Background: The effect of circumcision on feeding behavior in the newborn period is unknown. We hypothesized that circumcision would not have a significant effect on newborn feeding.
Methods: This prospective study analyzed the effect of circumcision on neonatal feeding behavior. Inclusion criteria were healthy male infants WHO were exclusively bottle-fed and underwent a circumcision before discharge from the newborn nursery. We collected data (N = 42) on gestational age, birth weight, Apgar scores, maternal age, gravid status, anesthesia used during delivery, analgesia used after circumcision, time of circumcision, and volume and frequency of feeding before and after circumcision. Data were analyzed by using paired t tests, multivariable regression analysis, and analysis of variance (with SPSS version 18). Significance was P < .05 (2-tailed α).
Results: Descriptive statistics for the entire group (N = 42) are as follows: mean ± SD gestational age: 38.7 ± 1.2 weeks; mean birth weight: 3.3 ± 0.4 kg; maternal age: 26.7 ± 6.3 years; baseline feeding (mean of first 2 feedings before circumcision): 24.5 ± 9.9 mL; mean first feeding after circumcision: 21.7 ± 11.9 mL; and mean second feeding: 26.7 ± 13.5 mL. Forty-eight percent of patients increased their feeding volume after circumcision compared with baseline, and 52% of patients decreased their feeding volume, which persisted with the second feeding. There was no statistical difference between the baseline and first feeding (P = .11) or second feeding (P = .22).
Conclusions: Our data suggest that circumcision does not alter feeding after circumcision. This information will be useful in counseling families regarding circumcision in the newborn period.
Literature exists that addresses both the positive and negative effects of circumcision. However, few studies have examined the specific effect of circumcision on neonatal feeding behavior. In the current literature, some evidence suggests that male circumcision has many health benefits, including protection against HIV, urinary tract infections, and penile cancer.1–7 Macke,8 however, reported that circumcision in the newborn period causes severe and persistent pain, often requiring analgesia for circumcision to prevent negative effects on mother–infant interaction. Anecdotal evidence suggests that circumcision before breastfeeding is associated with difficulty breastfeeding.9 Marshall et al10 studied 59 bottle-fed infants and found that the feeding behavior of the experimental group (circumcised on day 2) differed slightly from the control group (circumcised on day 3). However, the findings from this observational study were neither statistically significant (Wilcoxon P < .09) nor did they evaluate volumetric data.
Thus, there is a paucity of literature that measures the effect of circumcision on neonatal feeding behavior, and we are unaware of any literature that evaluates the feeding behavior on a volume-based scale. The purpose of our study was to evaluate the effects of male circumcision on neonatal feeding behavior by using volumetric data.
This was a prospective study that collected feeding data on healthy, bottle-fed males that were circumcised before discharge from the newborn nursery at CS Mott Children’s Hospital in Ann Arbor, Michigan, from June 2010 to April 2011. We elected to study newborns who were bottle-fed for the purposes of more accurate feeding volumes. Feeding volumes precircumcision and postcircumcision were compared. Exclusion criteria included breastfed male infants, female infants, uncircumcised male infants, infants with abnormal vital signs in the first 24 hours of life, and infants admitted to the NICU. This study was approved by the institutional review board at the University of Michigan Medical School (HUM00036627).
All newborn male patients were identified on the newborn daily patient census by using OB Tracevue, the electronic medical record (EMR) system for the newborn service. The newborns were identified by the resident assistant, who determined if the infant was bottle-fed or breastfed and subsequently e-mailed the study team with the names and medical record numbers of the newborn males who were bottle-fed. A study team member then determined eligibility of the patients based on the aforementioned formal exclusion criteria. During the 11-month study, ∼1880 male infants on the nursery census were screened for eligibility, and 1838 male infants were excluded due to breastfeeding. Written informed consent was obtained from the patient’s mother by a study team member. After obtaining consent, the newborn was entered (via name and medical record number) into our data collection file in Microsoft Excel (Microsoft Corporation, Redmond, Washington) and enrolled in the study. The following predictor variables were collected from the EMR: time of birth, birth weight, Apgar scores, gestational age, maternal age, delivery method, and record of any anesthesia given during delivery. After circumcision completion, we collected the date and time of circumcision, dose of 0.1% lidocaine (without epinephrine) administered for circumferential block, dose of acetaminophen administered, whether the patient received sucrose, type of clamp used, and description of any complications of the circumcision procedure. Date and time of patient discharge were also obtained from the EMR.
