CD003362

Male circumcision for prevention of heterosexual acquisition of HIV in men

Date of most recent substantive amendment: 3 July 2001

BACKGROUND

Male circumcision is defined as the surgical removal of all or part of the prepuce (foreskin) of the penis and may be practiced as part of a religious ritual usually conducted shortly after birth, as a medical procedure related to infections, injury or anomalies of the foreskin, or as part of a traditional ritual performed as an initiation into manhood (Horizons 00). For over a decade observational studies have pointed to an association between male circumcision and HIV infection in males. Most of these studies suggest a protective effect of male circumcision on HIV acquisition in men.

Five reviews (Moses 94; De Vincenzi 94; Moses 98; Van Howe 99; Weiss 00) and one meta-analysis (Weiss 00) of these observational studies have been published, reaching different conclusions on the association between circumcision and HIV infection. However, search strategies are not clearly described in all the reviews, several focused only on published studies and confounding was not always assessed adequately. The most rigorous is a systematic review and meta-analysis recently published by Weiss (Weiss 00), but this review was limited to published studies on HIV-1 infection in sub-Saharan Africa. Adjusted analyses produced a relative risk (RR) of 0.42 (95% CI: 0.34-0.54) for all studies combined (N=15) with a RR of 0.55 (95% CI: 0.42 - 0.72) for population-based cross-sectional studies (N=5) and a RR of 0.24 (95% CI: 0.18 - 0.31) for cross-sectional studies of high-risk groups (N=4). The authors conclude that there is compelling evidence that male circumcision is associated with a reduced risk of HIV infection in sub-Saharan Africa despite warning that residual confounding may exist in some studies due to behavioural or biological factors that are unknown or unmeasured.

Known sources of confounding identified by all the above reviews include sexual behaviour, penile hygiene and religion. Circumcision itself may be a proxy measure of the knowledge and behaviour learnt during the process of initiation during which time young men are taught about traditional sexual practices, including monogamy, and penile hygiene. A potential confounder that has not been measured in any study to date to our knowledge is the use of vaginal drying agents in female partners of the men. This practice is reportedly common in parts of Africa (Brown 93; Kun 98; Runganga 95) and may result in increased vaginal abrasions and micro-lacerations, possibly facilitating HIV transmission in both directions. Viral load is increasingly considered to be a crucial factor in HIV transmission (Quinn 00) and may be both an important confounder and an effect modifier. Misclassification of exposure is also an important source of bias given that some studies classify circumcision status by self-report rather than direct observation.

Biological theories to support the protective effect of circumcision on HIV exist. Researchers have noted that the inner aspect of the foreskin is well-supplied with Langerhans cells (Szabo 00) and that in vitro, HIV-1 demonstrates a specific tropism (attraction) for these cells (Soto-Ramirez 96), in particular the CD4 receptors (Hussain 95). CD4 and other HIV co-receptors have been shown to facilitate HIV entry into host cells, although this has yet to be demonstrated specifically for preputial Langerhans cells. According to this theory, circumcision would remove the potential entry site for HIV. However, not all Langerhans cells are removed during circumcision as even after the procedure, there is residual penile mucosa of the glans and there are also Langerhans cells in the penile shaft (Cold 99). In direct contradiction to the above theory, the inner prepuce contains apocrine glands which secrete lysozyme (Fleiss 98). Lysozyme reportedly kills HIV-1 in vitro (Lee-Huang 99), suggesting a protective effect of the foreskin. As the study of the immunological function of the prepuce is not well-developed (Cold 99), caution must be observed when assuming in vitro viral behaviour is equivalent in vivo.

The presence of a sexually transmitted infection (STI) enhances HIV infection and susceptibility (Cohen 98; Grosskurth 95; Fleming 99). The role circumcision plays in the transmission of STIs is less certain, although it has been reported that male circumcision is associated with a reduced risk of genital ulcer disease, particularly chancroid and syphilis (Moses 98). It would therefore be valuable to examine the relationship between circumcision and STIs. Should this relationship prove to be temporal, it seems likely that circumcision may protect against HIV infection indirectly via decreased STI transmission.

