This was a hospital based randomised double blind controlled study, conducted at the paediatric surgery unit of the hospital, from March 2015 to December 2015. It involved sequential recruitment of neonates that presented for circumcision at the clinic, and whose parents/guardians gave consent. Ethical approval was obtained from the hospital health research ethical board (NAUTH/CS/66/Vol.5/82).

An average of 230 circumcision procedures is done in infants per year in the centre. Fisher’s formula was used to determine the sample size of 121. Using the formula for a population less than 10,000:

Where, nf=the desired sample size (when population is less than 10,000)

n=sample size for population greater than 10,000=(Z²pq)/d²

N=estimate of population size=number of neonates who had circumcision at the centre yearly=233

z=the standard normal deviate, set at 1.96, which corresponds to the 95% confidence level.

p=proportion of male neonates in Nigeria (target population). Here, 84% was used as this is the prevalence for male neonates who underwent circumcision from previous studies [22-24]. Therefore the available prevalence=0.84

q=1-p (1-0.84)=0.16

d=degree of accuracy desired, 0.05

n=206.52

=(206.52)/{1+(206.52/233)}

=109.85

Making provision for attrition rate of 10% which is 10.99

=109.85+10.99

=120.84

=121

All term male neonates were included in the study. Excluded from the study were neonates delivered pre-term, with penile lesions like rashes/ulcers and congenital penile anomalies like hypospadias, or epispadias. Every week, recruited neonates [Table/Fig-1] were randomised into two groups, A and B by a simple ballot method, picking from a bag containing equal amount of tags A and B. Each was assigned to the group of its tag. One assistant was responsible for the balloting and another for administering the medications. These two personnel did not participate in the actual procedure and data recording. The group and medications were filled into the proforma at the end of the procedure.

[Table/Fig-1]:

Consort flowchart.

The neonates in group A received 10 mg/kg body weight of oral ketamine (10 mg/mL) [16]. The neonates in group B received a placebo, simple sucrose (66.7% w/w) at 1 mL/kg of body weight. This ensured that neonates of similar weight received same volume of “syrup”. The oral ketamine was prepared by the compounding section of the hospital pharmacy using simple sucrose, as a base and sweetener, without a preservative. The operating room was made conducive for the neonates by having the air-conditioner switched off and the cleansing solutions warmed to reduce confounding effect on parameters of outcome measures. There was no restriction on feeding.

Two senior registrars from the anaesthesia department of the hospital, who were conversant with the study, provided Monitored Anaesthesia Care (MAC). Part of their duty was to observe for the known side-effects of ketamine that may be life-threatening such as apnea, laryngeal spasm, and aspiration. They had facilities and instruments to manage such side-effects whenever the need arise. Such facilities included but not limited to: anaesthetic machine, piped oxygen, laryngoscope, various sizes of endotracheal tubes, patient monitor, suction machine, atropine, adrenaline, suxamethonium, pancuronium etc. The circumcision procedure was carried out 30 minutes after oral medication administration. The surgeon, parent or caregiver, the anaesthetists and the assistant that recorded the details were all blinded to the type of oral medication received by the neonates.

The neonates were placed on continuous pulse oximetry (MMED, MD300A, China; 12110/000189; 2012) monitoring before, during and after the procedure. The baseline SpO₂ and PR were recorded before the onset of the procedure. The SpO₂, PR and NIPS score were recorded at the end of adhesiolysis, dorsal slit, tying and excision. The neonates were also observed for complications such as laryngeal spasm, apnoea, aspiration, increased salivation, and vomiting. These were recorded on the proforma.

Data analysis was done using SPSS version 21. The statistical differences in the mean age, weight, pre-procedural SpO₂ and PR of both groups were compared using t-test for confounding factors. The significance of the difference in the mean SpO₂ and mean PR at the stages of the procedure for the different syrups were analysed using Mann-Whitney U test. Independent t-test was used to assess the significance of the difference in mean NIPS score of both groups. Fisher’s exact test was used to compare categorical variables.

A total of 121 subjects were recruited in this study, after 220 were assessed: 61 in group A and 60 in group B [Table/Fig-1]. The age and weight distributions of the neonates are shown in [Table/Fig-2].

The mean intra-procedural SpO₂ for the group A and group B were 95.5% and 95.6 (adhesiolysis); 95.3% and 94.4% (dorsal slit); 94.6% and 93.5% (tying); and 95.9% and 94.8% (excision), respectively. There were no significant differences in the mean SpO₂ levels between the two groups during the adhesiolysis, dorsal slit and excision stages of the procedure. However, the mean SpO₂ levels were significantly lower in the placebo group during the tying stage (p=0.022). The mean PR values were significantly higher in the placebo group during the dorsal slit, tying and excision stages of the operation (p-values were <0.001, <0.001 and <0.001, respectively). There was, however, no difference in the PR of the two groups during the adhesiolysis stage (p=0.077) [Table/Fig-3].

The mean NIPS were significantly higher in the placebo group compared to the oral ketamine group in all stages of the procedure (p-values <0.001, respectively) [Table/Fig-4].

There was no significant difference in the occurrence of side-effects between the two groups. The side-effects were all self-limiting, and neonates with excessive salivation were cleaned and re-positioned with satisfactory results [Table/Fig-5].

The study revealed that orally administered ketamine produced a significant intraoperative analgesia during neonatal circumcision when compared to a placebo. Majority of the procedures were carried out in neonates around the 15^th day of life. This was on account of the traditional timing of neonatal circumcision in the environment. The traditional thinking was that the timing of the procedures is best from the 8^th day and very importantly after the umbilical cord has fallen off [25]. This also was reflected in the average weight of the neonates of 3.67 kg and 3.65 kg. When the initial reduction in weight that occurs in the newborn, due to fluid shift is taken into consideration, it puts these values into a proper perspective.

