The Association Between Male Circumcision and HIV Infection in Sub-Saharan Africa: a Systematic Review of the Literature Gazimbi MM1*, Magadi MA2 and Kruger C3

Home , Circumcision and HIV, Southern Africa

ISSN: 2643-4512 Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 DOI: 10.23937/2643-4512/1710022 Volume 3 | Issue 1 International Archives of Open Access Public Health and Community Medicine

Research Article The Association between Male Circumcision and HIV Infection in Sub-Saharan Africa: A Systematic Review of the Literature Gazimbi MM1*, Magadi MA2 and Kruger C3

1Global Development Institute, University of Manchester, UK 2 Check for School of Education and Social Sciences, University of Hull, UK updates 3Centre for Sexualities, AIDS and Gender, University of Pretoria, South Africa

*Corresponding author: Martin Marufu Gazimbi, Global Development Institute, University of Manchester, M139Pl, United Kingdom

Abstract countries where the practice is not common, the concern that male circumcision might substitute other efforts such as Background: Although both traditional and medical male condom use and behavioural modification interventions must circumcision are now being promoted as part of strategies be addressed alongside ethical concerns such as conflict for HIV prevention in Africa, particularly in countries with low with traditional values. A purely biomedical approach to the circumcision prevalence, there are debates on the efficacy of HIV/AIDS epidemic is unlikely to be sufficient in addressing male circumcision in the prevention of HIV at both individual the continued spread of HIV in sub-Saharan Africa. and population levels. Keywords Methods and objective: In light of debates on the validity of male circumcision as the panacea for HIV prevention, Male circumcision, HIV infection, Systematic literature the objective of this paper is to systematically review review, Sub-Saharan Africa, Medical circumcision evidence on the association between male circumcision and acquisition of HIV infection in sub-Saharan Africa. Introduction Results: Thirty-six articles addressing the association between male circumcision and HIV transmission in sub-Sa- Male circumcision (MC) and female genital mutila- haran Africa were included in our assessment. Consistent tion are believed to be the oldest surgical procedures evidence was found showing the protective effect of male introduced over 65 million years ago [1-4]. Worldwide, circumcision on HIV infection for males at both individual and population level. In particular, evidence from three ran- appropriately 30-38% of males are circumcised [5]. domised control trials comparing the risk of HIV infection be- In the literature, male circumcision is understood to tween circumcised and uncircumcised males in Sub-Sahara have been started as a religious rite among Jews and Africa suggest that male circumcision is significantly asso- Muslims [2,6]. Other motives for circumcision includ- ciated with risk reduction of HIV acquisition from female to male by approximately 50-60%. However, evidence of the ed rites of passage, blood sacrifices, cultural markings, protective effect of male circumcision for females shows enhancement of masculine fearlessness and fecundity, mixed patterns and is inconclusive. Risky sexual behaviours preparation for marriage and adult sexuality, and the post circumcision (ie., inconsistent condom use, having hardening of boys for warfare [2]. Today, circumcision multiple sexual partnership); age at circumcision, surgical is near universal in the North and West African coun- safety, type of circumcision (medical versus traditional), resuming sexual intercourse before the healing of the wound tries, while it is not commonly practiced in Southern Af- have been shown to modify the efficiency of male circumci- rican countries. Male circumcision is being medicalised sion in protecting males against acquisition of HIV. in countries of sub-Saharan Africa where circumcision is Conclusion: Although evidence from existing research not commonly practiced and it has joined existing HIV supports promotion and scaled up of male circumcision in interventions programs such as HIV testing, Antiretro-

