Male circumcision and the risk of gonorrhoea,syphilis,HIV and human papillomavirus among men in Tanzania

Abstract

To examine the association between male circumcision and the risk of gonorrhoea, syphilis, HIV and high-risk (hr) human papillomavirus (HPV). We used data from a cross-sectional study conducted among 1902 men in Tanzania. Circumcision status was assessed at a clinical examination and history of gonorrhoea and syphilis was obtained from questionnaire data. Penile samples were tested for HPV using Hybrid Capture 2 and genotyped by the INNO-LiPA HPV Genotyping Extra test. Blood samples were tested for HIV. Using logistic regression the association between male circumcision and gonorrhoea, syphilis, HIV and hr HPV was assessed estimating odds ratios (ORs) and 95% confidence intervals (CIs). All analyses were adjusted for age and lifetime number of sexual partners. In the multivariable analysis, the odds of gonorrhoea were lower in circumcised men compared with uncircumcised men (OR = 0.52; 95% CI: 0.37–0.74). Likewise, the odds of HIV were considerably lower in circumcised men (OR = 0.42; 95% CI: 0.26–0.67). Furthermore, lower odds of hr HPV were seen in circumcised men compared with uncircumcised men, although not statistically significant (OR = 0.81; 95% CI: 0.56–1.17). Finally, the odds of HPV16 (OR = 0.48; 95% CI: 0.23–0.98) and multiple (≥2) hr HPV types (OR = 0.71; 95% CI: 0.44–1.12) were lower in circumcised men than in uncircumcised men. Circumcised men have a significantly lower risk of gonorrhoea, HIV and HPV16, compared with uncircumcised men.

Keywords

Male circumcision gonorrhoea syphilis human papillomavirus HIV Tanzania

Introduction

Three randomised controlled trials have shown that male circumcision reduces the risk of HIV acquisition in heterosexual men by approximately 60%.^1–3 Based on these findings, WHO/UNAIDS recommends that male circumcision be added to the current HIV prevention strategies in areas with high HIV prevalence.⁴ In addition, male circumcision may reduce the risk of acquiring other sexually transmitted infections (STIs). The data on other STIs are, however, variable. Recently, WHO has encouraged investigations in biomedical interventions including adult male circumcision as a potential tool to decrease the worldwide burden of STIs.⁵

Some studies,⁶ including a meta-analysis,⁷ have reported that male circumcision reduces the risk of syphilis, whilst others found no association.^8,9 For gonorrhoea, most studies found no association between male circumcision and gonorrhoea,^10–13 but not all.¹⁴ In addition, studies have reported that male circumcision is associated with a lower prevalence of high-risk (hr) human papillomavirus (HPV) both among HIV-negative and HIV-positive men,^8,15–17 and furthermore it has been found to reduce the acquisition and increase the clearance of hr HPV in HIV-negative men.¹⁸ Some studies have also shown a lower prevalence of multiple HPV types in circumcised men.^8–15

One of the suggested mechanism by which male circumcision may reduce the acquisition of STIs is by removal of the moist environment under the foreskin, which may favour pathogen survival and replication. Moreover, male circumcision may result in a reduction of pathogen entry through abrasions in the thinly keratinised inner mucosal surface of the foreskin.¹⁹

The association between male circumcision and reduced risk of HIV in men during heterosexual intercourse is well-established. The data on other STIs are, however, variable. We have previously described the prevalence of HPV and risk factors for HPV among men in Tanzania.^20,21 The aim of the present study was to examine the association between male circumcision and the risk of gonorrhoea, syphilis, HIV and hr HPV and specific hr HPV types among men in Tanzania.

Material and methods

Description of study

This study uses data from a cross-sectional study called ‘TaMas – Tanzanian male study’ which was carried out among men in Tanzania from February to June 2009. The study design and data collection are described in detail elsewhere.²⁰ Briefly, we recruited men from both urban (Dar es Salaam region) and rural areas (Pwani, Tanga and Kilimanjaro regions) in Tanzania. In the urban area, we invited men employed either at a public university or at one of two different factories and in the rural area, we recruited men who were visiting their relatives at different public health facilities. We aimed at including altogether 2000 men and were able to enrol 1933 men.

