PEDIATRICS Vol. 108 No. 2 August 2001, pp. 522-524

Circumcisions: Again

To the Editor.

The flurry of recent articles and commentaries addressing potential medical benefits of circumcision^1-4 provide much food for thought. I agree with the Academy's Task Force on Circumcision⁵ that the benefits of circumcision are not sufficient to recommend it routinely, that parents should be given accurate and unbiased information, and that procedural analgesia should be provided. It is unfortunate that the coverage in the lay press suggested that the AAP had changed its position to one opposing circumcision. A better headline would have read "AAP Task Force notes benefits, risks of circumcision; urges anesthesia."

Nonetheless, I agree with Schoen, Wiswell, and Moses⁴ that the language used by the Task Force is partly responsible for this misunderstanding. The Task Force statement says: "Existing scientific evidence demonstrates potential medical benefits; however, these data are not sufficient to recommend routine neonatal circumcision. In circumstances in which there are potential benefits and risks yet the procedure is not essential to the child's current well-being, parents should determine what is in the best interests of the child."

Although it is true that the benefits are "potential" in that most uncircumcised males will not develop conditions circumcision would have prevented, this is not the reason not to recommend it. We routinely recommend immunizations, car seat use, and safer sex, even though their benefits and risks are "potential" in the same way as those of circumcision and they also are "not essential to the child's current well-being."

The key difference between circumcision and preventive interventions that the AAP does recommend is neither the quality of the evidence nor the absolute magnitude of the benefits; it is the nature of the intervention. Circumcision involves surgery to remove a natural part of the boy's body and alter its appearance forever. It is a procedure with considerable nonmedical significance in many cultures. Thus, although there are definite medical benefits that seem to outweigh the medical risks, I think it is appropriate for the AAP to refrain from routinely recommending it, because it is not solely a medical decision.

This being said, it is important not to exaggerate the known benefit of preventing urinary tract infections (UTIs). The degree to which urinary tract infections in infancy lead to hypertension and end-stage renal disease (ESRD) in adulthood is not known, but is almost certainly less than suggested by combining the results of the long-term studies cited by Wiswell.¹ Using those figures, if 2.2% of uncircumcised boys got UTIs and 44% of them developed renal scarring and 10% of those with scarring developed ESRD over the next 30 years, the 30-year risk of ESRD from infancy would be about 2.2% × 44% × 10%, or about 1 in 1000. In fact, according to the US Renal Data Service, ESRD in children younger than 20 years old has an annual incidence of only about 13 million per year, of which only about 2.7% is attributed to chronic pyelonephritis or reflux nephropathy.⁶ Over the next 25 years the annual incidence rises to 109 million per year, but the proportion attributed to chronic pyelonephritis or reflux nephropathy declines to 0.7%. Even if all the cases in males were attributable to UTIs in those who were uncircumcised, the 45-year risk would still be an order or magnitude lower than would be projected from the numbers cited by Wiswell.

Thomas B. Newman, MD, MPH
Department of Epidemiology
University of California, San Francisco
San Francisco, CA 94143

REFERENCES

1. Wiswell TE The prepuce, urinary tract infections, and the consequences. Pediatrics. 2000; 105:860-862 [Free Full Text]
2. Schoen EJ, Colby CJ, Ray GT Newborn circumcision decreases incidence and costs of urinary tract infections during the first year of life. Pediatrics. 2000; 105:789-793 [Abstract/Free Full Text]
3. American Academy of Pediatrics, Task Force on Circumcision Circumcision debate. Pediatrics. 2000; 105:641-642 [Free Full Text]
4. Schoen EJ, Wiswell TE, Moses S New policy on circumcision-cause for concern. Pediatrics. 2000; 105:620-623 [Free Full Text]
5. American Academy of Pediatrics, Task Force on Circumcision Circumcision policy statement. Pediatrics. 1999; 103:686-693 [Abstract/Free Full Text]
6. US Renal Data Service. Annual Data Report. 1999, Chapter 2: Incidence and Prevalence of ESRD. Available at: http://www.usrds.org/chapters/ch02.pdf

To the Editor.

