Comparing Penile Problems in Circumcised vs. Uncircumcised Boys: Insights from a Large Commercial Claims Database with a Focus on Provider Type Performing Circumcision

Abstract

Objectives

To compare penile problems in circumcised relative to uncircumcised boys, and to determine which providers performing the circumcision have fewer post-circumcision problems.

Methods

CPT codes in the 2011-2020 MarketScan database were used to identify boys who had a circumcision. Uncircumcised control subjects of the same age, state of residence, and insurance type were selected. The primary outcome was a penile problem, defined as penis-specific infection, inflammation, and urethral stricture/stenosis, among others. The secondary outcomes were procedure-related complications limited to 28 days after circumcision, and whether post-circumcision problems varied by the clinician performing the procedure. ICD-9/10 diagnostic codes were used to identify these problems.

Results

We identified ~850,000 cases and ~850,000 matched controls. Overall, the rate of penile problems within the first five years of life was 1.7% in circumcised boys versus 0.5% in uncircumcised boys (p<0.05). Multivariable regression models showed that the risk of penile problems was 2.9-fold higher among circumcised compared to uncircumcised males (95%CI [2.8-3], p<0.001). Compared to males circumcised by pediatricians, those circumcised by surgeons had 2.1-fold higher penile problems in the year after circumcision (95% CI [2-2.3], p<0.001). Procedure-related complications within 28 days of circumcision were infrequent (0.5%), with the most common being penile edema (0.2%).

Conclusions

Penile problems are very infrequent in boys in the first five years of life. However, when they occur, they are 3x more likely to occur in circumcised boys relative to uncircumcised boys. Penile problems are more likely to occur in boys circumcised by surgeons.

Introduction

Male circumcision is one of the most common procedures in the world. Approximately 40% of the worldwide male population is circumcised, and the prevalence is even higher among male children born in the United States [1,2]. According to policy statements from the American Academy of Pediatrics and the Centers for Disease Control and Prevention, the prophylactic benefits of elective male circumcision at a young age outweigh the risks associated with this procedure [3,4]. Previous studies demonstrate that male circumcision has prophylactic benefits against several urinary tract infections (UTI), sexually transmitted infections (STI), malignancies of the male genitalia (e.g. penile cancer), and inflammatory disorders [5–9]. In addition, more than half of uncircumcised males will experience an adverse foreskin-related medical condition during their lifetime [5]. Male circumcision is the definitive choice of treatment for conditions such as phimosis in uncircumcised adolescent and adult males [5,10]. Conversely, it has been estimated that 13-65 million of the worldwide male population are enduring circumcision-related problems and complications including acute surgical adverse events and penile problems with later manifestations such as adhesions, fistulae, and meatal stenosis, as well as unsatisfactory cosmetic appearances ^10-15. However, a limitation to these findings arises from the potential bias in reporting, as certain penile problems may be more commonly documented in circumcised children while going unnoticed in uncircumcised individuals due to a lack of comparable encounters.

One of the earliest dilemmas parents of male neonates encounter is whether to circumcise their child. Much of the data purporting the benefits of circumcision occurs over the long term. While these data demonstrate the long-term benefits and risks of circumcision, there is little data investigating the incidence of penile problems in infants and young children. El Bcheraoui and colleagues demonstrate an early adverse event rate for circumcision in the U.S. that is less than 0.5% [11]. What is not known, however, is what the rate of penile problems in circumcised males is over the first five years of life. It is well known that male circumcision is performed in various settings and by different types of care providers (e.g. family practitioners, pediatricians, obstetricians, midwives, nurse practitioners, pediatric surgeons, and urologists) in the U.S. [12,13]. Given the diverse backgrounds of those who circumcise, are the risks and longer-term outcomes of this procedure different depending on which care provider performs the circumcision? Conversely, what is the likelihood of penile problems if parents choose not to circumcise their boy? That is, is an uncircumcised boy going to have more or less penile problems compared to a circumcised boy?

To the best of our knowledge, there are no reports comparing penile problems in circumcised and uncircumcised boys over the first five years of life. This study was designed to evaluate penile problems and circumcision-related complications in young males over the first several years of life and to determine whether the short and long-term circumcision-related complications varied by the type of provider performing the circumcision. We hypothesized that uncircumcised boys would have fewer penile problems compared to circumcised boys and that circumcision performed by a surgeon would have fewer penile problems and procedure-related complications.

