Effectiveness and tolerability of pharmacologic and combined interventions for reducing injection pain during routine childhood immunizations: Systematic review and meta-analyses
Immunization is the most common cause of iatrogenic pain in childhood. Despite the availability of various analgesics to manage vaccine injection pain, they have not been incorporated into clinical practice. To date, no systematic review has been published on the effectiveness of pharmacologic and combined interventions for reducing injection pain. The objectives of this article were to assess the effectiveness and tolerability of various pharmacologic and combined interventions for reducing the pain experienced by children during immunization. MEDLINE, EMBASE, CINAHL, and the Cochrane Central Register of Controlled Trials were searched to identify randomized controlled trials (RCTs) and quasi-RCTs pertaining to pharmacologic and combined interventions to reduce injection pain in children 0 to 18 years of age using validated child self-reported pain or observer-reported assessments of child pain and distress. We included trials that (1) investigated the effects of pharmacologic interventions (ie, topical local anesthetics, sweet-tasting solutions, vapocoolants, and oral analgesics [acetaminophen or ibuprofen]); (2) compared 2 different analgesic interventions; and (3) evaluated combinations of >or= 2 analgesic interventions, including breastfeeding. Meta-analyses were performed using a fixed-effects model. Thirty-two studies, involving 3856 infants and children 2 weeks to 15 years of age, were included in this systematic review; 23 of these trials were included in meta-analyses. Ten trials, including 1156 infants and children, evaluated topical local anesthetics. In a meta-analysis of 2 trials, including 276 children, child self-reported pain ratings were lower in children who received topical local anesthetics than in those who received a placebo. The standardized mean difference (SMD) was -0.25 (95% CI, -0.49 to -0.01; P = 0.04). The use of topical local anesthetics was associated with less pain than was placebo in 4 trials (527 infants) based on the difference between Modified Behavioral Pain Scale scores (range, 0-10) before and after vaccination: the weighted mean difference (WMD) was -0.79 (95% CI, -1.10 to -0.48; P < 0.001) and the SMD was -0.43 (95% CI, -0.60 to -0.26; P = 0.001). Observer-rated pain, using visual analog scale (VAS) scores (range, 0-100 mm), was significantly lower (WMD, -16.56 mm; 95% CI, -22.11 to -11.01; P < 0.001; and SMD, -0.75; 95% CI, -1.00 to -0.49; P < 0.001). The number needed to treat (NNT) to prevent 1 child from having clinically significant pain, measured using the Faces Pain Scale (FPS; score, >-3), was 3.7 (95% CI, 2.5 to 7.7) from 1 study. Eleven trials (1452 infants and children) evaluated sweet-tasting solutions. In a meta-analysis of 6 studies (665 infants), administration of sucrose with or without non-nutritive sucking (NNS; use of a pacifier) was associated with less pain than no intervention or sterile water with or without NNS; the SMD was -0.56 (95% CI, -0.72 to -0.40; P < 0.001). Total cry duration was lower in infants who received sucrose than in those who received sterile water (WMD, -9.41 sec; 95% CI, -13.18 to -5.64; P < 0.001; and SMD, -0.43; 95% CI, -0.61 to -0.25; P < 0.001). The NNT to prevent 1 child from having clinically significant pain, using the Neonatal Infant Pain Scale (score, >3), was 1.4 (95% CI, 1.0 to 2.5). In 3 trials that evaluated sweet-tasting solutions longitudinally, administration of sucrose or glucose (vs sterile water, with or without NNS) was associated with reduced pain based on cry duration or the University of Wisconsin Children's Hospital Pain Scale (all, P < 0.05). Data were pooled for 2 studies conducted in 100 children who received a spray with a vapocoolant or placebo at the injection site before the procedure. Child self-rated pain (4-point scale) was lower in the group treated with the vapo-coolant (SMD, -0.43; 95% CI, -0.83 to -0.02; P = 0.04); significant heterogeneity was reported for this outcome (chi(2) = 5.51; P = 0.02; I(2) = 82%). In 2 studies (117 children), no significant difference was found between vapocoolants and typical care (no treatment) based on child self-reports; significant heterogeneity was reported for this outcome (chi(2) = 9.89; P = 0.02; I(2) = 90%). None of the studies identified in the literature search evaluated oral analgesics (acetaminophen or ibuprofen). Four studies (318 infants and children) compared 2 different analgesic interventions; there was insufficient evidence to suggest superiority of 1 intervention over another. Combinations of >or=2 analgesic interventions were more effective than the individual interventions used alone. Child self-reported pain ratings were combined for 4 studies (350 children); the SMD was -0.52 (95% CI, -0.73 to -0.30; P = 0.001). Data on cry duration were pooled for 3 studies (229 infants and children); the WMD was -18.87 seconds (95% CI, -32.05 to -5.69; P = 0.005). Parent-rated child pain (VAS) scores were combined for 3 studies (365 infants and children); the WMD was -15.66 mm (95% CI, -19.74 to -11.57; P < 0.001). Nurse- or physician-rated child pain (VAS) scores were combined for 3 studies (368 infants and children); the WMD was -17.85 mm (95% CI, -21.43 to -14.28; P < 0.001). In a meta-analysis of 4 studies (474 infants), infants who were breastfed before, during, and after the procedure had less pain than did those who were not breastfed (SMD, -2.03; 95% CI, -2.26 to -1.80; P < 0.001). A meta-analysis of 3 studies (344 infants) found a shorter cry duration for infants who were breastfed than for those who were not breastfed (WMD, -38.00 sec; 95% CI, -42.27 to -33.73; P < 0.001; and SMD, -2.00; 95% CI, -2.27 to -1.73; P < 0.001). The NNT to prevent 1 infant from having clinically significant pain, using the Facial Pain Rating Scale (pain vs no pain), was 7.7 (95% CI, 4.5 to 25.0) from 1 study. Topical local anesthetics, sweet-tasting solutions, and combined analgesic interventions, including breastfeeding, were associated with reduced pain during childhood immunizations and should be recommended for use in clinical practice.