Circumcision practices are largely culturally determined and as a result there are strong beliefs and opinions surrounding its practice. It is important to acknowledge that researchers' personal biases and the dominant circumcision practices of their respective countries may influence their interpretation of findings.

Given the enormous mortality and morbidity associated with HIV/AIDS, it seems reasonable to fully explore potential prevention measures, including male circumcision. However, promoting or instituting mass circumcision may have profound cultural and social implications and represents a formidable public health challenge (Cohen 00). Of particular concern is the potential negative impact introduction of circumcision may have on current health promotion endeavours to promote sexual behavioural change. This may include promulgation of the belief that circumcision completely protects against HIV transmission, resulting in a lack of condom use. The following systematic review seeks to inform this debate by presenting the evidence from both published and unpublished studies from around the world which examine the association between circumcision and HIV-1 and HIV-2 infection.

OBJECTIVES

1) To examine the association between male circumcision and acquisition of HIV-1 and HIV-2 by men through heterosexual intercourse.
2) To examine the feasibility and value of performing individual person data (IPD) meta-analysis.

CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW

Randomised or quasi-randomized controlled trials. If insufficient data are available from randomised controlled trials to draw any meaningful conclusions, data from observational studies (e.g. cohort, case-control and cross-sectional studies) will be considered for inclusion in this review. Studies performed in general or specific populations and in hospitals or clinics will be included. Studies performed in any country and published in any language will be included. Studies with historical controls and ecological studies will be excluded.

Heterosexual men included in studies assessing the association between circumcision and HIV-1 and HIV-2. Men are defined as males 12 years or older.

Male circumcision is defined as removal of the foreskin of the penis either via surgical technique or via cultural practices which involve cutting off the foreskin. Circumcision status can be determined by self- or partner-report or by direct observation.

The outcome measures are:

HIV-1 or HIV-2 infection in men

Any adverse events associated with circumcision will be recorded if reported in the trials.

SEARCH STRATEGY FOR IDENTIFICATION OF STUDIES

See: HIV/AIDS Collaborative Review Group search strategy

Electronic searches will be undertaken using the following databases: MEDLINE, Embase, AIDSLINE, CINAHL, Scisearch, the Cochrane Controlled Trials Register, Database of Abstracts of Reviews of Effectiveness (DARE), the Cochrane HIV/AIDS and the Cochrane STD registers of studies. Hand searches of the reference lists of all pertinent reviews and studies found will also be undertaken, as well as abstracts from relevant conferences. Agencies, organizations and academic centers, as well as experts in the field of HIV prevention will be contacted to locate any further studies or relevant conference proceedings that may not be included in the databases and to ensure that unpublished studies are included.

The search strategy will be iterative.

METHODS OF THE REVIEW

The search for trials will be performed with the assistance of the Cochrane HIV/AIDS Group. The selection for potentially relevant studies will be undertaken by NS and MM. The titles, abstracts and descriptor terms of all downloaded material from the electronic searches will be read and irrelevant reports discarded to create a pool of potentially eligible studies. All citations identified will then be independently inspected by NS and MM to establish relevance of the article and whether or not the full article should be acquired. If there is uncertainty, the full article will be obtained.

1. Selection of studies
NS and MM will independently apply the inclusion criteria. Differences will be resolved by discussions with a third reviewer, JV. Studies will be reviewed for relevance based on study design, types of participants, exposures and outcome measures. Finally, where resolution is not possible because further information is necessary, the study will be allocated to the list of those awaiting assessment. Attempts will then be made to contact authors to provide further clarification of data.

2. Data extraction
Data will be extracted independently by NS and MM using a standardised data extraction form. The following characteristics will be extracted from each included study:

Administrative details:
Identification; author(s); published or unpublished; year of publication; number of studies included in paper; year in which study was conducted; details of other relevant papers cited.