Data collection

Information on socio-demographic and lifestyle factors, sexual behaviour and self-reported history of STIs (gonorrhoea and/or syphilis) was collected using a structured questionnaire. All study participants also underwent a clinical examination with collection of penile samples and assessment of circumcision status.

Penile samples were obtained using two pre-wetted (saline) Dacron swabs: In circumcised men, one sample was collected from the glans penis and the coronal sulcus and one sample from the shaft of the penis. In uncircumcised men, one sample was collected from the glans penis and the preputial cavity and one sample from the shaft of the penis. The penile swabs were placed in one tube containing 1 ml Digene specimen transport medium (Digene Corporation, Gaithersburg, Maryland, USA) and were subsequently stored at −20°C. At the end of the study, all samples were shipped to the Department of Medical Virology, Tuebingen University in Germany to be tested for HPV.

We asked the study participants if they would have a blood sample taken to be tested for HIV. Before HIV testing, the study participants were offered HIV counselling and this was done afterwards as well. This is in accordance with national and international guidelines.^22,23

The study participants all gave informed consent to participate in the study. Furthermore, the study participants received approximately US$5 for their participation. The National Health Research Ethics Review Committee of the National Institute of Medical Research in Tanzania approved the study (approval number NIMR/HQ/R.8c/Vol. 1/73).

HPV testing

Hybrid Capture 2 (HC2)

We used the HC2 method (Qiagen, Hildesheim, Germany) to test for HPV. This method uses a hr probe that can detect 13 hr-HPV types (HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68) and a low-risk (lr) probe that can detect five lr-HPV types (HPV6, 11, 42, 43, 44). The Food and Drug Administration recommended cut-off value for test-positive results was used. Here, a test result of ≥1.0 relative light units/Co is considered positive (equivalent to 1 pg HPV DNA per 1 ml of sampling buffer).

LiPA

The HC2-positive samples were genotyped using the PCR-based assay INNO-LiPA HPV Genotyping Extra test (LiPA) (Innogenetics Inc., Gent, Belgium). LiPA is a line probe assay which is based on the reverse hybridisation principle and can detect 14 hr HPV genotypes (HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59 and 68), 4 probable hr types (HPV26, 66, 73 and 82), 7 lr-HPV genotypes (HPV6, 11, 40, 43, 44, 54, 70) and 3 types with undefined risk (HPV69, 71 and 74). Altogether, 200 µl of the remaining denatured product from the HC2 test was used for genotyping. Using the MagnaPure device (Roche Systems, Indianapolis, IN) the DNA was isolated and subsequently, 5 µl of DNA solution was used for the LiPAV Extra SPF-PCR assay. PCR products were then denatured and a 10-µl aliquot hybridised to a HPV Genotype detection strip. The reading of the HPV Genotype detection strip was done using the LiRAS prototype software scanner (Innogenetics Inc., Gent, Belgium).

For five HC2-positive samples, no HPV could be detected using LiPA and these were classified as ‘LiPA negative’. Furthermore, in some cases the strip was not interpretable due to presence of too many reactive probe lines and these were categorised as ‘LiPA fail’ (n = 40). Lastly, LiPA-positive samples from which no specific HPV type could be identified or where the LiPA test could not discriminate between two HPV types were classified as HPV of undetermined type (‘HPV X’) (n = 10).

HIV testing

The WHO-approved HIV serial testing algorithm²² was used to test blood samples for HIV. Using this method is in accordance with the Tanzanian Ministry of Health guidelines.²³ The test consists of first, a rapid immunoassay (SD Bioline HIV-1/2 3.0 rapid test, Standard Diagnostics Inc., Republic of South Korea) and if this is found to be positive, an additional confirmatory immunological test (Determine HIV-1/2 test, Abbott Laboratories SA, South Africa) is applied. If there is discrepancy between these two tests, a third test is used (UNIGold method/Recombigen®HIV; Trinity biotech, USA). The study participants were informed about the results of their HIV test individually and in a separate room to ensure confidentiality. The participants that were found to be HIV-positive were referred to an HIV care and treatment clinic and were offered a voluntary and free-of-charge follow-up. The follow-up included determination of primary CD4 cell count as well as advice, information and psychological support. If relevant, immediate treatment was offered as well.

