Sucrose for analgesia in newborn infants undergoing painful procedures

Management of pain for neonates is less than optimal. The administration of sucrose with and without non-nutritive sucking (pacifiers) has been the most frequently studied non-pharmacological intervention for relief of procedural pain in neonates.

To determine the efficacy, effect of dose, and safety of sucrose for relieving procedural pain as assessed by validated individual pain indicators and composite pain scores.

Standard methods as per the Neonatal Collaborative Review Group. A MEDLINE search was carried out for relevant randomized controlled trials (RCTs) published from January 1966 - April 2001, EMBASE from 1993-2001 and search of the Cochrane Library, Issue 2, April 2001. Key words and (MeSH) terms included infant/newborn, pain, analgesia and sucrose. Language restrictions were not mposed. Bibliographies, personal files, the most recent relevant neonatal and pain journals and recent major pediatric pain conference proceedings were searched manually. Unpublished studies, or studies reported only as abstracts, were not included. Additional information from published studies was obtained.

RCTs in which term and/or preterm neonates (postnatal age maximum of 28 days after reaching 40 weeks corrected gestational age) received sucrose via oral syringe, NG-tube, dropper or pacifier for procedural pain from heel lance or venepuncture. In the control group, water, pacifier or positioning/containing were used. Studies in which the painful stimulus was circumcision were excluded.

Trial quality was assessed according to the methods of the Neonatal Collaborative Review Group. Quality measures included blinding of randomization, blinding of intervention, completeness of follow up and blinding of outcome measurement. Data were abstracted and independently checked for accuracy by the three investigators. Statistical Analysis The statistical package (RevMan 4.1) of the Cochrane Collaboration was used. For meta-analysis, a weighted mean difference (WMD) with 95% confidence intervals (CI) using the fixed effects model was reported for continuous outcome measures.

Thirty-eight studies were identified for possible inclusion in this review. Seven studies reported only as abstracts, and fourteen additional studies were excluded, leaving 17 studies included in this review. Sucrose in a wide variety of dosages was generally found to decrease physiologic (heart rate) and behavioural (the mean percent time crying, total cry duration, duration of first cry, and facial action) pain indicators and composite pain scores in neonates undergoing heel stick or venepuncture. When pain scores (Premature Infant Pain Profiles) were pooled across 3 studies (Gibbins 2001, Johnston 1999, Stevens 1999), they were significantly reduced in infants who were given sucrose (dose range 0.012 g to 0.12 g) compared to the control group, [WMD -1.64 (95% CI -2.47,- 0.81); p = 0.0001] at 30 seconds and [WMD -2.05, (95% CI -3.08, -1.02); p = 0.00010] at 60 seconds after heel lance. When results for change in heart rate were pooled for two studies (Haouari 1995, Isik 2000), there were no significant differences between changes in heart rate for infants given sucrose (dose range 0.5 g to 0.6 g) compared to the control group, [WMD 0.90 (95% CI -5.81, 7.61); p = 0.8] at one minute and [WMD -6.20 (95% CI -15.27, 2.88); p = 0.18] at three minutes after heel lance.

Sucrose is safe and effective for reducing procedural pain from single painful events (heel lance, venepuncture). There was inconsistency in the dose of sucrose that was effective (dose range of 0.012 g to 0.12 g), and therefore an optimal dose to be used in preterm and/or term infants could not be identified. The use of repeated administrations of sucrose in neonates needs to be investigated as does the use of sucrose in combination with other behavioural (e.g., facilitated tucking, kangaroo care) and pharmacologic (e.g., morphine, fentanyl) interventions. Use of sucrose in neonates who are of very low birth weight, unstable and/or ventilated also needs to be addressed.