Methods

In this retrospective case-control study, we used the Merative MarketScan database to identify male children who had circumcision [14]. MarketScan is a commercial claims database that provides de-identified longitudinal data from settled and paid claims for privately insured individuals with employment-based insurance. Institutional Review Board (IRB) approval was not required for this study due to the de-identified nature of the dataset.

The MarketScan database provides individual-level demographic information and inpatient and outpatient medical information including provider type, setting of care, and received care services. We accessed all available inpatient and outpatient claims data between January 2011 and December 2020 and used Current Procedural Terminology (CPT) codes for circumcision (54160, 54150, and 54161) to identify cases, who were circumcised male individuals below 18 years old. The age at which circumcision occurred as well as which provider type performed the circumcision were also recorded. Provider type was identified based on the STDPROV variable of the clinician who was billed for the procedure. We used the facility header (F) and selected among the options available for provider types using the MarketScan standardized provider type.

Excluded were children who had a diagnosis of prior urogenital conditions (i.e., hypospadias, chordee, epispadias bladder exstrophy, penile torsion) and those who were circumcised in conjunction with other penile procedures such as hypospadias repair or chordee correction. The control group was composed of uncircumcised male children of the same age, state of residence, and insurance type for each circumcision case. International Classification of Diseases 9th or 10th edition (ICD-9/ ICD-10) and CPT codes were utilized to capture penile problems, complications, and related urological procedures (Supplementary 1). We collected demographic information including geographic region and insurance plan type. For each case and control, penile problems, circumcision complications (for cases), and treatments received for these conditions were recorded. The primary outcome of this investigation was evidence of new penile problems. The secondary outcome was circumcision-related complications controlling for the type of provider performing the procedure.

We evaluated possible penile problems and complications based on a described method in a previously published study [11]. General penile problems not specific to circumcision included balanitis, urethral strictures, meatal stenosis, buried/hidden penis, fistulas, and the procedure received for these issues including meatotomy/meatoplasty, incision/drainage, lysis/excision of adhesions, plastic repair/reconstruction, and penis replantation. A 28-day post-circumcision timeframe was used for procedure-specific complications resulting from circumcision including hemorrhage, edema, amputations, laceration, wounds, gangrene, staphylococcal-scalded skin, cellulitis, and other inflammatory disorders of the penis as well as the procedures received for these issues including hemorrhage control, suture of vessels, and suture of laceration of the penis.

Statistical analysis

Descriptive statistics were generated for covariates of interest, and results were expressed as frequencies for the categorical variables and as median (with interquartile range) for continuous variables. We compared all demographic and healthcare variables between cases and controls using univariate tests of variation. Multivariable Cox proportional hazards models were built to compare penile problems in circumcised to uncircumcised children. Children were followed from birth until they developed a penile complication. Only complications not associated with a surgical procedure were considered. Patients were censored when their insurance lapsed or at five years of age. The main predictor of interest was circumcision which was treated as a time-varying covariate. Geographic region and rurality (i.e., whether the patient lived within a metropolitan statistical area or not) were included as additional covariates. The outcomes were reported as hazard ratios (HR) with 95% confidence intervals (95% CI).

For the secondary outcomes, additional models within the circumcised patients only were developed. In this model, we began following cases at the time of their circumcision until they developed a penile problem. Here we considered two outcome profiles. The first was the same complications as in the primary model described above as well as any procedure-related complications arising within 28 days of the procedure. The second only considered the procedure-related outcomes. Again, patients were censored when their insurance lapsed or at age 5. For this model, we also include two additional covariates: the timing of the procedure (defined as within the first 90 days of life or after) and the type of provider associated with the procedure. The level of significance in all models was defined as p<0.05.

Results

Our total cohort contained 1704102 subjects including 852051 (50%) circumcised cases and 852051 (50%) uncircumcised controls. The median (IQR) follow-up for cases and controls were 458 (183-1036) and 504 (188-1095) days. Obstetricians (n: 395153, 46%) and pediatricians (n: 277145, 33%) performed over three-quarters of all circumcisions. Surgeons including urologist providers performed slightly more than 3% (n: 26967) of all circumcisions. Circumcisions were carried out on average at 47 ± 89 days for surgeons, 2 ± 6 days for obstetricians, 5 ± 13 days for family practitioners, and 5 ± 11 days for pediatricians (p<0.001).