As described previously¹⁹ men who reported to be HIV-positive in the questionnaire (n = 30), and who were not tested for HIV in the present study, were categorised as HIV-positive. Furthermore, men tested for HIV, but with missing values on HIV result (n = 20) were categorised as not tested for HIV. Altogether, 353 men were not tested for HIV.

Self-reported STIs

Using a structured questionnaire, the study participants were asked ‘Has a doctor or other health care provider told you that you had one of the following sexually transmitted diseases?’ (at any time in the past) with the following three response categories ‘Yes’, ‘No’ and ‘Don’t know’ for the two STIs, namely gonorrhoea and syphilis. In the analyses of gonorrhoea and syphilis, the response category ‘Don’t know’ (n = 62 for gonorrhoea and n = 69 for syphilis) was set as missing.

Definition of study population and outcomes

Altogether, 1933 men participated in the study. We excluded men with missing value on circumcision status (n = 4), age at study entry (n = 5) or lifetime number of sexual partners (n = 22), resulting in 1902 men. For each of the different outcomes, men with missing value on that outcome were excluded (for the outcome variables gonorrhoea n = 80 and syphilis n = 103, respectively). Moreover, men not tested for HIV (n = 353) or with missing values (n = 1) were excluded when HIV was considered as an outcome. Additionally, men with inadequate penile samples (n = 113) due to insufficient volume or quality for HPV testing or lr-HPV types only were excluded (n = 75). This left a total of 1822 men with information on gonorrhoea, 1799 men with information on syphilis, 1548 with information on HIV and 1714 men with results on hr-HPV test.

Statistical analysis

The prevalence of gonorrhoea, syphilis, HIV, hr-HPV and multiple (≥2) hr-HPV in relation to circumcision status was assessed and the associations examined using Chi square tests.

Logistic regression analyses estimating odds ratios (ORs) and 95% confidence intervals (CIs) were performed to examine the association between male circumcision and gonorrhoea, syphilis, HIV and hr-HPV, respectively. Initially, age-adjusted analyses were conducted. Subsequently, in the multivariable analyses, we adjusted for a priori selected variables (lifetime number of sexual partners). Finally, among men tested for HIV, we further adjusted the analyses for HIV status; however, HIV may be considered an intermediate factor.

The prevalence of specific hr-HPV types and multiple (≥2) hr-HPV in relation to circumcision status was examined both overall and among HC2 hr-positive men. By means of logistic regression analyses, we estimated the OR and 95% CI of the specific hr-HPV types comparing circumcised and uncircumcised men whilst adjusting for age and lifetime number of sexual partners. In this analysis, no adjustment was made for HIV due to low numbers.

All analyses were performed using the statistical software SAS version 9.4.

Results

Overall, among the 1902 study participants, 1676 men (88.1%) were circumcised. Information on age at circumcision was available for 1632 men (97.3%). Among these, the median age at circumcision was 10 years (SD 4.3; range 1–34 years) and 1505 men (92.2%) were circumcised ≤15 years. The age distribution and the distribution of lifetime number of sexual partners at study entry are shown in Table 1 for the total study population and for circumcised and uncircumcised men, respectively. The median age at study entry was 30 years, and median lifetime number of sexual partners was 6 (SD 13.5; range 0–200 partners). These figures were similar irrespective of circumcision status (Table 1).

Table 1.

The distribution of age and lifetime number of sexual partners according to circumcision status among men in Tanzania.

			Circumcision status
	Total study population (N=1902)		Circumcised (n=1676)		Uncircumcised (n=226)
	n	(%)	n	(%)	n	(%)
Age at study entry (years)
≤19	78	(4.1)	70	(4.2)	8	(3.5)
20–29	768	(40.4)	689	(41.1)	79	(35.0)
30–39	573	(30.1)	506	(30.2)	67	(29.7)
40–49	262	(13.8)	233	(13.9)	29	(12.8)
≥50	221	(11.6)	178	(10.6)	43	(19.0)
Median	30 years		30 years		31 years
Lifetime number of sexual partners
Never had sex	141	(7.4)	124	(7.4)	17	(7.5)
1–3 partners	548	(28.8)	487	(29.1)	61	(27.0)
4–9 partners	699	(36.8)	622	(37.1)	77	(34.1)
10+ partners	514	(27.0)	443	(26.4)	71	(31.4)
Median	6 partners		6 partners		6 partners