Citation: Gazimbi MM, Magadi MA, Kruger C (2019) The Association between Male Circumcision and HIV Infection in Sub-Saharan Africa: A Systematic Review of the Literature. Int Arch Public Health Community Med 3:022. doi.org/10.23937/2643-4512/1710022 Accepted: June 06, 2019; Published: June 08, 2019 Copyright: © 2019 Gazimbi MM, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 1 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512 viral drug (ARV), condom use and abstinence programs There is also a concern that male circumcision might [7,8] following the successful randomised control trials substitute other efforts such as condom use andbe- [9-11] showing the efficacy of voluntary male circum- havioural modification [7,8,16,22-25] and no researcher cision in controlling the spread of HIV. For men who has ever advocated male circumcision as a natural con- are not part of traditionally circumcising communities, dom [7]. Not only are the debates on male circumcision voluntary medical MC presents an option as a HIV pre- and HIV ongoing within the Western medical commu- ventative measure. It is also presented, however, as a nity, Africans themselves are in discussions about male way to introduce more sterile practices to those men circumcision and HIV/AIDS [8]. Political and traditional who would already be set to undergo circumcision. De- leaders generally endorse the idea that male circumci- bates around medical male circumcision and traditional sion is protective against HIV [26]. The low prevalence initiation practices often situate these two practices as of HIV among circumcised societies has given the prac- binaries [12]. Often, however, there is very little discus- tice a stronger rationale or justification in a modern day sion about the ideological underpinnings of these two society than even traditional and religious demands can options. Campaigns promoting voluntary medical male provide. However, based on studies published to date, circumcision seem to position it as a safe alternative to recommending routine circumcision as a prophylactic traditional circumcision without paying much attention measure to prevent HIV infection maybe a counter- to the socio-cultural nuances attached to the larger con- productive and short-lived intervention, because there cept of the initiation process. are some evidence from trials [9,24] of serious adverse A scan on the literature on medical circumcision re- events from the procedure and of behavioural risk com- veals ethical questions regarding the implementation of pensation among circumcised men. Looking at the cor- medical male circumcision, and rolling out this program pus of scientific literature concerning the relation of without adequately addressing the ethical concerns male circumcision and HIV may be helpful in developing could be counterproductive. Male circumcision has long recommendations for HIV prevention. The objective of been perceived to promote health, but traditionalist this paper is therefore, to systematically review scientif- and cultural activities in Africa have not yet claimed that ic literature on the association between HIV and male they practice circumcision in order to reduce HIV infec- circumcision in SSA countries. The study will also eval- tion [13]. To them it is done to promote a cultural sense uate whether the impact of male circumcision on HIV of manhood and sexual enhancement. In fact, commu- acquisition in women, surgical safety or risky sexual be- nities practising traditional male circumcision were until haviours could explain the contradictory evidence in the recently discourage from practising circumcision and association between HIV and circumcision. warned that their traditional practises fuel the spread Methods of HIV [14]. The irony now in research is that circumcision is reported to have contributed to the reduction of The Academic Search Complete, CINAHL Complete, HIV among ethnic groups that practice traditional male MEDLINE, PsycINFO and PubMed databases were circumcision [13]. Evidence that societies that practice searched for papers published up to April 2018 that circumcision are less at risk of HIV epidemic is now be- included ‘circumcision’ and ‘HIV’ as keywords or text ing used as a basis to launch large scale male circum- in the abstract. Eight hundred and nine (n = 809) cision initiatives. The evidence should be disseminated original research studies were identified in this way. with caution to avoid creating a false sense of security A further search was carried out using keywords to among traditionally circumcised men who may believe search for all published studies of HIV risk factors in that they are not vulnerable to HIV infection. men in sub-Saharan Africa using the search condition While medical male circumcision is increasingly be- ‘(HIV-infections (epidemiology, transmission)’ OR ‘HIV- ing incorporated in comprehensive strategies for the sero-prevalence’ OR ‘HIV sero-positivity (epidemiology, prevention of HIV [15], there are reports suggestive of transmission)’ AND ‘sub-Saharan Africa’ AND (‘risk an increased risk of HIV infection due to modified sexual factors’ or ‘risk’). A total of hundred and seven (n = behaviour post-circumcision [16] and non-sexual activ- 107) articles remained after screening in this way. Of ities such as circumcision using unhygienic procedures these, abstracts of fifty-nine (n = 59) papers focused on [17]. Surgical safety of male circumcision is of great con- health economics of voluntary male circumcision and cern, particularly in traditional male circumcision [4]. It were excluded, yielding forty-eight (n = 48) potential was reported in Kenya that knives that are traditionally and eligible studies. Finally, we removed thirteen (n = used to remove the foreskin, are sometimes blunt and 13) papers that used a proxy for circumcision, such as can tear apart the male organ, leaving scars which ex- Muslim religion and those focussing on new-born and poses them to HIV and other sexually transmitted dis- infants circumcision living the final sample of 36 for eases [18]. Others have highlighted that the fragile fore- analysis. skin is susceptible to scratches and tears and it contains Description of studies specialized cells (e.g., Langerhans) that join readily with HIV and other pathogens [19-21]. The studies included in the review are shown in

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 2 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

Table 1, grouped by author/year of publication, study These quantitative studies included ten cross-sectional design and key findings of the study. The eligible stud- studies using DHS data [3,17,19,24,27-30], case-con- ies were published between 1989 and 2017. Twenty trol studies [22,23], two cohort studies [18,31]; and six studies used quantitative research methods to evaluate randomised control trials [8,9,11,31-33]. Eight studies the relative risk of male circumcision on HIV infection. used a systematic review approach [4,5,7,33-35]. The

Table 1: Studies addressing the association between male circumcision and HIV transmission in SSA countries.

Author/s, Year Study design, population Title Key findings Grund, et al. [7] Systematic reviews; 60 Association between male Although the evidence shows a protective publication (sub-Saharan circumcision and women’s association with HIV, it was categorised as Africa-studies up to April biomedical health outcomes: low consistency, because one trial showed 2016) a systematic review an increased risk to female partners of HIV- infected men resuming sex early after male circumcision. Makatjane, et al. [27] Cross-sectional Sample MC and HIV in Lesotho: Is Medically circumcised men have lower (n = 2283 sexually active the Relationship Real or odds of HIV infection than uncircumcised males), Lesotho Spurious? Analysis of the men. Odds of HIV infection among 2009 DHS traditionally circumcised men were similar to those of uncircumcised. Ediau, et al. [22] Case-control study Risk factors for HIV infection Being circumcised at adulthood >18 (HIV Infected = 155 vs. among circumcised men in years (AOR = 5.0, 95% CI = 2.4-10.2), Controlled group HIV Uganda: a case-control study resumption of sexual intercourse before infected = 155), Uganda would healing (AOR = 3.4, 95% CI = 1.6- 7.3), inconsistent condom use (AOR = 2.7, 95% CI = 1.5-5.1) were significantly associated with HIV infection post- circumcision. Ombere, et al. [23] Qualitative study Semi- Wimbo: implications for Men develop false belief that circumcision structured interviews (n = risk of HIV infection among provides protection against HIV infection 101) IDI (n = 10), Kenya circumcised fishermen in post-circumcision. The prevalence of Western Kenya condom use reported very low among migrant fishermen and that risky sexual behaviour was very high post circumcision. Mauughan-Brown, et Randomised control trial What Do people Actually Men randomly assigned to receive al. [8] Treatment group (n = 609), Learn from public Health information about the protective benefits of Control group (n = 619), Campaigns? Incorrect circumcision were more likely to adopt the Malawi Inferences about male erroneous beliefs that MC reduces HIV risk circumcision and Female HIV for men. infection Risk Among men and Women in Malawi Kim and Poulin, et Cross-sectional (N = 3400), Ethnic identity, region, Attitudes toward circumcision varied al. [28] Malawi and attitudes toward male by ethnicity and region. Acceptance of circumcision in a high HIV– circumcision as a tool for HIV prevention prevalence country divided by ethnoregional identities that also shape the practice of circumcision. Greely, et al. [16] Qualitative (FGD = 15), Traditional male circumcision Concerns raised post MC included: effects South Africa for reducing the risk of HIV of botched traditional circumcision, risky infection: perspectives of sexual young people in South Africa behaviour involving multiple sexual partners and non-condom use exhibited by some men after circumcision. Sakutukwa, et al. Qualitative FGD in Understanding and Concerns about the impact of HIV on [26] Zimbabwe (n = 4) FGD in addressing socio-cultural communities resulted in willingness to South Africa (n = 4) barriers to medical male consider adult male circumcision. Adult circumcision in traditionally MC-promotional messages that create a non-circumcising rural synergy between understandings of both communities in sub-Saharan traditional and medical circumcision will be Africa more successful in these communities. Ayiga, et al. [24] Cross-sectional (data from Impact of male circumcision Circumcised males less likely to use Botswana AIDS Impact on HIV risky compensation Condom at the last sexual intercourse than Survey N = 1257 men), through the impediment of uncircumcised males (AOR = 1.34, 95% CI: Botswana condom use in Botswana 0.82-2.15).