The percentage of patients experiencing penile problems during their first five years of life was 1.7% (n=14363) for the circumcised and 0.5% (n=4334) for the uncircumcised group (p<0.001). Balanoposthitis was the most common penile problem detected in both groups: 0.8% (n=6602) of the circumcised and 0.4% (n=3344) of uncircumcised boys p<0.001). Urethral strictures and/or meatal stenosis occurred in 0.3% (n=2918) of circumcised boys compared to 0.07% (n=605) of uncircumcised boys (p<0.001). The average time from circumcision to meatal stenosis and/or urethral stricture was 677 (150-1284) days. Penile cellulitis (n=890, 0.1% in circumcised vs. n=547, 0.06% in uncircumcised; p<0.001) and buried/hidden penis (n=1898, 0.2% in circumcised vs. n=394, 0.05% in uncircumcised; p<0.001) were more frequently observed among the circumcised boys (Table 1).

Table 1:

Long-term penile problems in circumcised versus uncircumcised males in 2011-2020 IBM MarketScan database.

Penile problems	Circumcised N = 852051	Uncircumcised N = 852051
TOTAL	14363 (1.69%)†	4334 (0.51%)†
Balanoposthitis*	6602 (0.77%)	3344 (0.39%)
Meatal stricture and stenosis*	2918 (0.34%)	605 (0.07%)
Buried/hidden penis*	1898 (0.22%)	394 (0.05%)
Other specified disorders of penis (e.g., chordee, cysts, pain)*	2181 (0.26%)	339 (0.04%)
Unspecified disorders of penis (e.g., penile adhesions, penile pain)*	1198 (0.14%)	152 (0.02%)
Urethrocutaneous fistula*	66 (0.01%)	0 (0%)

^* = p < 0.05

^†500 circumcised and 449 uncircumcised males had more than one penile problem

Furthermore, within the circumcised group, an additional 0.5% (n=3907) of children had at least one procedure-related complication such as penile edema and hemorrhage within a 28-day post-circumcision time frame. Procedure-related complications within 28 days of circumcision were very rare (total n=3907, 0.5%), with the most common being penile edema (n=1981, 0.2%). Inflammation/infection (n=1045, 0.1%) and bleeding (n=863, 0.1%) were the next two most common. Major penile injuries (e.g., penile amputation, penile laceration) and deaths were never coded (Table 2).

Table 2:

Procedure-related complications within 28 days of circumcision.

Penile problems	Circumcised N = 852051
TOTAL	3907 (0.46%)
Edema	1981 (0.23%)
Inflammation	1045 (0.12%)
Bleeding	863 (0.1%)
Gangrene	10 (0%)
Other (amputation, laceration, etc.)	8 (0%)

Additional procedures and surgeries on the penis were infrequent. The most common penile procedures were lysis or excision of penile adhesions (n: 4199, 0.5% in circumcised vs. n: 433, 0.05% in uncircumcised; p<0.001) and meatotomy/meatoplasty (n: 1052, 0.1% in circumcised vs. n: 111, 0.01% in uncircumcised; p<0.001). The average time from circumcision to lysis of penile adhesions and urethral meatotomy among those circumcised was 228 (73-497) and 1291 (973-1541) days, respectively. In addition, 3096 (0.4%) of the circumcised male children underwent a revision circumcision.

The estimated survival rate for all penile problems in circumcised children in the first year after the circumcision was 0.96 (95% CI [0.96-0.96]), and in the first year after birth in the uncircumcised boys was 0.99 (95% CI [0.98-0.99]). Using circumcision as a time-varying covariate and controlling for region and rurality, our models showed that the risk of any penile problem was 2.9 times (95% CI [2.8-3]; p<0.001) higher among circumcised boys compared to uncircumcised boys within five years after circumcision/birth (Figure 1, Table 3).

Figure 1.

Kaplan-Meier curves of time to penile problem for circumcised (CC) (blue) vs. uncircumcised boys (black).

Table 3.

