Circumcision, sexual function and penis sensitivity:


Letters in urology journal criticise Bossio study


A study by Bossio et al published in the Journal of Urology in June 2016 was widely misreported as showing that circumcision made no difference to the sensitivity of the penis. Even leaving aside the vagueness of this measure (do they mean sensitivity to pain or to pleasure? what about other measures of sexual functionality and satisfaction?) the reports seriously misrepresented the findings of the study, which actually showed that men with foreskins had a lower threshold of sensitivity and circumcised men a higher threshold – in other words, that men with foreskins were more sensitive to touch, and circumcised men less sensitive. As the authors admit in their reply to criticisms of their article published in a later issue of the journal, “the foreskin was observed to be most sensitive to fine touch pressure thresholds.” Of course, there are other issues to be taken into account, some of which are raised in the letters criticising the paper, but there is not the slightest basis for media and other reports that circumcision “makes no difference” to sexual experience.

The following letters criticising the Bossio et al study on circumcision and penis sensitivity were published in the December 2016 issue of the Journal of Urology, with a reply by the original authors.

1. Professor Morten Frisch

Bossio et al assessed penile sensitivity in circumcised and intact males using quantitative sensory measures, and concluded that infant circumcision likely entails only “minimal long-term implications for penile sensitivity.” However, this conclusion does not follow logically from their findings.

Despite a small sample size (62 sexually non-dysfunctional males 18 to 37 years old, of whom 30 were neonatally circumcised and 32 were intact), the study confirmed findings from a prior investigation by Sorrells et al, which revealed significantly lower tactile thresholds (i.e. greater tactile sensitivity) in the foreskin compared to other examined penile sites (the glans and 2 shaft positions). Additionally the current study found significantly greater warmth sensitivity at the foreskin than at the glans (p = 0.02). Finally, the authors assessed thresholds for tactile pain and heat pain, and failed to find any significant evidence that the foreskin is more or less pain sensitive than other parts of the penis.

These results suggest that 1) the foreskin is more sensitive to gentle touch than any other tested penile site, replicating prior research, 2) the foreskin may be more sensitive to subtle variations in temperature, at least compared to the glans, and 3) the foreskin appears to be no more or less sensitive to potentially uncomfortable tactile or temperature induced stimuli leading to sensations of penile pain. From these observations the logical conclusion would be that neonatal circumcision appears to be associated with reduced sensitivity in the adult penis. It is noteworthy that the findings of Bossio et al of greater foreskin sensitivity to tactile and temperature related stimuli emerged despite the fact that the examined outer cutaneous part of the foreskin may not even be the most sensitive part of this complex, double layered structure.

Figure 2 in the article combines individual sensitivity threshold scores from the 30 circumcised and 32 intact males. To better understand the relevance and meaning of the study findings, I kindly urge the authors to repeat their analysis and provide a supplementary figure, in which they divide the results depicted in figure 2 into 2 separate panels (1 for intact and 1 for circumcised males), illustrating each of the 4 sensory threshold graphs as demonstrated for the combined group of 62 males in this figure. An additional simplified 4-graph panel for the 32 intact males may help readers to judge whether and to what extent the foreskin is more sensitive than other parts of the intact penis. Due to the apparent lack of statistically significant differences in sensitivity among the various non-foreskin penile sites, I suggest that Bossio et al recombine the original 5 tested anatomical sites, i.e. the foreskin, other penile sites (including the glans and the 2 penile shaft sites) and the forearm, to gain statistical power. I also would be interested to see the associated p values provided in a manner similar to that used in figure 2.

Morten Frisch, Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark

2. Dr Alexandre Rotta

In a study fraught with methodological problems Bossio et al conclude that “circumcision is not associated with changes in penile sensitivity” and believe they have provided “preliminary evidence to suggest that the foreskin is not the most sensitive part of the penis.” However, neither of these assertions is supported by their data. Although nowhere explicitly stated in the article, we assume that all penile sensitivity tests were performed in penises in the flaccid state. Such a testing condition can only provide an indirect hint as to what would occur during sexual arousal, which is the relevant situation to consider when studying penile sensitivity in this context.