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 3 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

Gebremedhin S [40] Cross-sectional Male circumcision and its Uncircumcised status was significantly Demographic Health association with HIV infection associated with risk of HIV, with odds ratio Surveys (DHS) 18 and sexually transmitted (OR) of 4.12 (95% CI: 3.85 - 4.42). The countries (n = 70,554 diseases: evidence from association was even more significant males aged 15 - 59 years) 18 demographic and health 4.95 (95% CI: 4.57-5.36) after adjustment surveys in sub-Saharan for number of lifetime sexual partners and Africa socio-demographic variables. Bailey, et al. [15] Qualitative study FGD The acceptability of male Barriers to acceptance of male circumcision circumcision to reduce included cultural identification, fear of pain HIV infections in Nyanza and excessive bleeding and cost. To be Province, Kenya accepted, both men and women were eager for promotion of genital hygiene and male circumcision, and they prefer medical circumcision than traditional. But they believe clinicians lacked the knowledge and resources to offer safe circumcision counselling and services. Obure, et al. [25] Qualitative study 24 FDG Psychosocial Factors Perceived barriers to circumcision were with Luo men, Kenya Influencing Promotion of MC pain and healing complications, cultural for HIV Prevention in a Non- identity, and reduced sexual satisfaction; Circumcising Community in perceived facilitators were hygiene, HIV/ Rural Western Kenya. STI risk reduction, ease in condom use, cultural integration, and sexual satisfaction. Wawer, et al. [32] RCT Intervention; (n = 474) Circumcision in HIV-infected The trial was stopped early because of Control; (n = 448). men and its effect on HIV futility. 92 couples in the intervention group transmission to female and 67 couples in the control group were partners in Rakai, Uganda: a included in the modified ITT analysis. 17 randomised controlled trial. (18%) women in the intervention group and eight (12%) women in the control group acquired HIV during follow-up (p = 0.36). Siegfried, et al. [21] Systematic reviews; 24 Male circumcision for There was a strong epidemiological studies retrieved (Sub- prevention of heterosexual association between male circumcision and Saharan Africa- studies up acquisition of HIV in men prevention of HIV (a relative risk reduction to June 2007) of acquiring HIV of 50%), especially among high-risk groups. Bailey, et al. [41] Prospective study (n = Male circumcision for HIV Of 443 circumcised traditional, 156 1007 men interviewed post- prevention: a prospective experienced an adverse event compared to circumcision season) study of complications in 99 (559) circumcised clinically (OR, 2.53; clinical and traditional 95% CI. 1.89-3.38) settings in Bungoma, Kenya Londish, et al. [29] Mathematical model Significant reduction in HIV Complete coverage of MC could reduce simulation prevalence according to male HIV prevalence from 12 to 6% for an circumcision intervention in average population in SSA Sub-Saharan Africa Tussime [42] Systematic review; 13 Circumcision and HIV 13 studies were included: Circumcised men studies, from 1997-2007 infection: Assessment of had a reduced risk of HIV infection by (RR (Sub-Saharan Africa) causality = 0.42, 95% CI 0.33-0.53) Mattson, et al. [33] Qualitative study in-depth Risk compensation, male No evidence of differential risk reduction interviews about sexual circumcision, and HIV between circumcised and uncircumcised behavior post circumcision. prevention in Kisumu, men 6 or 12 months post-enrolment. Kenya. Thus, there was no evidence of risk compensation Lukobo, et al. [36] Qualitative study 24 FGD in Acceptability of male In communities where circumcision is four districts, Zambia circumcision for prevention of little practiced, the main facilitators for HIV infection in Zambia. acceptance were improved genital hygiene, HIV/STI prevention, and low cost. The main barriers were cultural tradition, high cost, pain, and concerns for safety. Acceptability of male circumcision for STI and HIV prevention appears to be high in Zambia. Bailey, et al. [9] RCT Intervention group Male circumcision for The protective effect of circumcision (circumcision; n = 1391) HIV prevention in young was 60% (32-77). No behavioural risk Control group (circumcision, men in Kisumu, Kenya: a compensation after circumcision was 1393), Kenya randomised controlled trial. observed.

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 4 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

