Timing for maximum anaesthetic effect of topical cream during early infant circumcision (EIC) in Rakai, Uganda

Abstract Objectives The objective of this study is to determine the optimal timing for device‐based infant circumcision under topical anaesthesia. Subjects/patients We include infants aged 1–60 days who were enrolled in a field study of the no‐flip ShangRing device at four hospitals in the Rakai region of south‐central Uganda, between 5 February 2020 and 27 October 2020. Methods Two hundred infants, aged 0–60 days, were enrolled, and EMLA cream was applied on the foreskin and entire penile shaft. The anaesthetic effect was assessed every 5 min by gentle application of artery forceps at the tip of the foreskin, starting at 10 min post‐application until 60 min, the recommended time to start circumcision. The response was measured using the Neonatal Infant Pain Scale (NIPS). We determined the onset and duration of anaesthesia (defined as <20% of infants with NIPS score >4) and maximum anaesthesia (defined as <20% of infants with NIPS score >2). Results Overall, NIPS scores decreased to a minimum and reversed before the recommended 60 min. Baseline response varied with age, with minimal response among infants aged 40 days. Overall, anaesthesia was achieved after at least 25 min and lasted 20–30 min. Maximum anaesthesia was achieved after at least 30 min (except among those aged >45 days where it was not achieved) and lasted up to 10 min. Conclusion The optimal timing for maximum topical anaesthesia occurred before the recommended 60 min of waiting time. A shorter waiting time and speed may be efficient for mass device‐based circumcision.


| INTRODUCTION
Male circumcision (MC) was recommended by the World Health Organization (WHO) and the Joint United Nations Program on HIV/AIDS (UNAIDS) as part of combination HIV prevention interventions in countries with low MC prevalence and high HIV burden 1 after three circumcision trials demonstrated 60% effectiveness in reducing female to HIV transmission. [2][3][4] National HIV programs and donor agencies have primarily targeted MC scale-up adolescents and adults for quicker impact 5 and focussed much less on infants. However, infant circumcision is technically more accessible and cheaper, healing is faster, there is no risk of early sex resumption, and it may be more sustainable in the long term. [6][7][8] Early infant circumcision (EIC) can be performed by non-physician health workers using devices and topical anaesthetics, 9,10 removing the need for an injection. When using topical anaesthetics, the WHO recommends circumcision 60 min after application of the anaesthetic. 11 However, there are reports of incomplete pain control during infant circumcision, 12 and there is minimal literature on the optimal timing of the circumcision after the administration of anaesthesia.
Also, although infant pain can be measured using standard tools such as the Neonatal Infant Pain Scale (NIPS), 13 pain responses can be variable and little is known about what determines the level of pain sensitivity or pain-like reactions in infants.
We assessed the optimal timing of infant circumcision after administration of topical anaesthesia. The success of infant circumcision programs is largely dependent on maintaining both the ease of the procedure and adequate pain control.

| Study design and setting
We conducted a short observational substudy during a noncomparative field study of no-flip ShangRing infant circumcisions performed by non-physician providers in routine clinical settings. This field study followed a clinical trial that comparing the safety and acceptability of infant circumcisions performed by non-physician health workers using the no-flip ShangRing device versus the Mogen clamp (clinicaltrials.gov identifier NCT03338699). The field study was simultaneously conducted at three sites at Iringa Regional Referral Hospital in Tanzania

| Selection of participants
We included all eligible infants whose parents or legally acceptable representatives (LAR) provided informed consent for their infants to participate in the field study at the four facilities in Rakai. Eligibility included healthy infant boys from 24 h up to 60 days of age, born after ≥ 37 weeks of gestation, weighing ≥ 2.5 kg, with no penile abnormality, no family history of bleeding disorders, and a history of maternal or infant Tetanus vaccination. Infants were enrolled between 5 February 2020 and 27 October 2020.

