Performance of non-treponemal serology tests for syphilis

Key Question

What are the performance characteristics, stratified by the stage of syphilis, for non-treponemal serologic tests?

Literature Search Terms

(syphilis OR Treponema pallidum) AND (genital ulcer disease OR primary syphilis OR secondary syphilis OR tertiary syphilis OR congenital syphilis OR ocular syphilis) AND (diagnosis OR lesions OR polymerase chain reaction OR PCR OR nucleic acid amplification test OR NAAT OR multiplex test OR silver stain OR silver staining OR immunohistochemistry OR IHC OR rabbit infectivity testing OR RIT OR direct detection OR dark field microscopy OR darkfield microscopy OR dark-field microscopy OR direct fluorescent antibody OR DFA OR direct fluorescent antibody for T. pallidum OR DFA-TP OR direct fluorescent antibody tissue test for T. pallidum OR DFAT-TP). Solely-based international studies were excluded from the literature search.

Grading

Four reviewers the evidence as high, medium, and low based on each study’s strengths and weaknesses. Case reports or small case studies were reviewed.

High quality publications: Studies using clinically characterized specimens, stratified by stage, larger sample size, prospective or a well-done cross sectional or retrospective study. Studies with large sample sizes, clinically characterized but not stratified by stage, or characterized but unclear exactly how it was done.

Medium quality publications: Studies with small sample sizes, moderate methodological issues, single lab test as gold standard, or descriptive.

Low quality publications: Studies with major methodological issues or small sample sizes.

I: Case reports or small case studies.

Key Question 3
Citation Description of study type, design, population, and setting Reported findings, quantitative results related to key question Overall quality; strengths/weaknesses or limitations Relevance to the key question and/or overall importance Rank
El Zaatari/ Incidence of the prozone phenomenon in syphilis serology/Obstet Gynecol/1994 Cross-sectional:

4328 patient sera from the U of Texas Galveston checked for syphilis by RPR and then rechecked  with serial two-fold dilutions up to 16-fold.

N=2065 pregnant women

N=1439 nonpregnant women

N=824 men

 

Gold standard: FTA-ABS was done on all sera to measure false positive rate

Only one prozone reaction was detected in a man (overall prozone 95% CI was 0-0.4%)

Female nonpregnant BFP: 1.1%

Female pregnant BFP: 0.6%

Male BFP: 1.3%

Large sample size, general population with low prevalence of disease. Relevant

 

High
El-Zaatari/False-negative syphilis screening due to change in temp/STD/1994 “Case” report:

2 labs screening 2232 pts for syphilis with RPR

One lab reported 64/1210 (5.3%) as positive

Other lab reported 78/1210 (6.4%) as positive.

 

The authors investigated the difference and found that it seemed to be related to a cold centrifuge.

 

Gold standard: NA

It was found that a refrigerated centrifuge might have caused false negative results. When temperature was adjusted from 4 to 27C an additional 1022 samples tested were consistent between the two laboratories. Few specific numbers, just a mention of the cold as a factor in false negatives. Relevant I
Liu/Incidence and risk factors for the prozone phenomenon in serologic testing for syphilis in a large cohort/CID/2014

 

Retrospective cross sectional:

N=46856 clinical samples at Zhongshan Hospital in China which had both RPR and TPPA assay.

 

1573  were neg by RPR but pos by TPPA were diluted from 1:1 to 1:32.

 

Gold standard: TPPA, CIA, then dilution

Overall incidence of prozone phenomenon of 0.83%

 

36/1573 RPR neg, TPPA pos were prozone.

 

Prozone reaction was most common in primary and secondary syphilis.

Neurosyphilis and pregnancy also increased the odds of prozone.

 

Interestingly nearly 31% of patients titers with the prozone reaction were <=1:16.

Those which were TP-PA neg were not diluted. Relevant High
Post/Case report and evaluation of the frequency of the prozone phenomenon in syphilis serology-an infrequent but important lab pheno/Sexual Health/ 2012 Case Report and Prospective Study:

Case report of HIV+ patient with secondary syphilis with 2 consecutive neg RPRs which when retested were shown to be neg due to prozone reaction.

 

Prospective study: n=3222 samples collected in Prince of Wales hospital in Australia for RPR testing. All diluted at 1:1 and 1:8.

 

Gold standard: negative at 1:1 but positive at 1:8

2/3222 showed a prozone reaction (initially NR at 1:1 dilution but reactive at 1:8)

 

Poor gold standard. Limited relevance I
Wuepper/false positive reaction to VDRL test with prozone phenomena/JAMA/ 1966 Case report:

A woman who was ultimately diagnosed with lymphosarcoma had a false positive VDRL after dilution (not initially positive)

 

Gold standard:TPI and FTA which were both negative

False positive VDRL after dilution: single case No follow-up testing after treatment, single patient Limited relevance I
Liu/ Characterization of the classical biological false-positive reaction in the serological test for syphilis in the modern era/ Intl Immunopharmacol /2014 Retrospective study:

N= 63,765 blood samples obtained at Zhongshan Hospital in the Medical College of Xiamen University were tested with RPR.

 

Gold Standard: TPPA and CIA

206 (0.32%) had the BFP reaction. In multivariate analysis, an increased likelihood

of the CBFP reaction was associated with female subjects, subjects ≥80 years old, and subjects between 16 and 35 years old

 

 

No clear association with diseases Relevant High
Geusau/ BFP tests comprise a high proportion of VDRL reactions in an analysis of 300,000 sera/ Int J STD AIDS/2005 Retrospective study from Austria on n=300,000 sera with age, sex, and stage of disease.

 

Gold Standard: TPHA

2799 patients (0.92%) of the study population were positive, of whom 736 (26%) were BFP. BFP reactivity was found in 0.24% and was   higher in women than in men (0.27% versus 0.20%, P<0.001) and in patients over 60 years of age (0.34%) as compared with those under 60 (0.25%, P<0.001). HIV-positive patients (n=1415)

had a 10-fold higher rate of BFP tests (2.1% versus 0.24)

Large sample size; reasonable data on population; limited clinical data Relevant High
Bala M, et al/ Evaluation of the usefulness of Tp hemagglutination test in the diagnosis of syphilis in weak reactive VDRL sera/ Indian Journal of Sexually Transmitted Diseases and AIDS/ 2012 Retrospective study:

Examined n=5785 sera screened by VDRL from multiple clinical sites in India;

 

Gold standard: TPHA

Comparator: VDRL

Out of 80 qualitative VDRL reactive sera, 68 had <1:8 titer on quantitation and

TPHA was positive in 59 samples, indicating BFP reactivity in 0.2% in all the subject groups.

There were no BFPs among sera with VDRL titers of >=1:8.  The male-to-female ratio of BFP reactions was 2:1

No stage of syphilis; background prevalence of non-syphilitic Treponematoses Relevant High
Brede HD/ Detection of BFP Syphilis Serum Reactions/SA Medical Journal/ 1974 Cross sectional study:

Sera from N=5271 persons from SA

2493 pregnant women, 1 130

healthy prospective employees, 1345 newborn babies and

303 leprosy patients

 

Gold Standard: FTA-ABS

Comparator: RPR and VDRL

VDRL: 16% BFP

RPR: 20% BFP

The group of male and female employees showed roughly a third less false reactivity than pregnant women; higher BFP among those with leprosy.

No syphilis staging; background prevalence of non-syphilitic Treponematoses Moderately relevant High
Sichy/Syphilis serology in patients with primary syphilis and non-treponemal STDS in southern Africa/GU Med/ 1991 Cross sectional study:

N= 1170 black mine workers presenting with acute urethritis or genital ulceration had physical exam and RPR and FTA performed

 

Primary syphilis:

Gold standard: DF on ulcers: n=21 primary syphilis cases DF+ on ulcers.

 

Others: Gold standard: BFP defined as positive RPR and negative FTA with “no other signs or symptoms of syphilis”

Of the 1149 patients without primary syphilis, 178 were positive by RPR, 2 of these were felt to be biologic false positive (0.01%, though they report as 0.02%) as the FTA was negative. Little clinical information on previous treatment. Moderately relevant. Medium
Walker/RPR card test a screening method for treponemal disease/BJVD/ 1971 Cross sectional study:

N=6225 sera tested routine hospital patients. If RPR positive, was tested with VDRL, FTA and TPI

 

 

Gold standard: BFP: Positive VDRL or RPR and negative FTA and TPI.

BFP:

95/6225 tests done were BFP’s with RPR

 

86 BFP’s with VDRL-but note that VDRL was tested only if RPR was positive

Not initially tested with VDRL. Only if RPR pos. Moderately relevant. Medium
Omer/Evaluation of the VDRL test in Sudanese blood donors/ Tropical doctor/1982 Cross sectional retrospective:

 

N=2201 blood donors and N=199 pts with STDs tested with VDRL and FTA

 

Gold standard BFP: VDRL + and FTA neg

30/2201 BFPs (all of the pos VDRLs) in blood donor group: 1.36% overall prevalence BFP

 

10/11 VDRL pos in STD pt group were FTA neg, therefore  10/199 tested in  STD group were felt to be BFP: 5% overall prevalence

No clinical data Moderately relevant Medium
Goldman/FTAABS and VDRL slide test reactivity in a population of nuns/ JAMA/1971 Nuns from two orders were asked to participate if they had never had sexual contact and did not have self-reported hx of acquired or congenital syphilis.

 

Gold standard: Nun status, self-reported hx of no syphilis and no syphilis risk factors.

0/250 had a positive VDRL

 

(of note 3/250 had a positive FTA)

Gold standard: Self report, also relying on assumption that nuns were life long celibates as they self reported.. Moderately relevant Medium
Castro/Evaluation of an enzyme immunoassay technique for detection of antibodies against T. pallidum/JClinMicro/ 2003 Cross-sectional study:

N=107 patients with no clinical history of syphilis.

 

Gold standard: FTA-Abs positive, plus clinical hx.

 

Divided into 5 groups based on clinical hx:

25 primary syphilis

25 secondary syphilis

179 latent syphilis

105 “hx syphilis that had been correctly treated”

107 no clinical hx syphilis.

Specificity:

“No clinical history of syphilis”, FTA negative:

RPR Negative:  95/107 (88.8)

 

12/107 were felt to be BFP

Gold standard was FTA abs alone rather than two treponemal tests. Moderatelyrelevant Medium
Ashkar/Serologicial tests for syphilis in diseases of the thyroid/JAMA/ 1970 Cross sectional study:

N=1102 patients with thyroid disease had a VDRL and FTA done.

 

Gold standard: FTA

7 were positive for VDRL, all had a positive FTA and so all were felt to be true positives

 

No false positives in thyroid disease

No comparison group. Limited relevance Medium
Smikle/ BFP serological tests for syphilis in the Jamaican population/GU med/ 1990 Cross sectional study:

N=19067 sera screened with VDRL.

N=441 general pop,

N=145 pregnant women VDRL+ with a titer <1:8 were confirmed with FTA

 

Gold standard: BFP: VDRL <1:8 with a negative FTA

General pop:

94/347 + VDRL with titer<1:8 were FTA neg (27%)

Pregnant women:

22/71 + VDRL with titer<1:8 were FTA neg

(31.0%)

Only tested if titer <1:8. There could have been false positives with a higher titer that would have been missed. Limited relevance Medium
Johansson/ Serological syphilis tests in the elderly/ Annals of Clin Res/1970 Cross sectional study:

N=484 geriatric inpatients tested with VDRL.

Gold standard: FTA-ABS

5.4% BFP No control arm Limited relevance Medium
Johansson/3 Lipoidal tests in screening for syphilis and BFPs in a dermatological series/ Annals of Clin Res/1970 Cross sectional study:

N=6737 dermatological inpatients tested with VDRL.

Gold standard: FTA-ABS

28/6737 BFP using the VDRL

 

No control arm Limited relevance Medium
Lee/The Significance of Syphilis Serology Tests on Long-Term Hemodialysis patients/Chang Gung Med J/1998 Prospective cross sectional study:  N=556 patients on dialysis who were tested with an RPR

Gold standard:TPHA

1.8% BFP No control arm Limited relevance Medium
Wiwanitkit/ BFP VDRL Tests: When to re-Test/ Southeast Asian J of Trop Med/2002 Prospective study: N=30 patients with BFP VDRL tested every 2 weeks

Gold Standard: TPHA

Seroreversion occurred between 9.25 and 10.49 weeks; 25 returned to nonreactive by 10 weeks; 2 cases within 14 weeks. No clinical data on cause of BFP Limited relevance Medium
Tuffanelli/ Ageing and false positive reactions for syphilis/BJVD/1966 Cross sectional:

There were several groups, but reasonable gold standard for  the 58 aged persons randomly selected from a jewish old age home.

 

Gold standard: FTA-ABS

6/58 (9%) had persistently positive RPR for syphilis but negative FTA. No control arm.

 

No clinical information.

Limited relevance Low
Kostant/False positive seroreactions for syphilis/Report of two families, one with 3 generations affected/ JAMA/1970 Case report of 2 families with positive RPR and negative treponemal tests.

 

Gold standard: positive RPR, negative treponemal test persistent over time.

Family 1: 4/7 members

Family 2: 3/5 members

Very small numbers. Little clinical data.  

Limited relevance

I
Moore/Sensitivity and Specificity in Syphilis Serology/Southern Medical Bulletin/1965 Cross sectional retrospective study:

 

Samples tested with KRP (Kolmer test) and VDRL

 

Gold standard:

Primary syphilis: clinical dx with + DF (N=76)

 

Secondary syphilis: Clinical diagnosis with + DF (N=100)

 

“Non-infected”

1.”Presumeably healthy individuals who underwent a careful screening procedure (N=74)

 

2. Non syphilitic controls from the original Tuskegee study (N=36)

 

3. “Patients diagnosed as being acute or chronic biologic false positive reactors” (N=38)

 

Unclear stage syphilis:

“Syphilitic” patients from the Tuskegee study (N=46)

 

 

 

“Normals” from “presumably healthy individuals”

VDRL+ 0/0

KRP + 0/0

 

 

Nonsyphilitic controls from Tuskegee:

VDRL+ 0/36

KRP + 0/36

 

“Biologic false positives”

VDRL+ 38/38

KRP+  0/38

 

“Syphilitic” patients from Tuskegee

VDRL+ 54.4% (out of 46.)

Poorly defined gold standards

 

Definitions quite unclear for non-syphilitic controls

 

“Biologic false positives” unclear gold standard

 

 

Limited relevance Low
Knight/the clinical significance of the biological false positive serologic reactor/Canada Med Ass J/1963 Case series

N=113

 

Gold standard: TPI

Described the characteristics of 113 “BFP’s”. IN most of them some sort of abnormality (malignant tumor, collagen disease, chronic liver disease, skin disease, malaria, chronic infection and gastrectomy) was found.  

 

No comparison group

Limited relevance Medium
Constable/Positive serological tests for syphilis and administration of IVIG/STI /2007 Case report of one patient who had a positive treponemal ab but neg RPR, after getting IVIG, subsequent Trep tests negative. In this case, appears to be a false positive based on treponemal test Single case report, focused on treponemal false pos Not relevant I
Marangoni/ Laboratory dx of syphilis with automated immunoassays / J Clin Lab Analysis/ 2009 Cross sectional:

N=244 patients suffering from different stages of syphilis. The staging of the disease was done following clinical and

laboratory criteria

 

Gold standard: unclear, but appears to be + architect syphilis TP, and some clinical criteria

 

Second group n=74 sera from 1.culture-confirmed

Lyme disease (n=10), 2. clinical diagnosis of infectious mononucleosis detected

as positive by the Paul–Bunnel–Davidsohn agglutination

(n=10), 3.sera from pregnant women (n=14), 4.sera

from patients with cytomegalovirus acute infection (IgM

positive and low avidity) (n=520),  sera from patients

with acute toxoplasmosis (IgM positive and low avidity)

(n=20).

Gold standard : TPHA

 

Third group: N=129

sera selected from 9,210 samples submitted for routine screening for syphilis. All  were positive by ARCHITECT Syphilis TP, but negative by

TPHA and WB.

Gold standard: WB and TPHA.

 

For the Syphilitics: Gold standard unclear: clinical and lab

 

 

Lyme n=10

RPR spec: 10/10 neg: 100%

 

Mono n=10

RPR spec: 9/10 neg: 90%

 

Preg women n=14

RPR spec: 12/14 neg:86%

 

Toxo n=20

RPR spec 19/20 neg:95%

 

CMV n=20

RPR spec 19/20 neg:95%

 

False pos CMIA results:

RPR spec 129/129 neg: 100%

Unclear gold standard

 

No stage of syphilis-no clinical data presented.

 

Small numbers

 

No comparison between groups

Limited relevance Medium
*De Lemos/Characterization of the western blotting IgG reactivity patterns in the clinical phases of acquired syphilis/Diag Microbio and ID/ 2007. Retrospective cross sectional study: Tested serum from 3 groups of patients in Brazil

 

1)”Syphilis group” N=122 sera from patients classified as having primary, secondary, early latent, late latent and tertiary syphilis.