Brewer, et al. [17] Cross-national DHS Male and Female Among adolescents, regardless of sexual Sample, (n = not known) Circumcision Associated experience, circumcised males and adolescence, Kenya, with Prevalent HIV infection females were substantially more likely to Tanzania, Lesotho in Virgins and adolescents be infected than uncircumcised virgins. in Kenya, Lesotho, and HIV transmission may occur through Tanzania circumcision-related blood exposures in eastern and Southern Africa Agot, et al. [18] Prospective cohort study- Male Circumcision in No difference in risky sex behaviours comparison of sexual Siaya and Bondo Districts, between circumcised and uncircumcised behaviours of (n = 324 Kenya: Prospective Cohort men circumcised) (n = 324 Study to Assess Behavioural uncircumcised men), Disinhibition Following Kenya Circumcision Drain and Halperin Cross-sectional study (Data Male circumcision, religion, Male circumcision was strongly associated [43] from 118 DHS) and infectious diseases: with lower HIV prevalence among countries an ecologic analysis of 118 with primarily heterosexual transmission developing countries Dowsetter, et al. [44] Systematic reviews (Sub- MC and HIV prevention: is No much evidence found of the association Saharan Africa) there really enough of the between male circumcision and reduced right kind of evidence? HIV infection Williams, et al. Dynamic simulation models The potential impact of male Male circumcision was associated [15] National data circumcision on HIV in sub- with decreased HIV infection among Saharan Africa circumcised compared with uncircumcised males, MC may increase the proportion of infected people among women. Auvert, et al. [11] RCT with three follow- Randomised Controlled Male circumcision offers some protection up visits (n = 3274 Intervention Trial of Male against HIV infection, by 61% (95% CI: uncircumcised men) N = Circumcision for Reduction 34-77%) after controlling for risk sexual 1582 control group N = of HIV Infection Risk: The behaviour and non-condom use. Other 1546 intervention group, ANRS 1265 Trial findings: complications reported in the South Africa trial were pain, excessive bleeding, and infection, damage to penis, excessive skin removal, insufficient skin removal, delayed healing, and problems with urination. Kiwanuka, et al. [39] Randomized control trial Circumcision for HIV of No male circumcision complications Sample (n = 136 men), a pilot study in Rakai, reported Uganda Uganda. Auvert, et al. [19] Cross-sectional population Male circumcision and HIV In Cotonou and in Yaoundé, the two low based study in Cotonou infection in four cities in sub- HIV prevalence cities, 99% of men were (Benin) and in Yaoundé Saharan Africa circumcised. In Kisumu 27.5% of men were (Cameroon), Kisumu circumcised, and in Ndola this proportion (Kenya) and in Ndola was 9%. In Kisumu, the prevalence of HIV (Zambia) infection was 9.9% among circumcised men and 26.6% among uncircumcised men. Gray, et al. [31] Cohort study Sample (n = Male circumcision and HIV Circumcision significantly associated with 5507), Uganda acquisition and transmission: reduced HIV acquisition (RR 0.53, CI 0.33- cohort studies in Rakai, 0.87). Prepubertal circumcision significantly Uganda. Rakai Project reduced HIV acquisition (RR 0.49, CI Team. 0.26-0.82), but post pubertal circumcision did not. In discordant couples with HIV- negative men, no seroconversion occurred in 50 circumcised men. In couples with HIV-positive men, HIV transmission was significantly reduced in circumcised men. Weiss, et al. [4] Systematically review of Male circumcision and risk of 27 studies were included. Of these 21 studies published up to HIV infection in sub-Saharan showed a reduced risk of HIV infection April 1999 sub-Saharan Africa: a systematic review among circumcised men (RR = 0.55, 95% Africa (n = 27 studies) and meta-analysis. CI 0.40-0.68). The strongest association is seen among men at high risk of HIV (RR = 0.27, 95% CI 0.20-0.41).

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 5 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

O’Farrell, et al. [35] Systematic review- 33 Circumcision in men and the This re-analysis support the contention diverse studies (Sub- prevention of HIV infection: a that male circumcision offer protection Saharan Africa) `meta-analysis’ revisited against HIV infection- particularly in high- risk groups where genital ulcers and other STDs `drive’ the HIV epidemic. Van Howe, et al. [45] Systematic review (Sub- Circumcision and HIV A meta-analysis was performed on Saharan Africa) infection: review of the the 29 published articles where data literature and meta-analysis were available. When the raw data are combined, a man with a circumcised penis is at greater risk of acquiring and transmitting HIV than a man with a non- circumcised penis OR =1.06, 95% CI, (1.01-1.12). Based on the reviewed studies, recommending male circumcision as prevention strategy of HIV is unfounded. Moses, et al. [3] Cross-sectional DHS, 41 Geographical patterns of In locations where male circumcision is countries male circumcision practices practised, HIV sero-prevalence was con- in Africa: association with siderably lower than in areas where it is not HIV sero-prevalence. practised. This study supports the hypoth- esis that lack of circumcision in males is a risk factor for HIV transmission. Kelly, et al. [30] Cross-sectional (n = 6821) Age of male circumcision and Prepubertal circumcision is associated with risk of prevalent HIV infection reduced HIV risk, whereas circumcision in rural Uganda. after age 20 years is not significantly protective against HIV-1 infection. Bongaarts, et al. [46] Literature review (Sub- The relationship between In 5 countries where more than 3 Saharan Africa) male circumcision and quarters of males were estimated to be HIV infection in African uncircumcised, the average HIV prevalence populations was 16.4%. Among the 20 countries where more than 90% of males were circumcised, the average sero-prevalence was 0.9%

Table 2: Male circumcision prevalence in SSA countries.

Male Circumcision Prevalence Low (< 20%) Intermediate (20-80%) High (> 80%) Botswana Central African Morocco Mali Namibia Cote d'Ivoire Republic Angola Mauritania Rwanda Ethiopia Liberia Niger Burundi Lesotho Madagascar Nigeria Cape Verde Mozambique Benin Senegal Swaziland South Africa Burkina Faso Sierra Leone Malawi Sudan Cameroon Somalia Zambia Tanzania Chad Togo Zimbabwe Uganda Comoros Equatorial Guinea Congo (Brazzaville) Eritrea Dem Rep of the Congo Gabon Djibouti Gambia Kenya Ghana Guinea Guinea-Bissau

Source: Morris, et al. 2016 [5]. systematic review papers covered a period from 1989 factors associated with male circumcision such as lack to 2017. Another eight studies were qualitative studies of condom use post circumcision [16,22,23-25,33,37]. using focus groups and in-depth interviews. The focus Four qualitative studies included reported complica- groups were predominantly focusing on the acceptabili- tions after circumcision such as excessive bleeding, in- ty and attitude towards male circumcision among males fection, redundant foreskin and amputation of glands and females [9,25,28,36], while others focused on risk [9,11,38,39]. Geographically, all studies included in the