We determined that the average hours of life that males are circumcised at CS Mott Hospital is ∼19 hours, necessitating 36 to 48 hours of feeding data for most patients. Staff nurses record all newborn feedings each shift in the patient’s EMR; therefore, this source was used to retrieve feeding data on each newborn precircumcision and postcircumcision.
The means and frequencies of the predictor variables were determined. The primary outcome of this study was the difference in feeding volume measured before and after circumcision. The baseline was determined to be the average of the first 2 feedings before circumcision to reduce the variability of the baseline value. Baseline was compared with both the first and second feeding volumes after circumcision. The primary outcome was determined by using the paired t test. Mean feeding volume postcircumcision was categorized according to patients who increased or decreased feeding volumes postcircumcision. Lastly, a multivariable regression analysis was performed to analyze these 2 trends. The significance was P < .05 (2-tailed α).
We enrolled 39 full-term infants and 3 preterm infants; we did not exclude infants whose gestational age was between 35 and 37 weeks due to the limited availability of bottle-fed subjects. Therefore, 42 male infants who met the inclusion criteria were studied. Table 1 presents a description of the study group.
The baseline was determined as the average of the first and second feeding volumes before circumcision (24.5 ± 9.9 mL). The first and second feeding volumes after circumcision were compared with the baseline control. The first feeding volume was 21.7 ± 11.9 mL. The second feeding volume was 26.7 ± 13.5 (Fig 1). There was no significant difference between the first feeding volume and baseline (P = .11) or second feeding volume and baseline (P = .22). Paired t tests revealed that, compared with baseline, 48% of patients showed an increase in feeding volume for the first feeding after circumcision, persisting with the second feeding. Furthermore, 52% of patients showed a decrease in feeding volume for the first feeding after circumcision, with a slight increase in the second feeding (Fig 2).
A multivariable analysis was performed on all the variables listed in Table 1 to determine if any of the predictor variables influenced the volume of infant feedings postcircumcision. Variables categorized as dichotomous included maternal anesthesia, whether the patient received sucrose, and Gomco clamp use. Variables categorized as continuous included gestational age, birth weight, maternal age, Apgar scores, previous children, age at circumcision, dose of lidocaine, and dose of acetaminophen. No statistically significant relationship was found between these 2 trends.
Male neonatal circumcision is a common procedure, with limited research on all the risks and benefits. Our study results suggest that circumcision does not have a significant effect on neonatal feeding behavior. Our prospective study design is unique because we incorporated volume measurements of feedings before and after circumcision. There was no significant difference between baseline and the first or second feeding after circumcision (Fig 1). Furthermore, some infants had an increase in feeding after circumcision and others had a decrease. None of the collected variables correlated to either trend.
We did not exclude the 3 infants who were not given acetaminophen for pain management after circumcision due to sample size availability. Acetaminophen use was a variable we collected and included in the multivariable regression analysis, and it did not seem to affect whether a newborn initially had an increase or decrease in the volume of feeding postcircumcision.
The current study addresses a novel clinical question and provides a platform for future research, although it does have limitations. The first limitation is its small sample size due to the limited number of newborns bottle-fed at the time of hospital discharge. Although the overall rate of successful breastfeeding in the state of Michigan is low at 3, 6, 9, and 12 months of age, the vast majority of newborns evaluated for our study were being breastfeed at the time of discharge. According to the number of male infants excluded due to breastfeeding, the breastfeeding rate at our hospital was ∼98%. Therefore, the number of eligible patients for our study was limited to 42. Secondly, we are assuming that our results are generalizable to the bottle-fed population. Lastly, an a priori power analysis was not performed. A post hoc power analysis suggests that an n value of 17 will detect a 5-mL change in feeding volume at the 0.05 level (2-tailed test) with 80% power. The means used for the post hoc power analysis were estimated by using previously observed feeding volumes.
Our study suggests that male circumcision performed before discharge had no significant effect on neonatal feeding behavior among bottle-fed male infants. Further research is required to study the volumetric effect among breastfed male infants and to continue expanding our knowledge in this area.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- electronic medical record
- Baeten JM,
- Richardson BA,
- Lavreys L,
- et al
- Singh-Grewal D,
- Macdessi J,
- Craig J
- Schoen EJ
- Auvert B,
- Taljaard D,
- Lagarde E,
- Sobngwi-Tambekou J,
- Sitta R,
- Puren A
- Copyright © 2013 by the American Academy of Pediatrics