Literature review > Issue 1 > Review Scoular et al. 

Sometimes when I lecture on genital herpes, I illustrate the attitude of the public health establishment toward the disease with a photograph of an unclothed person with his head literally buried in the sand, his nether anatomy bared to the chill winds of an unseen world. Belatedly, some public health experts and agencies are coming to understand the direct and indirect impacts of genital herpes and neonatal herpes, the substantial psychosocial morbidity for many infected persons, and the potent influence of herpes simplex virus type 2 (HSV-2) infection on sexual transmission of human immunodeficiency virus (HIV) [1-3]. Still, standardized recommendations for diagnosis and prevention of genital herpes remain few and far between, despite the availability of effective therapy and substantially effective prevention strategies, and until last year, the U.S. Centers for Disease Control and Prevention (CDC) advised that persons with genital ulcer disease (GUD) undergo diagnostic testing for Treponema pallidum and Haemophilus ducreyi but not for HSV, far and away the most likely cause. Anecdotal evidence suggests that few public health agencies, health care providers, or health maintenance organizations systematically attempt to diagnose, treat, or prevent genital herpes. Head in the sand, indeed.

CDC's 2002 STD treatment guidelines [4] are an important step forward. Routine testing of lesions for HSV is advised for all patients with GUD, as is determination of virus type in all patients with genital HSV infection, because the risks of clinical recurrences and subclinical transmission are substantially greater for HSV-2 than HSV-1 infection [5,6]. Culture has been the standard virologic test; it is relatively inexpensive and widely available in industrialized countries, and virus type is readily determined [7]. Direct fluorescent antibody tests for HSV are faster and less expensive than culture but not as sensitive, particularly in healing or recurrent lesions, and not all assays distinguish HSV-1 from HSV-2 [7].

Polymerase chain reaction (PCR) for detection of HSV DNA is substantially more sensitive than culture for the diagnosis of HSV infections of the central nervous system and for detection of subclinical viral shedding [5,7-9], and several studies have documented improved sensitivity over culture in detecting HSV in genital lesions [9-13]. A. Scoular, G. Gillespie, and W. F. Carman of the Glasgow Royal Infirmary and associated institutions in Glasgow, Scotland, evaluated an HSV PCR using the LightCycler model LC32 (Idaho Technologies) for rapid specimen processing; the results were compared to those of virus culture among patients with GUD attending an STD clinic [14]. Virus type was determined by restriction endonuclease digestion of PCR-amplified HSV DNA (i.e., restriction fragment length polymorphism, RFLP). During 6 months in 1998 and 1999, the investigators evaluated 236 consecutive patients with clinical features suggestive of genital herpes, defined as vescicular, ulcerative, or crusting lesions of the genitals, anus, perineum, buttocks, or thighs [14].

HSV was identified by both PCR and culture in the lesions of 88 patients (37%). In an additional 21 patients, HSV was identified by PCR but not culture, raising the total yield to 109 (44%) of the patients. No specimen was positive only by culture. It is probable that PCR for HSV DNA has a specificity of 100% or close to that figure, so the investigators' conclusion is reasonable that the sensitivity of culture was 81% (88 of 109) compared to PCR, equivalent to a 24% higher yield (21/88) for PCR. Not surprisingly, higher PCR yield and improved sensitivity over virus isolation was associated with duration of disease, especially lesions of >5 days duration, and the increment in yield was greater for ulcerative or crusting lesions than vesicles. These results are similar to those of earlier reports in patients with GUD, in which the sensitivity of virus isolation was 67% to 81% compared with PCR [10-13].

Although PCR clearly outperforms virus isolation, there are concerns about its actual sensitivity. In the largest, recent, well-controlled study of the etiology of GUD, a multiplex PCR assay for HSV, T. pallidum, and H. ducreyi was performed in 516 patients at 10 STD clinics in the U.S., and HSV infection was diagnosed in 65% [15]. The study sites were selected because they had especially high rates of syphilis and patients with clinically typical genital herpes were excluded, so it is likely that the true prevalence of HSV infection in GUD patients in similar STD clinics is still higher, perhaps 80% or more. Even though the prevalence of HSV-2 infection is substantially lower in the U.K. than in the U.S. [16], the 44% yield found in the Glasgow study is low for patients with the clinical presentations described. Testing for HSV antibody with a glycoprotein G-based assay or Western blot [7] could have been used to distinguish initial from recurrent genital herpes and to help discern the extent to which both culture and PCR were falsely negative.

Despite promoting RFLP to determine virus type, including display of an electrophoretic gel that clearly differentiates HSV-1 and HSV-2, Scoular et al. do not present data on HSV type for the 21 specimens in which the virus was identified only by PCR [14]. Among the 88 culture-positive specimens, HSV-1 was isolated in 46 patients (52%) and HSV-2 in the remainder. It would have been useful to know the performance of RFLP compared with traditional methods in determining virus type and whether PCR might have differential sensitivity for HSV-1 and HSV-2. Thus, it is not possible to determine the absolute sensitivity of the investigators' HSV PCR assay from the data presented. Wald et al. recently reported that a noncommercial PCR developed in their laboratory is up to fourfold more sensitive than culture [9], further casting doubt on the sensitivity of the assay described by Scoular et al.

These concerns notwithstanding, the central conclusion of the Glasgow investigators-that the PCR run on the LightCycler instrument is more sensitive than culture for the virologic diagnosis of HSV in genital ulcers-is valid, and there is a special advantage of PCR when patients present with more prolonged or more mature lesions. Scoular et al. go on to promote PCR using the LightCycler, plus RFLP for type determination, as the routine method for diagnosis of herpes in persons with GUD [14]. Here I have concerns.