The associations between male circumcision and gonorrhoea, syphilis, HIV and hr-HPV are shown in Tables 2 to 5. Circumcised men were less likely to report having had gonorrhoea than uncircumcised men (266/1599, 16.6% versus 67/223, 30.0%; p < 0.01). In the multivariable analyses adjusting for age and lifetime number of sexual partners, the odds of gonorrhoea were almost half in circumcised men compared with uncircumcised men (OR = 0.52; 95% CI: 0.37–0.74) (Table 2). Additional adjustment for HIV status did not change the estimates substantially (OR = 0.61; 95% CI: 0.39–0.95) (data not shown).

Table 2.

The association between male circumcision and gonorrhoea among men in Tanzania.

Male circumcision and gonorrhoea (N=1822)
	Prevalence of gonorrhoea				Age-adjusted model^a		Multivariable model^b
	n	N	%	p-value^c	OR	95% CI	OR	95% CI
				p < 0.0001
Uncircumcised	67	223	30.0		1	–	1	–
Circumcised	266	1599	16.6		0.53	(0.38–0.74)	0.52	(0.37–0.74)

CI: confidence interval; OR: odds ratio.

^aAdjusted for age as a continuous variable.

^bAdjusted for age and lifetime number of sexual partners (with an indicator variable for ever/never had sex) as continuous variables.

^cp-value from Chi square test.

The odds of syphilis were slightly lower in circumcised men (OR = 0.92; 95% CI: 0.53–1.61) than in uncircumcised men adjusting for age and lifetime number of sexual partners, although not statistically significant (Table 3). When additionally adjusted for HIV status, the odds remained statistically non-significant (OR = 1.08; 95% CI: 0.51–2.29) (data not shown).

Table 3.

The association between male circumcision and syphilis among men in Tanzania.

Male circumcision and syphilis (N = 1799)
	Prevalence of syphilis				Age-adjusted model^a		Multivariable model^b
	n	N	%	p-value^c	OR	95% CI	OR	95% CI
				p = 0.18
Uncircumcised	18	223	8.1		1	–	1	–
Circumcised	91	1576	5.8		0.85	(0.49–1.46)	0.92	(0.53–1.61)

CI: confidence interval; OR: odds ratio.

^aAdjusted for age as a continuous variable.

^bAdjusted for age and lifetime number of sexual partners (with an indicator variable for ever/never had sex) as continuous variables.

^cp-value from Chi square test.

The prevalence of HIV was considerably lower in circumcised men (251/1514, 8.4%) than in uncircumcised men (29/150, 19.3%) (p < 0.01). This was also seen in the multivariable analyses, where the OR of HIV was 0.42 (95% CI: 0.26–0.67) after adjustment for age and lifetime number of sexual partners (Table 4).

Table 4.

The association between male circumcision and HIV among men in Tanzania.

Male circumcision and HIV (N=1548)
	Prevalence of HIV				Age-adjusted model^a		Multivariable model^b
	N	N	%	p-value^c	OR	95% CI	OR	95% CI
				p < 0.0001
Uncircumcised	29	150	19.3		1	–	1	–
Circumcised	118	1398	8.4		0.42	(0.27–0.67)	0.42	(0.26–0.67)

CI: confidence interval; OR: odds ratio.

^aAdjusted for age as a continuous variable.

^bAdjusted for age and lifetime number of sexual partners (with an indicator variable for ever/never had sex) as continuous variables.

^cp-value from Chi square test.

Table 5.

The association between male circumcision and hr-HPV among men in Tanzania.

Male circumcision and hr-HPV (hr-HPV versus HPV-negative) (N = 1714)
	Prevalence of hr HPV				Age-adjusted model^a		Multivariable model^b
	n	N	%	p-value^c	OR	95% CI	OR	95% CI
				p = 0.17
Uncircumcised	41	200	20.5		1	–	1	–
Circumcised	251	1514	16.6		0.81	(0.56–1.18)	0.81	(0.56–1.17)

CI: confidence interval; HPV: human papillomavirus; hr: high-risk; OR: odds ratio.

