I am writing in response to a recent article that appeared in the journal Pediatrics that purports to establish the possibility of a causal relationship between the practice of oral suctioning of blood (metzizah bepeh) as part of a religious circumcision and neonatal Herpes infection. I am a Board Certified Infectious-Disease specialist practicing in New York City and Westchester County. I have been in practice since 1987. I am currently the Chief of Infectious-Disease at New York Westchester Square Hospital and the former Chief of Infectious- Disease at White Plains Hospital Medical Center.

 First, it should be known that Herpes simplex infection involving the lips, cold sores, was first described by the Roman physician, Herodotus, in the year 100 A.D. He named it Herpes Febrilis. The virus was grown in a laboratory in 1925. Metzizah bepeh, as part of the circumcision procedure, was first described in the Talmud, which was compiled in Babylonia in the sixth century. Most of the laws described in the Talmud had been practiced for hundreds of years prior to that time. It is likely that Herpes simplex infection and metzizah bepeh have coexisted for the past 2000 years! Until this article appeared in Pediatrics in August of 2004, there had never been an observation of any relationship between neonatal Herpes infection and metzizah bepeh. In fact, there has never been an observation of transmission of any virus in this procedure, including HIV, hepatitis A, hepatitis B, hepatitis C, Epstein-Barr virus, and Cytomegalovirus.

Despite the perfect record of 2000 years, if one were concerned and wished to investigate the possibility of transmission of Herpes simplex through metzizah bepeh, one could do so in several ways. The best approach would be to follow a cohort of babies born. One would examine them daily from birth to be sure that there was no rash prior to the circumcision and then conduct an epidemiologic study of all those infants who developed neonatal Herpes (Herpes infection in children less than six weeks old). One would study all of the contacts of these babies, including parents, caretakers, siblings, other relatives, and friends. One would investigate what type of contact these individuals had with the infant. One would look into whether these babies had circumcision, whether metzizah bepeh was part of the procedure, and then study the individual who did the circumcision, whether it be a physician or mohel. All contacts would be studied in a uniform manner, using antibody testing and PCR testing of saliva. Ultimately, the way to link anybody to the infant with neonatal Herpes would be through DNA analysis of the virus.

 A weaker method to investigate this possibility would be to study retrospectively all cases of neonatal Herpes in a certain area or institution. One would then conduct the same detailed epidemiologic analysis of all the contacts of this infant. The deficiency in this method is that one would not be sure as to exactly when the infection began. If the diagnosis was made after a circumcision, one could not of be sure that the illness did not develop beforehand. Again, DNA matching would be the way to link any contact to the infant.

 The least effective method would be to study retrospectively the case histories of all infants in a certain time period who were discharged from the hospital with a diagnosis of neonatal Herpes. This kind of information can be obtained from the Information Services department of hospitals. In this kind of study, the epidemiologic information would be far more difficult to obtain, as the infants would have to be tracked down long after their discharge from the hospital.

 This study does not use any of these methods. In fact, it is not clear at all how this study was conducted. It is stated in the "Methods" section, "the 8 cases were collected from personal communication and the experience of the authors from 1997 to 2003. Clinical data from all patients were collected, and follow-up was conducted during hospitalization and after the discharge of the infants from the hospital."

 How were these eight cases collected? Were there any other cases of neonatal Herpes during this time? How does "personal communication" differ from "the experience of the authors." Were the data collected retrospectively or prospectively? None of this is clear from the study.

 In all other studies analyzing transmission of viruses from one human to another, the method has been to prove by DNA analysis that the virus is shared by the two individuals. Examples of this can be found in other viruses. For example, in the 1980s, there was an outbreak of HIV associated with one dentist in Florida. The Center for Disease Control (CDC) in the United States determined through DNA analysis that the virus was shared by the dentist and a group of patients who were infected. There was a case of possible transmission of HIV from an orthopedic surgeon to a patient in France published in the Annals of Internal Medicine in 1999 . Again, the relationship was established with DNA sequence analysis. There was a case of hepatitis B transmission from a thoracic-surgery resident to multiple patients in 1992 reported in the New England Journal of Medicine in 1996. Likewise, the CDC found that the DNA was identical in the hepatitis virus of the surgeon and in that of the patients who acquired it from him. This is the method of establishing transmission of viruses.

 In the literature of transmission of Herpes virus infection, this also has been the method of establishing transmission. There was a case reported in the Journal of Pediatrics in 1983 of a neonate who acquired HSV-1 from his father. In this case, the father developed a large lesion on his lip soon after his son developed HSV-1 infection. Through DNA analysis, it was determined that the virus acquired by the son was identical to that of his father.

 In all of the studies, another essential piece was to exclude the possibility that the virus could have been contracted in some other way. This was done through detailed epidemiologic analysis.