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 6 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512 review were conducted in SSA with the majority coming tive studies conducted in SSA countries among circum- from Kenya (n = 7), South Africa (n = 4), Uganda (n = 3), cised and uncircumcised males from 1989 to 2017, we and no more than two from Zimbabwe, Zambia, Leso- found sufficient evidence from 26 studies showing the tho, Nigeria and Cameroon. protective effect of male circumcision against HIV trans- Results mission. In the late 1980s, small-scale studies conducted in SSA countries suggested an association between Prevalence of male circumcision having a foreskin and a greater risk of contracting HIV [3,43,46]. It was first reported by Bongaats, et al. [46] The global prevalence of male circumcision is ap- that in 5 countries where more than three-quarters of proximately 38% [5]. The general population prevalence males were uncircumcised, the average HIV prevalence of male circumcision in SSA has been reported by three studies [5,16,26]. The population estimates show that was 16.4% and that among the 20 countries where male circumcision is practised in many parts of Africa more than 90% of males were circumcised, the average and there is a spatial variation of the practice as shown sero-prevalence was 0.9%. Moses, et al. [3] examined in Table 2. Twenty-eight African countries are catego- the geographical patterns of male circumcision practic- rized as having a high (> 80%) male circumcision prev- es in Africa and HIV sero-prevalence and concluded that alence, nine countries have an intermediate (20-80%); in locations where male circumcision was practised, HIV and the other nine are categorized as having a low (< sero-prevalence was considerably lower than in areas 20%) prevalence. There is also a general correlation be- where it was not practised. Conclusive evidence of the tween areas where there is low circumcision and those protective effect of male circumcision on HIV infection with high HIV prevalence (see Appendix: Figure 1). How- came from three randomised trial (RCTs) conducted ever, this ecological association may mean little without in Uganda, Kenya and South Africa between 2000 and looking at individuals within populations, and taking into 2006. In total, 10,908 uncircumcised men were random- account other factors associated with circumcision sta- ized to immediate circumcision (intervention group) or tus. There may be other factors associated with HIV risk circumcision at the end of the trial (control group), and that may be more prevalent among non-circumcising were followed up for up to 2 years [9-11]. The findings than circumcising populations [34,41]. demonstrated that voluntary medical male circumcision reduced male HIV acquisition by 50-60%, a finding Male circumcision and HIV infection consistent with that of ten observational studies later It has been pointed out that removal of a foreskin [5,9,21,22,27,29,40,44,49]. Most recently, a systematic may reduce the risk of HIV transmission from women review of observational studies by Grund, et al. [7] con- to men [47,48]. In a systematic review of 36 quantita- firmed these findings.

HIV Prevalence % MMC Prevalence %

≤ 0.5 0.6-1.0 1.1-5.0 5.1-15.0 15.1-28.0 ≤ 20 21-40 41-60 61-80 81-100 Figure 1: Map of HIV prevalence (left) and male circumcision prevalence in Africa. Sources: Morris, et al. [5].

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 7 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

However, casting a doubt on the validity of the than uncircumcised virgins, suggesting that HIV trans- studies outlined above, eleven studies [8,16,18, mission may occur through circumcision-related blood 23,24,33,34,36,50,51] have noted that the RCT and ob- exposures [17]. Ten studies reported common surgi- servational studies might not have been adjusted effec- cal complications associated with male circumcision tively for confounding. These studies have argued that which included bleeding, infection, pain and swelling the impact of male circumcision on HIV acquisition in [9-11,34,38,41,55]. It has been revealed that compli- women, sexual behaviour after circumcision and surgi- cations are likely to be higher outside clinical settings, cal safety have not been controlled for, and therefore where the quality and quantity of training, supervision results are unreliable. However, the increase in HIV in- and resources vary [41]. However, in three randomised fections in Southern Africa where circumcision is less control trials, approximately 3-8% of surgeries were prevalent, has been faster than in all other parts of Afri- associated with an adverse event and most were mild ca, a suggestive of benefits accrued from male circumci- [10,11,55]. While there are reports suggesting that sion in reducing HIV cases, particular in the nations with complications are likely to be higher outside clinical set- near universal male circumcision rates in West Africa. tings, a systematic review of complications associated with male circumcision in sub-Saharan Africa revealed Male circumcision and HIV infection in women low prevalence of complications [34] for both medical Seven studies reported on the impact of male cir- and traditional circumcision. cumcision on HIV acquisition in women [7,15,31,32,52] Another major concern of being circumcised is that and they have all reported higher proportion of part- unsafe sex may occur if men believe they are complete- ners of circumcised men becoming HIV infected during ly protected from HIV infection following circumcision. the trials. Four randomised trials studies confirm simi- Different patterns of sexual behaviour may be expect- lar results [15,31,49,52]. A RCT was conducted in Rakai, ed during the scale-up of male circumcision programs. Uganda, to evaluate the impact of male circumcision on Seven out of thirteen observational studies reported HIV acquisition in women [32]. In total, 922 HIV-infect- sexual behavioural changes involving multiple sexual ed men aged 15-49 years with high CD4 counts (≥ 350 partners, resuming sex early; and non-condom use after cells/microlitres) were randomized to immediate or de- circumcision [7,8,10,22-25,55] and six studies reported layed circumcision and 163 uninfected female partners no behavioural risk compensation after circumcision [9- were enrolled and followed for up to 24 months. A high- 11,18,33,37]. er proportion of partners of circumcised men became HIV infected during the trial with a hazard ratio of 1.58 The timing of male circumcision is crucial. Medical (95%CI: 0.68, 3.66). Conversely, three studies reported practitioners point to the neonatal period as the ideal time [10]. Failure to circumcise early in infancy means that male circumcision, by reducing HIV prevalence in loss of the benefit of protection against diseases and heterosexual men, will help reduce HIV prevalence and later circumcision at adult age is a more substantial, other sexual transmitted diseases that exacerbate HIV more expensive operation, and carries a higher risk of risk in women and children [31,53,54]. One recent study, complications, and longer healing time [10,55]. If an after adjustment for the male partner’s viral load, the adolescent or adult male normally engages in sexual risk of HIV transmission was lower in female partners of activity, temporary sexual abstinence for 6 weeks is circumcised men than in partners of uncircumcised men required, which some males and their sexual partners [52]. One possible reason for the contrasting results be- find challenging, putting themselves at risk ofHIV tween observational studies and the RCTs is that males transmission. Four studies revealed that circumcised in the observational study were likely to have been cir- men resume sexual intercourse before the wound is cumcised in childhood, which would have allowed for completely healed [55]. Another claim is that male complete wound healing which have reduced their risk circumcision diminishes sexual function, sensitivity and of exposure to HIV infection. Although it is possible that pleasure [56]. However, medical male circumcision no direct impact of circumcision on male-to-female does not adversely affect sexual function as shown transmission exists, there will be an indirect benefit to by a detailed systematic review and meta-analysis of women since HIV prevalence is likely to fall in male part- common forms of sexual dysfunction [5,57]. ners as circumcision services are expanded [29]. Traditional authority and medical circumcision Surgical safety and sexual behaviour after circum- Conflict between the custodian of traditional culture cision and medical practitioner in relation to transforming tra- One of the major concerns about the male circum- ditional circumcision into medical circumcision may act cision is the surgical safety and procedures of the skin as a barrier. Medical male circumcision may be perceived removal which can exacerbate the risk of HIV infec- as a replacement for traditional male practice under the tion through blood-sharing. Analysis of DHS data from excuse of HIV prevention [14]. This could be worsened Kenya, Lesotho and Tanzania revealed that circum- by the perception of some political figures, tradition- cised male virgins were more likely to be HIV infected al and religious leaders that medical professionals are