Details of study:
Study design; type, duration and completeness of follow-up in the case of cohort studies; country and location of the study; setting in which the study was performed (e.g. urban or rural; population or hospital/clinic based); background HIV prevalence of the selected study sample (high-risk or low-risk); method(s) of recruitment; number of participants

Characteristics of participants:
Age; religion; socio-economic status; dominant cultural practices regarding circumcision; number of sexual partners; condom use; other identified risk factors (e.g. presence of sexually transmitted infections; use of vaginal drying agents in female partner)

Details of intervention:
Circumcision based on self- or partner-report or direct observation; age at circumcision; circumcision procedure used (surgical procedure or traditional); reason for circumcision (traditional beliefs or medical treatment)

Details of outcomes:
HIV infection (incidence and prevalence); number of HIV positive men in the circumcised and uncircumcised groups (n/N) (except for case-control studies, where number of men circumcised in the case and control groups will be recorded); types of tests used to determine HIV status; method of surveillance of adverse effects associated with circumcision; number of men with specific adverse effects associated with circumcision (n/N).

3. Quality assessment
Quality assessment will be done using a standardised quality assessment form.

Randomised Controlled Trials:
The methodological quality of the included trials will be evaluated independently by NS and MM in terms of method of randomisation, adequacy of allocation concealment, blinding, intention-to-treat and loss-to-follow-up.

Observational Studies:
The methodological quality of the included cohort, case-control and cross-sectional studies will be evaluated independently by NS and MM. One further reviewer JD will check the use of quality assessment. In all types of studies, particular emphasis will be placed on the method of selection of the sample (criteria specific to the study design), method of ascertainment of circumcision status, and method of ascertainment of HIV status. In cohort studies, quality assessment will include the type, length and completeness of follow-up and in case-control studies, case definition and control selection will be included.

Reviewers will not be blinded to the names of the authors, institutions, journal of publication or results of the studies.

4. Data synthesis
Incomplete data:
If data are incomplete, attempts will be made to contact the authors for clarification of important information.

Outcome measures:
All outcomes included in this review will be binary. Measures of association for each study will be expressed as crude and adjusted odds ratios together with their 95% confidence intervals. Odds ratio will be used because they are the only valid estimator from case-control studies, and from adjusted analyses obtained from logistic regression. To assist interpretation the overall results will be translated into measures of absolute benefit (attributable risk percent) for both population-based and high-risk groups.

5. Sub-group analysis, investigation of heterogeneity and sensitivity analysis
Meta-analysis will be stratified according to study design (RCT versus cohort studies versus case-control versus cross-sectional studies). Within each study design strata we will further stratify for population studies versus studies in groups perceived to be at high risk of HIV.

Where possible meta-analysis will be conducted combining odds ratios using a random effects (DerSimonian and Laird) model as between study heterogeneity is anticipated. Heterogeneity will be tested by REVMAN software using the Chi square test for heterogeneity. Between-study heterogeneity (p <0.1) will be explored by subgroup analysis and meta-regression depending on the data available. The following study factors will be examined:

- Circumcision: studies which assess status by self-report versus direct observation; medical procedure versus traditional procedure
- Background prevalence of HIV (low or high) in the general population of the study country or study region
- HIV type: HIV 1 or HIV 2

The effect of continuous variables such as number of sexual partners in a specified time will be further explored using random effects meta-regression (STATA) if data allow. In cases where a meta-analysis is not possible and/or appropriate, narrative data synthesis will be provided.

We will conduct sensitivity analysis based on study quality using meta-regression if appropriate. The following effects will also be investigated using sensitivity analysis:

- Adjusted and unadjusted analyses
- Published and unpublished studies

POTENTIAL CONFLICT OF INTEREST

None known.

REFERENCES

Brown 93

Brown J.E., Ayowa O.B., Brown R.C. Dry and tight - sexual practices and potential AIDS risk in Zaire. Social Science and Medicine 1993;37:989-994.

Cohen 00

Cohen M.S. Preventing sexual transmission of HIV - new ideas from sub-Saharan Africa The New England Journal of Medicine 2000;342:970 - 972.

Cohen 98

Cohen M.S. Sexually transmitted diseases enhance HIV transmission: no longer a hypothesis. The Lancet 1998;351 Suppl:S5 - S7.

Cold 99

Cold C.J., Taylor J.R. The prepuce. British Journal of Urology 1999;83:34-44.