^aAdjusted for age as a continuous variable.

^bAdjusted for age and lifetime number of sexual partners (with an indicator variable for ever/never had sex) as continuous variables.

^cp-value from Chi square test.

We found that a lower proportion of circumcised men (251/1514, 16.6%) than uncircumcised men (41/200, 20.5%) tested positive for hr-HPV (Chi square test; p = 0.17). In the multivariable analyses, the odds of hr-HPV tended to be lower in circumcised men (OR = 0.81; 95% CI: 0.56–1.17) than in uncircumcised men, although the association was not statistically significant (Table 5). After adjusting for HIV status, the estimates were virtually unchanged (OR = 0.82; 95% CI: 0.52–1.28) (data not shown).

In Table 6 the distribution of specific hr-HPV types is displayed among circumcised and uncircumcised men. In the total study population, HPV16 was significantly less common in circumcised men than in uncircumcised men (37/1577, 2.4% versus 10/212, 4.7%, p = 0.04) (Table 6). Among hr-HPV-positive men, HPV16 was detected in 14.3% (36/251) of circumcised men compared to 24.4% (10/41) of uncircumcised men. In the analyses where age and lifetime number of sexual partners were taken into account, the odds of HPV16 were lower in circumcised men than in uncircumcised men (OR = 0.48; 95% CI: 0.23–0.98). Among hr-HPV-positive men, the estimate was virtually identical, but the association was not statistically significant (OR = 0.49; 95% CI: 0.22–1.10). The odds of the other hr-HPV types did not differ significantly among circumcised and uncircumcised men, whilst the odds of multiple hr-HPV types tended to be lower in circumcised men (OR = 0.71; 95% CI: 0.44–1.12) than in uncircumcised men (Table 6).

Table 6.

The distribution of HPV types and multiple (≥2) hr-HPV in relation to circumcision status among men in Tanzania.

	Circumcised men				Uncircumcised men				Analysis of the prevalence of hr-HPV types in circumcised men compared with uncircumcised men
	Number of positives	Prevalence in the total study population	Number of positives	Prevalence among hr-HPV-positive	Number of positives	Prevalence in the total study population	Number of positives	Prevalence among hr-HPV-positive	Multivariable model^a (in the total study population)		Multivariable model^a (among hr-HPV-positive)
hr-HPV types		(n=1577)		(n=251)		(n=212)		(n=41)
hr-HPV types	n	%	n	(%)	n	%	n	%	OR	(95% CI)	OR	(95% CI)
16	37	(2.4)	36	(14.3)	10	(4.7)	10	(24.4)	0.48	(0.23–0.98)	0.49	(0.22–1.10)
18	38	(2.4)	31	(12.4)	7	(3.3)	6	(14.6)	0.79	(0.34–1.80)	0.87	(0.34–2.26)
31	11	(0.7)	11	(4.4)	3	(1.4)	3	(7.3)	0.51	(0.14–1.85)	0.59	(0.16–2.21)
33	4	(0.3)	4	(1.6)	0	(0.0)	0	(0.0)	–	–	–	–
35	39	(2.5)	35	(13.9)	2	(0.9)	2	(4.9)	2.93	(0.70–12.28)	3.52	(0.80–15.5)
39	24	(1.5)	16	(6.4)	1	(0.5)	1	(2.4)	3.40	(0.46–25.36)	2.54	(0.33–19.8)
45	10	(0.6)	9	(3.6)	1	(0.5)	1	(2.4)	1.23	(0.16–9.69)	1.34	(0.16–10.9)
51	72	(4.5)	63	(25.1)	14	(6.6)	13	(31.7)	0.67	(0.37–1.21)	0.67	(0.33–1.40)
52	88	(5.6)	72	(28.7)	17	(8.0)	16	(39.0)	0.71	(0.41–1.23)	0.64	(0.32–1.28)
56	18	(1.1)	18	(7.2)	4	(1.9)	4	(9.8)	0.65	(0.22–1.97)	0.72	(0.23–2.26)
58	9	(0.6)	9	(3.6)	0	(0.0)	0	(0.0)	–	–	–	–
59	3	(0.2)	3	(1.2)	0	(0.0)	0	(0.0)	–	–	–	–
68	21	(1.3)	20	(8.0)	3	(1.4)	3	(7.3)	0.98	(0.29–3.31)	1.12	(0.32–3.99)

Multiple (≥2)hr HPV types	127	(8.1)	113	(45.0)	24	(11.3)	23	(56.1)	0.71	(0.44–1.12)	0.62	(0.32–1.22)

CI: confidence interval; HPV: human papillomavirus; hr: high-risk; OR: odds ratio.