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 8 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512 sceptical to traditional practices, authority and values fant circumcision is simpler, more convenient and less [7]. Traditionalist argue that the medical circumcision risky than male circumcision performed later at adult model to HIV prevention has long condemned tradition- age; and it confers immediate protection against urinary al circumcision and has been urging governments to ban infections. Male circumcision also protects the female the traditional practice because it is regarded as retro- partners indirectly through reduced risk for male part- gressive and prevent the achievement of health goals. ners, as confirmed in randomised control trials [8,32]. The perceived replacement of traditional authority and Disputing the value of male circumcision's protec- values could result in community resistance to the use tion against HIV, some studies argue that less invasive of medical circumcision in HIV prevention. Community HIV prevention strategies should instead be promoted, defensiveness can ethically be justified in the context such as encouraging safe sex practices [24]. We argue of freedom of choice and right to practice what they that public health messages should include all effec- believe in. In traditional sense, medicalisation of male tive measures for protection against HIV, and that male circumcision tends to alienate people from their tradi- circumcision should therefore complement current tional practices. Program managers should emphasise safe sex messages. The effectiveness of each approach that even though people have a right to their traditional should, moreover, be considered in real-world settings. practices, the risk of HIV infection should be reduced by The benefits of medical procedures should always, of practising safer sex. In the context of HIV, both tradi- course, be weighed against the potential risks. Other tional and medical circumcision does not protect 100% studies question whether the potential benefits of male individuals against HIV infection. Medical research has circumcision are worth the risk [12,58] pointing to po- demonstrated that male circumcision reduces risk of tential risks of surgical accidents and supposed adverse HIV transmission by up to 60%. However, we argue psychological or sexual effects. The risk of major surgical that the remaining 40% is still high risk. Medical male mishaps with MC, however, seven out of ten studies re- circumcision as a new HIV prevention strategy can be viewed indicated that the risk is extremely low and the made more effective by the use of condoms in order to benefits gained from male circumcision far outweigh the offer full protection against HIV. risks. Although some studies argue against downplaying Discussion and Conclusions the procedural pain that can occur during circumcision, the evidence cited by Morris, et al. [5] suggest that with Data from randomised control trials, meta-analyses, the use of local anaesthesia, pain is negligible and three large observational studies in SSA countries and high studies revealed that the procedure is well tolerated quality systematic reviews were compiled and examined when performed by trained professionals under sterile in order to determine the HIV risk reduction conferred conditions with appropriate pain management. Com- by male circumcision. Overall, there is strong evidence plications are infrequent, most are minor; and severe of an association between male circumcision and risk complications are rare. reduction in HIV transmission from female to male. Al- though male circumcision is not as effective as other HIV Although health benefits are not great enough to interventions such as condom use and abstinence, its recommend routine circumcision for all males, the immediate and lifelong protection against HIV could be benefits of circumcision are sufficient to justify access greater than these interventions. For instance, condoms to this procedure for males choosing it. It is important are 80% protective against HIV infection, but they must that clinicians routinely inform individuals on the health be used consistently and correctly [7]. Unlike condoms, benefits and risks of male circumcision in an unbiased male circumcision is a one-off procedure that does not and accurate manner. Individuals ultimately should de- require future voluntary compliance each time a man cide whether circumcision is in their best interests. They has sexual intercourse. Nevertheless both male circum- will weigh medical information in the context of their cision and condom use should be advocated. While crit- own religious, ethical and cultural beliefs and practices. ics of male circumcision argue against undue reliance The medical benefits alone may not outweigh these and on findings from randomised trials conducted in Kenya, other considerations for individuals in specific societies. Uganda and South Africa concerning male circumcision Various socio-political and cultural meanings attached and HIV, current data and most recent country-specif- to traditional circumcision practices and how biomedi- ic figures (UNAIDS, 2016) show strong correlation be- cal approaches to circumcision are read through tradi- tween low male circumcision prevalence and higher tional understandings of circumcision as a ritual passage HIV prevalence in other countries. Therefore, there is must also be taken into consideration. There is a greater strong support for the conclusion conferred by the ran- need to understand how traditionally circumcised soci- domised control trials that male circumcision is protec- eties understand and introduce aspects of biomedicine tive against HIV infection at least in males. However, the into the process of traditional circumcision and how timing of circumcision is crucial [30]. It has been shown such pluralities could aid an understanding of the rela- that the cumulative lifeline benefits is greatest if male tionship between the cultural repertoires and HIV-inter- circumcision is performed early in infancy since early in- ventions

