Eleven studies (3 double-blind, placebo-controlled RCTs; 5 placebo-controlled RCTs; 1 RCT; 2 non-randomised controlled trials) including a total of 662 participants.
Circumcision (1 double-blind RCT and 2 RCTs; 138 participants): All three studies demonstrated that EMLA diminishes pain responses. Two studies were pooled to give a mean increase (WMD) in heart rate (compared to baseline values). The heart rate with EMLA varied as compared to placebo depending on the stage of the procedure from WMD=-11.67 (95% CI: -19.93, -3.42) for clamp removal to WMD=-27.21 (95% CI: -35.98, -18.45) when applying the clamp (P<0.05).
Heel lancing (1 double-blind RCT, 2 placebo controlled RCTs and non-randomised controlled study; 225 participants):
The four studies used various outcome measures to assess the level of pain (heart rate, blood pressure, crying, respiration rate, Premature Infant Pain Profile (PIPP), transcutaneous oxygen tension and carbon dioxide tension). EMLA was not shown to significantly diminish pain in any of the studies.
Venipuncture (1 double-blind RCT and 1 non-randomised controlled trial; 127 procedures):
Both studies demonstrated EMLA was effective at decreasing pain. In the RCT both heart rate and cry duration were lower in the EMLA as compared to the placebo group, however no data or significance levels were provided. In the cohort study significant changes in favour of EMLA (vs control) were observed for both heart rate and the behavioural pain score.
Arterial puncture (1 non-randomised controlled trial, 90 procedures): The study showed a higher frequency of low pain scores (41% vs 21.5%, P<0.05) for the EMLA vs the control group. Heart rate and oxygen saturation also favoured the EMLA group, however these findings were not significant (P>0.05).
Lumbar puncture (1 RCT, 49 participants):
The study showed physiologic parameters (heart rate, blood pressure, and oxygen saturation) and behavioural pain measures as compared to baseline values did not differ significantly between the EMLA and control groups. Therefore EMLA appeared to have no effect on pain. However, the nature of the behavioural pain score and the observed values were not reported.
PVC placement (1 RCT, 13 participants):
The heart rates of EMLA-treated neonates were significantly lower than controls and the respiration rate responses attenuated, but only during skin puncture. Blood pressure and oxygen saturation rates were not significantly altered. EMLA therefore appeared to have some efficacy in terms of decreasing pain.
Safety data (12 studies, >355 participants):
MetHb concentrations were compared in infants before and after administration of EMLA, or between infants exposed to EMLA and placebo, with no significant differences observed in seven studies. Data from four of the RCTs was combined and the results showed that mean MetHb concentrations did not differ between EMLA-treated and placebo-treated individuals (WMD=-0.11%, 95% CI: -0.31%, 0.10%). Overall, the studies suggested that the risk of methemoglobinemia is low after a single dose. In full-term neonates, single doses ranging from 0.5-2g applied for 10-180min were not been reported to cause methemoglobinemia. Concentrations of lidocaine and prilocaine were considerably lower than those considered toxic (>5micrograms/ml) using these dose regimens. There was insufficient data however to comment on the safety of repeated administration of EMLA.