| Primary outcome
The primary outcome of interest was the NIPS score measured at 5-min intervals before circumcision, that is, between 10 min after application of a topical analgesic cream containing lidocaine and prilocaine (Eutectic Mixture of Local anaesthetics) and before performing the circumcision with the ShangRing at 60 min. Sensitivity to pain was tested by gentle application of toothed artery forceps at the foreskin.
The pain was measured using the NIPS scale, 13 a standard tool commonly used for pain measurement in infants. The NIPS scale is a behavioural scale composed of six indicators of infant pain or distress.
These include facial expression, cry, breathing patterns, arms, legs, and state of arousal. Each is scored 0 or 1, except cry, which may be scored 0, 1, or 2. A total score between 0 and 2 is considered mild pain to no pain, a score of 3-4 as mild to moderate pain, and 4-7 as severe pain (Table 1). We also retrospectively obtained and appended the NIPS scores during circumcision to the NIPS scores before circumcision. For example, intraoperative measurements at time points between 60 and 65 min after application of topical anaesthesia were assigned to the 65th minute, and so on (each infant had one intraoperative measurement). The circumcisions lasted a median of 11 min (IQR 10-13).

| Explanatory variables
The main explanatory variable was time after application of topical anaesthetic cream. We also captured infant age to assess differential pain responses/pain experience by age, if any and the recruiting health facility.

| Analysis
Infant characteristics were described using proportional distributions for categorical variables. We assessed for baseline variations in pain responses with respect to age using a generalized additive model to determine the need for stratification by age in case of a non-linear relationship. Baseline measurements were the first two NIPS scores at 10 and 20 min after the topical cream application. After observing non-linear variations (Figure 1), we grouped infants into those aged 0-15 days, 16-30 days, 31-45 days, and 46-60 days. In each age group, we calculated and plotted the prevalence and 95% confidence interval of pain (NIPS score > 2) and the prevalence and confidence interval of severe pain (NIPS score > 4) at each 5-min interval before and during circumcision (all infants included had a measurement at each of these time points).
We then determined the time points at which anaesthesia was achieved (defined as less than 20% of infants with NIPS scores > 4; the confidence interval also entirely falling below 20% prevalence). We also determined the time points at which maximum anaesthesia was achieved (defined as less than 20% of infants with NIPS scores > 2; the confidence interval also entirely falling below 20% prevalence, which we considered as 'virtual elimination' of pain). In secondary analyses, we assessed the temporal variation in median NIPS scores, which are not sensitive to outliers if present.

| Variations in baseline response by age
We observed a nonlinear relationship with a bimodal distribution of NIPS scores taken between 10 and 20 min after EMLA application, a peak at about 30 days and a minimum at about 40 days of age ( Figure 1). F I G U R E 2 Proportions with NIPS score > 2 or NIPS score > 4 following application of topical anaesthetic cream, all ages combined. The error bars represent 95% confidence intervals.

| Onset and duration of maximum anaesthesia
Overall (all ages combined), NIPS scores > 2 were detectable in a few infants (<20%) after at least 30 min following topical anaesthetic application, with maximum anaesthesia lasting <25 min (Figure 2).

| Variation in median NIPS scores (secondary analyses)
Similar patterns to those reported above were observed, with minimum scores recorded before 60 min after topical cream application.
F I G U R E 3 Proportions with NIPS score > 4 following application of topical anaesthetic cream, stratified by age. The error bars represent 95% confidence intervals.
F I G U R E 4 Proportions with NIPS score > 2 following application of topical anaesthetic cream, stratified by age. The error bars represent 95% confidence intervals.
While several studies have measured pain control in infants undergoing circumcision, 12  The bimodal distribution of baseline responses to pain stimuli (i.e., before anaesthesia takes effect) was an unexpected finding. A transition seems to happen around 40 days of age that results in relatively lower pain scores around that time, which subsequently reverses. We hypothesize this may be due to changes in blood supply to the foreskin and its thickness, resulting in slower wearing off of the anaesthesia, which subsequently reverses.
In the analyses stratified by age, unlike in pooled analyses, the numbers are smaller therefore we are underpowered to conclusively detect pain in at least 20% of infants. However, the patterns observed in these age stratifications could be confirmed in future studies with larger sample sizes.
To our knowledge, this is the first study to report on the variability of pain responses in infants with respect to age and time after applying a topical anaesthetic and to suggest an earlier timing and optimal window of circumcision after application of topical anaesthetics during infant circumcision. The hypotheses generated, including actual mechanisms, may be explored in future studies.