The 1999 Circumcision Policy Statement by the Task Force on Circumcision of the American Academy of Pediatrics does not recommend routine neonatal circumcision. Notwithstanding, the need for routine neonatal circumcision continues as a controversial topic in the American pediatric literature. We read with interest the study by Schoen et al¹ and the commentary by Wiswell² on circumcision; both authors are long-time proponents of routine neonatal circumcision and have contributed much to this debate.

Wiswell continues to discuss the risk of end-stage renal disease (ESRD) as a justification for routine neonatal circumcision.² Current research does not support this position.^3-7

Sreenarasimhaiah and Hellerstein³ reported on 102 patients with ESRD assessed in Kansas City, Missouri, between 1986 and 1995. Urinary tract infection (UTI) was considered an important contributing factor (our italics) in only 1 patient!³

Wennerström et al⁴ from Goteborg, Sweden, a center that has provided decades of excellent research on UTI, recently reported a long-term prospective follow-up on 1221 children (232 boys) with a first recognized symptomatic UTI during the years 1970 and 1979. Circumcision is not common in Sweden. Only 21 boys were found to have scarring and, of these, 18 (86%) were considered to have primary or congenital scarring, rather than acquired resulting from UTI. The authors note that "chronic renal failure caused by pyelonephritic renal scarring in Swedish children has decreased during the last few decades. In fact, in a population of 8.5 million, no child has been registered in this category during the last decade."⁴ Based on Swedish epidemiologic data, we estimate that >500 000 circumcisions are necessary to prevent end-stage renal failure in one boy.^5-7

We suggest it is time to put the specter of ESRD as a consequence of failure to circumcise into the history books as an anachronism previously fed on well-intentioned speculation but less enlightened data.

Wm. Lane, and M. Robson, MD, FRCP(C), FAAP, FRCP (Glasgow)
Pediatric Nephrology
The Children's Hospital
Greenville Hospital System
Greenville, SC 29605-4490

Robert S. Van Howe, MD
Department of Pediatrics
Marshfield ClinicLakeland Center
Minocqua, WI

REFERENCES

1. Schoen EJ, Colby CJ, Ray GT Newborn circumcision decreases incidence and costs of urinary tract infections during the first year of life. Pediatrics. 2000; 105:789-793
2. Wiswell TE The prepuce, urinary tract infections, and the consequences. Pediatrics. 2000; 105:860-862
3. Sreenarasimhaiah S, Hellerstein S Urinary tract infections per se do not cause end-stage kidney disease. Pediatr Nephrol. 1998; 12:210-213 [CrossRef][Medline]
4. Wennerström M, Hansson S, Jodal U, Stokland E Primary and acquired renal scarring in boys and girls with urinary tract infection. J Pediatr. 2000; 136:30-34 [CrossRef][Medline]
5. Esbjörner E, Aronson S, Berg U, Jodal U, Linne T Children with chronic renal failure in Sweden 1978-1985. Pediatr Nephrol. 1990; 4:249-252 [CrossRef][Medline]
6. Esbjörner E, Berg U, Hansson S Epidemiology of chronic renal failure in children: a report from Sweden, 1986-1994. Pediatr Nephrol. 1997; 11:438-442 [CrossRef][Medline]
7. Helin I, Winberg J Chronic renal failure in Swedish children. Acta Paediatr (Scand). 1980; 69:607-611

To the Editor.

I was intrigued by the article in the April 2000 issue of Pediatrics entitled "Circumcision and Urinary Tract Infections During the First Year of Life" and its accompanying editorial. I have another study in which you may be interested entitled "Do Moustaches Cause Staphylococcal Pharyngitis: (Implications for UTIs and Circumcision).

Abstract. 1000 teenage males with fever had throat cultures performed; 500 had moustaches and 500 did not. Culture technique included swabbing the throat, first running the swab through the moustache (if present).

Results. Statistically significant differences in staphylococcal recovery rates were seen in those with moustaches, with a 20-fold increase in culture-positive "pharyngitis" seen in those with moustaches.

Conclusion. Moustaches cause a drain on our medical expenses, causing unnecessary episodes of fever and potentially more serious sequelae.

