning, and private practice clinics in U.S.	99.6% initial agreement between the two methods; 9 were discrepant. Hybrid Capture® 2: sensitivity=100%, specificity=99.7% PACE® 2: sensitivity=87.1%, specificity=100%	Some of subjects from STD clinics; too few men to report results by sex.	Fair
Moncada, 2004⁷⁷	1,569 women	1. TMA (APTIMA™ Combo 2) 2. LCR (LCx) Urine (used as part of specimen standard) Endocervical swabs (used as single specimen diagnosis)	Culture plated to Thayer Martin media; oxidase-positive colonies yielding gram-negative diplococci were subcultured to chocolate agar plates. Isolates were confirmed as N. gonorrhoeae by either sugar utilization tests, fluorescent antibody, or Haemophilus-Neisseria identification (HNID). Infected=culture positive or both LCR and TMA positive.	Another TMA assay	8.7% endocervical 7.9% urine	Symptomatic and asymptomatic 18-35 year old women from 7 geographically diverse STD, family planning, and OB/GYN clinics with wide prevalence rates of N. gonorrhoeae.	Endocervical swab: TMA: sensitivity=99.2%, specificity=98.6% LCR: sensitivity=96.1%, specificity=99.7% Culture: sensitivity=85.9%, specificity=100%	Examined differences in single specimen vs infected patient standard; performance results for endocervical swab only; unknown symptom status.	Fair
Palladino, 1999⁶²	73 men	PCR (Amplicor® CT/NG) Urethral swabs Urine	Culture plated to heated blood agar and modified Thayer Martin agar and confirmed by morphology, gram-stain, positive oxidase test, lack of growth on nutrient agar, and positive Gonocheck II test.	In-house PCR assay, if positive then additional confirmation by an alternative PCR assay. Further 16S rRNA PCR testing was done on specimens positive by Amplicor PCR, but negative for all other tests.	52.1%	Men in Western Australia seeking medical attention for STDs or sexual partners of confirmed cases.	Urine PCR: sensitivity=100%, specificity=100%, PPV=100%, NPV=100% Urethral culture: sensitivity=86.8%, specificity=100%, PPV=100%, NPV=87.5%	High risk population; small sample size; unknown symptom status; questionable generalizability.	Poor
Roymans, 1999⁶⁸	358 women 71 men	1. PCR (In house) 2. DNA probe (PACE® 2) Cervical or urethral swabs	None	Retesting by PCR and DNA probe, then with another PCR method	3%	Symptom status not known, patients visiting OB/GYN and dermatology clinics of 5 hospitals in the Netherlands.	PCR: sensitivity=100%, specificity=99.5% DNA probe: sensitivity=61.5%, specificity=100%	Low prevalence; symptoms not known; results not given by sex.	Fair
Schachter, 1999⁶⁴	1370 women	DNA probe (Hybrid Capture® 2 CT/GC) Cervical brush (nonpregnant women) or swab (pregnant women)	Culture plated to Thayer Martin media. Oxidase-positive gram-negative diplococci were confirmed as gonococci by using sugar utilization tests, except at one site where Gonochek-II was used.	PCR if negative culture, positive DNA probe	6.9%	Symptom status not known for women seen in STD or family planning clinics at several U.S. sites.	For cervical brush: of 218 positive by the initial test, 106 were positive by the confirmation test. DNA probe: sensitivity=92.6%, specificity=98.5% Results similar for swab specimens (data not provided).	Unknown symptom status.	Fair
Schwebke, 1996⁷³	453 women 546 men	DNA probe (PACE® 2) Women: endocervical swab Men: urethral swab	Culture plated to modified Thayer Martin agar. Presence of N. gonorrhoeae was determined by presumptive identification using growth on Thayer Martin agar, gram stain morphology, and positive oxidase reaction.	Gram stain	17% women 28% men	Symptom status not known for patients seen at 5 public STD clinics in Chicago, IL.	Disparate results occurred among 55 specimens. Using culture as gold standard, there were 31 false-positive and 24 false-negative DNA probe assays. DNA probe after discrepancy resolution Urethral: sensitivity=91.7%, specificity=96.4% Endocervical: sensitivity=86.2%, specificity=96.8%	Some subjects from STD clinics; gram stain used to resolve discrepancies.	Poor
Tabrizi, 2004⁵⁴	174 women	PCR (Cobas Amplicor® CT/NG) Stored patient samples: 99 tampons 14 urine 4 urethral swabs 7 cervical swabs	Consensus results: positivity defined as any sample positive by at least 2 out of the 4 methods used besides PCR (Cobas Amplicor® CT/NG) (MB, FRET, 16S rRNA, or LCx).	In-house developed PCR confirmatory assays (LightCycler using MB and FRET probes), 16S rRNA, and LCR.	All initially positive	122 positive samples (by PCR Cobas Amplicor® CT/NG) were selected from stored samples of >3000 women who had participated in studies in Australia and Pacific islands; 50 negative samples also randomly selected.	Using the consensus algorithm, 73/122 (59.8%) were confirmed positive; sensitivity and specificity were not reported.	Symptom status not known; no sensitivity/specificity data.	Poor