Gold standard: “clinical, lab (FTA-ABS, TPHA, VDRL) and epidemiologic criteria.”

 

2) Blood donor group: N=250 sera samples from individuals screened by VDRL (We did not include any analysis from this group here as VDRL was part of the definition for diagnosis).)

 

3) “Other infections group”N= 135 sera samples from pts with other pathologies that might cause false pos (viruses, toxo, Chagas, malaria, toxocariasis, hansen’s, SLE)Gold standard for BFP: TPHA, FTA, WB TP-IgG negative but VDRL pos.

In the “other infections” group 7/134 patients were positive by VDRL but negative by TPHA, FTA and WB TP-IgG and were felt to be biological false positives. These were in patients with CMV infection, malaria and leprosy.

(Note there were 135 pts positive by VDRL but one was felt to be true positive because all treponemal tests were positive.)

 

 

 

Strength: Gold standard included 3 trep tests.

 

Small numbers in each disease group, so difficult to say what specific contribution CMV, malaria and leprosy might have.

Limited Relevance Medium
Harrison et al/An evaluation of the CSL RPR Test for Syphilis/ Aust J Med Technol/1976 Retrospective study:

N=200 syphilitic sera of various stages and treatment and N=500 sera from antenatal patients

Gold Standard: FTA-ABS and TPHA

Comparator: RPR and VDRL

 The RPR was more sensitive than the VDRL (174/200 vs. 167/200)

 

The RPR was more specific than VDRL (1 FP in 500 vs. 2 FP in 500)

 

No clinical stage data available Relevant High
Siletti/Comparison of CAPTIA Syphilis G Enzyme Immunoassay with RPR test for detection of syphilis/JClinMicro /1995 Cross sectional study, 2 groups:

N=646 blood samples submitted to laboratory which were not from STD clinics and N=265 blood samples from pregnant women on obstetrics unit taken immediately after delivery

 

Gold Standard:

“Active syphilis” diagnosed if:

1.Chart c evidence of diagnosis/specific therapy,

OR

2. RPR and FTA positive

OR

3. RPR negative and CAPTIA Syph M and CAPITIA Syph G positive

 

Excluded: Those which were RPR neg, CAPTIA Syph G neg and Capitia Syph M positive

 

Overall combining both populations: RPR sensitivity: 96.4%, specificity 97.5%

 

For the 646 blood samples not from STD clinics: RPR sensitivity: 95.6% (44/46 patients ‘with syphilis’ pos), specificity (585/600 patients ‘without syphilis’ neg) 97.5%

 

For the 265 samples from the obstetric patients: RPR sensitivity: 100% (10/10 patients ‘with syphilis’ positive), specificity 97.6% (249/255 patients ‘without syphilis’ neg)

Gold standard criteria were focused on evaluating CAPTIA Syph G (a treponemal test), not RPR (see excluded column 2)

 

Little clinical data on syphilis stage provided (though they say it was considered, however appears (though unclear from gold standard) if positive serologies were required)

 

Gold standards somewhat inexact.

Moderately relevant. Medium
Hare/ Serological Tests for Treponemal Disease in Pregnancy/J Obstet Gynecol/1973 Retrospective study: over 10 years in the UK in the antenatal setting.

N=42904 pregnant women tested

 

Gold standard: TPI or FTA

Of 42904 women tested 219 were consistently reactive; 205 were initially reactive but nonreactive thereafter (BFP) The sera of 49 women (22.4 per cent of patients

with positive tests) gave BFP reactions.

Data are poorly presented and difficult to extrapolate Limited relevance Medium
Brede HD/ Detection of BFP Syphilis Serum Reactions/SA Medical Journal/ 1974 Cross sectional study:

Sera from N=5271 persons from SA

2493 pregnant women, 1130

healthy prospective employees, 1345 newborn babies and

303 leprosy patients

Gold Standard: FTA-ABS

Comparator: RPR and VDRL

VDRL: 16% BFP

RPR: 20% BFP

The group of male and female employees showed roughly a third less false reactivity than pregnant women; higher BFP among those with leprosy.

No syphilis staging; background prevalence of non-syphilitic Treponematoses  

Limited relevance

Medium
Angue/ Syphilis serology testing: A comparative study of Abbot Determine RPR card test and VDRL methods/ PNG Med J/ 2005 Cross sectional study:

N=2100 women attending antenatal clinic in Papua New Guinea

Gold Standard: VDRL

Comparator Tests: Abbot Determine (treponemal test) and Abbot Syfacard-

R (Rapid Plasma Reagin (RPR) card test)

Determine versus VDRL: sensitivity 92.0%; specificity 94.6%; PPV 42.6%; and NPV 99.6%. RPR card versus VDRL: sensitivity 56.3%;

specificity 96.5%; PPV 41.2%; and NPV 98.1%

No clinical information; no treponemal testing; prevalence of non-syphilitic Treponematoses Limited relevance Low
Parco/Public banking of umbilical cord blood or storage in a private bank: testing social and ethical policy in NE Italy/J Blood Med/ 2013 Cross sectional study:

 

N=3450 pregnant patients screened for syphilis

 

Gold standard somewhat unclear,appears to be IgG/IgM western blot

VDRL BFP: 12/3450 tested. 0.3% Unclear gold standard

No comparator arm

Limited relevance Medium
Smikle/ BFP serological tests for syphilis in the Jamaican population/GU med/ 1990 Cross sectional study:

19067 sera screened with VDRL.

N=441 general pop,

N=145 pregnant women VDRL+ with a titer <1:8 were confirmed with FTA

 

Gold standard BFP: VDRL <1:8 with a negative FTA

General pop:

94/347 + VDRL with titer<1:8 were FTA neg (27%)

Pregnant women:

22/71 + VDRL with titer<1:8 were FTA neg

(31.0%)

Only tested if titer <1:8. There could have been false positives with a higher titer that would have been missed. Moderately relevant Low
Marangoni/ Laboratory dx of syphilis with automated immunoassays / J Clin Lab Analysis/ 2009 Cross sectional:

N=244 patients suffering from different stages of syphilis. The staging of the disease was done following clinical and

laboratory criteria

 

Gold standard: unclear, but appears to be + architect syphilis TP, and some clinical criteria

 

Second group n=74 sera from 1.culture-confirmed

Lyme disease (n=10), 2. clinical diagnosis of infectious mononucleosis detected

as positive by the Paul–Bunnel–Davidsohn agglutination

(n=10), 3.sera from pregnant women (n=14), 4.sera

from patients with cytomegalovirus acute infection (IgM

positive and low avidity) (n=520),  sera from patients

with acute toxoplasmosis (IgM positive and low avidity)

(n=20).

Gold standard : TPHA

 

Third group: N=129

sera selected from 9,210 samples submitted for routine screening for syphilis. All  were positive by ARCHITECT Syphilis TP, but negative by

TPHA and WB.

Gold standard: WB and TPHA.

 

For the Syphilitics: Gold standard unclear: clinical and lab

 

Syphilis n=244

RPR sens: 140/244: 57.4%

 

Lyme n=10

RPR spec: 10/10 neg: 100%

 

Mono n=10

RPR spec: 9/10 neg: 90%

 

Preg women n=14

RPR spec: 12/14 neg:86%

 

Toxo n=20

RPR spec 19/20 neg:95%

 

CMV n=20

RPR spec 19/20 neg:95%

 

False pos CMIA results:

RPR spec 129/129 neg: 100%

Unclear gold standard

 

No stage of syphilis-no clinical data presented.

 

Small numbers

 

No comparison between groups

Limited relevance Medium
Achimastos/Occurence of biologic false positive reactions with RPR card test on leprosy patients/PH reports /1970 Cross sectional study: from Athens, Greece/

N=50 leprosy patients without a self-reported history of syphilis tested with RPR (circle card test), the Kahn test and the FTA.

 

Gold standard: FTA test.

14/50 reactive by RPR card

13/50 reactive by Kahn

None reactive by FTA—all considered BFP.

Limitations: small sample size, no comparator group. Relevant Medium
Garner/The BFP reaction to serological tests for syphilis J Clin Path/ 1970 Cross-sectional retrospective study:

 

Several populations, but most data difficult to interpret. Relevant: N=270 patients with lepromatous leprosy from the phillipines were tested

 

Gold standard:FTA-ABS

25/270 samples tested were positive by a non treponemal test (not clear whether VDRL or RPR) and negative by FTA. 15/270 were positive by both treponemal and non treponemal tests. Unclear which non treponemal test used.

 

No comparator group.

Limited relevance Low
Buck et al. Comparative studies of the RPR Card Test for syphilis and the VDRL Slide Test in Ethiopia/Am J Hyg/1964 Cross sectional study:

N=50 Ethiopian men with a clinical diagnosis of early syphilis (6 with primary and 44 with secondary) with positive darkfield (gold standard)

Gold standard: clinical and positive darkfield

N=62 with lepromatous leprosy

N=46 with tuberculoid leprosy

Gold standard: FTA (for the above two)

Sensitivity RPR/VDRL Early syphilis: 96.8% both

Lepromatous leprosy RPR/VDRL: 47%/31% (FTA was 30.5%)

Tuberculoid leprosy RPR/VDRL: 8.3/9.2% (FTA was 21.3%)

While the data for early syphilis had a reasonable gold standard, the data for the leprosy patients are very limited Limited relevance Medium
Tuffanelli /Narcotic Addiction and False Positive Reaction for Syphilis/ Acta Dermato Venereologica/1968 Retrospective study:

Hereof 111 ex narcotic users, 29 had history of  BFP. 54 of the 111 participated in the study.

Gold Standard: FTA ABS

18/54 had repeatedly false positive VDRLs

Average duration of false positive was 25 months

Max BFP VDRL titer 1:64

Longitudinal data Relevant High
Cushman/Biologic false positive reactions in serologic tests for syphilis in narcotic addiction/AJCP/1974 Cross sectional study:

N=69 patients from methadone maintenance clinic tested initially and then over time—Retested at a mean of 23 +/- 7 mos of methadone maintenance

 

Unclear N: Controls using IVD: adolescent narcotic detox patients who were assumed to be actively using

 

Controls Normal: N=875 blood donors to blood bank

 

Gold standard BFP: FTA neg, VDRL pos.

24/69 positive by VDRL initially and of these 8 had positive FTA.

 

16/24 positive tests were considered BFP by VDRL.

 

Overall 23% (16/69) with BFP, in the initial using group

 

6/875 (0.7%) with BFP in the normal control group, P<0.001

 

They then report data for BFPs after methadone tx that are lower and compared to the adolescent narcotic users , but unclear if true positives were treated and how this affected subsequent data.

Small numbers, little clinical data.

 

Unclear numbers for controls.

Limited relevance Medium
Kaufman/biological false positive serological tests for syphilis among drug addicts/BJVD/1974 Cross sectional study:

Two groups:

Group 1:

N=1646 drug addicts from Kentucky-VDRL first, then if pos confirmed with FTA

Gold standard: FTA

Group 2:

N=197 drug addicts tested with VDRL, RPR, unheated serum reagin (USR) and automated reagin (ART) and FTA, auto FTA and MPHA

Gold standard: positive by any non treponemal test, and non-reactive by all 3 treponemal tests

Group 1:

VDRL:

143/1646 tests done were BFP’s

143/244 VDRL positives were BPF’s.

 

Group 2:

6.1% BFP’s

Group 2 data combines  non treponemal tests, Making data difficult to interpret.

 

Little clinical data.

 

No comparison group

Limited relevance Medium
Maves/False positive rapid RPR testing in pts with acute P. vivax malaria: A Case Control Study/Travel Med and ID 2014 Cross sectional study:

N=73 patients with P. vivax malaria and N=76 controls with other febrile illnesses.

Gold standard: TPHA

8.2% (6/73) of patients with malaria due to Plasmodium vivax in northern Peru had a BFP RPR. Range was up to 1:16; 0% BFP in controls. Small sample size Relevant High
Chuthanondh/ VDRL in Malaria/Annals of Int Medicine/ 1972 Longitudinal study:

N=31 patients with acute malaria had 3 blood draws each.

 

Gold standard: FTA Abs

1/31 patients had a positive VDRL with neg FTA abs. All three lab draws for this patient were VDRL+/FTAabs – Limited clinical information on this patient other than that they had acute malaria. Small sample size

No comparator group

Limited relevance Medium
Hernandez-Aguado/ False-Positive Tests for Syphilis Associated with HIV  and Hepatitis B Virus Infection among Intravenous Drug Abusers/ Eur J Clin Microbiol Inf Dis/1998 Prospective study:

N=5532 IVDA and N=820 gay men.

 

Gold standard: FTA-ABS or TPHA.

 

Study team obtained either serum RPR or VDRL

HIV: 10.7% BFP vs. 4.2%

HCV: 4.5% vs. 3.8%

HBV: 8.3% vs. 3.7%

HIV and HBV were both statistically associated with increased risk of BFP

Of the 229 (12.3%) IVDAs who had BFPs at their

first visit, only 47 of those 229 (20.5%) yielded a

BFP result again at the subsequent visit a median of 18 months later

Different gold standards and both RPR and VDRL were used- no data on differences Relevant High
Rompalo/Association of biologic false positive reactions for syphilis with HIV/JID/ 1992 Cross sectional study: all patients attending STD clinics for a new problem were asked to complete a questionnaire and blood was collected for testing. N=4863Gold standard: FTA After excluding 317 patients with reactive FTA-ABS tests,

 

BFP RPR tests were seen in 6 (4%) of 159 HIV-seropositive patients and 34 (0.8%) of 4387

HIV-seronegative patients (odds ratio, 5.0; 95% confidence interval, 1.9-12.7).

No clinical data Moderately relevant High
Augenbraun M, et al/Biological False-Positive Syphilis Test Results for Women Infected with HIV/ CID 1994 Retrospective study:  N=156 HIV+ women vs. N=633 HIV negative women in WIHS cohort

 

Gold Standard: MHA TP and FTA-ABS

Comparator test: RPR

6.9% and 0.2% of HIV-seropositive and HIV-seronegative women, respectively, had BFPs

(P < .001; odds ratio, 39.45; 95%confidenceinterval, 6.4-879.0). An association was found between injection drug use and BFPs for the HIV+ population

No longitudinal follow-up data; no syphilis  staging; Relevant Medium
Joyanes/The association of false positive RPR results and HIV infection/STD/1998 Prospective study:

N=2533 serum samples from spain, all tested with RPR, and if positive confirmed with FTA.

N=222 HIV+, N=2311 HIV-

Gold standard: RPR pos, FTA neg

BFP based on RPR: 15% in HIV positive, 1.2% in HIV neg group

 

Attributable risk for HIV: 14.97

 

 

 

Lacks a direct measure of whether the differences in the groups is statistically significant.

 

Moderately relevant. Medium
Glatt/High-titer positive nontreponemal tests with negative specific treponemal serology in patient with HIV infection or IV substance use/JAIDS/1991 Case series:  N=7 IVDUs (of which 6 were HIV+ ) and 1 HIV positive patient with high titer false positive VDRL with negative FTA

 

Gold standard: FTA

Illustrates that you can have a high titer false positive VDRL.

 

All patients had a titer >=1:16

Small case series, only one treponemal test as gold standard. Limited relevance I
Rusnak/ False pos rpr in hiv and relation to anti cardiolipin ab and serum immunoglobulin levels/JID/1994 Cross sectional retrospective study:

N=3371 periodic syphilis serology results from 1077 HIV seropositive patients

Gold standard: FTA

16/3371 tests found to be biologic false positive. No info on other conditions which might have contributed to BFP

No comparisongroup

Limited relevance Low
Thomas/Association of HCV infection with false positive tests for syphilis/JID 1994 Retrospective cross sectional study:

N=2672 patients attending an STD clinic

 

Of 400 HCV Ab +, 254 were RPR positive, and 231 were also positive by FTA-ABS

Gold standard: FTA-ABS

After excluding 231 patients with positive FTA-ABS tests, False positive RPR tests were found in 9/330 (2.7%) of HCV Ab positive patients and 14/2154 HCV negative patients (0.6%) p 0.0017. No clinical data. Relevant High
Sonmez/ False Positive Reaction Between Syphilis and Hepatitis C Infection/Isr J Med Sci /1997 Cross-sectional study:

N=21 syphilitic patients

N=50 HCV+ patients

N=50 “healthy controls”

 

Gold Standard: MHA-TP

10% (5/50) of patients with HCV had a positive VDRL with negative treponemal testing vs 0% in the 50 controls. Small number of patients Relevant

 

Medium
Dorwart/Comparison of RPR and VDRL in detection of BFP in SLE/Brit J VD/ 1974 Retrospective cross sectional study:

N=74 patients with connective tissue diseases, 41 of whom had SLE. Also tested 19 healthy blood donors and 23 who had syphilis were tested

All 74 connective tissue pts and healthy blood donors were FTA neg.