^aAdjusted for age and lifetime number of sexual partners as continuous variable.

Discussion

In the present study among more than 1900 men in Tanzania, we found that the majority (88.1%) were circumcised. The prevalence of STIs was lower in circumcised men compared with uncircumcised men (gonorrhoea 16.6% versus 30.0%, syphilis 5.8% versus 8.1%, HIV 8.4% versus 19.3% and hr-HPV 16.6% versus 20.5%). In the multivariable analyses, the risk of gonorrhoea and HIV was lower in circumcised men than in uncircumcised men. In addition, the risk of hr-HPV was somewhat lower in circumcised men and in particular the risk of HPV16 was significantly lower in circumcised men compared with uncircumcised men.

We found a reduced OR of gonorrhoea when comparing circumcised men with those uncircumcised. In contrast, most previous studies did not find a difference in the risk of gonorrhoea between circumcised and uncircumcised men.^10–13 These studies used urine samples for the assessment of gonorrhoea, whilst our data were self-reported, which may partly explain the difference in findings. Urine samples may represent a more accurate measure, though only assessing current infections, whereas self-reported data reflect the accumulated lifetime prevalence of gonorrhoea. Gonorrhoea is assessed to be asymptomatic in around 10% of men²⁴ and it is therefore unlikely that gonorrhoea is substantially underreported in our data due to this. Gonorrhoea predominantly infects the urethral mucosa¹⁰ and whether male circumcision has a protective effect against urethral infections is uncertain.

We found no strong association between male circumcision and syphilis, which could partly be due to a low number of cases with that particular outcome. Previous studies have provided diverging results. A meta-analysis of observational studies found that male circumcision reduced the risk of syphilis by 33%, with similar results for both lifetime and more recent infections.⁷ In line with this, a prospective cohort from sub-Saharan Africa showed that male circumcision was significantly associated with a reduced risk of incident syphilis (adjusted hazard ratio [aHR] = 0.58; 95% CI: 0.37–0.91), in particular among HIV-positive men (aHR = 0.38; 95% CI: 0.18–0.81).⁶ In contrast, two randomised controlled trials from Uganda and Kenya, respectively, showed that there was no significant difference in the incidence of syphilis between circumcised and uncircumcised men although low power to detect differences could be a possible explanation.^8,9 The results on the association between male circumcision and syphilis are, however, still not conclusive.

We found a 58% reduction in the odds of HIV in circumcised men compared with uncircumcised men after adjusting for age and lifetime number of sexual partners. In the present study, the age of circumcision lies before the age of first sexual intercourse in the vast majority of men. So even though our data are cross-sectional, our findings match with those of the randomised trials from sub-Saharan Africa, which have shown that male circumcision reduces the acquisition of HIV by approximately 60%.^1–3

The risk of hr-HPV tended to be lower, although not statistically significant, in circumcised men (OR = 0.81; 95% CI: 0.56–1.17) compared with uncircumcised men. In agreement with this, some previous studies have shown that male circumcision is associated with a lower prevalence of hr-HPV both among HIV-negative and HIV-positive men.^8,15–17 In addition, it has been suggested to reduce acquisition and increase clearance of hr-HPV in HIV-negative men.¹⁸ Based on a meta-analysis, Larke et al.¹⁷ reported that circumcised men were less likely to have prevalent genital HPV infection than uncircumcised men (summary OR = 0.57; 95% CI: 0.45–0.71).