This farce is simply to illustrate the point that the presence of the foreskin can make statistical analysis of culture results difficult, if these cultures are derived using a technique potentially contaminated by the very foreskin implicated in the conclusions. I believe that good evidence exists to support the claim that the foreskin allows bacterial adherence, and so the concentration of bacteria locally will be higher in the region compared with a circumcised individual. Going through this region to obtain cultures used to make a diagnosis of urinary tract infection (UTI) is an obvious complicating factor in any study looking at potential infectious consequences of the presence of the foreskin.

I would be very interested indeed in reviewing a study of UTI and circumcision in which all of the culture data are collected via suprapubic aspiration, and in which the presence of inflammatory changes in the urinalysis (white cells, protein) are used in conjunction with culture results to define a UTI. I have not been able to find a large study that satisfies these criteria. Realistically, to state that the foreskin harbors more bacteria than a circumcised penis, and then to go through the region for culture collection, is asking for technical difficulties in data interpretation.

Mark F. Mangano, MD
Radnor, PA 19087

In Reply.

Dr Newman addresses the problematic semantics of the 1999 AAP Circumcision Task Force Statement. If similar wording were applied to routine health care ("not essential to a child's current well being"), virtually all preventive medicine strategies would be eliminated (immunizations, etc). Although there are ample data detailing the short-term consequences (bacteremia, renal scarring, etc) of urinary tract infection (UTI), long-term data concerning the subsequent development of hypertension and end-stage renal disease (ESRD) are scant. Published information varies considerably. Unfortunately, there are no large, comprehensive, prospectively accumulated databases that denote the incidence of ESRD after UTI. Although Newman quotes only 2.7% of ESRD in children to be attributable to the adverse effects of such infections, others suggest that as much as 22% to 43% of ESRD may be attributable to pyelonephritis.^1-3 Similarly, although Newman states that among adults aged 20 to 45 years with ESRD the proportion attributed to childhood infection is only 0.7%, others put the number as high as 13% to 21%.^1,4

Dr Mangano intimates that all studies assessing UTI in uncircumcised males have erroneously demonstrated infections when there were none present. He speculates that urine culture specimens obtained via bladder catheterization are contaminated by uropathogenic organisms. His speculation is erroneous. The latter method of urine collection is reliable in both circumcised and uncircumcised boys. Isolation of multiple organisms, as well as colony counts <10 000 to 50 000 colony-forming units/mL, are typical manifestations of contaminated urine cultures. These were not the findings in the 15 studies I referenced in my commentary. I remind Dr Mangano that when one obtains urine via bladder catheterization in uncircumcised boys, it is not obtained "through" the prepuce. Rather, the distal glans and urethral meatus are initially gently exposed and aseptically cleaned.⁵ A sterile catheter is then directly inserted into the urethral meatus, with care taken not to touch surrounding tissue. Apparently, Dr Mangano is unfamiliar with the largest investigation to date in which suprapubic bladder aspiration was the preferred method of urine collection.⁶ In that study, of 468 uncircumcised boys with UTIs, 92% of the urine culture specimens were obtained by suprapubic aspiration. Additionally, as most pediatricians recognize, urinalysis findings are notoriously unreliable in helping to substantiate UTIs in infants. The "50-50 rule" is a well-known mnemonic in this regard: 50% of young children with UTIs do not have increased numbers of white blood cells in their urine, while 50% of those with elevated white blood cell levels in their urine do not have UTIs.

Robson and Van Howe assert that ESRD is an exceedingly rare consequence of UTI that pediatricians should ignore. Unfortunately, they choose to reference only a few selected publications concerning childhood chronic renal failure (CRF), while they ignore the substantial literature attributing adult CRF to childhood UTI. Interestingly, in the very references they do provide, Sreenarasimhaiah and Hellerstein⁷ note that the severe consequences of UTI (hypertension and ESRD) are late events that often manifest 20 years or more after the initial UTI. Helin and Winberg⁸ similarly acknowledge the role of childhood UTI as the precursor of hypertension and CRF in adults. Additionally, Wennerström and colleagues⁹ comment that in most parts of the world, the detection rate of UTI in small children is low (ie, few such infections are diagnosed and treated), while the long-term consequences of these infections are considerable. In the recent publication of Craig and colleagues,¹⁰ their review of the world literature suggests that pyelonephritis is the second leading cause of ESRD in childhood, with approximately 20% to 25% of ESRD cases being attributable to infection.