 

Gold standard: FTA Abs

7/74 with connective tissue disease RPR+

6/74 VDRL +

0/19 for blood donors RPR+, 0/19 VDRL +

23/23 with syphilis RPR and VDRL +

Small sample size Moderate relevance Medium
Jessop/False-positive test results for syphilis in relatives of a patient with SLE/BJVD 1979 Case series:

Gold standard: “no history or physical signs of syphilis”, also FTA

(Though authors did not consider as such)

In a family of 10, 2 patients had lupus and 4 had a positive VDRL with no history of lupus, 2/4 with positive VDRL also had a positive FTA, and an additional 1 /4 had a borderline positive FTA Poor gold standard, small numbers, no comparator group  

Limited relevance

I
Salo/low frequency of biologic false positive reactors to serological tests for syphilis in RA and AS/ Ann Rheum Dis /1968 Retrospective study:

 

N=14676 patients treated at a rheumatism hospital had Rhuematoid Arthritis, Ankylosing Spondylosis, SLE, DJD, rheumatic fever or some other joint disease

 

Gold standard: TPI or FTA neg and/or clinical history

Data is difficult to interpret, separating results for different diseases difficult

.

 

Out of 36 patients with + VDRL with RA or AK, none had both neg TPI and FTA.

 

SLE: 1/150 with SLE was a definite biologic false positive. (only five had a positive VDRL

test.

The TPI test was performed in four of them,

and two were TPI positive, one with no information

about syphilis and the other with congenital syphilis treated as a child. One of the two TPI-negative patients had syphilis treated at the primary stage and the other denied having had syphilis. The patient in whom the TPI test was not done had a serologically-positive husband.

Loose gold standard Limited relevance Medium
Chi/Molecular Differentiation of Treponema pallidum Subspecies in Skin Ulceration Clinically Suspected as Yaws in Vanuatu Using Real-Time Multiplex PCR and Serological Methods/AmJTropMed 2015 Cross-sectional study:

Conducted RTPCR to detect T. pallidum pallidum, T pallidum pertenue (yaws) and T pallidum endemicum (bejel) in skin lesions from N=155  children <15yo thought to have yaws in Vanuatu. RPR and TPPA were also done. 24 were positive for T. pertenue

 

Gold standard: RTPCR for different Treponema pallidum subsp. All children so lower risk for syphilis.

 

None tested positive for T. endemicum or syphilis

23/24 positive for TP pertenue PCR were RPR positive (interestingly these were all TPPA positive as well)

 

Overall 55/155 were positive for RPR.

Not clear exactly what sensitivity/specificity is for the RT-PCR. Relevant High
Schueller/ Immunizations and FP VDRL Tests in Military Recruits/ Military medicine /1976 Prospective study:

Gold Standard: FTA and clinical exam

Looked at VDRL changes following Influenza, meningococcal, Adenovirus, smallpox, tetanus, polio, and typhoid vaccines

1 of 263 healthy young recruits developed a BFP VDRL and it reverted to nonreactive 6 months later Important because they followed participants longitudinally Relevant High
Grossman/Biologically false positive serologic tests for syphilis due to smallpox vaccination/Am J Clin Path/ 1969 Prospective study:

Staff of a hospital got smallpox vaccination.

 

N=575 patients without a history of prior syphilis

 

Gold standard: Biological false positive if : (1) serologic tests prior to vaccination were negative, (2) there was no evidence of recent syphilis, and (3) serologic reactions reverted spontaneously from positive to

negative while under observation, or confirmatory tests (Reiter’s complement fixation) for syphilis gave negative results.

 

Samples with positive RPR had a repeat RPR, and if that was still positive a VDRL and Reiter’s complement fixation test.

10/575 false positives on RPR. All of thesethen either had a subsequent negative RPR or negative confirmatory test by Reiter’s complement fixation test. Reiter’s complement fixation no longer in use.

 

No modern treponemal test used as part of gold standard

Moderately relevant High
Harrison et al/An evaluation of the CSL RPR Test for Syphilis/ Aust J Med Technol/1976 Retrospective study: N= 200 syphilitic sera of various stages and treatment and 500 sera from antenatal patients

Gold Standard: FTA-ABS and TPHA

 The RPR was more sensitive than the VDRL (174/200 vs. 167/200)

The RPR was more specific (1 FP in 500 vs. 2 FP in 500)

 

No clinical stage data available Relevant High
Marangoni/ Laboratory dx of syphilis with automated immunoassays / J Clin Lab Analysis/ 2009 Cross sectional:

N=244 patients suffering from different stages of syphilis. The staging of the disease was done following clinical and

laboratory criteria

 

Gold standard: unclear, but appears to be + architect syphilis TP, and some clinical criteria

 

Second group n=74 sera from 1.culture-confirmed

Lyme disease (n=10), 2. clinical diagnosis of infectious mononucleosis detected

as positive by the Paul–Bunnel–Davidsohn agglutination

(n=10), 3.sera from pregnant women (n=14), 4.sera

from patients with cytomegalovirus acute infection (IgM

positive and low avidity) (n=520),  sera from patients

with acute toxoplasmosis (IgM positive and low avidity)

(n=20).

Gold standard : TPHA

 

Third group: N=129

sera selected from 9,210 samples submitted for routine screening for syphilis. All  were positive by ARCHITECT Syphilis TP, but negative by

TPHA and WB.

Gold standard: WB and TPHA.

 

For the Syphilitics: Gold standard unclear: clinical and lab

 

Syphilis n=244

RPR sens: 140/244: 57.4%

 

Lyme n=10

RPR spec: 10/10 neg: 100%

 

Mono n=10

RPR spec: 9/10 neg: 90%

 

Preg women n=14

RPR spec: 12/14 neg:86%

 

Toxo n=20

RPR spec 19/20 neg:95%

 

CMV n=20

RPR spec 19/20 neg:95%

 

False pos CMIA results:

RPR spec 129/129 neg: 100%

Unclear gold standard

 

No stage of syphilis-no clinical data presented.

 

Small numbers

 

No comparison between groups

Limited relevance Medium
Castro/Evaluation of an enzyme immunoassay technique for detection of antibodies against T. pallidum/JClinMicro 2003 Prospective cross-sectional

 

Gold standard: FTA-Abs positive, plus clinical hx.

 

Divided into 5 groups based on clinical hx:

25 primary syphilis

25 secondary syphilis

179 latent syphilis

105 “hx syphilis that had been correctly treated”

107 no clinical hx syphilis.

RPR Sensitivity in “past treated syphilis” : 55/95: 57.9%

 

Specificity:

“No clinical history of syphilis”, FTA negative:

RPR Negative:  95/107 (88.8)

Gold standard was FTA abs alone rather than two treponemal tests. Moderately relevant High
Wilkinson/A comparison of the FTAABS test and other screening tests for treponemal disease in patients attending a VD clinic/Jclinpath 1972 Cross sectional study:

N=1922 patients attending VD clinic were tested with FTAABS, RPCFT, VDRL, automated reagin and cardiolipin Wasserman

 

Gold standard: FTA-ABS

(N=107 were pos by FTA ABS)

VDRL +: 36/107 (33.6%)

RPR + 44/107 (41.1%)

No clinical data

No syphilis stage

FTA was gold standard, but later reported that specificity of FTA-ABS might be 73% in those which were only positive by FTA, which is somewhat confusing

Moderately relevant Medium
Sharma/comparision of tpha test and rpr card test, vdrl test and fta-abs test in diagnosis of syphilis/IJPM/ 1977 Cross sectional study:N=50 patients “suspected to be suffering from syphilis”

 

Of these 42 were positive with FTA and were “diagnosed as syphilitics”

 

Gold standard: FTA

VDRL+: 36/42 Sensitivity 89.0%

RPR tear drop card+: 40/42 Sensitivity 95.2%

 

Little clinical information, no syphilis stage  

Moderately relevant

Medium
White/Visuwell Reagin, a non-treponemal enzyme linked immunosorbent assay for the serodiagnosis of syphilis/JClinMicro /1989 Large cross-sectional study of patients from 3 different sites evaluating an automatable (but not automated) non treponemal test: Visuwell Reagin

 

Site A: n=1948 serum samples without clinical hx tested with RST, Visuwell Reagin and positives or discordants tested by MHA-TP.

 

Gold standard: MHA-TP (but only those positive or discordant by RST and Visuwell were tested)

 

Site B: n=1406 samples without clinical data tested with Visuwell, VDRL, and positives confirmed with FTA

Gold standard: FTA (but only the positives were tested)

 

Site C: N=311 “clinically characterized”samples:

Group 1: untreated syphilis: dx with prim, sec, or latent syphilis

Group 2: Treated syphilis (not using this group here)

Group 3: non syphilitic: “presumed non syphilitic” from i.patients with conditions other than syphilis (eg lupus, yaws), ii.BFPs that were pos by VDRL but neg by FTA, and iii.sera from donors with no hx prev or present syphilis infection.

Gold standard for site C “clinical data” not otherwise defined, except for Group 3 ii where FTA was gold standard.

Site A: (Sens and spec based on MHA-TP)

Visuwell: Sensitivity 25/26: 96.2%, Specificity: 1883/1891: 99.6%

 

RST (Reagin screening test): Sensitivity: 24/26: 92.3%, Specificity: 1878/1891: 99.3%

*But note that sensivity here is not truly based on MHA-TP as only those pos by Visuwell or RST were tested.

 

Site B: (Sens and Spec based on FTA)

Visuwell: Sensitivity 67/79: 84.8%, Specificity: 1311/1327: 98.8%

 

VDRL: Sensitivity: 57/79 72.2%, Specificity: 1320/1327 99.5%

*But note that sensitivity here is not truly based on FTA as only those pos by Visuwell or VDRL were tested.

 

 

Site C: the “non syphilitics”

Diseases other than syphilis :

VDRL+ 11/50

RPR + 7/50

Visuwell+ 4/50

 

BFP:

VDRL+ 44/44

RPR + 39/44

Visuwell+ 33/44

 

Normals:

VDRL+ 0/78

RPR + 0/78

Visuwell+ 0/78

 

 

Overall specificity for all these groups:

VDRL: 68%

RPR: 73.3%

Visuwell 78.5%

“Clinical data” loosely defined.

 

Sensitivity for Site A and B is not truly based on gold standard of MHA-TP or FTA but actually on conglomerate of Visuwell and RPR or Visuwell and VDRL.

Moderately Relevant. Medium
Stone/Capture-S, a Nontreponemal solid-phase erythrocyte adherence assay for serological detection of syphilis/JClinMicro /1997 Retrospective study: for Capture-S (non treponemal test)

Several populations:

Site 1:

N=366 reactive syphilis samples from treated and untreated patients in different stages of disease. Gold Standard criteria unclear other than this.

 

Site 2: n=1917 serum samples wth no clinical info tested with RPR and Capture S: all reactives were confirmed with FTA.

Gold standard: FTA (but note  that only positives were tested)

 

Site 3: 1932 blood donor samples tested with Capture S, all pos confirmed with FTA.

Gold Standard for spec: FTA (note that only positives were tested with FTA)

 

Site 4: plasma center n=1193 donors tested by RPR and Capture S. All reactive confirmed with FTA.

Gold standard: FTA (but note that only positives were tested)

 

Site 5: 1425 blood donors tested with RPR and Capture S, reactive sera tested with FTA.

Gold standard: FTA (but note that only positives were tested)

Site 1: Note poorly defined gold standard for this site, difficult to interpret.

Cap S

Sens 80.2

Spec 96.2

 

RPR

Sens 79.9

Spec 98.7

 

 

Site 2:

Cap-S

Sens 94.6

Spec 98.7

RPR

Sens 94.1

Spec 98.1

 

Site 3

Cap-S

Sens 100

Spec 99.3

RPR

Sens 100

Spec 99.5

 

Site 4

Cap-S

Sens 93.8

Spec 98.6

RPR

Sens 43.8

Spec 99.7

 

Site5

Cap-S

Sens 83.3

Spec 99.9

RPR

Sens 100

Spec 99.9

Somewhat poorly defined gold standards, very poor for Site 1 Mod relevant Medium
Angue/ Syphilis serology testing: A comparative study of Abbot Determine RPR card test and VDRL methods/ PNG Med J/ 2005 Cross sectional study:

N=2100 women attending antenatal clinic in Papua New Guinea

Gold Standard: VDRL

Comparator Tests: Abbot Determine and Abbot Syfacard-

R (Rapid Plasma Reagin (RPR) card test)

Determine versus VDRL: sensitivity 92.0%; specificity 94.6%; PPV 42.6%; and NPV 99.6%. RPR versus VDRL: sensitivity 56.3%;

specificity 96.5%; PPV 41.2%; and NPV 98.1%

No clinical information; no treponemal testing; prevalence of non-syphilitic Treponematoses Limited relevance Medium
Fowler E/ A comparison of 4 screening tests for the detection of syphilis/Canadian Journal of Public Health/1976 Cross sectional study:

N=6488 sera from Canadian STD clinics were screened: 354 had syphilis. No stage provided

Gold Standard: FTA and RPCF

Sensitivity of RPR: 87.8%

Sensitivity of VDRL: 89.8%

The VDRL was the least specific test compared to RPR but we could not calculate an exact value

Limited data available; unclear how syphilis cases were defined; no data on syphilis stage Limited relevance Medium
Stevens et al. Evaluation of Reagin Card Tests for Syphilis/HLS / 1978 Cross sectional study:

Compared RPR, VDRL

N=2300 samples

Gold Standard a reactive FTA-ABS.

Sensitivity of RPR: 71.9%

Sensitivity of VDRL: 64.6%

No clinical data available Limited relevance Medium
Williams/ Reproducibility of Syphilis Serology Results/AJCP/ 1980 Longitudinal study: Repeat testing of RPR and VDRL sera  used by CAP proficiency testing

 

 

Gold Standard: Repeat testing with RPR and VDRL

Good agreement for 9 months for specimens nonreactive and strongly reactive by VDRL and RPR. Those that were weakly reactive showed a drop in agreement by 3 months.

 

No clinical data Relevant High
Harrison et al/An evaluation of the CSL RPR Test for Syphilis/ Aust J Med Technol/1976 Retrospective study:

N=200 syphilitic sera of various stages and treatment and N=500 sera from antenatal patients

Gold Standard: FTA-ABS and TPHA

Comparator: RPR and VDRL

 The RPR was more sensitive than the VDRL (174/200 vs. 167/200)

 

The RPR was more specific than VDRL (1 FP in 500 vs. 2 FP in 500)

 

No clinical stage data available Relevant High
Malm/ Analytical evaluation of nine serological assays for

Diagnosis of syphilis/ JEADV/2015

Cross sectional:

Serum samples from Guinea-Bissau and Sweden were examined, as well as two performance panels and samples from blood donors

N=595 samples tested with VDRL and RPR

Gold Standard: Macro-Vue RPR

Card test

Compared to the MacroVue RPR the VDRL was less sensitive and specific: Sensitivity 76.1% (70.0-82.2) and Specificity 93.0% (90.8-95.2) No clinical data Relevant Medium
Muic/ Limitations and consequences of basing late syphilis seroassessments on VDRL Test/ Acta Dermato-venereologica/1997 Partial Mathematical modeling approach to estimating the potential miss of a VDRL test compared to the Gold Standard TPHA in missing cases of untreated Late or Late Latent syphilis

N=265,

Gold standard: TPHA and modelling

False negative VDRLs in the clinical sample of 265 patients: 15.5%

Model Estimate of FN VDRL: 21%

Modeling data may be limited Relevant Medium
Saral/Serologic diagnosis of syphilis: comparison of different diagnostic methods/acta dermatovernerol Croat/ 2012 Cross sectional study:

N=4226 serum specimens, 117 patients diagnosed with syphilis:

 

Gold standard: “Diagnosed with syphilis based on clinical, epidemiological, and laboratory methods”, (presumably referring to early syph cases) “the dx was confirmed by examining the material from a chancre using a darkfield microscope or blood test”

Later says that TPHA may be gold standard.

Sensitivity of RPR “against TPHA” 58%. Overall for all stages of syphilis. Specificity of RPR is reported as 0% but how this was calculated was not reported. Unclear gold standard definition, no clear clinical information

 

 

Limited relevance Low
Cate/sensitivity and specificity of automated serologic tests for syphilis/AJCP /1971 Retrospective cross sectional study examining VDRL and an automated reagin (VDRL based) (nontreponemal) test.

N=139 sera reactive by FTA-ABS

N=315 sera non reactive by FTA-ABS

 

Gold Standard for late or latent syphilis: FTA-ABS + some clinical (not well defined) available for 90% of patients.

 

Gold standard for specificity was FTA negative.

 

.

 

 

Sensitivity for “late or latent syphilis”:

VDRL+102/139 (73%)

Automated reagin using VDRL + 98/139: 71%

 

General:

Specificity : VDRL neg:  303/315 (96%)

 

Specificity: Automated reagin (VDRL) neg: 305/315: 97%

Unclear how they distinguished late vs latent syphilis.