For unclear reasons the authors tested only a single site on the foreskin – a location on the dorsal unretracted external skin, somewhat proximal to the more sensitive tip of the penis - that previous research has already indicated may be less sensitive than other parts of the foreskin. Specifically if the authors wanted to determine the sensory thresholds of the foreskin of intact men compared to other penile sites, why would they not test its most sensitive point, or at least include a site representing the (widely considered to be) more sensitive transitional or internal surfaces of the foreskin that become exposed when the structure is pulled back?

Due to differences in erection mechanics in the intact and circumcised penis, the tested sites on the penile shaft may not be comparable between the 2 groups during sexual activity. The penile shaft skin of circumcised males does not move back and forth a great deal during sex, so the penile shaft sites tested by Bossio et al in flaccid circumcised males may serve as a reasonable proxy for what those same subjects will experience while erect but not for their intact counterparts. During intercourse a considerable part of the penile shaft of intact males is covered intermittently by the everted mucosal portion of the foreskin, which rolls back and forth over the glans. During penetration the skin that covered the penile shaft in the flaccid state moves proximally closer to the pubis. Consequently the penile shaft skin measurements by Bossio et al are unlikely to represent analogous penile sites in sexually aroused intact and circumcised males. To provide a more meaningful comparison of penile shaft sensitivities in intact and circumcised males that might serve as a plausible proxy for what occurs in the erect state, penile shaft sensitivity tests in the flaccid state should have been carried out with the foreskin pulled back (as typically will be the case during sexual arousal) in intact males.

Even with its underpowered sample and inadequate choice of cutaneous testing sites the conclusion of the authors that the foreskin is not the most sensitive part of the penis is certainly puzzling in light of figure 2, part A in the article, which shows the foreskin to be significantly more sensitive than any other tested site. The inference that surgical removal of this most sensitive penile segment would not decrease penile sensitivity seems logically and anatomically incoherent.

Alexandre T. Rotta, Department of Pediatrics, Case Western Reserve University School of Medicine, Cleveland, Ohio

3. Professor Robert Van Howe et al

This study examining penile sensitivity left us with several concerns. The report fails to adequately describe the main outcomes of the study, namely measurements broken down by circumcision status. Instead of average scores and standard deviations, the reader is provided with p values, statements about failing to find a statistically significant difference and a figure that collapses the data from the 2 groups into 1 group. Of interest is whether fine touch sensitivity was reduced in the glans in the flaccid penis of circumcised men, as has been reported in 3 previous studies. However, these results were not provided. Instead, the authors provide what could be characterized as pilot data, along with the number of participants needed for a more definitive study, while failing to report the results that are needed to design such a study.

Several of the conclusions of Bossio et al do not follow from their findings. Although several of their tests were underpowered, and the maximum participant age was only 37 years (mean 24.2), the authors make sweeping generalizations regarding long-term implications of circumcision on penile sensitivity. Similarly they state that “this study challenges past research suggesting that the foreskin is the most sensitive part of the penis,” while at the same time reporting that the “foreskin of intact men was more sensitive to tactile stimulation.” Finally, the authors conclude that they “failed to consistently replicate the findings by Sorrells et al across stimuli” when they did, in fact, replicate our findings along the only dimension that was consistent - and hence even potentially replicable - between the 2 studies, namely assessment of fine touch thresholds.

The authors struggle to explain some of their results but the histology may help. The glans is innervated mainly by free nerve endings, which primarily sense deep pressure and pain, so it is not surprising that the glans was more sensitive to pain. By contrast, the foreskin has a paucity of free nerve endings and is primarily innervated by fine touch neuroreceptors, so it was comparatively less sensitive to pain. On a positive note, it is heartening that the term “intact” has replaced “uncircumcised,” which is considered by many to be a pejorative term.

Robert S. Van Howe, Department of Pediatrics, Central Michigan University College of Medicine; with Morris L. Sorrells, James L. Snyder, Mark D. Reiss and Marilyn F. Milos

Reply by authors

Based on a careful review of these letters, we believe that our conclusions have been misconstrued in different ways by the authors (and others, as apparent in media reports and on social media). We will attempt to clarify what our article indicates, what it does not indicate and how the responses to it highlight the need to improve the empirical rigor of research on the impact of circumcision on the sexual lives of men and their partners.