For the large majority of laboratories worldwide, the conclusion that the authors' method is cost effective is questionable at best. Despite the manifest importance of accurate diagnosis of genital herpes, it is not clear that the 24% increment in sensitivity is sufficient to warrant routine implementation of a substantially more expensive technology, especially by laboratories that do not already have PCR capability, the resources to purchase a LightCycler, or a testing volume that might permit negotiation of lower reagent costs or donation of machinery by the manufacturer. It would be unfortunate if laboratories and clinicians were to forego inexpensive testing by culture for want of more sensitive but substantially more expensive testing by PCR. In addition, accurate type-specific serological testing is now readily available in industrialized countries, and staged testing by virus isolation followed by a glycoprotein G-based serological assay in culture-negative patients might be an equally sensitive and more cost-effective approach.

PCR is crucial as a research tool (e.g., to detect subclinical HSV shedding) [5,9] and is the only suitable virologic test for HSV in some clinical settings, such as the diagnosis of neonatal herpes or suspected central nervous system HSV infection [7,8]. Where validated PCR assays are offered at reasonable cost, they should supplant virus isolation as the test of choice in patients with GUD, and HSV PCR should be routinely available in reference laboratories [7-9]. The PCR assay described by Wald et al. has been studied only in patients with known HSV infection, including persons with subclinical viral shedding [9]; it remains to be seen whether the test offers a similar enhancement in sensitivity in unselected patients with GUD. Accordingly, for some years to come culture is likely to remain the primary virologic test for HSV in most settings, and direct fluorescent antibody tests that distinguish HSV-1 from HSV-2 also have acceptable performance. Finally, all clinicians and laboratories should have access to and routinely use the glycoprotein G-based type-specific serological tests as a diagnostic adjunct and to screen selected persons at risk.

The bottom line is that virologic and serologic testing for HSV infection are grossly underutilized worldwide in patients with genital ulcer disease. From both clinical and public health perspectives the top priority is to assure that accurate diagnostic testing is readily accessible and liberally used. It is time to pull our heads out of the sand.

References

1.  Corey L, Handsfield HH.  Genital herpes and public health:  addressing a global problem. JAMA 2000;283:791-5.

2.  Brown ZA, Wald A, Ashley-Morrow R, Selke S, Zeh J, Corey L.  Effect of serologic status and cesarean delivery on transmission rates of herpes simplex virus from mother to infant. JAMA 2003;389:203-9.

3.  Wald A, Corey L. How does herpes simplex virus type 2 influence human immunodeficiency virus infection and pathogenesis?  J Infect Dis 2003;187:1509-12.

4.  Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2002.  MMWR 2002;51(No. RR-6):11-7.

5.  Wald A, Zeh J, Selke S, et al.  Reactivation of genital herpes simplex virus type 2 infection in asymptomatic seropositive persons.  N Engl J Med 2000;342:844-50.

6.  Engelberg R, Carrell D, Krantz E, Corey L, Wald A.  Natural history of genital herpes simplex virus type 1 infection.  Sex Transm Dis 2003;30:174-7.

7.  Jerome KR, Ashley RL.  Herpes simplex viruses and herpes B virus.  In  Murray P, et al (ed), Manual of Clinical Microbiology, 8th edition.  Washington, DC, ASM Press, 2003:1291-303.

8.  Lakeman FD, Whitley RJ.  Diagnosis of herpes simplex encephalitis:  application of polymerase chain reaction to cerebrospinal fluid from brain-biopsied patients and correlation with disease.  J Infect Dis 1995;171:857-63.

9.  Wald A, Huang M-L, Carrell D, Selke S, Corey L.  Herpes simplex virus DNA polymerase chain reaction for detection of HSV on mucosal surfaces:  comparison with HSV isolation in cell culture.  J Infect Dis, in press.

10. Cullen AP, Long CD, Lorincz AT.  Rapid detection and typing of herpes simplex virus DNA in clinical specimens by the hybrid capture II signal amplification probe test. J Clin Microb. 1997;35:2275-8.

11. Safrin S, Shaw H, Bolan G, Cuan J, Chiang CS.  Comparison of virus culture and the polymerase chain reaction for diagnosis of mucocutaneous herpes simplex virus infection.  Sex Transm Dis 1997;24:176-80.

12. Slomka MJ, Emery L, Munday PE, Moulsdale M, Brown DW.  A comparison of PCR with virus isolation and direct antigen detection for diagnosis and typing of genital herpes.  J Med Virol 1998;55:177-83.

13. Waldhuber MG, Denham I, Wadey C, Leong-Shaw W, Cross GF.  Detection of herpes simplex virus in genital specimens by type-specific polymerase chain reaction.  Int J STD AIDS 1999;10:89-92.

14. Scoular A, Gillespie G, Carmen WF.  Polymerase chain reaction for diagnosis of genital herpes in a genitourinary medicine clinic.  Sex Transm Inf 2002;78:21-5.

15. Mertz KJ, Trees D, Levine WC, et al.  Etiology of genital ulcers and prevalence of human immunodeficiency virus coinfection in 10 US cities.  J Infect Dis 1998;178:1795-8.

16. Smith JS, Robinson NJ.  Age-specific prevalence of infection with herpes simplex virus types 2 and 1:  a global review.  J Infect Dis 2002;186 (Suppl 1):S3-28.

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001