 

 

 

 

 

 

 

 

 

Limited relevance

 

 

 

Low
Siletti/Comparison of CAPTIA Syphilis G Enzyme Immunoassay with RPR test for detection of syphilis/JClinMicro /1995 Cross sectional study, 2 groups:

N=646 blood samples submitted to laboratory which were not from STD clinics and N=265 blood samples from pregnant women on obstetrics unit taken immediately after delivery

 

Gold Standard:

“Active syphilis” diagnosed if:

1.Chart c evidence of diagnosis/specific therapy,

OR

2. RPR and FTA positive

OR

3. RPR negative and CAPTIA Syph M and CAPITIA Syph G positive

 

Excluded: Those which were RPR neg, CAPTIA Syph G neg and Capitia Syph M positive

 

Overall combining both populations: RPR sensitivity: 96.4%, specificity 97.5%

 

For the 646 blood samples not from STD clinics: RPR sensitivity: 95.6% (44/46 patients ‘with syphilis’ pos), specificity (585/600 patients ‘without syphilis’ neg) 97.5%

 

For the 265 samples from the obstetric patients: RPR sensitivity: 100% (10/10 patients ‘with syphilis’ positive), specificity 97.6% (249/255 patients ‘without syphilis’ neg)

Gold standard criteria were focused on evaluating CAPTIA Syph G (a treponemal test), not RPR (see excluded column 2)

 

Little clinical data on syphilis stage provided (though they say it was considered, however appears (though unclear from gold standard) if positive serologies were required)

 

Gold standards somewhat inexact.

Moderately relevant. Medium
Theodoropoulos/Improving Syphilis Screening in Deceased Organ Donors/Transplantation/ 2015 Retrospective cross sectional study:

N=32 RPR positive donors were matched with N=61 RPR negative donors.

All RPR positive donors and matched negative donors were retested with two alternate RPR tests, an FTA, and MHA, a TPPA and a CLIA.

 

Gold standard for syphilis: positive TPPA.

*For RPR the sensitivity and specificity were based on having one of three RPR tests positive. (Initial evaluation was with the Macro-Vue RPR Card Test(BD), they were retested with the REMEL RPR card and the ASI RPR Card test.)

RPR sensitivity (again based on one of 3 RPR tests being positive) 79.2% (CI 57.2-92.8), Specificity

RPR Specificity  Specificity 81.2%(69.9-89.6)

No clinical data.

Somewhat unclear gold stanadards  for specificity and sensitivity.

Moderately relevant Medium
Franken/Comparison of a combined non trep (VDRL) and trep immunoblot to traditional nontrep and trep assays/ JClinLabAnal /2015 Cross sectional study:

Tested 97 RPR positive and 101 RPR negative samples with a combined IgM/IgG immunoblot assay that detects VDRL and treponemal antibodies.

 

Gold standard: RPR

Sensitivity of VDRL in this combo assay compared to RPR:

77.3% (70.2-83.4)

Specificity of VDRL in this combo assay compared to RPR: 71.3% (64.4-77.1)

 

Assay not in use, no clinical information. Limited relevance NR
Harris/Comparative reactivity of the VDRL slide and other tests for syphilis in random population groups/ /1954

 

Cross sectional study:

The interpretable data in this study is from N= 466 sera from random blood collections in south carolina on  which TPI and VDRL were done

 

Gold standard: TPI

VDRL:

Sensitivity for any stage:

VDRL+ or Weakly pos: 177/183 which were pos by TPI

 

Specificity:

VDRL neg: 199/ 211 which were neg by TPI.

 

No clinical data Limited relevance Medium
Gibowski/Detection of specific IgM-CLASS antitreponemal antibodies in blood

serum of patients with syphilis with the use of CAPTIA Syphilis-M

reaction and comparing it with VDRL, FTA-ABS and TPHA reactions/ Med Sci Monit/1998

Cross sectional study:

N=200 Polish patients clinically characterized with syphilis. None of the early syphilis subjects were treated

 

Gold standard: Clinical staging + FTA-ABS, TPHA, Captia Syphilis M

Primary (18) sensitivity: VDRL: 55.6%

Recent secondary (17): 100%

Recurrent Secondary (44): 100%

Early Latent (34): 100%

Late Latent (44): 63.6%

ASN (29): 89.7%

Symptomatic neurosyphilis (17): 77.8%

Good gold standard; small numbers Relevant High
Glicksman/Instant Syphilis Screening/Texas Medicine/1967 Cross sectional study:

N=575 patients seen at an STD clinic in TX.

Gold Standard: Clinical staging + VDRL

Primary (19): RPR 79%

Secondary (31): RPR 100%

Untreated latent (87): 94%

Previously Treated Latent (69%): 78%

Untreated Congenital (2): 100%

Previously treated Late (10): 100%

Limited data on timing of treatment; no confirmatory treponemal testing Relevant High
*Creegan/An evaluation of the relative sensitivities of the VDRL and The TPPA test among pts diagnosed with primary syphilis/STD/ 2007 Retrospective cross sectional study: on patients in San Francisco, N=106 cases diagnosed by positive DF test.

 

Excluded 62 not diagnosed with positive DF test, 19 with a prior hx of syphilis, and 28 because either the VDRL or TPPA test was not done on day of dx.

 

31% HIV +

 

Gold standard: DF microscopy.

77/106 had +VDRL test (sensitivity of 72.6%)

 

RPR tests were available for 51 cases.  37/51 were positive. (sensitivity 72.5%)

 

 

If a strategy of VDRL confirmed by TPPA was used, sensitivity was 71%. In HIV positive patients using this strategy, sensitivity was 55%. (No information on how many HIV + were + by VDRL unconfirmed by TPPA)

Relativelyrestrictive definition of primary syphilis (Darkfield positive). Relevant High
*Bossak/assay of tests for syphilis on unheated serum/PHR/1960 Retrospective cross-sectional study: testing done on serum bank of clinically categorized donors. N=119 primary syphilis

 

Note: data was available on BFP and other stages but these other stages were treated.

Gold standard: Darkfield examination

Conducted unheated RPR (USR), Rapid RPR, and VDRL.

 

Sensitivity: primary syphilis untreated patients

USR: 85/119: 71.4%

RPR-US: 86/119: 72.3%

VDRL: 79/119: 66.4%

Primary syphilis defined by darkfield positivity Relevant High
*Dyckman/Evaluation of Reagin Screen, a new serological test for syphilis/JCM 1976 Cross-sectional study:

Serum from N=243 patients with untreated primary, secondary, and latent syphilis and N=354

patients previously treated for syphilis were examined by the four tests.(VDRL, RPR, Reagin Screen “RST”, FTA-ABS)

*note also tested “normals” and those with other conditions thought to evoke a false positive RPR, but not possible to abstract from the tables which were positive by VDRL and not by FTAABS.

 

Gold standard:

(i) primary syphilis:primary, dark-field positive chancre, and no symptoms of secondary

syphilis;

(ii) secondary syphilis:-dark-field positive secondary lesions or multiple symptoms

of secondary syphilis such as condyloma lata, alopecia, and lymphadenopathy; and

(iii)latent syphilis: no physical signs of syphilis, no history of treatment for syphilis, AND one reactive

nontreponemal test, AND a reactive FTA-ABS.

Sensitivity: Primary syphilis:

VDRL+: 70/111: 63.1%

RPR+: 72/111: 64.8%

RST+: 64/111: 57.7%

 

 

Note: non treponemal test is part of the gold standard definition for latent syphilis

Included clinical information. Relevant High
*Dyckman/Clinical evaluation of a new screening test for syphilis/AmSocClinPath 1977 Cross sectional. Tested serum from presumed healthy individuals and syphilitic patients Tested with RPR, VDRL, Syphlacheck (lipid ag based)

N=141 with untreated syphilis of various stages reported on here.

Gold standard:

(1) primary syphilis—primary, darkfield-positive

chancre and no symptom of secondary syphilis; (2)

secondary syphilis—darkfield-positive secondary lesions

or two or more symptoms of secondary syphilis such as

condylomata lata, alopecia, and lymphadenopathy; (3)

latent syphilis—no physical sign of syphilis, no history

of treatment for syphilis, AND one reactive nontreponemal

test, AND a reactive FTA-ABS test

Sensitivity Primary syphilis:

VDRL+: 50/80 62.5%

RPR+: 50/80 62.5%

SyphlaCheck+: 53/80 66.3%

 

 

Included clinical information Relevant High
*Dyckman/Comparison of Serum and Plasma Specimens for Syphilis Serology Using the Reagin Screen Test/J Clin Micro /1980 Cross sectiona study:

N=140 patients with treated/untreated syphilis reported on here.

 

Gold Standard Criteria for Primary and Secondary syphilis are as above in the first two studies.

Sensitivity: Primary syphilis:

VDRL+ 48/63= 76.2%

RST+ w/o anticoag 49/63= 77.8%

RST+ w/ EDTA 50/63= 79.4%

RST+ w/ citrate 50/63= 79.4%

RST+ w/heparin 49/63=77.8%

 

Same caveats as above study apply to the latent syphilis category and the ‘healthy’ individuals

Clinical data included Relevant High
*Dyckman/Reactivity of MHA-TP, FTA &VDRL in primary syphilis/JCM /1980 Cross sectional study:

N=130 cases of primary syphlis

Gold standard:

DF positive primary chancre

Sensitivity Primary syphilis

VDRL: 89/130=68.5%

Clinical data included Relevant High
*Falcone/ Evaluation of the RPR card test/PHreports/ 1964 Cross sectional study:

 

Gold standard:

Presumed normal N=260. Donors with no history of previous or present infection with syphilis.

However this data was not verified with a treponemal test and not included in this review)

 

Syphilis Primary (N=134) and Secondary(N=217)

 

Gold standard: primary and second¬

ary syphilis: “proved by darkfield examination, who had received no treatment.”

 

Donors with

latent, late, and congenital syphilis who were untreated, inadequately treated, or adequately treated. ( However these were not included in this review as they were not verified with a treponemal test)

 

Biologic false positive (BFP). Donors with reactive nontreponemal tests with no clinicalevidence of syphilis and who had no previous

Treponema pallidum immobilization (TPI)test (Again not included as not verified with treponemal test

Primary syphilis sensitivity:

RPR+ 102/134=76.1%

VDRL+105/134=78.4%

Clinical data included. Relevant High
*Greaves/ A comparative study of serologic tests in early syphilis/arch of derm/1962 Cross sectional study

N=13 primary, N=16 secondary syphilis cases

Gold standard: Primary: dark field positive primary Secondary: darkfield positive

Sensitivity primary syphilis

VDRL+ 10/13: 76.9%

 

Small sample size Moderately relevant Medium
*Huber/Reactivity of microhemagglutination, FTA, VDRL and RPR in primary syphilis J clin micro/ 1983 Cross sectional study

N=109 primary syphilis

Gold standard: primary syphilis: positive darkfield

Sensitivity primary syphilis

RPR 101/109 (92.7%)

VDRL 79/109 (72.5%)

Clinical data included Relevant High
*Lassus/ the order of appearance of reactivity to treponemal and lipoidal tests in early syphilis/ Acta path et microbial scandinav 1967 Cross sectional study

N=62 Primary

Gold standard:

Primary syphilis: DF positive

N=14 Secondary syphilis: but unclear definition, not included in this review.

Primary syphilis:

VDRL+ ( Sensitivity) 63%

Clinical data included Relevant. High
*Moore/Sensitivity and Specificity in Syphilis Serology/Southern Medical Bulletin/ 1965 Cross sectional retrospective.

N=76 primary syphilis

 

Gold standard:

Primary syphilis: clinical dx with +

 

Secondary syphilis: Clinical diagnosis with + DF (N=100)

 

Non-infected: two groups

1.”Presumeably healthy individuals who underwent a careful screening procedure” N=74)

 

2. Non syphilitic controls from the original Tuskegee study (N=36)

 

3. “Patients diagnosed as being acute or chronic biologic false positive reactors” (N=38)

 

Unclear stage syphilis:

Syphilitic patients from the Tuskegee study (N=46)

 

 

 

Primary syphilis: Sensitivity:

VDRL+ 50% (n=76 cases primary syphilis)

KRP: 48.7%

 

Clinical data included Relevant High
*Wende/The VDRL Slide Test in 322 cases of DF positive primary syphilis/1971 Retrospective cross sectional study:

N=322 cases of primary syphilis

 

Gold standard: DF+ of lesions of primary syphilis

VDRL +: 236/322: Sensitivity 73.3% Clinical data included Relevant High
*Sischy/Syphilis serology in patients with primary syphilis and non-treponemal stds in southern Africa/GUMed /1991 Cross sectional study:

N=1170 black mine workers presenting with acute urethritis or genital ulceration had physical exam and RPR and FTA performed

 

Primary syphilis:

Gold standard: DF on ulcers: n=21 primary syphilis cases DF+ on ulcers.

 

Others: BFP defined as positive RPR and negative FTA with “no other signs or symptoms of syphilis”

Of n=21 patients with DF+ lesion, 15/21 + by RPR,

RPR sensitivity 71%.

Small numbers Moderately relevant High
*Talwar/VDRL titres in early syphilis before and after treatment/ GUmed /1992 Cross sectional study:

N=1008 cases of primary syphilis

 

Gold standard: unclear, but for 85 seronegative was +DF for primary syphilis.

 

Gold standard unclear for secondary and latent, so did not include, but all of these were VDRL positive.

Sensitivity primary syphilis:

VDRL+: 923/1008: 91.6%

 

 

Gold standard somewhat unclear. Moderately relevant High
*Gibowski/Detection of specific IgM-CLASS antitreponemal antibodies in blood

serum of patients with syphilis with the use of CAPTIA Syphilis-M

reaction and comparing it with VDRL, FTA-ABS and TPHA reactions/ Med Sci Monit/1998

Cross sectional study:

N=200 Polish patients clinically characterized with syphilis. None of the early syphilis subjects were treated

 

Gold standard: Clinical staging + FTA-ABS, TPHA, Captia Syphilis M

Primary (18) sensitivity: VDRL: 55.6%

Recent secondary (17): 100%

Recurrent Secondary (44): 100%

Early Latent (34): 100%

Late Latent (44): 63.6%

ASN (29): 89.7%

Symptomatic neurosyphilis (17): 77.8%

Good gold standard; small numbers Relevant High
*Glicksman/Instant Syphilis Screening/Texas Medicine/1967 Cross sectional study:

N=575 patients seen at an STD clinic in TX.

Gold Standard: Clinical staging + VDRL

Primary (19): RPR 79%

Secondary (31): RPR 100%

Untreated latent (87): 94%

Previously Treated Latent (69%): 78%

Untreated Congenital (2): 100%

Previously treated Late (10): 100%

Limited data on timing of treatment; no confirmatory treponemal testing Relevant High
*Gratzer/Evaluation of Diagnostic Serological Results in Cases in Suspected Primary Syphilis infection/STD /2014 Retrospective case series:

 

Gold standard: clinical signs or symptoms of primary syphilis AND at least 1 positive/reactive syphilis serological result (either positive FTA or both TS-EIA and RPR positive) AND no reported and/or known history of syphilis infection

Primary syphilis: RPR: 40/52 positive (Sensitivity 76.9%)

 

Two cases had an equivocal TS-EIA, if these were excluded sensitivity was 76%.

Limitation: a positive serology (whether EIA, TS-EIA or RPR ) part of gold standard definition (and inclusion in study—partially circular definition).

 

Strength: clinical info.

Relevant High
*Singh/Characteristics of primary and late latent syphilis cases which were initially non-reactive with the RPR as screening test/Int J STD AIDS/2008 Retrospective case series querying Alberta Canada STI notification database. All RPR tests were confirmed with FTA-ABS or MHA-TP

 

N=863 primary cases  Excluded those in which initial serology occurred prior to study period, those whose initial test was not an RPR, those who did not have an RPR titer reported, those who had a prior diagnosis of syphilis and those who had incomplete demographic or clinical information.

 

Gold standard: positive FTA-ABS or MHA-TP, diagnosis of “primary syphilis” based on clinician’s judgement.

639/863 cases were positive by RPR. (sensitivity: 74%) Restrictive definition of primary syphilis: all cases had to have a positive non-treponemal test.

 

Little clinical information

Relevant High
*Castro/Evaluation of an enzyme immunoassay technique for detection of antibodies against T. pallidum/JClinMicro 2003 Prospective cross-sectional

Divided into 5 groups based on clinical hx:

25 primary syphilis

25 secondary syphilis

179 latent syphilis

105 “hx syphilis that had been correctly treated”

107 no clinical hx syphilis.

 

Gold standard: FTA-Abs positive and clinical history

 

 

Sensitivity RPR+ Primary syphilis: 23/25 (92) Small sample size, FTA-ABS positive may be too restrictive. Moderately relevant High
*Ballard RC et al./The influence of concomitant HIV infection on the serological diagnosis of primary syphilis in Southern Africa/ SAMJ /2007 Cross sectional study

868 patients with GUD enrolled in S Africa.

 

Gold Standard: Multiplex PCR for T. pallidum, H. ducreyi, CT, and HSV

T. pallidum (TP)  was detected in 163 patients by RT-PCR

Sensitivity of RPR HIV+: 81.8%

Specificity of RPR HIV+:90.6%

Sensitivity of RPR HIV-: 78.6%

Specificity of RPR HIV-: 87.3%

The sensitivity of the gold standard use is problematic. Relevant High
*McMillan/ Qualitative and quantitative aspects of the serological

diagnosis of early syphilis/ Int J STD AIDS/2008

Prospective study: of untreated patients with early syphilis.