Hazard ratio (HR) of penile problems in circumcised vs. uncircumcised boys during the study period controlling for region (relative to Northeast) and rurality (relative to rural).

Variables	HR	Low95	High95	p-value
Region (Midwest)	0.78	0.75	0.82	p<0.001
Region (South)	0.91	0.87	0.95	p<0.001
Region (West)	1.1	1.04	1.15	p<0.001
Region (Other)	1.14	1.01	1.3	p=0.038
Rurality (urban)	1.23	1.2	1.3	p<0.001
Study group (circumcised)	2.9	2.8	3	p<0.001

Controlling for region, rurality, and the timing of the procedure (relative to <= 90 days of age), we found that boys who were circumcised by obstetricians (HR: 1, 95% CI [1-1.1]; p=0.1) and family practitioner (HR: 1, 95% CI [0.9-1.1]; p=0.1) showed no significant differences in penile problems in comparison with children who pediatricians circumcised. In comparison with those circumcised by pediatricians, penile problems were 1.3 times more prevalent (95% CI [1.1-1.6]; p= 0.008) among male children circumcised by nurse practitioners, while surgical complications were 0.9 times less frequent (95% CI [0.8-1]; p= 0.03). Boys circumcised by surgeons had 2.1-fold (95% CI [2-2.3]; p<0.001) higher penile problems at any time during the study period and 2.9-fold (95% CI [2.5-3.5]; p<0.001) higher procedure-related complications within 28 days following the procedure compared to those circumcised by pediatricians. Children circumcised after the first 90 days of life had 3.4 times (95% CI [3.1-3.7]; p<0.001) higher penile problems and 5.1 times (95% CI [4.4-5.9]; p<0.001) higher procedure-related complications compared to those circumcised within the first three months of life.

Discussion

Evaluation of 2011-2020 commercial claims data for male children demonstrated that relative to uncircumcised males, circumcised males had nearly 3-fold more penile problems and received more interventions related to these problems in the first five years of life. Additionally, children circumcised by surgeons had higher rates of penile problems and procedure-related complications compared to those circumcised by other provider types. Surgeons performed circumcisions at an older age compared to other provider types. Therefore, the higher rate of penile problems and complications in surgeon-performed circumcisions could be partly due to the older age of the patients at the time of the procedure and the increased complexity of cases handled by surgeons.

The overall complication rate for non-therapeutic circumcision varies between 2-14% in previous investigations [15]. Our findings showed that the overall rates of common penile problems during the first five years of life were 1.5% for circumcised and 0.5% for uncircumcised boys, with higher rates of penile cellulitis and balanoposthitis in circumcised boys. These findings are contrary to previously reported protective roles of circumcision on penile problems [5,10,16,17]. Although previous studies indicated the most common late complication of circumcision as meatal stenosis affecting between 2-10% of the circumcised males [10,15,18], our findings showed that meatal stenosis rates were detected in less than 1% of circumcised males. This complication occurred at an average of nearly two years after the circumcision. Also, in contrast to other studies that reported higher rates of penile infection in uncircumcised males, we found no evidence of such differences in comparison with circumcised boys [5,11].

The risk of procedure-related complications within a 28-day timeframe following the procedure in our study (0.5%) agrees with previous reports on the risk of male circumcision adverse events ranging from 0.0008% to 3.6% in infants and from 0.9% to 8.8% in adults [11]. The most common procedure-related complications reported in other investigations were hemorrhage, edema, and inflammation which were also consistent with our findings [15]. The outcomes of our study demonstrated that penile problems and procedure-related complications were more common among boys who were older at the time of the circumcision. These findings are also in accordance with previous reports on higher circumcision complications at older ages [19–21]. Several studies have shown that neonatal circumcision is associated with the lowest complication rates among all age groups with complication rates ranging between 0.2% and 0.6% [5,15]. The complication rate is considerably higher if circumcision is performed at an older age [10,15]. In an investigation by El Bcheraoui et al., the incidence rates of adverse events were 20-fold and 10-fold higher in boys circumcised at age 1-9 and at 10 years or older in comparison with those who underwent circumcision at younger than 1 year old [11]. Late circumcision is also more expensive, entails risks of general anesthesia, and is associated with a longer healing period[5].