Finally, we found that HPV16 was significantly less common in circumcised men than in uncircumcised men. There was no significant difference in the distribution of any other of the examined HPV types in relation to circumcision status. In the randomised controlled trials among HIV-negative men in Uganda, Gray et al.¹⁸ found a lower acquisition of HPV18 and 33 and an increased clearance of HPV39, 51 and 58 in circumcised men compared with uncircumcised men. On the other hand, Serwadda et al.¹⁶ did not find any statistically significant differences in the type-specific acquisition or clearance of HPV infections among HIV-positive men in Uganda. With regards to multiple hr-HPV types, Auvert et al.¹⁵ found that the prevalence of multiple hr-HPV was lower in circumcised men (4.2%) than in uncircumcised men (9.9%) (PRR = 0.43; 95% CI: 0.28–0.66). Similarly, Tobian et al.⁸ found a lower prevalence of multiple hr-HPV in circumcised men (4.3%) than in uncircumcised men (12.2%) (RR = 0.35; 95% CI: 0.17–0.71). Finally, Gray et al.¹⁸ found that the acquisition of multiple hr-HPV was markedly lower in circumcised men than in those uncircumcised (IRR = 0.45; 95% CI: 0.28–0.73). In the present study, the prevalence of multiple hr-HPV types tended to be lower in circumcised men than in those not circumcised. The association was, however, not statistically significant.

The foreskin mucosa contains a high density of dendritic (Langerhans) cells, CD4+ T cells and CD8+ T cells, which are all targets for HIV infection.^19,25 Removal of the foreskin may therefore explain reductions in HIV acquisition in circumcised men, and in addition, male circumcision may also create a less favourable environment for other STIs to occur at the male genital area.

The strengths of the present study include the large sample size and the large number of circumcised men. Furthermore, information on circumcision status and HIV status was assessed at, respectively, the clinical examination and at laboratory testing, which improves the accuracy of the data. The limitations of the study comprise the cross-sectional design, which does not allow for the assessment of temporality, i.e. the timing of male circumcision and of the acquisition of STIs. However, median age at circumcision (10 years) preceded median age at first sexual intercourse (19 years) by several years and thus, most men probably have contracted STIs after eventual male circumcision taking place. The data on gonorrhoea and syphilis were self-reported and therefore misclassification of these data cannot be ruled out. Nonetheless, this will most likely result in non-differential misclassification and therefore underestimate the associations. Furthermore, a non-random convenience sample was used, thus the sample may not be entirely representative of the general male population in Tanzania.

In conclusion, we found reduced odds of gonorrhoea and HIV in circumcised men compared with uncircumcised men. The odds of syphilis and especially hr-HPV also tended to be lower in circumcised men, although they did not reach statistical significance. Finally, HPV16 was statistically significantly lower in circumcised men than in uncircumcised men. These data add to the current understanding of the association between male circumcision and STIs. This could aid decision-makers if considering biomedical interventions including male circumcision as a tool to prevent the growing burden of STIs.

Footnotes

Authors’ contribution

SKK and JM designed the study; SKK, JM, CM and CK were responsible for the data collection in the original study; TBO, KF, SKK and CM analysed the data; all authors contributed to the interpretation of data; TBO drafted the manuscript; all authors revised the manuscript critically for important intellectual content and approved the final version of the article to be submitted for publication. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: SKK has received speaker’s fee from Sanofi Pasteur MSD and Merck, scientific advisory board fee from Merck, and unrestricted research grant through her institution from Merck. TI received speaker honoraria from Hologic GmbH and Becton Dickinson, and an unrestricted research grant to his institution from Hologic. CM received lecture fees and support for conference participation from Sanofi Pasteur MSD. The remaining authors declared no conflicts of interest.

Funding

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was investigator initiated and financially supported by Merck (ClinicalTrials.gov NCT00932009). The study sponsors had no role in study design, data analysis, data interpretation, writing of the report or the decision to submit the paper for publication.

References

Bailey

Moses

Parker

, et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007; 369: 643–656.

Gray

Kigozi

Serwadda

, et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007; 369: 657–666.

Auvert

Taljaard

Lagarde

, et al. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 trial. PLoS Med 2005; 2: e298.

New data on male circumcision and HIV prevention: policy and programme implications: conclusions and recommendations. Montreux: WHO/UNAIDS, 2007.

Sexually transmitted infections. Evidence brief. WHO/RHR/14.10. 2019. Geneva: WHO, https://www.who.int/reproductivehealth/publications/stis-fact-sheet/en/ (accessed 18 July 2019).