N=89 primary, 68 secondary, 72 early latent

Gold standard: Clinical staging (occasional darkfield or PCR confirmation for primary syphilis) and “treponemal tests.”

Early latent syphilis stage defined as infection within two years.

Primary (n=89): VDRL reactive in 67%

Secondary (n=68): VDRL reactive in 100%

Early Latent (n=72): VDRL reactive in 85%

Some clinical data. Relevant High
*De Lemos/Characterization of the western blotting IgG reactivity patterns in the clinical phases of acquired syphilis/Diag Microbio and ID/ 2007. Retrospective cross sectional study: Tested serum from 3 groups of patients in Brazil

 

1)”Syphilis group” N=122 sera from patients classified as having primary, secondary, early latent, late latent and tertiary syphilis.

Gold standard: “clinical, lab (FTA-ABS, TPHA, VDRL) and epidemiologic criteria.”

 

2) Blood donor group: N=250 sera samples from individuals screened by VDRL (We did not include any analysis from this group here as VDRL was part of the definition for diagnosis).)

 

3) “Other infections group”N= 135 sera samples from pts with other pathologies that might cause false pos (viruses, toxo, Chagas, malaria, toxocariasis, hansen’s, SLE)Gold standard for BFP: TPHA, FTA, WB TP-IgG negative but VDRL pos.

In the “syphilis group” primary syphilis

VDRL+ 9/11 sensitivity of 81.8%

 

 

Limitations: small sample size, unclear gold standard definition—may be circular if VDRL is included in definition.

 

Strength: does include some clinical information, but little detail (e.g. darkfield information)

Limited relevance Medium
Thakar/Seroprevalence of syphilis by TPHA test/Ind J Pathol Microbiol /1996 Cross sectional study:

N=1074 sera tested in india by TPHA and VDRL, 408 from “clinically suspected” cases of syphilis, 302 from patients with primary syphilis, 48 secondary syphilis 58 tertiary syphilis

 

Also: 111 specimens from other STD patients, 237 from pregnant women and 318 from “apparently healthy”

 

Gold standard: unclear  definitions for the “other std” pregnant and “apparently healthy”

 

Gold standard for primary, secondary and tertiary is “clinically suspected”. No other details provided.

 

% (including weakly reactive with reactive)

 

Primary syphilis

Sensitivity VDRL+: 231/302=76.5%

 

Secondary syphilis

Sensitivity VDRL+: 35/48=72.9%

(*note that 26/48 were TPHA pos)

 

Tertiary syphilis

Sensitivity VDRL+: 37/58-63.8%

Poorly defined gold standard.

 

Limited relevance Medium
*Backhouse/T. pallidum western blot: comparison with the FTA-ABS test as a confirmatory test for syphilis/ Diag micro and ID/ 2001 Retrospective cross-sectional study.

 

N=278 including 98 reactive sera from patients with documented treponemal infection.

 

Gold standard: unclear: appears to be “clinical plus serologic.”

 

N=11 untreated patients with primary syphilis, N=10 untreated patients with secondary syphilis, N=6 untreated patient with early latent syphilis, N=12 patients with untreated late latent syphilis.

Primary syphilis:

11/11 VDRL+: Sens 100%

Unclear gold standard, may include VDRL.

 

Data for BFPs not included at this study was mostly focused on treponemal tests.

Limited relevance Medium
*Brown/Evaluation of RPR card test for syphilis screening in field investigations/ PHReports /1964 Cross-sectional study: from tests done at VD clinics in Texas.

N=121 primary

N=148 secondary

Gold standard:  somewhat unclear but for primary and secondary syphilis appears to be physical exam, DF was not mentioned to be necessary

 

Data was also included on Latent and “non syphilitic” patients but this was not included as gold standard was unclear.

Primary syphilis untreated

Sensitivity:

RPR+: 102/121: 84.3 %

VDRL+ 73/121: 60.3%, but 13 more were WR. If include these, 86/121: 71.0%

 

Secondary syphilis untreated

Sensitivity:

RPR+: 148/148: 100%

VDRL+: 147/148: 99.35 but 1 more was WR if include that one 148/148: 100%

Little data on what constituted clinical definition, DF was not mentioned to be required.

Unclear gold standard

Moderately relevant. Medium
*Dandoy/Initial serological reactions in infectious syphilis/BJVD /1967 Retrospective cross sectional of diagnosed cases of primary, secondary and early latent syphilis from VD clinic in Los Angeles.

N=94 primary syphilis

 

Gold standard: somewhat unclear, but for primary and secondary syphilis included physical exam and DF (though some cases of both were DF negative, so unclear if dx in these cases was based solely on physical exam or if serologies played a role.

 

Gold standard early latent: too unclear: appears to be Kolmer, but one case was Kolmer neg so may include VDRL—circular, won’t use here.

Sensitivity VDRL primary syphilis: VDRL+

67/94: 71.2%

 

 

Gold standard unclear here—may have included serologies, so somewhat circular, though there were definitely DF pos primary syphilis cases which were VDRL neg. Limited relevance Medium
*Dang/Evaluation of specific ab’s for early dx and management of syphilis/ Int J Derm/ 2006 Retrospective cross sectional

N=67 patients, 20 pts with primary syphilis, 47 with secondary syphilis

 

Gold standard:

Primary syphilis: “in whom the diagnosis of syphilis was made by

clinical, epidemiological and laboratorial methods,”

 

 

 

Sensitivity RPR primary syphilis:

12/20 (60%)

Poorly defined gold standard, may include RPR. Limited relevance Low
*Portnoy: RPR with unheated serum and new improved antigen suspension/PHR/1961 Retrospective cross sectional

Done on serums with “known clinical categories”

 

Tested with VDRL slide test, also tested with the RPR using unheated serum rather than plasma

N=48 primary syphilis

 

Gold standard: loosely defined/ unclear: “known clinical categories”

Did not include data onlatent, presumed non syphilitic and BFP, the as the definitionswere too unclear.

Here we include all those which were weakly reactive as reactive:

 

Sensitivity primary syphilis untreated:

VDRL + 29/48: 60.4%

 

RPR + A: 25/48: 52.1%

RPR + B: 25/48:52.1%

RPR+ C: 26/48:54.2%

 

Sensitivity secondary syphilis untreated:

VDRL + 25/25 100%

RPR+ A:24/25: 96%

RPR+ B:24/25:96%

RPR+ C:24/25: 96%

Poorly defined gold standard. Limited relevance Low
*Stone/Capture-S, a Nontreponemal solid-phase erythrocyte adherence assay for serological detection of syphilis/JClinMicro/ 1997 Retrospective for Capture-S (non trep)

Cross sectional in several populations:

Site 1:

N=366 reactive syphilis samples from treated and untreated patients in different stages of disease. Gold Standard criteria unclear other than this.

 

Site 2: n=1917 serum samples wth no clinical info tested with RPR and Capture S: all reactives were confirmed with FTA.

Gold standard: FTA (but only tested positives)

 

Site 3: 1932 blood donor samples tested with Capture S, all pos confirmed with FTA.

Gold Standard for spec: FTA (but only tested postivies)

 

Site 4: plasma center n=1193 donors tested by RPR and Capture S. All reactive confirmed with FTA.

Gold standard: FTA, but only tested positives

 

Site 5: 1425 blood donors tested with RPR and Capture S, reactive sera tested with FTA.

Gold standard: FTA but only tested positives.

Site 1:

 

Primary untreated n=29

CaptureS Sens: 88.5%

RPR: 96.2%

 

Secondary untreated: n=50

CaptureS Sens: 91.8%

RPR: 97.9%

 

Latent untreated: n=41

CaptureS Sens: 94.9%

RPR: 92.3%

 

Poorly defined gold standards. Moderately relevant. Medium
*Moore/Sensitivity and Specificity in Syphilis Serology/Southern Medical Bulletin/ 1965 Cross sectional retrospective.

N=100 patients with secondary syphilis

 

Samples tested with KRP (Kolmer test) and VDRL

 

 

Gold standard:

Primary syphilis: clinical dx with + DF (N=76)

 

Secondary syphilis: Clinical diagnosis with + DF (N=100)

 

Non-infected: two groups

1.”Presumeably healthy individuals who underwent a careful screening procedure”N=74)

 

2. Non syphilitic controls from the original Tuskegee study (N=36)

 

3. “Patients diagnosed as being acute or chronic biologic false positive reactors” (N=38)

 

Unclear stage syphilis

 

Syphilitic patients from the Tuskegee study (N=46)

 

 

 

Secondary syphilis: Sensitivity:

VDRL+: 100%

KRP: 91%

 

Clinical information included Relevant High
*Bossak/assay of tests for syphilis on unheated serum/PHR/1960 Retrospective crosssectional: testing done on serum bank of clinically categorized donors.

 

Gold standard: Darkfield examination

Conducted unheated RPR (USR), Rapid RPR, and VDRL.

 

Sensitivity: secondary syphilis untreated patients

USR: 98/98: 100%

RPR-US: 98/98: 100%

VDRL: 98/98: 100%

 

**They also have some info on bio false pos and lateent and late but latent and late were treataed. Also on primary treated and secondary treated but I did not include those. High
*Castro/Evaluation of an enzyme immunoassay technique for detection of antibodies against T. pallidum/JClinMicro 2003 Prospective cross-sectional study:

Divided into 5 groups based on clinical hx:

25 primary syphilis

25 secondary syphilis

179 latent syphilis

105 “hx syphilis that had been correctly treated”

107 no clinical hx syphilis.

 

Gold standard: FTA-Abs positive plus clinical

 

Secondary syphilis:  RPR+ :25/25 (100) Small Sample size. Moderately relevant High
*Dyckman/Evaluation of Reagin Screen, a new serological test for syphilis/JCM 1976 Cross-sectional study:

Serum from N=243 patients with untreated primary, secondary, and latent syphilis and N=354

patients previously treated for syphilis were examined by the four tests.(VDRL, RPR, Reagin Screen “RST”, FTA-ABS)

*note also tested “normals” and those with other conditions thought to evoke a false positive RPR, but not possible to abstract from the tables which were positive by VDRL and not by FTAABS.

 

Gold standard:

(i) primary syphilis:primary, dark-field positive chancre, and no symptoms of secondary

syphilis;

(ii) secondary syphilis:-dark-field positive secondary lesions or multiple symptoms

of secondary syphilis such as condyloma lata, alopecia, and lymphadenopathy; and

(iii)latent syphilis: no physical signs of syphilis, no history of treatment for syphilis, AND one reactive

nontreponemal test, AND a reactive FTA-ABS.

Sensitivity: Secondary syphilis:

VDRL+:56/56: 100%

RPR+: 56/56: 100%

RST+: 56/56:100%

 

Note: we do not report the latent here because non treponemal test is part of the gold standard definition.

Serologies not part of definition, may have missed patients with more subtle symptoms Relevant High
*Dyckman/Clinical evaluation of a new screening test for syphilis/AmSocClinPath 1977 Cross sectional study:. Tested serum from presumed healthy individuals and syphilitic patients (Tested with RPR, VDRL, Syphlacheck (lipid ag based)

N=29 with secondary syphlilis

Gold standard:

(1) primary syphilis—primary, darkfield-positive

chancre and no symptom of secondary syphilis; (2)

secondary syphilis—darkfield-positive secondary lesions

or two or more symptoms of secondary syphilis such as

condylomata lata, alopecia, and lymphadenopathy; (3)

latent syphilis—no physical sign of syphilis, no history

of treatment for syphilis, one reactive nontreponemal

test, and a reactive FTA-ABS test

Sensitivity Secondary syphilis:

VDRL+: 29/29 100%

RPR+: 29/29 100%

SyphlaCheck+: 29/29 100%

 

 

Serologies not part of definition, may have missed patients with more subtle symptoms Relevant High
*Dyckman/Comparison of Serum and Plasma Specimens for Syphilis Serology Using the Reagin Screen Test/J Clin Micro 1980 Cross sectional study:

Gold Standard Criteria for Primary and Secondary syphilis are as above in the first two studies.

Sensitivity: Secondary syphilis:

VDRL+ 23/23= 100%

RST+ w/o anticoag 23/23= 100%

RST+ w/ EDTA 23/23= 100%

RST+ w/ citrate 23/23= 100%

RST+ w/heparin 23/23= 100%

 

Same caveats as above study apply to the latent syphilis category and the ‘healthy’ individuals

Serologies not part of definition, may have missed patients with more subtle symptoms Relevant High
*Falcone/ Evaluation of the RPR card test/PHreports /1964 Cross sectional study:

 

Gold standard:

Presumed normal N=260. Donors with no history of previous or present infection with syphilis.

However this data was not verified with a treponemal test and not included in this review)

 

Syphilis Primary (N=134) and Secondary(N=217)

 

Gold standard: primary and second¬

ary syphilis: “proved by darkfield examination, who had received no treatment.”

 

Donors with

latent, late, and congenital syphilis who were untreated, inadequately treated, or adequately treated. ( However these were not included in this review as they were not verified with a treponemal test)

 

Biologic false positive (BFP). Donors with reactive nontreponemal tests with no clinicalevidence of syphilis and who had no previous

Treponema pallidum immobilization (TPI)test (Again not included as not verified with treponemal test

Secondary syphilis sensitivity:

RPR+ 211/217: 97.2%

VDRL+ 217/217 : 100%

Clinical data included Relevant High
*Greaves/ A comparative study of serologic tests in early syphilis/arch of derm/1962 Cross sectional study

N=13 primary

N=16 secondary

Gold standard: for primary and secondary: dark field positive

Sensitivity secondary syphilis

VDRL+16/16: 100%

 

Small sample size, restricted to darkfield positive. Moderately relevant Medium
*Gibowski/Detection of specific IgM-CLASS antitreponemal antibodies in blood

serum of patients with syphilis with the use of CAPTIA Syphilis-M

reaction and comparing it with VDRL, FTA-ABS and TPHA reactions/ Med Sci Monit/1998

Cross sectional study:

N=200 Polish patients clinically characterized with syphilis. None of the early syphilis subjects were treated

 

Gold standard: Clinical staging + FTA-ABS, TPHA, Captia Syphilis M

Primary (18) sensitivity: VDRL: 55.6%

Recent secondary (17): 100%

Recurrent Secondary (44): 100%

Early Latent (34): 100%

Late Latent (44): 63.6%

ASN (29): 89.7%

Symptomatic neurosyphilis (17): 77.8%

Good gold standard; small numbers Relevant High
*Glicksman/Instant Syphilis Screening/Texas Medicine/1967 Cross sectional study:

N=575 patients seen at an STD clinic in TX.

Gold Standard: Clinical staging + VDRL

Primary (19): RPR 79%

Secondary (31): RPR 100%

Untreated latent (87): 94%

Previously Treated Latent (69%): 78%

Untreated Congenital (2): 100%

Previously treated Late (10): 100%

Limited data on timing of treatment; no confirmatory treponemal testing Relevant High
*McMillan/ Qualitative and quantitative aspects of the serological

diagnosis of early syphilis/ Int J STD AIDS/2008

Prospective study: of untreated patients with early syphilis.

N=89 primary, 68 secondary, 72 early latent

Gold standard: Clinical staging (occasional darkfield or PCR confirmation for primary syphilis) and “treponemal tests.”

Early latent syphilis stage defined as infection within two years.

Primary (n=89): VDRL reactive in 67%

Secondary (n=68): VDRL reactive in 100%

Early Latent (n=72): VDRL reactive in 85%

Relatively small sample size Relevant High
*De Lemos/Characterization of the western blotting IgG reactivity patterns in the clinical phases of acquired syphilis/Diag Microbio and ID/ 2007. Retrospective cross sectional study: Tested serum from 3 groups of patients in Brazil

 

1)”Syphilis group” N=122 sera from patients classified as having primary, secondary, early latent, late latent and tertiary syphilis.

Gold standard: “clinical, lab (FTA-ABS, TPHA, VDRL) and epidemiologic criteria.”

 

2) Blood donor group: N=250 sera samples from individuals screened by VDRL (We did not include any analysis from this group here as VDRL was part of the definition for diagnosis).)

 

3) “Other infections group”N= 135 sera samples from pts with other pathologies that might cause false pos (viruses, toxo, Chagas, malaria, toxocariasis, hansen’s, SLE)Gold standard for BFP: TPHA, FTA, WB TP-IgG negative but VDRL pos.

In the “syphilis group” secondary syphilis

VDRL+ 21/22 sensitivity of 95.5%

 

 

Limitations: small sample size, unclear gold standard definition may be circular if VDRL is included in definition.

 

Limited relevance Medium
Thakar/Seroprevalence of syphilis by TPHA test/Ind J Pathol Microbiol /1996 Cross sectional study:

N=1074 sera tested in india by TPHA and VDRL, 408 from “clinically suspected” cases of syphilis, 302 from patients with primary syphilis, 48 secondary syphilis 58 tertiary syphilis

 

Also: 111 specimens from other STD patients, 237 from pregnant women and 318 from “apparently healthy”

 

Gold standard: unclear  definitions for the “other std” pregnant and “apparently healthy”

 

Gold standard for primary, secondary and tertiary is “clinically suspected”. No other details provided.