With an average SpO₂ and PRs above 96% and 140 for both groups [Table/Fig-2], the neonates were fairly stable and comparable for the procedure. In this study, neonates that received oral ketamine had a significantly higher SpO₂ during the tying stage of the circumcision. This shows that oral ketamine provided significant analgesia for the procedure. This is significant considering that the tying is the most painful stage. Pain reduces the blood arterial oxygen saturation which is reflected in the SpO₂ measurements [10]. The precise mechanism for the reduction of blood oxygen saturation by pain may be due to breath hold, hypoventilation, among others during crying. The analgesic effect provided by oral ketamine during the plastibell circumcision ensured that the blood oxygen saturation is higher in the neonates of group A (oral ketamine), than of group B (placebo). It means less stress and therefore safety for those neonates that received oral ketamine [10,12].

The mean PRs were lower at all the stages of the procedure in the oral ketamine group relative to the placebo group. These were significant in the dorsal slit, tying and excision stages. Pain increases the heart rate and hence PR. This suggests that the neonates that received oral ketamine syrup felt less pain relative to the neonates who received placebo. It suggests also that oral ketamine provided analgesia during the procedure. Pain is a stressor that stimulates increased cardiac activity, among other metabolic responses, with its attendant demands on the body. Neonates, that received oral ketamine mount less metabolic response to trauma and are therefore, safer during and after the procedure [9-12].

PRs readings at the 3 stages of dorsal slit, tying, and excision significantly reflected the analgesic properties of oral ketamine over placebo, while the SpO₂ reading did so in only stage (tying). The differences in the measurements of SpO₂ and PR could be attributed to the workings of the pulse oximeter. There are variations in sensitivity of pulse oximeter. The manufacturers alluded that it may take a maximum of five seconds for a change in SpO₂ reading to reflect. These may mean that PR readings are more sensitivity than SpO₂ within the procedure. This is coupled with the fact that there were no observed strict time-intervals between the stages in order to take readings. All these may have compromised the accuracy and sensitivity of the SpO₂ readings.

The NIPS scores were quite informative. The mean NIPS score for the oral ketamine group were less than the placebo group in all the stages and were statistically significant in all. This portrays the capacity of oral ketamine to provide analgesia. The sedative effect of oral ketamine may also have contributed to the low NIPS score of the group. The maximum score of the scale is 7. The mean NIPS score for the placebo ranged from 4.8 to 4.9, approaching the maximum score of 5. The mean NIPS score for the oral ketamine group in all stages ranged from 3.7 to 4.4. The implication was that oral ketamine besides its analgesic effect appeal to the sense of the “modern mother” and doctor who do not want the neonates to cry out in pain while undergoing the procedure.

Oral ketamine has been shown to provide a measure of analgesia for those in acute and chronic pain [16,17,19,21]. Sareenmma E et al., found that ketamine has no effect in attenuating pain associated with suctioning in newborn infants on ventilator [21]. They used heart rate, arterial blood pressure and pain scoring systems of Children’s Hospital of Eastern Ontario Pain Scale (CHEOPS) and NIPS. However, they used intravenous ketamine in doses ranging from 0.5 mg/kg to 2 mg/kg. Ketamine administered orally is converted to nor-ketamine (four-fifth) by the GIT, with minimal unaltered ketamine (16%) entering the system. The nor-ketamine is thought to produce the major analgesic effect of ketamine. Parenterally administered ketamine will produce lesser analgesic effect for procedural pain, compared to orally administered ketamine. Ogboli-Nwasor E et al., found that oral ketamine is very effective in the emergency room as analgesia for fracture manipulations [19]. This was similar to the findings in another study by Buvanendran A et al., [26]. The study by Ogboli-Nwasor E et al., was among adults and soda water was used as an elixir to sweeten the ketamine [19]. This study was on neonates and simple sucrose was used to make ketamine syrup.

In this study, seven out of 61 neonates in the oral ketamine group had complications, while in the placebo group one out of 60 neonates had complications. This was statistically not significant. The complications noted in the oral ketamine group were increased salivation, nonbilous vomiting and hiccough. These were handled by cleaning and repositioning. Studies on ketamine usage had recommended using anti-sialogogue [27]. This was not used in this study as ketamine administered orally delivers more of norketamine into the system. Moreover, the neonates were not anaesthetised and their swallowing reflexes were not altered, hence they were able to swallow their saliva [16]. In the placebo group, one patient presented with increased salivation during the procedure, and this was handled in a similar fashion like the oral ketamine group. The cause of increased salivation in this group was not clear, but may be from pain. These adverse effects did not continue in the immediate post-procedural period.

Neonates are conversant with the mother’s body and cuddling. When separated from such, they may cry or become restless. These may influence the outcome measures. Attempts were made to eliminate them. However, it was not possible to do so completely. Again the overlapping effect of the pains in one stage to the next stage could not be averted during the procedure. These were limitations of this study.

Oral ketamine provides effective and safe analgesia for neonatal Plastibell^® circumcision, in comparison with placebo. It is recommended for use in neonatal circumcision. It is suggested that studies be conducted for its use in other procedures and in other age groups. Further study is necessary to compare the efficacy of oral ketamine and other forms of procedural analgesia during neonatal circumcision in order to draw better conclusion and to properly place oral ketamine in its rank among other analgesic agents.