Page 9 - Canadian Urological Association guideline on the care of the normal foreskin and neonatal circumcision in Canadian infants (Full Version)

dave et al

able and cannot be extrapolated to Canada from the lower in the circumcised group (adjusted PRR 0.66; 95%
African RCTs. CI 0.39‒1.12; p=0.12). 90
2. Male to male transmission: Based on current evi- The third trial evaluated HPV acquisition and clearance in
dence, MC does not provide protection for men who HIV-negative men using glans and coronal sulcus samples.
91
have sex with men (Level 2a). The incidence rate of HR HPV infection was statistically
3. Women partners: Based on current evidence, MC significant at one year (IRR 0.61; 95% CI 0.44‒0.85) in the
is not protective for female partners (Level 2a‒b). uncircumcised group, but at two years’ followup, the effect
4. Universal infant circumcision cannot be recom- was not statistically significant (IRR 0.64; 95% CI 0.38‒1.07).
mended to prevent HIV infection based on current The incidence of type-specific HR HPV was statistically sig-
evidence (Grade B). nificant for only the 18 and 33 genotypes and not 16. The
acquisition of new multiple infections was lower in the cir-
2. Circumcision and HPV prevention cumcised group (IRR 0.45; 95% CI 0.28‒0.73) compared to
the non-circumcised group. Clearance rates per 100 person
HPV is the commonest STI worldwide and of the more than years were statistically significant only for types 39, 51, and
100 types, about 40 can infect the ano-genital area. In the 58 and overall clearance rates were higher in the circum-
absence of vaccination, up to 75% of Canadians would have cised group (RR 1.36; 95% CI 1.13‒1.63). The fourth study
at least one lifetime HPV infection (www.phac-aspc.gc.ca). in the Ugandan trial evaluated HR HPV prevalence, acquisi-
High- risk oncogenic types, like 16 and 18, are implicated tion/incidence and clearance in HIV-positive married men
92
in cervical, penile, vulval, vaginal, anal, and some oropha- with or without a circumcision. A random sample compris-
ryngeal cancers, while low-risk non-oncogenic subtypes, ing 22% of those enrolled was tested at enrollment and at 24
like 6 and 11, cause genital warts. The effect of circumcision months’ followsup. MC provided partial protection, with the
on HPV is difficult to interpret, as HPV infection can be circumcised group (55% positive) showing lower HR HPV
transient, affect multiple genital areas outside the foreskin, prevalence at 24 months (PRR 0.77; 95% CI 0.62‒0.97).
include several high-risk and non-high-risk types, and is sig- The incidence rate for one or more new infections after
nificantly associated with other behaviuoral confounders. In adjustment was not statistically significant between the cir-
addition, HPV prevalence, incidence, clearance, and viral cumcised and non-circumcised group (IRR 0.74; 95% CI
load are all potential outcomes that can be studied and have 0.54‒1.01); however the proportion of men acquiring mul-
differing health implications. tiple new HR HPV infections was lower in the circumcised
group (IRR 0.40; 95% CI 0.19‒0.84). The clearance rate of
HPV in men HPV infections was not significant different between the two
arms. In a more recent analysis of 999 men (HIV-positive
Level 1–2 evidence and -negative) from the Ugandan trial, Tobian et al showed
an increased HPV clearance in HIV-negative circumcised
Auvert et al showed a reduction in the prevalence of urethral men (adjusted RR 1.48; 95% CI 0.55‒0.89) and lower inci-
high- risk HPV infection following male circumcision, with dence of HR HPV acquisition in HIV-positive men (IRR 0.70;
93
a prevalence rate ratio of 0.68 (95% CI 0.52‒0.89; p=0.002) 95% CI 1.67‒2.44). The final study evaluated HPV viral
in circumcised men as compared to uncircumcised men. load in circumcised and uncircumcised HPV-infected men
88
Significantly, the prevalence differences between the two at 24 months. MC decreased HPV viral load in circum-
94
groups were not significant for HPV type 16, but were for cised compared to uncircumcised men for new infections
HPV type 18. acquired after enrollment, but the results were statistically
Six secondary trials analyzed HIV-positive and -nega- significant for only serotype 16 (p=0.001).
tive men enrolled in the Ugandan HIV trial with regards to Two meta-analyses were conducted evaluating the effect
HPV prevalence, acquisition/incidence, clearance, and viral of MC on HPV. 96,97 Albero et al in 2012 analyzed data from
load. 89-94 The first trial investigated the prevalence of HPV 14 observational studies and two RCTs conducted between
97
in a subgroup of participants and only included samples 1971 and 2010. Accepting heterogeneity in MC report-
from the glans and coronal sulcus, a factor for possible bias ing status, sites sampled and methods of detection, in the
due to differential infection sites in circumcised males. 89,95 1784 participants analyzed with data from the two RCTs,
The adjusted risk ratio for prevalence of high risk HPV at the authors detected a strong inverse association between
two years’ followup in the circumcised group was 0.65 circumcision and high-risk HPV prevalence, with an OR of
(95% CI 0.46‒0.90; p=0.0009) and 0.66 for low-risk HPV 0.67 (95% CI 0.54‒0.82). The 14 prevalence studies showed
genotypes (95% CI 0.49‒0.91; p=0.01). In a subsequent a similar pooled result of overall HPV prevalence (OR 0.57;
report, the same group showed that the one-year penile shaft 95% CI 0.42‒0.77). HPV prevalence remained lower in cir-
HPV prevalence after MC was not statistically significantly cumcised men, even pooling studies where the penile shaft

E84 CUAJ • February 2018 • Volume 12, Issue 2