 

% (including weakly reactive with reactive)

 

Primary syphilis

Sensitivity VDRL+: 231/302=76.5%

 

Secondary syphilis

Sensitivity VDRL+: 35/48=72.9%

(*note that 26/48 were TPHA pos)

 

Tertiary syphilis

Sensitivity VDRL+: 37/58-63.8%

Poorly defined gold standard.

 

Limited relevance Medium
*Backhouse/T. pallidum western blot: comparison with the FTA-ABS test as a confirmatory test for syphilis/ Diag micro and ID/ 2001 Retrospective cross-sectional study.

 

N=278 including 98 reactive sera from patients with documented treponemal infection.

 

Gold standard: unclear: appears to be “clinical plus serologic.”

 

N=11 untreated patients with primary syphilis, N=10 untreated patients with secondary syphilis, N=6 untreated patient with early latent syphilis, N=12 patients with untreated late latent syphilis.

Primary syphilis:

11/11 VDRL+: Sens 100%

Unclear gold standard, may include VDRL.

 

Data for BFPs not included at this study was mostly focused on treponemal tests.

Limited relevance Medium
*Dandoy/Initial serological reactions in infectious syphilis/BJVD/ 1967 Retrospective cross sectional of diagnosed cases of primary, secondary and early latent syphilis from VD clinic in los angeles.

 

Gold standard: somewhat unclear, but for primary and secondary syphilis included physical exam and DF (though some cases of both were DF negative, so unclear if dx in these cases was based solely on physical exam or if serologies played a role.

 

Gold standard early latent: unclear: appears to be Kolmer, but one case was Kolmer neg so may include VDRL—circular,not included in this review.

Sensitivity VDRL secondary syphilis: VDRL+

83/84: 98.8%

 

 

Gold standard unclear here—may have included serologies, so somewhat circular, though there were definitely DF pos prim syphilis cases which were VDRL neg. Limited relevance Medium
*Dang/Evaluation of specific ab’s for early dx and management of syphilis/ Int J Derm/ 2006 Retrospective cross sectional

N=67 patients, 20 pts with primary syphilis, 47 with secondary syphilis

 

Gold standard:

Primary syphilis: “in whom the diagnosis of syphilis was made by

clinical, epidemiological and laboratorial methods,”

 

 

 

Sensitivity RPR secondary syphilis:

47/47 (100%)

Very poorly defined gold standard, may have included RPR, unclear. Limited relevance Medium
*Portnoy: RPR with unheated serum and new improved antigen suspension/PHR/1961 Retrospective cross sectional

Done on serums with “known clinical categories”

 

Tested with VDRL slide test, also tested with the RPR using unheated serum rather than plasma

N=48 primary syphilis

 

Gold standard: loosely defined/ unclear: “known clinical categories”

Did not include data onlatent, presumed non syphilitic and BFP, the as the definitionswere too unclear.

Here we include all those which were weakly reactive as reactive:

 

Sensitivity primary syphilis untreated:

VDRL + 29/48: 60.4%

 

RPR + A: 25/48: 52.1%

RPR + B: 25/48:52.1%

RPR+ C: 26/48:54.2%

 

Sensitivity secondary syphilis untreated:

VDRL + 25/25 100%

RPR+ A:24/25: 96%

RPR+ B:24/25:96%

RPR+ C:24/25: 96%

Poorly defined gold standard. Limited relevance Low
*White/Visuwell Reagin, a non-treponemal enzyme linked immunosorbent assay for the serodiagnosis of syphilis/JClinMicro/ 1989 Large cross-sectional study of patients from 3 different sites:

Site A: n=1948 serum samples without clinical hx tested with RST, Visuwell Reagin and positives or discordants tested by MHA-TP.

Gold standard: MHA-TP.

 

Site B: n=1406 samples without clinical data tested with Visuwell, VDRL, and confirmed with FTA

Gold standard: FTA

Site C: 311 serum samples from clinically characterized samples:

Group 1: untreated syphilis: dx with prim, secondary, or latent syphilis

 

Gold standard for secondary syphilis: unclear: only that some clinical data was used

 

Note:  Did not report data for primary syphilis as only 2 patients. Did not report for latent as unclear GS and only 8 patients.

 

 

Group 2: Treated syphilis (not reporting on this data in this review)Group 3: non syphilitic: “presumed non syphilitic” from i.patients with conditions other than syphilis (eg lupus, yaws), ii.BFPs that were pos by VDRL but neg by FTA, and iii.sera from donors with no hx prev or present syphilis infection.

Secondary syphilis:

Site C Group 1

Visuwell+: Sensitivity 26/27: 96.2%,

RPR+: 27/27: 100%

VDRL: 27/27: 100%

RST (Reagin screening test): Sensitivity: 24/26: 92.3%

 

Note: Visuwell is a modified verison of the VDRL.

Small numbers.

Unclear gold standard.

 

Limited relevance Medium
*Stone/Capture-S, a Nontreponemal solid-phase erythrocyte adherence assay for serological detection of syphilis/JClinMicro/ 1997 Retrospective for Capture-S (non trep)

Cross sectional in several populations:

Site 1:

N=366 reactive syphilis samples from treated and untreated patients in different stages of disease. Gold Standard criteria unclear other than this.

 

 

Site 1:

 

Primary untreated n=29

CaptureS Sens: 88.5%

RPR: 96.2%

 

Secondary untreated: n=50

CaptureS Sens: 91.8%

RPR: 97.9%

 

Latent untreated: n=41

CaptureS Sens: 94.9%

RPR: 92.3%

 

Poorly defined gold standards. Limited relevance Medium
*Buck et al. Comparative studies of the RPR Card Test for syphilis and the VDRL Slide Test in Ethiopia/Am J Hyg/1964*Buck et al. Comparative studies of the RPR Card Test for syphilis and the VDRL Slide Test in Ethiopia Cross sectional study:

N=50 Ethiopian men with a clinical diagnosis of early syphilis (6 with primary and 44 with secondary) with positive darkfield (gold standard)

62 with lepromatous leprosy

46 with tuberculoid leprosy

Gold standard is FTA for the above two.

Sensitivity RPR/VDRL

Early syphilis: 96.8% both

 

Lepromatous leprosy RPR/VDRL: 47%/31% (FTA was 30.5%)

Tuberculoid leprosy RPR/VDRL: 8.3/9.2% (FTA was 21.3%)

While the data for early syphilis had a reasonable gold standard, the data for the leprosy patients are very limited

 

Note that it is difficult to draw any conclusions for primary syphilis due to only 6 cases.

Relevant High
*Portnoy/RPR card for syphilis and other treponematoses/ PHR/1962 Cross sectional study:

N=600 syphilis patients, also included patients “without syphilis” and those with treated yaws and pinta but gold standard not defined and those data were not included in the review.

Gold standard for syphilis: not described, though apparently had some clinical data for primary.

For latent unclear

For late syphilis all were treated.

 

 

Sensitivity:

In untreated primary and secondary syphilis:

RPR card: 54/57: 94.7%

VDRL slide test: 52/57: 91.2%, if include weakly reactive: 54/57:94.7%

 

Uncleargold standard. Limited relevance Low
*Castro/Evaluation of an enzyme immunoassay technique for detection of antibodies against T. pallidum/JClinMicro 2003 Prospective cross-sectional

Divided into 5 groups based on clinical hx:

25 primary syphilis

25 secondary syphilis

179 latent syphilis

105 “hx syphilis that had been correctly treated”

107 no clinical hx syphilis.

Gold standard: FTA-Abs positive plus clinical

 

 

Sensitivity RPR+ “Latent syphilis” 174/179 (97.2) No information on early vs late latent. Moderately relevant High
*Glicksman/Instant Syphilis Screening/Texas Medicine/1967 Cross sectional study:

N=575 patients seen at an STD clinic in TX.

Gold Standard: Clinical staging + VDRL

Primary (19): RPR 79%

Secondary (31): RPR 100%

Untreated latent (87): 94%

Previously Treated Latent (69%): 78%

Untreated Congenital (2): 100%

Previously treated Late (10): 100%

Limited data on timing of treatment; no confirmatory treponemal testing Relevant High
Cate/sensitivity and specificity of automated serologic tests for syphilis/AJCP /1971 Retrospective cross sectional study examining VDRL and an automated reagin (VDRL based) (nontreponemal) test.

N=139 sera reactive by FTA-ABS

N=315 sera non reactive by FTA-ABS

 

Gold Standard for late or latent syphilis: FTA-ABS + some clinical (not well defined) available for 90% of patients.

 

Gold standard for specificity was FTA negative.

 

.

 

 

Sensitivity for “late or latent syphilis”: VDRL+

102/139 (73%)

Automated regain using VDRL + 98/139: 71%

 

General Specificity: VDRL -:  303/315 (96%)

 

Specificity Automated regain (VDRL)  -: 305/315: 97%

Unclear definition of late vs latent syphilis was and no distinguishing between early and late latent.

“ Review of hospital records disclosed that

no patient with FTA-ABS reactive sera

had clinically apparent primary or congenital

syphilis. When syphilis was diagnosed

clinically, it was classified as late or

latent.”

Limited relevance Medium
*Stone/Capture-S, a Nontreponemal solid-phase erythrocyte adherence assay for serological detection of syphilis/JClinMicro 1997 Retrospective for Capture-S (non trep)

Cross sectional in several populations:

Site 1:

N=366 reactive syphilis samples from treated and untreated patients in different stages of disease. Gold Standard criteria unclear other than this.

 

Site 2: n=1917 serum samples wth no clinical info tested with RPR and Capture S: all reactives were confirmed with FTA.

Gold standard: FTA (but note that for sensitivity, really based on conglomerate of RPR and Capture S)

 

Site 3: 1932 blood donor samples tested with Capture S, all pos confirmed with FTA.

Gold Standard for spec: FTA

 

Site 4: plasma center n=1193 donors tested by RPR and Capture S. All reactive confirmed with FTA.

Gold standard: Sens: conglom of RPR and Capture S, Spec: FTA

 

Site 5: 1425 blood donors tested with RPR and Capture S, reactive sera tested with FTA.

Gold standard: Sens: conglom of RPR and Capture S, Spec: FTA

Site 1:

 

Primary untreated n=29

CaptureS Sens: 88.5%

RPR: 96.2%

 

Secondary untreated: n=50

CaptureS Sens: 91.8%

RPR: 97.9%

 

Latent untreated: n=41

CaptureS Sens: 94.9%

RPR: 92.3%

 

Poorly defined gold standards. Limited relevance Medium
*Gibowski/Detection of specific IgM-CLASS antitreponemal antibodies in blood

serum of patients with syphilis with the use of CAPTIA Syphilis-M

reaction and comparing it with VDRL, FTA-ABS and TPHA reactions/ Med Sci Monit/1998

Cross sectional study:

N=200 Polish patients clinically characterized with syphilis. None of the early syphilis subjects were treated

 

Gold standard: Clinical staging + FTA-ABS, TPHA, Captia Syphilis M

Primary (18) sensitivity: VDRL: 55.6%

Recent secondary (17): 100%

Recurrent Secondary (44): 100%

Early Latent (34): 100%

Late Latent (44): 63.6%

ASN (29): 89.7%

Symptomatic neurosyphilis (17): 77.8%

Good gold standard; small numbers Relevant High
*McMillan/ Qualitative and quantitative aspects of the serological

diagnosis of early syphilis/ Int J STD AIDS/2008

Prospective study: of untreated patients with early syphilis.

N=89 primary, 68 secondary, 72 early latent

Gold standard: Clinical staging (occasional darkfield or PCR confirmation for primary syphilis) and “treponemal tests.”

Early latent syphilis stage defined as infection within two years.

Primary (n=89): VDRL reactive in 67%

Secondary (n=68): VDRL reactive in 100%

Early Latent (n=72): VDRL reactive in 85%

Relatively small sample size Relevant High
*De Lemos/Characterization of the western blotting IgG reactivity patterns in the clinical phases of acquired syphilis/Diag Microbio and ID/ 2007. Retrospective cross sectional study: Tested serum from 3 groups of patients in Brazil

 

1)”Syphilis group” N=122 sera from patients classified as having primary, secondary, early latent, late latent and tertiary syphilis.

Gold standard: “clinical, lab (FTA-ABS, TPHA, VDRL) and epidemiologic criteria.”

 

2) Blood donor group: N=250 sera samples from individuals screened by VDRL (We did not include any analysis from this group here as VDRL was part of the definition for diagnosis).)

 

3) “Other infections group”N= 135 sera samples from pts with other pathologies that might cause false pos (viruses, toxo, Chagas, malaria, toxocariasis, hansen’s, SLE)Gold standard for BFP: TPHA, FTA, WB TP-IgG negative but VDRL pos.

In the “syphilis group” early latent syphilis

VDRL+ 23/28 sensitivity of 82.1%

 

 

Limitations: small sample size, unclear gold standard definition–may be circular if VDRL is included in definition.

 

Limited relevance Medium
*Backhouse/T. pallidum western blot: comparison with the FTA-ABS test as a confirmatory test for syphilis/ Diag micro and ID/ 2001 Retrospective cross-sectional study.

 

N=278 including 98 reactive sera from patients with documented treponemal infection.

 

Gold standard: unclear: appears to be “clinical plus serologic.”

 

N=11 untreated patients with primary syphilis, N=10 untreated patients with secondary syphilis, N=6 untreated patient with early latent syphilis, N=12 patients with untreated late latent syphilis.

Early latent

6/6 VDRL+: Sens 100%

Very unclear gold standard, may include VDRL.

 

We did not use data for BFPs because this was mostly focused on treponemal tests.

Limited relevance I
Gibowski/Detection of specific IgM-CLASS antitreponemal antibodies in blood

serum of patients with syphilis with the use of CAPTIA Syphilis-M

reaction and comparing it with VDRL, FTA-ABS and TPHA reactions/ Med Sci Monit/1998

Cross sectional study:

N=200 Polish patients clinically characterized with syphilis. None of the early syphilis subjects were treated

 

Gold standard: Clinical staging + FTA-ABS, TPHA, Captia Syphilis M

Primary (18) sensitivity: VDRL: 55.6%

Recent secondary (17): 100%

Recurrent Secondary (44): 100%

Early Latent (34): 100%

Late Latent (44): 63.6%

ASN (29): 89.7%

Symptomatic neurosyphilis (17): 77.8%

Good gold standard; small numbers Relevant High
*Singh/Characteristics of primary and late latent syphilis cases which were initially non-reactive with the RPR as screening test/Int J STD AIDS/2008 Retrospective case series querying Alberta Canada STI notification database. All RPR tests were confirmed with FTA-ABS or MHA-TP

 

N=863 primary cases  Excluded those in which initial serology occurred prior to study period, those whose initial test was not an RPR, those who did not have an RPR titer reported, those who had a prior diagnosis of syphilis and those who had incomplete demographic or clinical information.

 

Gold standard: positive FTA-ABS or MHA-TP, diagnosis of “primary syphilis” based on clinician’s judgement.

512/1303 negative on RPR, (sensitivity: 61%) Limitation: relied on health department records, not clear from description how carefully patients were screened and if some might have had previous treatment. Relevant High
*De Lemos/Characterization of the western blotting IgG reactivity patterns in the clinical phases of acquired syphilis/Diag Microbio and ID/ 2007. Retrospective cross sectional study: Tested serum from 3 groups of patients in Brazil

 

1)”Syphilis group” N=122 sera from patients classified as having primary, secondary, early latent, late latent and tertiary syphilis.

Gold standard: “clinical, lab (FTA-ABS, TPHA, VDRL) and epidemiologic criteria.”

 

2) Blood donor group: N=250 sera samples from individuals screened by VDRL (We did not include any analysis from this group here as VDRL was part of the definition for diagnosis).)

 

3) “Other infections group”N= 135 sera samples from pts with other pathologies that might cause false pos (viruses, toxo, Chagas, malaria, toxocariasis, hansen’s, SLE)Gold standard for BFP: TPHA, FTA, WB TP-IgG negative but VDRL pos.

In the “syphilis group” late latent syphilis

VDRL+ 29/44 sensitivity of 65.9%

 

 

Limitations: small sample size, unclear gold standard definition– may be circular if VDRL is included in definition.

 

Relevant Medium
*Backhouse/T. pallidum western blot: comparison with the FTA-ABS test as a confirmatory test for syphilis/ Diag micro and ID/ 2001 Retrospective cross-sectional study.

 

N=278 including 98 reactive sera from patients with documented treponemal infection.

 

Gold standard: unclear: appears to be “clinical plus serologic.”

 

N=11 untreated patients with primary syphilis, N=10 untreated patients with secondary syphilis, N=6 untreated patient with early latent syphilis, N=12 patients with untreated late latent syphilis.

Late latent syphilis:

9/12 VDRL+: Sens 75%

Unclear gold standard, may include VDRL.

 

 

Limited relevance I
*De Lemos/Characterization of the western blotting IgG reactivity patterns in the clinical phases of acquired syphilis/Diag Microbio and ID/ 2007. Retrospective cross sectional study: Tested serum from 3 groups of patients in Brazil

 

1)”Syphilis group” N=122 sera from patients classified as having primary, secondary, early latent, late latent and tertiary syphilis.