Uhrin, 1997⁶⁶	199 women	PCR Endocervical swab	Culture plated to JEMBEC agar.	Not described (references provided)	3.5%	Symptom status not known; women patients seen in public clinics in Pittsburgh, PA.	Culture: sensitivity 28.6%, specificity 100% PCR: sensitivity 100%, specificity 100%	Discrepant testing not described, symptoms not known.	Poor
Van der Pol, 2001⁵⁶	2,109 men and women	1. SDA (BD ProbeTec™ ET) 2. LCR Women: endocervical swabs and urine Men: urethral swabs and urine	Culture plated to modified Thayer Martin agar or Martin Lewis media; gram-negative, oxidase-positive specimens were confirmed using a biochemical method and at least one other method (fluorescent antibody staining, Gonogen, or GenProbe direct probe). Infected=culture positive or LCR swab and urine positive (women); culture positive or LCR urine positive (men).	None	8% women 11% men	Symptomatic and asymptomatic patients attending STD, family planning and OB/GYN clinics at several U.S. sites.	SDA among asymptomatic patients: Endocervical swab: sensitivity=97.4%, specificity=99.6% Urine (women): sensitivity=86.5%, specificity=99.3% Urethral swab: sensitivity=100%, specificity=99.5% Urine (men): sensitivity=100%, specificity=100% SDA among symptomatic patients: Endocervical swab: sensitivity=96.1%, specificity=99.3% Urine (women): sensitivity=83.7%, specificity=99.6% Urethral swab: sensitivity=98.4%, specificity=94.8% Urine (men): sensitivity=97.9%, specificity=94.4%	STD populations included.	Good
Van Doornum, 2001⁶⁰	1,001 503 women 498 men	1. LCR (LCx Probe) 2. PCR (Cobas Amplicor® CT/NG) Women: endocervical swabs and urine Men: urethral swabs and urine	Culture plated to GC-lect for Gram staining and oxidase testing. Infected=3 of 4 tests positive, 2 swabs positive, or 2 urine specimens positive.	Discrepant results were repeated; discrepant samples that were solitary and repeatedly positive by only one assay were further studied by 16s analysis.	1.2% women 4.2% men	Visitors of an STD clinic in Amsterdam in 1998; < 30 years of age.	Significant differences in PCR performances for female swab and urine specimens (p<0.05). PCR Female swab: sensitivity=100%, specificity=97.4%, PPV=31.6%, NPV=100% Female urine: sensitivity=66.7%, specificity=98.6%, PPV=36.4%, NPV=99.6% Male swab: sensitivity=100%, specificity=99.2%, PPV=84.0%, NPV=100% Male urine: sensitivity=95.2%, specificity=99.4%, PPV=87.0%, NPV=99.8%	STD population; unknown symptom status.	Fair
Whiley, 2002^,74	152 specimens	PCR (LightCycler real-time assay using FRET) vs. in-house PCR (ELAHA based) Urine	Not reported	Not reported	19-20%	Specimens submitted to the Queensland health Pathology Service.	N. gonorrhoeae was detected in 29 (19%) specimens by LightCycler PCR and in 31 (20%) specimens by in-house PCR. Negative results were obtained from 121 (80%) by both assays; clinical sensitivity for LightCycler PCR was 94%.	No information on population, gold standard, or discrepant tests.	Poor
Young, 1997⁶⁷	161 homosexual men	DNA probe (PACE® 2) Pharyngeal and rectal swabs	Culture plated to modified New York City medium; suspect culture colonies were tested by the oxidase test: oxidase-positive Gram-negative diplococci were confirmed as N. gonorrhoeae by immunological and biochemical tests.	Discrepant results were retested by PACE 2 using the original specimen, and repeated reactive specimens from culture negative patients were tested by the Gen-PCA assay.	14.3-16.7%	Homosexual men attending the Dept. of Genitourinary Medicine, Edinburgh Royal Infirmary (1995-1996).	DNA probe (PACE® 2) after resolution of discrepant results Rectal: sensitivity=94.1%, specificity=100%, PPV=100%, NPV=99.3% Pharyngeal: sensitivity=86.4%, specificity=100%, PPV=100%, NPV=97.9% Culture Rectal: sensitivity=88.2%, NPV=98.6% Pharyngeal: sensitivity=59.0%, NPV=93.9% Overall agreement between rectal culture and Gen-Probe was 97.5% with 3/4 discrepancies due to negative cultures. Overall agreement between pharyngeal culture and Gen-Probe was 91.3% with 10/14 discrepancies due to negative cultures. Gen-Probe was significantly more sensitive than throat culture (p<0.05) but not rectal culture (p>0.2).	Unknown symptom status.	Fair

Notes: CT/NG, Chlamydia trachomatis N. gonorrhoeae; ELAHA, enzyme-linked amplicon hybridization assay; FDA, U.S. Food and Drug Administration; FRET, fluorescence resonance energy transfer; LCR, ligase chain reaction; NPV, negative predictive value; OB/GYN, obstetrics and gynecology; PCR, polymerase chain reaction; PPV, positive predictive value; SDA, strand displacement amplification; STD, sexually transmitted disease; TMA, transcription mediated amplification.

Return to Document