Pintye

Baeten

Manhart

, et al. Association between male circumcision and incidence of syphilis in men and women: a prospective study in HIV-1 serodiscordant heterosexual African couples. Lancet Glob Health 2014; 2: e664–e671.

Weiss

Thomas

Munabi

, et al. Male circumcision and risk of syphilis, chancroid, and genital herpes: a systematic review and meta-analysis. Sex Transm Infect 2006; 82: 101–109.

Tobian

Serwadda

Quinn

, et al. Male circumcision for the prevention of HSV-2 and HPV infections and syphilis. N Engl J Med 2009; 360: 1298–1309.

Mehta

Moses

Parker

, et al. Circumcision status and incident herpes simplex virus type 2 infection, genital ulcer disease, and HIV infection. AIDS 2012; 26: 1141–1149.

10.

Reynolds

Shepherd

Risbud

, et al. Male circumcision and risk of HIV-1 and other sexually transmitted infections in India. Lancet 2004; 363: 1039–1040.

11.

Forbes

Doyle

Maganja

, et al. Rapid increase in prevalence of male circumcision in rural Tanzania in the absence of a promotional campaign. PLoS One 2012; 7: e40507.

12.

Mehta

Moses

Agot

, et al. Adult male circumcision does not reduce the risk of incident Neisseria gonorrhoeae, Chlamydia trachomatis, or Trichomonas vaginalis infection: results from a randomized, controlled trial in Kenya. J Infect Dis 2009; 200: 370–378.

13.

Sobngwi-Tambekou

Taljaard

Nieuwoudt

, et al. Male circumcision and Neisseria gonorrhoeae, Chlamydia trachomatis and Trichomonas vaginalis: observations after a randomised controlled trial for HIV prevention. Sex Transm Infect 2009; 85: 116–120.

14.

Diseker

III Peterman

Kamb

, et al. Circumcision and STD in the United States: cross sectional and cohort analyses. Sex Transm Infect 2000; 76: 474–479.

15.

Auvert

Sobngwi-Tambekou

Cutler

, et al. Effect of male circumcision on the prevalence of high-risk human papillomavirus in young men: results of a randomized controlled trial conducted in Orange Farm, South Africa. J Infect Dis 2009; 199: 14–19.

16.

Serwadda

Wawer

Makumbi

, et al. Circumcision of HIV-infected men: effects on high-risk human papillomavirus infections in a randomized trial in Rakai, Uganda. J Infect Dis 2010; 201: 1463–1469.

17.

Larke

Thomas

Dos Santos Silva

, et al. Male circumcision and human papillomavirus infection in men: a systematic review and meta-analysis. J Infect Dis 2011; 204: 1375–1390.

18.

Gray

Serwadda

Kong

, et al. Male circumcision decreases acquisition and increases clearance of high-risk human papillomavirus in HIV-negative men: a randomized trial in Rakai, Uganda. J Infect Dis 2010; 201: 1455–1462.

19.

Szabo

Short

RV.

How does male circumcision protect against HIV infection?

BMJ 2000; 320: 1592–1594.

20.

Olesen

Iftner

Mwaiselage

, et al. Prevalence and type distribution of human papillomavirus among 1813 men in Tanzania and the relationship to HIV status. Sex Transm Dis 2013; 40: 592–598.

21.

Olesen

Mwaiselage

Iftner

, et al. Risk factors for genital human papillomavirus among men in Tanzania. J Med Virol 2017; 89: 345–351.

22.

Joint United Nations Programme on HIV/AIDS (UNAIDS)-WHO. Revised recommendations for the selection and use of HIV antibody tests. Wkly Epidemiol Rec 1997; 72: 81–87.

23.

National comprehensive guidelines for HIV testing and counseling. The United Republic of Tanzania, Ministry of Health and Social Welfare (National AIDS Control Program), 2013.

24.

Global prevalence and incidence of selected curable sexually transmitted infections. Overview and estimates. Geneva: World Health Organization, 2001.

25.

McCoombe

Short

RV.

Potential HIV-1 target cells in the human penis. AIDS 2006; 20: 1491–1495.