Gold standard: “clinical, lab (FTA-ABS, TPHA, VDRL) and epidemiologic criteria.”

 

2) Blood donor group: N=250 sera samples from individuals screened by VDRL (We did not include any analysis from this group here as VDRL was part of the definition for diagnosis).)

 

3) “Other infections group”N= 135 sera samples from pts with other pathologies that might cause false pos (viruses, toxo, Chagas, malaria, toxocariasis, hansen’s, SLE)Gold standard for BFP: TPHA, FTA, WB TP-IgG negative but VDRL pos.

In the “syphilis group” tertiary syphilis

VDRL+ 8/17 sensitivity of 47%

 

 

Limitations: small sample size, unclear gold standard definition– may be circular if VDRL is included in definition.

 

Limited relevance Medium
Thakar/Seroprevalence of syphilis by TPHA test/Ind J Pathol Microbiol /1996 Cross sectional study:

N=1074 sera tested in india by TPHA and VDRL, 408 from “clinically suspected” cases of syphilis, 302 from patients with primary syphilis, 48 secondary syphilis 58 tertiary syphilis

 

Also: 111 specimens from other STD patients, 237 from pregnant women and 318 from “apparently healthy”

 

Gold standard: unclear  definitions for the “other std” pregnant and “apparently healthy”

 

Gold standard for primary, secondary and tertiary is “clinically suspected”. No other details provided.

 

% (including weakly reactive with reactive)

 

Primary syphilis

Sensitivity VDRL+: 231/302=76.5%

 

Secondary syphilis

Sensitivity VDRL+: 35/48=72.9%

(*note that 26/48 were TPHA pos)

 

Tertiary syphilis

Sensitivity VDRL+: 37/58=63.8%

Poorly defined gold standard.

 

Limited Relevance Medium
*Castro/Nontrep tests in the diagnosis of neurosyphilis/J clin Lab analysis/ 2008 Retrospective cross sectional study.

Cases: 25 pts with NS

(of which 8 were symptomatic, also 24 NS pts were HIV+)

 

Controls:

163 pts with + syphilis serologies but no e/o NS

126 pts with no syphilis but other neurologic disorders

 

 

Gold standard : positive serologic test for syphilis, reactive CSF FTA-ABS, increased CSF protein >=45 mg/dL and CSF pleocytosis >=10cell/mm3.

 

 

Of the controls, one CSF VDRL and RPR were positive

 

CSF VDRL:

Reported specificity 99% overall (AS and S NS),

Sensitivity 70.8%

 

Symptomatic NS (n=8): 7/8: Sensitivity 87.5%

Asymptomatic NS (n=16): 10/16

Sensitivity: 62.5%

 

 

CSF RPR: Reported specificity 99.3% overall (AS and S NS)

Sensitivity 75%

 

Symptomatic NS (n=8): 8/8: Sensitivity: 100%

 

Asymptomatic NS (n=16): 11/16

Sensitivity:68.8%

Limitations: Specificity of CSF FTA

 

Strength: two control groups

Some clinical info.

 

Contamination of CSF with blood was assessed and contaminated samples discarded (but in real world clinical setting many samples with CSF blood contamination may be tested).

Relevant High
*Marra/CSF TPPA assay for NS diagnosis/JCM /2017 Retrospective case study

Training data set  (n=191)

(45 with T pallidum PCR pos

40 had symptoms)

Validation data set (n=380)

(41 with T pallidum PCR pos, 95 had symptoms)

 

Gold standard:

1) CSF VDRL positive or

2) CSF PCR detection of T pallidum or

3) New vision or hearing loss (with clinical or serologic evidence of syphilis)

 

 

 

Validation data set: more were previously treated, far fewer had a positive CSF VDRL and far fewer Tpallidum pcr positive.

 

Training dataset

Sensitivity of CSF VDRL:

1) comp with PCR: 48.9%(34.3-63.5)

2) comp with symptoms 67.5% (53-82)

 

Specificity of CSF VDRL

1) comp with PCR: 74.0(66.9-81.1)

2) comp with symptoms 78.2 (71.4-85)

 

Validation dataset

Specificity of CSF VDRL

1) comp with PCR: 91.2(88.1-94.2)

2)comp with symptoms 90.2(86.7-93.6)

 

No difference in sensitivity or specificity based on HIV status

Limitations: Retrospective, partially circular definition as CSF VDRL positive was part of Gold standard, though we took the comparison with PCR. However unclear relationship between T. pallidum PCR and NS, also new vision or hearing loss is a limited definition for NS.

 

Focused on the treponemal test

 

 

 

 

Relevant High
*Marra/The RPR cannot replace the VDRL for neurosyphilis dx/STD 2012 Retrospective cross sectional study conducted on samples from a study of CSF abnormalities in patients with syphilis

Referred to study: neurologic findings (esp hearing or vision loss), serum RPR>1:32 and in HIV+ CD4<=350.

 

N=149, 39 c “Lab defined”, 33 c “Symptomatic”

 

Gold standard:

1)      “Lab defined”:Positive CSF FTA and CSF WBC>20

2)      “Symptomatic”:Vision or hearing loss (with clinical or serologic evidence of NS)

 

 

RPRV technique: modified to be more similar to CSF VDRL technique, to adjust for lower conc of immunoglob in csf vs serum

“Lab defined”:

Sensitivity

CSF-VDRL: 71.8%(57.7-85.9)

CSF-RPR: 56.4%(40.8-72.0)

CSF-RPRV 59.0%(43.6-74.4)

Specificity

CSF-VDRL: 98.3%(95-100)

CSF-RPR: 100%(100-100)

CSF-RPRV 98.3%(95-100)

 

“Symptomatic:”

Sensitivity

CSF-VDRL: 66.7(50.6-82.8)

CSF-RPR:51.5(34.4-68.6)

CSF-RPRV:57.6(40.7-74.5)

 

Specificity

CSF-VDRL: 80.2(72.9-87.5)

CSF-RPR: 89.7(84.2-95.2)

CSF-RPRV: 84.5(77.9-91.1)

 

Limitations: “Symptomatic”: New vision or hearing loss a very limited definition for NS. Relevant High
*Zhu/ Comparison of the CSF TRUST and CSF RPR with the CSF VDRL for dx of NS among HIV- syphilis pts in china/JCM /2014 Prospective cross sectional study: pts >18yo c new syphilis diagnosis or serofast status (<4 fold decline in titer after tx and a constant low rpr titer of<=1:8 after standard syph tx at 2 yrs followup, excluded pregnant or HIV+ or refused LP

 

N=210 NS patients, 56 asymptomatic , 154 symptomatic

 

All had to have positive serum syphilis serologies

 

Gold standard:

“Symptomatic”: clinical signs and symptoms with a positive CSF TPPA in absence of blood contamination.

 

“Asymptomatic”: CSF WBC>-10 AND positive CSF TPPA in absence of blood contamination.

 

 

Combined for Symp NS and Asymp NS

Sensitivity:

VDRL: 81.4 (75.4–87.4)

RPR: 76.2 (70.2–82.2)

RPR-V: 79.5 (73.5–85.5)

TRUST: 76.2 (70.2–82.2)

Specificity:

VDRL: 90.3 (88.3–92.3)

RPR: 93.4 (91.4–95.4)

RPR-V: 92.7 (90.7–94.7)

TRUST: 93.1 (91.1–95.1)

 

Symp NS:

Sensitivity:

VDRL: 85.7 (79.7–91.7)

RPR: 81.8 (75.8–87.8)

RPR-V: 83.1 (77.1–89.1)

TRUST: 82.5 (76.5–88.5)

 

Specificity:

VDRL: 86.7 (84.7–88.7)

RPR: 90.2 (88.2–92.2)

RPR-V: 89.1 (87.1–91.1)

TRUST: 90.1 (88.1–92.1)

 

Asymp NS

Sensitivity:

VDRL: 69.6 (59.6–79.6)

RPR: 60.7 (50.7–70.7)

RPR-V: 69.6 (59.6–79.6)

TRUST: 58.9 (48.9–68.9)

 

Specificity:

VDRL: 79.4 (77.4–81.4)

RPR: 82.6 (80.6–84.6)

RPR-V: 81.8 (79.8–83.8)

TRUST: 82.1 (80.1–84.1)

CSF TPPA (part of gold standard) may lack specificity. Relevant High
Delaney/ False pos serology in CSF assoc with a spinal cord tumor/1975 Case report, one individual In this pt CSF was reactive to VDRL and fluorescent treponemal antibody absorption (FTA-ABS) tests, but became nonreactive with removal of the spinal cord tumor. Serum serologies were persistently negative. Single patient Moderately relevant. I
Izzat/Validity of the VDRL test on CSF contaminated by blood/BrJVD/ 1971 A laboratory experiment in which syphilitic blood was added to csf to find at what titer and what amount it had to be added to get a positive CSF VDRL. 50 lambda* for 1: 1 titre blood

to 3 lambda for 1: 256 titre blood. 3 lambda of whole blood per ml. of cerebrospinal fluid caused a definite blood colour, whereas 1 lambda did not.

With VDRL titres of 1: 256 or below, sufficient whole blood to cause a false positive cerebrospinal fluid VDRL test produces visibly bloody cerebrospinal fluid.

Useful to demonstrate the principle. Relevant High
Madiedo/False positiveVDRL and FTA in CSF Case report Man with meningeal carcinomatosis

Blood FTA and VDRL were neg, but CSF VDRL was positive and FTA neg.One week later had both vdrl and fta pos in CSF (though blood stayed neg)

Single case Relevant I
*Burke/Neurosyphilis in the antibiotic era/Neurology 1985

 

Retrospective cross sectional:

N=64 pts were diagnosed with NS during study period but 34 excluded because they did not have CSF exams or had normal CSF, positive CSF FTA with no other CSF abns, no medical records available, or one pt with pulm TB did not have CSF cx for mycobacteria. N=30 included.

 

 

Gold Standard: positive serum FTA or microhemagglutination assay for T. pallidum with one of the following CSF abns: 1) >=5wbc/ml, OR 2) at least 46mg% protein OR 3) pos VDRL

4 had no neurologic signs or symptoms but had + CSF VDRL

24/28 patients (2 did not have serum VDRL done) had a positive serum VDRL. Sens: 86%.

 

15/28 (53%) had a positive CSF VDRL.

 

Not possible to identify from the data presented sensitivity in symptomatic vs asymptomatic.

Difficult to assess sens/spec of CSF VDRL due to this being part of the gold standard diagnostic criteria—

 

While reporting on serum VDRL is fairly good quality, CSF VDRL is only moderate quality.

Limited relevance Medium
Davis/Clinical significance of CSF tests for NS/Ann Neurol 1989 Retrospective cross sectional study.

 

N=15

 

Gold standard: “Likely active NS” 1) recent onset <6mos or progression of 1 or more decreased cognition, unexplained personality

change, seizures, hemiparesis, cranial nerve palsies, optic

nerve dysfunction, and chorioretinitis AND 2) CSF containing more than 40 mg/dl glucose, sterile routine bacterial and fungal cultures plus more than 5 white blood cells per cubic

millimeter or more than 45 mg/dl protein or both; AND (3) a

reactive CSF-FTA-ABS test and a reactive serum ETA-ABS

test; AND (4) no other recognized cause for the neurological illness.

 

 

“Unlikely active neurosyphilis” If the patient had a reactive CSF-FTA-ABS test but did not fulfill the above criteria.

4/15 patients with “likely active NS” had a positive CSF VDRL—sensitivity only 26.7%, felt to have essentially 100% specificity.

 

13/14 patients had a positive serum RPR: sensitivity: 92.9%

 

All symptomatic

Small sample size. Relevant. Medium
Chen/Clinical and laboratory characteristics in patients suffering from general paresis in the modern era/Journal of the neurological sciences/2015 Retrospective chart review: N= 82 patients from China diagnosed clinically with General Paresis

 

Gold Standard : Positive syphilis serologies and: CSF positive RPR or TPPA or elevated protein or >10 WBCs (plus clinical syndrome consistent with general paresis)

3/85 patients had negative serum RPR (3.53%)

68% were CSF RPR positive and 82 (97%) had positive CSF TPPA

 

All symptomatic

RPR performance characteristics in the CSF are problematic- but the serum RPR data and the CSF TPPA data are useful Relevant High
Hooshmand/Neurosyphilis: A study of 241 patients/JAMA/ 1972 Retrospective case series

N=241 neurosyphilis diagnoses

 

Gold standard:

(1) positive blood FTA-ABS and ophthalmic or neurologic findings

suggestive of neurosyphilis; OR (2) positive blood and cerebrospinal

fluid (CSF) FTA-ABS with abnormal blood cell counts in CSF (more than five white blood cells per cubic millimeter with no evidence of bacterial or viral meningitis. In these cases, the spinal tap was done in search of

an unexplained neurologic illness; OR (3) positive blood and CSF FTA-ABS in patients with progressive neurologic symptoms in whom other etiologic factors were ruled out.

Sensitivity NS

CSF STS + 100/176 (56.8%)

 

117/241 had a + serum STS.

 

All had symptoms of some sort.

Gold standard for NS may have been too liberal. Moderately relevant High
Jiang/The usefulness of the TRUST serum test in the diagnosis of HIV negative NS/STD 2011 Retrospective cross sectional study: N=36 cases of symptomatic neurosyphilis. HIV negative.

Also N=20 cases of non neurologic syphilis

Not all NS got the TRUST test.

Gold standard: NS

CSF VDRL+ or CSF-TPPA plus CSF WBC>5

The patients with NS all also had symptoms.

 

Unclear exactly what the gold standard is for non-neurologic syphilis

Sensitivity of CSF-TRUST; Positive in 18/19 patients with symptomatic NS tested: 94.7%,

 

Specificity of CSF-TRUST: 0/20 patients with non-neurologic syphilis were positive= 100%

 

All had symptoms

Small sample size. Unclear exactly what the gold standard is for non neurologic syphilis. Limited relevance Medium
Lukehart/invasion of the CNS by T. pallidum: implications for diagnosis and treatment/ Ann Intern Med 1988 Prospective study:

 

N=58 patients with asymp untreated syphilis, N=12 had T pallidum isolated from CSF

 

Gold standard: T pallidum (rabbit infectivity test) from CSF

Sensitivity CSF-VDRL: 4/12 who had T. pallidum isolated from the CSF via the rabbit infectivity test had a positive CSF VDRL.

 

All patients were asymptomatic.

However, this may represent neuroinvasion rather than neurosyphilis.

Unclear significance of neuroinvasion. Limited relevance Medium
Dharmasaroja/ Serum and cerebrospinal fluid profiles for syphilis in Thai patients with acute ischaemic stroke/ Int J STD AIDS Retrospective case review from Thailand. N=7 patients

Gold standard: The diagnosis of symptomatic neurosyphilis was

based on three aspects: the clinical syndrome, positive serology and analyses of neurosyphilis CSF markers (CSF RPR, or protein, or >20WBCs).

 

Overall, of 7 symptomatic patients:

85.7% had a reactive serum RPR

57.1% had a reactive CSF RPR

100% had a reactive CSF FTA-ABS

Small numbers; CSF RPR not reliable Moderately relevant I
Gibowski/Detection of specific IgM-CLASS antitreponemal antibodies in blood

serum of patients with syphilis with the use of CAPTIA Syphilis-M

reaction and comparing it with VDRL, FTA-ABS and TPHA reactions/ Med Sci Monit/1998

Cross sectional study:

N=200 Polish patients clinically characterized with syphilis. None of the early syphilis subjects were treated

 

Gold standard: Clinical staging + FTA-ABS, TPHA, Captia Syphilis M

Primary (18) sensitivity: VDRL: 55.6%

Recent secondary (17): 100%

Recurrent Secondary (44): 100%

Early Latent (34): 100%

Late Latent (44): 63.6%

ASN (29): 89.7%

Symptomatic neurosyphilis (17): 77.8%

Clinical data included Relevant High
Russow/ Usefulness of CSF tests for neurosyphilis/ SAMJ/1994 Cross sectional study:N= 60 specimens of patients diagnosed as having NS

Gold Standard:   positive CSF VDRL test; or (if) negative VDRL test, but positive TPHA or FTA-Abs test, together with a raised cell count (more than 5 cells/ml),

protein concentration (> 0,45 g/l) or IgG index (> 0,6).

Compared to the CSF TPHA/FTA-ABS the sensitivity of the CSF VDRL was 52% (31/60)

 

Unclear whether all symptomatic or not

Clinical diagnosis of neurosyphilis was not clear; small study sample; Relevant High
Timmermans/Neurosyphilis in the Modern Era/ J Neurol Neurosurg Psychiatry/2004 Cross sectional study:

N=149 patients with NS

 

Gold Standard for NS Diagnosis: Clinical symptoms plus: reactive CSF VDRL or reactive CSF FTA with either pleocytosis (>5), or CSF protein >0.45 g/l or IgG index >0.6

73% CSF VDRL positivity in 149  patients with neurosyphilis

 

All symptomatic

Used the CSF VDRL as an entry criterion Relevant High
Wohrl/Neurosyphilis is Unlikely in patients with late latent syphilis and a negative blood VDRL test/2006 Retrospective study:

N=265 patients with syphilis who had paired CSF and serum available.