Is imperative to consider the significance of studying these issues in children, particularly within the first five years of life. Childhood circumcision remains a common practice in many societies, and understanding the early-life outcomes of this procedure is crucial due to the long-term implications of implications of this ubiquitous procedure on the individual’s health and well-being.

In contrast to our initial assumption, surgeon-performed circumcisions were associated with higher penile problems and complications. This finding confirms the outcomes of a systematic review by Muula et al. which revealed that no firm evidence of lower complication was found in studies where a surgeon performed the circumcision [22]. Since surgeon-performed circumcisions were generally performed at an older age compared to other care providers, we believe the increased risk of complication could be partly due to the older age of the children at the time of the circumcision, as well as increased patient complexity in surgeon-performed circumcisions. Complex cases usually undergo therapeutic circumcisions which result in nearly 2-fold complications compared with non-therapeutic circumcisions [19]. The most common penile complications in therapeutic circumcisions were adhesions and meatal stenosis [15,19].

We acknowledge the important distinction between therapeutic and elective/religious/social-cultural circumcisions. Therapeutic circumcisions are typically performed by surgeons on older males with pre-existing penile conditions. These conditions often necessitate surgical intervention, which inherently involves higher risks and complexities compared to elective circumcisions performed on younger, otherwise healthy males. The underlying disease processes, such as phimosis, recurrent balanoposthitis, or other penile abnormalities, significantly contribute to the higher complication rates observed in therapeutic circumcisions. Moreover, while we intentionally attempted to exclude males with congenital penile problems, there still may be contamination by said patients in the circumcision group. Lastly, the patients who received their circumcision from a surgeon likely had a good reason why the procedure was not done in the neonatal period, and this includes increased patient complexity (e.g., a former premature infant, a patient with another major congenital condition whose circumcision was appropriately delayed due to other more pressing medical needs like non-genital surgery).

This study should be viewed in light of its limitations. First, the study relied on an administrative database and thus is potentially subject to inherent coding mistakes and other limitations of existing diagnostic and procedure codes. For example, phimosis and redundant prepuce share the same ICD-9 code 605. In uncircumcised males, it is challenging to distinguish a pathologic phimosis causing dysuria and ballooning of the foreskin during urination from otherwise physiologic phimosis that is a common and expected finding. Similarly, circumcised males could have an acceptable circumcision but somewhat redundant prepuce, or they could have aggressive post-circumcision scarring and inflammation leading to ‘phimosis’. In either case, the same ICD-9 code could be selected. Thus, given the nonspecific nature of this code, and because we were unable to differentiate problematic from benign usage of this code, we excluded it from our analyses. This major exclusion likely leads to an underestimate of the rate of penile problems in both groups. In addition, there is a possibility that our control group includes individuals who have undergone circumcision, yet this information may not be fully captured in our dataset. The Merative MarketScan database captures only commercially insured individuals, so this analysis does not include any boys with Medicaid or Medicare. This major exclusion might also lead to an underestimation of penile problems in both groups.

Circumcisions performed outside conventional medical procedures, such as during religious ceremonies or immediately after birth, might not be adequately documented, leading to an incomplete representation of circumcision status in our control group.

Conclusion

Our findings demonstrated an increased risk of penile problems in circumcised male children associated with older age at the time of the circumcision. Considering that this study was performed in the U.S.-born population, the negative outcomes of circumcision can be even higher in other regions where traditional procedures are still practiced.

We also found more penile problems and procedure-related complications in children circumcised by surgeon providers. This could be due to older age or increased patient complexity in surgeon-performed circumcisions. Facilitating surgeon and urologist visits during the neonatal period for children with complexities can potentially reduce the risks and require further evaluation.

Highlights.

• It is unclear which-circumcised or uncircumcised-boys have more penis problems within their first 5 years and if the clinician performing the circumcision affects complications.

• Penile problems are 3x more likely to occur in a circumcised boy with 2x higher risk for boys circumcised by a surgeon compared to a pediatrician.

Abbreviations:

CI

Confidence intervals

CPT

Current Procedural Terminology

HR

Hazard ratios

ICD-9/ ICD-10

International Classification of Diseases 9th or 10th edition

SD

Standard deviation

STI

Sexually transmitted infection

UTI

Urinary tract infection