Esbjörner et al¹¹ describe the remarkable differences in ESRD attibutable to pyelonephritis in Sweden compared with that of multiple other countries. The International Reflux Study in Children in the 1980s revealed that 3 times as many boys with CRF attributed to UTI were enrolled from Europe compared with the United States (J. Duckett, personal communication). I know of no reason for this difference other than the higher circumcision rates in the United States and the 10- to 20-fold decreased risk for UTI among these boys. The major prerequisites for renal scarring are infection and inflammation, not reflux. Wennerström et al's data⁹ are from individuals with UTIs that occurred more than 20 years ago, long before the widespread use of nuclear scanning to evaluate such children. The latter techniques are several times more likely to reveal renal scarring than were the intravenous pyelography studies described by these authors. Thus, Robson and Van Howe mislead the readers by using the Wennerström data to infer that most renal scarring in males is congenital in nature. Furthermore, they misrepresent my commentary, stating that it is a continuation of efforts using the risk of ESRD to justify routine neonatal circumcision. I have never suggested ESRD should be the basis for routine neonatal circumcision. I have only discussed it as one of many potential sequelae of childhood UTI.

The developing kidney of early childhood is much more likely to be injured as a consequence of UTI compared with later in life. The current published data¹² assessing the incidence of long-term consequences of childhood UTIs are worrisome, noting relatively high rates of altered renal function, hypertension, and ESRD. What is clearly needed are large prospective trials in which cohorts of children with UTIs are followed well into adulthood to assess for sequelae of such infections.

Thomas E. Wiswell, MD
Department of Pediatrics
SUNY Stony Brook
Stony Brook, NY 11794-8111

REFERENCES

[Medline]
1. Chantler C. Kidney diseases in children. In: Schrier RW, Gottschalk CW, eds. Diseases of the Kidney. 5th ed. Boston, MA: Little, Brown and Company; 1993:2394-2403
2. Batisky D. Pediatric urinary tract infections. Pediatr Ann. 1996;25:266, 269-272, 274, 276
3. Altemeier WA. A backward look at urinary tract infections. Pediatr Ann. 1996;25:255-256, 268, 280
4. Shaw KN, Gorelick MH Urinary tract infection in the pediatric patient. Pediatr Clin North Am. 1999; 46:1111-1124 [CrossRef][Medline]
5. Sanchez JL, Walsh RF Procedures in the evaluation of the febrile pediatric patient. Pediatr Ann. 1996; 25:686-692 [Medline]
6. Wiswell TE, Roscelli JD Corroborative evidence for the decreased incidence of urinary tract infections in circumcised male infants. Pediatrics. 1986; 78:96-99 [Abstract/Free Full Text]
7. Sreenarasimhaiah S, Hellerstein S Urinary tract infections per se do not cause end-stage kidney disease. Pediatr Nephrol. 1998; 12:210-213
8. Helin I, Winberg J Chronic renal failure in Swedish children. Acta Paediatr Scand. 1980; 69:607-611
9. Wennerström M, Hansson S, Jodal U, Stokiland E Primary and acquired renal scarring in boys and girls with urinary tract infection. J Pediatr. 2000; 136:30-34
10. Craig JC, Irwig LM, Knight JF, Roy LP Does treatment of vesicoureteric reflux in childhood prevent end-stage renal disease attributable to reflux nephropathy? Pediatrics. 2000; 105:1236-1241 [Abstract/Free Full Text]
11. Esbjörner E, Aronson S, Berg U, Jodal U, Linne T Children with chronic renal failure in Sweden, 1978-1985. Pediatr Nephrol. 1990; 4:249-252
12. Wiswell TE The prepuce, urinary tract infections, and the consequences. Pediatrics. 2000; 105:860-862