 

Gold standard: reactive treponemal test in serum and the diagnosis of neurosyphilis was based either on a reactive

CSF VDRL test according to the CDC guidelines, and/or if

the patient met additional criteria defined in the European STD

guidelines

265 (72 were HIV positive) patients with LL syphilis who had paired CSF and serum available

43 had neurosyphilis; no neurosyphilis cases detected among those whose VDRL titers in serum was nonreactive. In patients with neurosyphilis, the median blood-

VDRL titer was ≥1:32, In none of the 43 patients with neurosyphilis was a correlation found between blood- and CSF-VDRL-titers.

 

10 patients were asymptomatic, but not possible to tell sensitivity and specificity from the data presented for these versus those who were symptomatic.

Retrospective study; exact reasons for LP are not clear Relevant High
Tuddenham/Neurosyphilis and Ophthalmic Syphilis in Persons with Negative Rapid Plasma Reagin and Positive Treponemal Antibody Test Results/STD 2015 Retrospective cross sectional.

N=48, Abstracted results for all patients in hospital system with positive treponemal and negative non treponemal test who had a lumbar puncture.

 

Gold standard: Neurosyphilis defined by clinician judgement as well as “definite” by positive CSF VDRL and “suspected” by CSF WBC >5 and CSF protein >50.

Of 48 serodiscordant patients with an LP, only 2 were treated for neurosyphilis, and this diagnosis was doubtful even in these patients.

 

Of 48 serodiscordant patients with an LP, 2 were treated for ocular syphilis and responded to treatment

 

Neurosyphilis seems rare with a negative serum nontreponemal test, but ocular syphilis may be more likely to occur in this setting.

Retrospective study, small sample size Moderately relevant. Medium
Zhang/A Clinical Study of New Cases of Parenchymal

Neurosyphilis: Has Tabes Dorsalis Disappeared or

Been Missed?/ J Neuropsychiatry Clin Neuroscience/2015

Retrospective case series of 11 patients with general paresis and 7 with tabes dorsalis

 

Gold standard: Clinical signs and symptoms with reactive serum VDRL and TPHA

Median RPR 1:32 in GP vs 1:2 in TD

CSF RPR percent reactive 90.9 in GP vs 28.6% in TD

 

All Symptomatic

Small retrospective study Relevant I
Merins/ Syphilis and Neurosyphilis: HIV co-infection and the value of diagnostic parameters in CSF/European J Med Research /2015. Retrospective cross-sectional in Germany.

N=67

 

All had positive serum treponemal tests.

 

Gold standard: “Definite” NS: Clinical symptoms AND/ OR: ITPA index >2 AND positive CSF FTA or CSF RPR OR CSF pleocytosis OR “blood-CSF barrier disturbance”

 

Gold standard: “Probable NS”: Chronically progressive course of neurologic-psychiatric symptoms with phases of aggravation and partly remission AND CSF with >4 cells per microliter, OR CSF protein >0.5g/l or albumin quotient >7.8 and/or IgG dominant immune response in CNS

HIV+:

Sensitivity CSF RPR: 5/23

 

HIV-:

Sensitivity CSF RPR: 4/15

 

HIV+:

Sensitivity Serum RPR: 21/23

 

HIV-:

Sensitivity Serum RPR: 13/15

 

No asymptomatic patients

Poorly defined gold standard, did not distinguish Definite from Probable NS.

 

CSF RPR part of gold standard dx

Limited relevance Low
Muic/ Role of Routine Syphilis serotest in tertiary neurosyphilis detection/ Neurologia Croatica/1998 4 labs in Croatia that did combined VDRL and (Gold standard=TPHA) testing of a total of  109 patients with a clinical diagnosis tertiary neurosyphilis Mean sensitivity of serum VDRL in detecting patients with tertiary neurosyphilis= 72.48 % (range 71.01-77.77%)

 

Unclear if all symptomatic, limited data on how diagnosis of NS was made.

Limited data on how the diagnosis of NS was made Relevant Medium
Feraru/Neurosyphilis in AIDS patients: initial CSF VDRL may be negative/ Neurology/1990 Case report of 2 HIV + patients with suspected neurosyphilis whose initial CSF VDRL was negative but became positive after treatment Both symptomatic Very small sample size Limited relevance I
Gran/False Positive VDRL in the CSF of a Patient with Meningeal Carcinomatosis/ Infect Dis Clin Pract/2014 47 year old with metastatic AdenoCA of lung; treated with erlotinib CSF VDRL reactive

Serum VDRL and treponemal test nonreactive

One case report; did not repeat CSF VDRL- could be a lab error. Limited  relevance I
Spoor et al/ Ocular Syphilis/ J Clin Neuroopthalmology/1986 Retrospective case series: N=  50 patients with clinically defined ocular syphilis

Gold Standard: ocular findings and positive serum FTA-ABS

24% had a reactive serum VDRL

0% had reactive CSF VDRLs

The diagnosis of ocular syphilis was, as is expected, not certain Relevant High
Tuddenham/Neurosyphilis and Ophthalmic Syphilis in Persons with Negative Rapid Plasma Reagin and Positive Treponemal Antibody Test Results/STD 2015 Retrospective cross sectional.

N=48 Abstracted results for all patients in hospital system with positive treponemal and negative non treponemal test who had a lumbar puncture.

 

Gold standard:Neurosyphilis defined by clinician judgement as well as “definite” by positive CSF VDRL and “suspected” by CSF WBC >5 and CSF protein >50.

Of 48 serodiscordant patients with an LP, only 2 were treated for neurosyphilis, and this diagnosis was doubtful even in these patients.

 

Of 48 serodiscordant patients with an LP, 2 were treated for ocular syphilis and responded to treatment. Neither had a positive CSF VDRL.

 

Neurosyphilis seems rare with a negative serum nontreponemal test, but ocular syphilis may be more likely to occur in this setting.

Retrospective study, small sample size Moderately relevant Medium
Kunkel / Ocular syphilis–indicator of previously unknown HIV-infection/J infect/2008 Retrospective case series: N= 24 patients with ocular syphilis, 11 HIV positive, 13 HIV negative.

 

Gold standard:

(1) inflammatory disease of the eye, the optic nerve or orbital

tissue; AND

(2) serological evidence for syphilis: reactive Treponema

pallidum particle agglutination (TPPA) or hemagglutination

(TPHA) test combined with either a venereal

disease research laboratory (VDRL) test >1:4 or detection

of FTA-Abs-IgM; AND

(3) improvement following adequate antimicrobial therapy.

Sensitivity CSF VDRL:

7/23 were positive by CSF VDRL

* one patient not tested

30.4 % sensitive

 

Could not eval serum VDRL as this was part if definition and not done in all pts.

 

3/13 HIV negative + CSF VDRL

4/10 HIV positive + CSF VDRL

Well defined gold standard, particularly with response to treatment, though TPHA combined with VDRL >1:4 may be too restrictive

 

However, small sample size.

Relevant High
Parc/ Manifestations and treatment of ocular syphilis during an epidemic in France/STD /2007 Retrospective case series: N= 10 patients identified to have syphilitic uveitis

 

8/10 HIV+

 

Gold standard:active uveitis and positive serum MHA-TP

Sensitivity CSF VDRL: 9/10 had an LP, 2/9 had a positive CSF VDRL: 22.2%

 

Sensitivity serum VDRL:

10/10 had a positive Serum VDRL: 100%

Very small sample size Moderately Relevant I
Ormerod/ syphilitic posterior uveitis: correlative findings and significance/CID /2001 Retrospective case series: N= 21 patients with a diagnosis of syphilitic posterior uveitis

 

5/21 HIV+

 

Gold standard: positive serum RPR AND FTA, AND “evidence of the appropriate pattern of active ocular inflammation”

Sensitivity CSF VDRL:

7/19 that had test done= 36.8%

 

‘acute’ (sx began within last 3 mos) posterior uveitis: 5/8

 

‘chronic’ posterior uveitis (sx longer than 3 mos) 2/11 that had test done 18%

Small sample size Relevant Medium
Browning/Posterior segment manifestation of active ocular syphilis, their response to neurosyphilis regimen of PCN therapy and the influence of HIV status on response/Ophthalmology/2000 Retrospective case series: N= 14 patients, only 9 of whom had an LP with CSF VDRL testing

 

5/14 HIV +

 

Gold standard:

Active disease of the vitreous, retina, retinal pigment epithelium, choroid, or optic nerve and a positive serum treponemal test (FTA or MHATP)

Sensitivity CSF VDRL:

CSF VDRL positive: 2/9 that had testing.  22.2%

 

Sensitivity Serum RPR:

Serum RPR positive

12/14: 85.7%

Very small sample size Moderately relevant I
Bollemeijer/Clinical manifestations and outcomes of syphilitic uveitis/I O V S/ 2016 Retrospective case series: N=85 patients diagnosed with syphilitic uveitis

28/85 HIV +

 

Gold standard: positive TPPA or TPHA and/or positive FTA AND “agreement on the diagnosis of syphilitic uveitis between ophthalmologist, dermatologist, ID specialist and neurologist.

62 patients had LPs, but only 31 had VDRL done on CSF

 

Sensitivity CSF VDRL: 12/31: 38.7%

 

 

 

Sensitivity serum VDRL: 69/85 positive 91.2%

Many patients did not have LP or CSF VDRL tested Relevant Medium
Villaneuva/ Posterior uveitis in patients with positive serology for syphilis/CID /2000 Retrospective case series N= 20 patients with syphilitic posterior uveitis

 

Only 9 patients underwent HIV testing, 3/9 were positive

 

Gold standard: posterior uveitis and positive serum treponemal test.

Sensitivity CSF VDRL:

LP was done on 15/20

2/15 had positive CSF VDRL. 13.3%

 

Sensitivity serum RPR: 17/20: 85%

Small sample size Relevant Medium
Li/Posterior syphilitic uveitis: clinical characteristics, co-infection with HIV, response to treatment/JJO/2011 Retrospective case series N=13 patients with active posterior syphilitic uveitis

 

10/12 with data were HIV positive

 

Gold standard: positive serum RPR AND treponemal test AND a clinical diagnosis of ocular syphilis with posterior segment findings eg retinitis choroiditis attributable to syphilis.

8 patients had LP

Sensitivity CSF VDRL:

2/8 25% positive

 

None were serum RPR negative, but this was part of the diagnosis.

Small sample size Relevant I
Restivo /Uvieitis heralding previously unknown lueitic and HIV infection/ AISS / 2013 Retrospective case series N= 14 patients

 

Gold Standard: Serological evidence of syphilis + exclusion of other causes

All patients had reactive serum VDRL (part of inclusion criteria)

11/14 had early syphilis; 6 were HIV+

Sensitivity CSF VDRL:

3/7 (1 HIV+ 2 HIV-) had  reactive CSF

VDRL (6 CSF specimens were not sent for VDRL Testing) 42.9%

Low quality; small numbers Relevant I
Dai / Clinical manifestations and cerebrospinal fluid status in ocular syphilis in HIV-Negative patients/ BMC ID/2016 Retrospective case series N= 25 HIV uninfected persons

 

Gold Standard: Serological evidence of syphilis + ocular manifestations + CSF exam

25/25 reactive serum TRUST (part of inclusion criteria)

Sensitivity CSF VDRL

9/25 had reactive CSF VDRL: 36%

 

Low quality; small numbers Relevant Medium
Kim/ Non-human Immunodeficiency Virus-related Ocular Syphilis in a Korean Population: Clinical Manifestations and Treatment Outcomes/ K J Opth/2016

 

Retrospective series N= 39 patients and 45 eyes;

Gold Standard: Serological evidence of syphilis + ocular manifestations

32/39 had reactive serum VDRL/or RPR

82%

Sensitivity CSF VDRL:

6/39 patients had CSF examination; of those, 0/6 had a reactive CSF VDRL

0%

Low quality; small numbers; limited CSF data Relevant Medium
Lee/Clinical and laboratory characteristics of ocular syphilis: a new face in the era of HIV co-infection.” Journal of Ophthalmic Inflammation & Infection/2015 Retrospective case series N= 16 patients and 29 eyes (10 HIV+)

 

Gold Standard: Serological evidence of syphilis + ocular manifestations

All had reactive serologies (inclusion criterion)

2/9 HIV+ reactive CSF VDRL 22.2%

1/6 HIV- reactive CSF VDRL 16.7%

Low quality; unclear if all patients had a CSF VDRL performed Relevant I
 Shen, J., et al.Ocular syphilis: an alarming infectious eye disease./ International journal of clinical and experimental medicine /2015 Retrospective case seriesN= 13 patients and 21 eyes (1 HIV+)

Gold Standard: Serological evidence of syphilis + ocular inflammation

Serum RPR reactive in 12/13 patients 92.3%

CSF RPR reactive in 3/11 patients 27.2%

Low quality; small sample size Relevant I
Mathew, R. G., et al. “British Ocular Syphilis Study (BOSS): 2-year national surveillance study of intraocular inflammation secondary to ocular syphilis.”Investigative Ophthalmology & Visual Science/2014 Prospective study: N= 41 cases (63 eyes); 13 HIV+

Gold Standard: Serological evidence of syphilis (both treponemal and nontreponemal reactive) + ocular inflammation + early syphilis stage

All had RPR/VDRL titers > 1:16 (but part of definition)

HIV+ median VDRL titer 1:256

HIV- median VDRL titer 1:128 (not statistically significant)

Low quality; no data on CSF; small sample size; Relevant Low
Rodrigues, R. A./”Yellowish dots in the retina: a finding of ocular syphilis?” Arquivos Brasileiros de Oftalmologi/2015 Retrospective case series: N= 11 patients and 19 eyes (3 HIV+)

Gold Standard: Serological evidence of syphilis + ocular inflammation

8/11 had a reactive serum VDRL 72.7%

2/9 had a reactive CSF VDRL 22.2%

 

Low quality; small sample size; Relevant I
Yap, S. C., et /Syphilitic uveitis in a Singaporean population/ ocular imm and inflam/2014 Retrospective case series N= 12 patients and 18 eyes (8 HIV+)

Gold Standard: Serological evidence of syphilis + active uveitis

10/12 had a reactive serum RPR 91.7%

3/6 had a reactive CSF VDRL 50%

Low quality; small sample size; Relevant I
Puech/; ocular manifestations of syphilis: recent cases over a 2.5 year period/GACEO/2010 Retrospective case series N= 8 patients (5/8 HIV+)

Gold Standard: Serological evidence of syphilis + ocular manifestations

8/8 Serum VDRL Reactive (part of inclusion criteria)

1/5 CSF VDRL reactive

Low quality; small sample size Relevant I
Hughes/Predictive value of serologic tests for syphilis in otology/AORL/1986 Prospective study

 

N=5349 , 25 found to have otologic syphilis.

Gold standard otosyphilis: 1) active inner ear dysfunction not explained by other causes, with or without evidence for systemic syphilis, and 2) positive FTAABS serology with or without positive RPR serology.

 

Control: 1) no evidence by history or physical examination

of previous or present syphilis, and 2) a presenting complaint of hearing loss characterized by bilateral symmetric sensorineural hearing loss and age consistent with a diagnosis of presbycusis, with or without associated acoustic trauma, but without dizziness or symptoms of hydrops. The use of presbycusis control patients is supported in the otolaryngology literature

Sensitivity of serum RPR based on prevalence of 570/100,000: 55%

 

Specificity of serum RPR based on prevalence of 570/100,000: 97%

 

Methodology on sensitivity and specificity a little confusing. Moderately relevant High
Abuzeid/Spirochetes in otology: are we testing for the right pathogens?/otolaryngology/H&NSurg/2007 Retrospective case series: N= 181 patients with idiopathic progressive sensorineural hearing loss (SNHL) 9 cases of otosyphilis identified with Gold standard: positive TPA and SNHL Sensitivity serum RPR:

8/9 positive: 88.9%

 

No CSF results reported

Small numbers, diagnosis not certain Minimally relevant Low
Gleich/otosyphilis: a diagnostic and therapeutic dilemma/Laryngoscope/1992 Retrospective case series N= 18 patients

Gold standard: positive serum FTA and sensorineural hearing loss, tinnitus or vertigo WITH a normal brainstem auditory response or posterior fossa MRI

Sensitivity CSF VDRL:

1/18+ 5.6% positive

 

Sensitivity: Serum VDRL:

4/18 positive 22.2%

 

 

Small numbers diagnosis not certain Moderately relevant. Medium
Yimtae/otosyphilis: a review of 85 cases/O-H&Nsurg/2007 Retrospective case series N= 85 cases of otosyphilis from Thailand

 

Gold standard: Positive serum VDRL AND treponemal test WITH cochleovestibular symptoms

LP done in 37 patients

Sensitivity CSF VDRL: 2/37 5.4%

 

Serum RPR part of definition

Diagnosis not certain Limited relevance Medium