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 9 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

Competing Interests 17. Brewer DD, Potterat JJ, Roberts JM, Brody S (2007) Male and female circumcision associated with prevalent HIV The authors declare that they have no competing infection in virgins and adolescents in Kenya, Lesotho, and interests. Tanzania. Ann Epidemiol 17: 217-226. 18. Agot KE, Ndinya-Achola JO, Kreiss JK, Weiss NS (2004) References Risk of HIV-1 in rural Kenya: A comparison of circumcised 1. Cox G, Morris BJ (2012) Why circumcision: From pre- and uncircumcised men. Epidemiology 15: 157-163. history to the twenty-first century. In: Bolnick DA, Koyle 19. Auvert B, Buve A, Lagarde E, Kahindo M, Chege J, et al. MA, Yosha A, Surgical Guide to Circumcision. Springer, (2001) Male circumcision and HIV infection in four cities in London, 243-259. sub-Saharan Africa. Aids 15: S31-S40. 2. Silverman EK (2004) Anthropology and Circumcision. Annu 20. Bailey RC, Muga R, Poulussen R, Abicht H (2002) The Rev Anthropol 33: 419-445. acceptability of male circumcision to reduce HIV infections 3. Moses S, Bradley JE, Nagelkerke NJ, Ronald AR, Ndinya- in Nyanza Province, Kenya. AIDS Care 14: 27-40. Achola JO, et al. (1990) Geographical patterns of male 21. Siegfried N, Muller M, Deeks JJ, Volmink J (2009) Male circumcision practices in Africa: Association with HIV circumcision for prevention of heterosexual acquisition of seroprevalence. Int J Epidemiol 19: 693-697. HIV in men. Cochrane Database of Systematic Reviews 4. Weiss HA, Quigley MA, Hayes RJ (2000) Male circumcision 2009, Issue 2, CD003362. and risk of HIV infection in sub-Saharan Africa: A systematic 22. Ediau M, Matovu JK, Byaruhanga R, Tumwesigye NM, review and meta-analysis. Aids 14: 2361-2370. Wanyenze RK (2015) Risk factors for HIV infection among 5. Morris BJ, Wamai RG, Henebeng EB, Tobian AR, D circumcised men in Uganda: A case-control study. J Int Klausner FD, et al. (2016) Estimation of country-specific AIDS Soc 18: 19312. and global prevalence of male circumcision. Population 23. Ombere SO, Nyambedha EO, Bukachi SA (2015) Wimbo: Health Metrics 14: 4. Implications for risk of HIV infection among circumcised 6. Weiss GN (1997) Prophylactic neonatal surgery and fishermen in Western Kenya. Cult Health Sex 17: 1147- infectious diseases. Pediatr Infect Dis J 16: 727-734. 1154. 7. Grund JM, Bryant TS, Jackson I, Curran K, Bock, N, et 24. Ayiga N, Letamo G (2011) Impact of male circumcision on al. (2017) Association between male circumcision and HIV risk compensation through the impediment of condom women’s biomedical health outcomes: A systematic review. use in Botswana. Afr Health Sci 11: 550-559. Lancet Glob Health 5: e1113-e1122. 25. Obure EO, Nyambedha O, Hezborn M, Kodero N (2009) 8. Maughan-Brown B, Godlonton S, Thornton R, The Qualitative Report. Volume 14, Issue 4, 665-687. Venkataramani AS (2015) What do people actually learn 26. Sakutukwa GK, Lane T, van-Rooyen H, Chingono A, from public health campaigns? Incorrect inferences about Humphries H, et al. (2013) Understanding and addressing male circumcision and female HIV infection risk among socio-cultural barriers to medical male circumcision in men and women in Malawi. AIDS Behav 19: 1170-1177. traditionally non-circumcising rural communities in sub- 9. Bailey RC, Moses S, Parker CB, Agot K, Maclean I, et al. Saharan Africa. Cult Health Sex 15: 1085-1100. (2007) Male circumcision for HIV prevention in young men 27. Makatjane T, Hlabana T, Letete E (2016) Male Circumcision in Kisumu, Kenya: A randomised controlled trial. Lancet and HIV in Lesotho: Is the Relationship Real or Spurious? 369: 643-656. Analysis of the 2009 Demographic and Health Survey. DHS 10. Gray RH, Kigozi G, Serwadda D, Makumbi F, Watya S, et WORKING PAPERS, 125. al. (2007) Male circumcision for HIV prevention in men in 28. Dionne KY, Poulin M (2013) Ethnic identity, region, Rakai, Uganda: A randomised trial. Lancet 369: 657-666. and attitudes toward male circumcision in a high HIV- 11. Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Prevalence country. Glob Public Health 8: 607-618. Sitta R, et al. (2005) Randomized, controlled intervention 29. Londish GJ, Murray JM (2008) Significant reduction in HIV trial of male circumcision for reduction of HIV infection risk: prevalence according to male circumcision intervention in The ANRS 1265 trial. PLoS Med 2: e298. sub-Saharan Africa. Int J Epidemiol 37: 1246-1253. 12. Vincent L (2008) “Boys will be boys”: Traditional Xhosa 30. Kelly R, Kiwanuka N, Wawer MJ, Serwada D, Sewankambo male circumcision and sexual socialisation in contemporary NK, et al. (1999) Age of male circumcision and risk of South Africa. Cult Health Sex 10: 431-446. prevalent HIV infection in rural Uganda. AIDS 13: 399-405. 13. Kepe T (2010) ‘Secrets’ that kill: Crisis, custodianship and 31. Gray RH, Kiwanuka N, Quinn TC, Sewankambo NK, responsibility in ritual male circumcision in the Eastern Serwadda D, et al. (2000) Male circumcision and HIV Cape Province, South Africa. Soc Sci Med 70: 729-735. acquisition and transmission: Cohort studies in Rakai, 14. Gwandure C (2011) The ethical concerns of using medical Uganda. Rakai Project Team. AIDS 14: 2371-2381. male circumcision in HIV prevention in Sub-Saharan Africa. 32. Wawer MJ, Makumbi F, Kigozi G, Serwadda D, Watya S, South African Journal of Bioethics and Law 4: 89. et al. (2009) Circumcision in HIVinfected men and its effect 15. Williams BG, Lloyd-Smith JO, Gouws E, Hankins C, Getz on HIV transmission to female partners in Rakai, Uganda: A WM, et al. (2006) The potential impact of male circumcision randomised controlled trial. Lancet 374: 229-237. on HIV in sub-Saharan Africa. PLoS Med 3: e262. 33. Mattson C (2008) Risk compensation, male circumcision, 16. Greely P, Maharaj P, Letsoalo T, Miti A (2013) Traditional and HIV prevention in Kisumu, Kenya Sciences and male circumcision for reducing the risk of HIV infection: Engineering. Volume 68, 5112. Perspectives of young people in South Africa. Cult Health 34. Muula AS (2008) HIV Infection and AIDS among Young Sex 15: 148-159. Women in South Africa. Croat Med J 49: 423-435.