 In this study, there is absolutely no evidence linking the DNA of Herpes virus in any mohel to any infant. In fact, there is no live virus obtained from any mohel. The authors admit that the mouth cultures obtained from mohels were all negative for HSV. Despite the author's own observation in the introduction that, "postnatal transmission usually results from non-genital infection of a caregiver, including parent or nursery personnel with oral lesions," there is also no discussion at all as to what other individuals came into contact with these infants and how it was determined that they could not have transmitted the virus to these infants.

 A simple way to link the infants to the mohels was available in the case of the two infants who were circumcised by the same mohel five weeks apart. As it appears from the Methods section that clinical data were collected from the patients during hospitalization, there should have been a comparison of the DNA of the Herpes virus of the two infants. If the viruses matched, and it was determined that the mohel was the only person who came into contact with both infants, even in the absence of Herpes virus beeing cultured from the mohel, one would have to suspect seriously that the mohel was the source. It would not be conclusive evidence; as in every other study I described, DNA was obtained from the person suspected of transmitting the virus. However, it would be worrisome. Why was this not done? If it was done, why was it not mentioned?

 With regard to sero-positivity, the Results section of the study states, "most infants and their mothers were sero-negative for HSV; only patient 7 and his mother were sero-positive with a titer of 1:16." It does not state how many of the mothers and infants were actually tested. It states that, " four mohels could be tested, and they were found to be sero-positive for HSV." It does not state whether they were sero-positive for HSV-1 or HSV-2. If they were seropositive for HSV-2, it would obviously have no significance. In any case, the authors fail to mention that 90% of adults are sero-positive for HSV.

 Several other points of the study need to be addressed. In describing the procedure of metzizah bepeh, the authors state in the Introduction, "this procedure was repeated several times until bleeding stopped." I have interviewed several mohels. The actual process is to do one instant of suctioning accomplished so quickly that eyewitnesses looking to observe are often unable to even see that it was done.

 Also in the Introduction, the authors state, "the ancient procedure of metzizah, also carries a risk of infection, and currently most mohels use an appropriate suction device, such as a mucous extractor." This is stated as a fact without any citation. There is no information anywhere that this procedure carries a risk of infection.

 In the Discussion, the authors state, "because shedding of HSV-1 in the saliva of both the symptomatic and asymptomatic individuals has been documented repeatedly, the act of metzizah represents a potential source of oral genital transmission...". In fact, little is known about shedding of HSV-1 in the saliva of asymptomatic individuals. In a recent review of HSV shedding, (online publication called antiviral research www.sciencedirect.com volume 6351) the authors state, "there are few [case] data on the transmission of HSV-1 from patients who were sero-positive for HSV-1 but have a negative history of herpes labialis."

 The authors also stated in their Discussion, "there is, however, the possibility that some previous cases were not reported for cultural reasons." This comment is pure speculation and is of no value.

 The authors conclude that "our findings provide evidence that ritual Jewish circumcision with oral metzizah may cause oral-genital transmission of HSV infection." No evidence at all has been provided. The authors also state, "furthermore, oral suction may not only endanger the child but also may expose the mohel to human immunodeficiency virus or hepatitis B from infected infants." There is absolutely no data to establish such transmission.

 We have a history of 2000 years of metzizah bepeh and Herpes infections running parallel courses but never crossing paths. Thousands and thousands of babies over the years have had metzizah bepeh without a single case of documented transmission of herpes virus or any other virus. In thinking of infections, one would worry that there might be a higher incidence of postcircumcision wound infections caused by bacteria from the mouth of the mohel doing metzizah bepeh. No such observation has ever been made.

 Why is it that viral transmission doesn't occur? Why don't bacterial infections happen? First, as I stated before, we do not know what the incidence of shedding is in an adult who never has had a reactivation of Herpes infection. It may be nonexistent. It also may be that even if such an individual were to shed, the virus that he has may be weaker in quality and less in quantity and thus unable to transmit virus. It may be the explanation as to why such a person never developed reactivation in the form of cold sores. Furthermore, it is known that saliva is not an efficient way of transmitting virus. Saliva is known to have components that weaken viruses. In fact, HIV, a highly contagious virus, has never been shown to be transmitted through saliva. An individual has never been shown to transmit HIV to another individual with intimate kissing. It also may be possible that by putting wine in the mouth, the virus, if it were to be present, would be further diluted. Finally, the time of contact between the mohel and the baby is so brief that the possibility of transmission is further reduced. Some of these factors would also explain why no bacterial infections have been associated with this procedure.

 In summary, metzizah bepeh has a perfect safety record for 2000 years with regard to bacterial infections and viral infections. This study is a feeble attempt to taint this record.

 I hope this information is of value to you.

 Respectfully yours, Daniel S. Berman, M.D.,F.A.C.P.