Despite the widespread global practice of circumcision, as well as the public beliefs about the procedure, there is limited objective peer-reviewed research assessing the sexual correlates of circumcision. The purpose of our study was to use objective measures to assess penile sensitivity across circumcision status (intact, circumcised). We examined sensitivity on certain penile areas, as well as on the forearm, with a focus on the glans penis and the foreskin, using touch (punctate or fine touch pressure, pain) and heat (warmth detection, heat pain). The stimulus modalities we used are expected to activate penile nerve fibers more likely associated with sexual pleasure than measuring fine touch pressure thresholds alone. Statistical analyses revealed that the sensory thresholds of intact and circumcised men were remarkably similar - hence, the data were collapsed across the 2 groups. Although presented in graphic format in the full article, we have included a breakdown of descriptive statistics in this response, as requested by Frisch and Van Howe et al (see table).

To reiterate the findings of the study, the foreskin was observed to be most sensitive to fine touch pressure thresholds. Indeed, this finding replicated the results reported by Sorrells et al. However, fine touch pressure, which was only 1 of 4 stimulus modalities assessed, activates nerve fibers that are likely less relevant for sexual pleasure than fibers activated by the other stimuli used in this study (stimuli that did not exhibit significant between group differences). Therefore, we maintain that we “failed to consistently replicate the findings by Sorrells et al across stimuli” (emphasis added). We also urge caution in the over-interpretation of this result (that the fine touch pressure threshold at the foreskin was significantly less than in other areas). This finding alone does not prove that circumcision reduces penile sensitivity, and it also does not prove that circumcision has no impact on the sexual lives of men. Insisting that this finding supports either the pro or anti-circumcision “camp” is not warranted, as it does not take into account the other study findings and ignores the limitations (e.g. small sample size). Instead, we consider the outcome of this study an indication of the need for further examination.

In light of the misinterpretations of our findings we believe it necessary to further clarify what the results of this particular study do not indicate. The results of this series are not to be taken as the definitive answer concerning the circumcision debate. As we discuss in the article, the results do not address the role of the foreskin during sexual activity, nor was the study, as Rotta suggests, an exploration of the biomechanics of the foreskin during intercourse (an interesting question, indeed, but one that would be incredibly difficult to measured and one that we did not examine). This study was not an exploration of the direct impact of circumcision status on sexual function or on the experience of sexual pleasure. Furthermore, and importantly, we do not state whether the findings offer support to either side of the circumcision debate, but instead highlight the need for more empirically rigorous research.

The strong reactions to this self-described preliminary study highlight the need for more research and serve as a “call to arms” for researchers interested in examining the sexual correlates of circumcision. This body of research is plagued by weak study design, such as the inclusion of non-random samples, equating outcomes of adult and neonatal circumcision without evidence to suggest that the 2 are comparable, failure to control for participant expectations of study outcomes and reliance on self-report to the exclusion of objective measures. These shortcomings represent a serious problem in this contentious field because they allow room for participant and author bias. One does not have to search far for these biases in the circumcision literature, such as frequent references to non-peer-reviewed articles and author involvement in anti or pro-circumcision advocacy groups. Are we, as scientists, not responsible for stepping back from questions about which we have a personal stake in the outcome?

Research in this area should be conducted by those who are not personally invested in the circumcision debate. That was our initial goal. But following the severe reactions we have received, we wonder about who would want to continue (or even start) working on this topic without biased motivations. However, that is not what science is about. As comedian John Oliver stated in a recent commentary on the current state of science, “In science you don’t just get to cherry pick the parts that justify what you were going to do anyway.” We will continue to report what the data show, regardless of whether that reflects the popular vote, and hope that others in this field do the same.

Journal of Urology, Vol. 196 (December 2016), 1821-26. References and the table referred to in the authors’ reply have been omitted.

The original article, Examining Penile Sensitivity in Neonatally Circumcised and Intact Men Using Quantitative Sensory Testing by J.A. Bossio, C.F. Pukall and S.S. Steele, was published in Journal of Urology, Vol. 195 (June 2016) 1848-53.

Further information on this site


Previous report on this site, with links to further critiques

Where does the pleasure of penile stimulation come from?

[ Back to Top ]