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 10 of 11 • DOI: 10.23937/2643-4512/1710022 ISSN: 2643-4512

35. O’Farrell N, Egger M (2000) Circumcision in men and the 47. McCoombe SG, Short RV (2006) Potential HIV-1 target prevention of HIV infection: A ‘meta-analysis’ revisited. Int J cells in the human penis. AIDS 20: 1491-1495. STD AIDS 11: 137-142. 48. Szabo R, Short RV (2000) How does male circumcision 36. Lukobo MD, Bailey RC (2007) Acceptability of male protect against HIV infection? BMJ 320: 1592-1594. circumcision for prevention of HIV infection in Zambia. 49. Weiss HA, Halperin D, Bailey RC, Hayes RJ, Schmid G, AIDS Care 19: 471-477. et al. (2008) Male circumcision for HIV prevention: From evidence to action? Aids 22: 567-574. 37. Naidu M, Khumalo S (2016) I am circumcised so HIV/AIDS can’t touch me!? Young Black African University Men and 50. De Camargo KR Jr, de Oliveira Mendonça AL, Perrey C, Narratives of Masculinity. The Oriental Anthropologist 16: Giami A (2013) Male circumcision and HIV: A controversy 163-181. study on facts and values. Glob Public Health 8: 769-783. 51. Boyle GJ, Hill G (2011) Sub-Saharan African randomised 38. Okeke LI, Asinobi AA, Ikuerowo OS (2006) Epidemiology of clinical trials into male circumcision and HIV transmission: complications of male circumcision in Ibadan, Nigeria. BMC Methodological, ethical and legal concerns. J Law Med 19: Urology 6: 21. 316-334. 39. Kiwanuka N, Gray R, Serwadda D, Kigozi G, Watya S, et 52. Baeten JM, Donnell D, Kapiga S, Ronald A, John-Stewart al. (2013) Circumcision for HIV of a pilot study in Rakai, G, et al. (2010) Male circumcision and risk of male-to- Uganda. XV International AIDS Conference 2004, Abstract female HIV-1 transmission: A multinational prospective no. TuPe c4771 19. study in African HIV-1 serodiscordant couples. AIDS 24: 40. Gebremedhin S (2011) Male circumcision and its 737-744. association with HIV infection and sexually transmitted 53. Hallett TB, Alsallaq RA, Baeten JM, Weiss H, Celum C, et al. diseases: Evidence from 18 demographic and health (2011) Will circumcision provide even more protection from surveys in sub-Saharan Africa. SAHARA J 8: 13-18. HIV to women and men? New estimates of the population impact of circumcision interventions. Sex Transm Infect 87: 41. Bailey RC, Egesah O, Rosenberg S (2008) Male circumcision 88-93. for HIV prevention: a prospective study of complications in clinical and traditional settings in Bungoma, Kenya. Bull 54. Wawer MJ, Tobian AA, Kigoz G, Kong X, Gravitt PE, et World Health Organ 6: 669-677. al. (2011) Effect of circumcision of HIV-negative men on transmission of human papillomavirus to HIV-negative 42. Tusiime B (2008) Circumcision and HIV infection: women: A randomised trial in Rakai, Uganda. Lancet 377: Assessment of causality. AIDS Behav 12: 835-841. 209-218. 43. Halperin DT, RC Bailey (1999) Male circumcision and HIV 55. Krieger JN, Bailey RC, Opeya J, Ayieko B, Opiyo F, et al. infection: 10 years and counting. Lancet 354: 1813-1815. (2005) Adult male circumcision: Results of a standardized procedure in Kisumu District, Kenya. BJU Int 96: 1109-1113. 44. Dowsett W, Couch M (2007) Male circumcision and HIV prevention: Is there really enough of the right kind of 56. Lang DP (2013) Circumcision, sexual dysfunction and the evidence? Reprod Health Matters 15: 33-44. child’s best interests: Why the anatomical details matter. J Med Ethics 39: 429-431. 45. Van Howe RS (1999) Circumcision and HIV infection: 57. Tian Y, Liu W, Wang JZ, Wazir R, Yue X, et al. (2013) Effects review of the literature and meta-analysis. Int J STD AIDS of circumcision on male sexual functions: a systematic 10: 8-16. review and meta-analysis. Asian J Androl 15: 662-666. 46. Bongaarts J, Reining P, Way P, Conant F (1989) The 58. Mfecane S (2016) “Ndiyindoda” [I am a man]: Theorising relationship between male circumcision and HIV infection Xhosa masculinity. Anthropology Southern Africa 39: 204- in African populations. Aids 3: 373-377. 214.

Gazimbi et al. Int Arch Public Health Community Med 2019, 3:022 • Page 11 of 11 •