De Vincenzi 94

De Vincenzi I.D., Mertens T. Male circumcision: a role in HIV prevention?. AIDS 1994;8:153-160.

Fleiss 98

Fleiss P.M., Hodges F.M., Van Howe R.S. Immunological function of the human prepuce. Sexually Transmitted Infections 1998;74:364-367.

Fleming 99

D.T. Fleming, J.D. Wasserheit. From epidemiological synergy to public health policy and practice: the contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sexually Transmitted Infections 1999;75:3-17.

Grosskurth 95

Grosskurth H., Mosha F., Todd J., Mwijarubi E., Klokke E., Senkoro K., Mayaud P., Changalucha J., Nicoll A., ka-Gina G., Newell J., Mugeye K., Mabey D., Hayes R. Impact of improved treatment of sexually transmitted diseases on HIV infections in rural Tanzania: randomised controlled trial. The Lancet 1995;346:530-536.

Horizons 00

Horizons. Report: Male circumcision and HIV prevention: Directions for future research.. Washington DC: Office of Health and Nutrition, Global Bureau, U.S. Agency for International Development. 2000.

Hussain 95

Hussain L. A., Lehner T. Comparative investigation of Langerhans' cells and potential receptors for HIV in oral, gentourinary and rectal epithelia. Immunology 1995;85:465-484.

Kun 98

Kun K.E. Vaginal drying agents and HIV transmission. Family Planning Perspectives 1998;24:93-94.

Lee-Huang 99

Lee-Huang S., Huang P.L., Sun Y., Kung H.F., Blithe D.L., Chen H.C. Lysozyme and RNases as anti-HIV components in Beta-core preparations of human chorionic gonadotropin. Proceedings of the National Academy of Science (USA). 1999;96:2678-2681.

Moses 94

Moses S., Plummer F.A., Bradley J.E., Ndiya-Achola J.O., Nagelkerke N.J.D., Ronald A.R. [The association between lack of male circumcision and risk for HIV infection: a review of the epidemiological data]. Sexually Transmitted Diseases 1994;21:201-210.

Moses 98

Moses S., Bailey R.C., Donald A.R. Male circumcision: assessment of health benefits and risks. Sexually Transmitted Infections 1998;74:368-373.

Quinn 00

Quinn T.C., Wawer M.J., Sewankambo N., Serwadda D., Li C., Wabwoire-Mangen F., Meehan M.O., Lutalo T., Gray R.H. Viral load and heterosexual transmission of human immunodeficiency virus type 1. The New England Journal of Medicine 2000;342:921-929.

Runganga 95

Runganga O.A., Kasule J. The vaginal use of herbs/substances: an HIV transmission facilitatory factor?. AIDS CARE 1995;7:639-645.

Soto-Ramirez 96

Soto-Ramirez L. E., Renjifo B., McLane M. F., Marlink R., O'Hara, C., Sutthent, R., Wasi, C., Vithayasi, P., Vithayasai, V., Apichartpiyakul, C., Auewarakul, P., Cruz, V. P., Chui, D.-S., Osathanondh, R., Mayer, K., Lee, T.-H., Essex, M. HIV-1 Langerhans' cell tropism with heterosexual transmission of HIV. Science 1996;271:1291-1293.

Szabo 00

Szabo R., Short R.V. How does male circumcision protect against HIV infection?. British Medical Journal 2000;320:1592-1594.

Van Howe 99

R.S. Van Howe. Circumcision and HIV infection: review of the literature and meta-analysis. International Journal of STD and AIDS 1999;10:8-16.

Weiss 00

Weiss H.A., Quigley M.A., Hayes R. Male circumcision and risk of HIV infection in sub-Saharan Africa: a systematic review and meta-analysis. AIDS 2000;14:2361-2370.

COVER SHEET

SOURCES OF SUPPORT

• HRP-UNDP/UNFPA/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction, Geneva SWITZERLAND
• Liverpool School of Tropical Medicine UK
• Effective Health Care Alliance UK

• Medical Research Council SOUTH AFRICA

ADDITIONAL TABLES

Additional tables are not available for this protocol