Preterm birth is defined as birth occurring at age <37 weeks, with extremely preterm birth occurring at <26 weeks. (1) Children born prematurely have been shown to have significantly lower cognitive scores compared with children born at term by age 5 years, with scores being significantly positively correlated to both birthweight and gestational age. (2) In addition, preterm children are significantly more likely to develop attention deficit hyperactivity disorder and other behavioral problems at school age than children born at term.Complementary and alternative medicine therapies are used to help alleviate long-term consequences of preterm birth. In this article, we review commonly used complementary and alternative medicine therapies (long-chain polyunsaturated fatty acid [LCPUFA] supplementation, massage therapy, and music therapy) and their effects on the neurodevelopment of preterm infants.The following electronic bibliographic databases were searched: MEDLINE, EMBASE, Cochrane’s Database of Systematic Reviews and Register of Controlled Trials (CENTRAL), and PsycInfo. The search was limited to systematic reviews and clinical trials. Search terminology included neurodevelopment, premature infant, nutritional supplementation, massage therapy, music therapy, and their respective synonyms according to database. No language limits were imposed. Included studies were restricted to clinical studies.The n-3 LCPUFA, docosahexanoic acid (DHA), is a major brain lipid believed to play a role in brain development. It is thought that preterm infants may receive less DHA than full-term infants due to a shortened third trimester. (3)(4)A systematic review included 32 randomized controlled trials (RCTs) of which 10 studies were meta-analyzed. The RCTs examined the effect of enteral feeds of LCPUFA-supplemented formulas on neurodevelopment and diseases associated with prematurity in premature infants. (5) By using the Bayley Scales of Infant Development, which measures three components of neurodevelopment (cognitive, motor, and behavioral functioning), the authors found no significant differences between groups.Although supplemented infants scored an average of 3.4 points higher than controls in cognitive development (weighted mean difference [WMD]: 3.44 [95% confidence interval (CI): 0.56–6.31]; P = .02; n = 879), the authors concluded that this difference was mainly attributable to two trials with large effect sizes and wide CIs. Psychomotor development was significantly lower in supplemented infants than controls (WMD: −7.99 [95% CI: −14 to −1.99]; P = .009; n = 87).A Cochrane systematic review identified 19 RCTs that assessed the effects of LCPUFA formulas on the visual development, general development, and growth of preterm infants. (6) Pooled data from 15 RCTs did not indicate any long-term benefits of LCPUFA supplementation over control for any of these outcomes.An RCT, published after the two preceding systematic reviews, compared supplementation of high-dose DHA (∼1% total fatty acids) to standard-dose DHA (∼0.3% total fatty acids) in premature infants (<33 weeks). (7) Six hundred fifty-seven infants were randomly assigned to receive either treatment. If infants were breastfed, mothers were asked to take a supplement containing either DHA or placebo (because human milk contains a base level of DHA) to achieve human milk concentrations of 1% or 0.3% DHA, respectively. If infants were fed formula, they were given either high-dose or low-dose DHA formula to achieve the same DHA levels as above. Infants received the intervention two to four times per day until their expected delivery date.The Mental Developmental Index (MDI) of the Bayley Scales of Infant Development (BSID-II) at corrected age 18 months was chosen as the primary outcome. The authors did not find any difference in the MDI scores between groups (mean difference, 1.9 points [95% CI: −1.0 to 4.7]), but after stratifying by gender, the study revealed that girls on high dose supplementation had a significantly higher score than those on standard dose (adjusted mean difference, 4.5 [95% CI: 0.5–8.5]); no such difference was found among boys. They postulated that this difference between genders may be due to the higher rate of endogenous synthesis of DHA from its precursor fatty acid, α-linoleic acid, in girls over boys. There was no significant difference in adverse clinical outcomes between groups. This study provides evidence that gender may be a potential confounder and that higher doses of DHA supplementation may be beneficial for improving mental development, at least in girls. No harms were found with high-dose DHA supplementation.Underlying neurobiologic mechanisms of the effects of massage are poorly understood, but it is postulated to involve mechanisms that increase insulin, insulin growth factor, vagal activity, and gastric motility. (8)A Cochrane systematic review included 14 RCTs that compared systematic tactile stimulation (consisting of massage or “still gentle touch”) versus standard care in preterm (<37 weeks) and low birthweight (LBW; <2500 g or 5.5 lb) infants. (9) Primary outcomes of the review were weight gain and length of stay; developmental and behavioral outcomes were measured by using the Brazelton Scale or the Bayley scale.Infants who received massage experience gained clinically insignificantly more weight (WMD, 5.1 g [95% CI: 3.5–6.7]), had a 4.5-day (95% CI: 2.4–6.5) shorter length of stay in hospital, but showed unclear benefits in subscales of the Brazelton scale. Studies were found to be highly heterogeneous, and an analytic approach to account for such differences revealed that massage was significantly better than routine care in one subscale: motor maturity (WMD, 0.85 [95% CI: 0.17–1.53]). Methodologic concerns about blinding and selective reporting of some outcomes in included studies of this review may diminish the credibility of its findings.A controlled clinical trial (CCT) conducted in Russia evaluated the effects of massage given from age 2 months to 8 months. (10) The group exposed to massage were 40 LBW infants (<2.5 kg birthweight), 90% of whom were premature. The unexposed group was 40 LBW infants matched for age, gender, gestational age, birthweight, and geographic distribution. The Minnesota Infant Development Inventory was used to assess social, self-help, gross motor, fine motor, and language skills. Authors reported significant improvement in favor of massage in all five areas.To determine whether massage accelerates brain development, a CCT in Italy evaluated the effects of three daily massages lasting 15 minutes each over a period of 10 days in 10-day-old infants (±1 day). (11) Ten healthy preterm infants of gestational age 30 to 33 weeks received massage and were compared with 10 similarly aged, gender- and birthweight-matched infants. The authors studied electrophysiologic outcomes (EEG, flash visual evoked potentials [fVEPs], and brainstem auditory evoked potentials) as an index of brain maturation. Behavioral acuity was assessed by using the Bébé Vision Tropique acuity card procedure. Both massaged and control infants were exposed to classical music (Johannes Brahms’ Wiegenlied) diffused in the environment in the daytime.The authors found an increase in EEG maturation after 3 weeks, as evidenced by transition from discontinuous to continuous activity with a progressive reduction of the duration of intervals between bursts of activity. There was a significant reduction (Mann–Whitney U test P = .011; n = 10 for both groups; Cohen d = 1.33) in maximum interburst interval in massaged infants (7 seconds, interquartile range [IQR] = 5.3–9) compared with control infants (2.8 seconds, IQR = 2–3.8). N300, the most prominent peak of fVEPs, normally reveals progressive lengthening during the first months after birth. N300 latency measured before and after treatment revealed a significantly larger reduction in the massaged infants than in the control group (Mann–Whitney U test P = .013; n = 10 in both groups; Cohen d = 1.4). These differences were not evident at 7 and 12 months of observation. No acceleration was seen in the brainstem auditory evoked potentials. Visual acuity at 3 months’ postterm was significantly better in massaged infants than in controls (t test, P < .001, Cohen d = 2.2, n = 10 for both groups).A systematic review and meta-analysis evaluated the effects of music on physiologic parameters and length of hospital stay in premature infants. (12) Ten studies of unclear design were included, four of which were conducted by review authors. The authors report the overall size (d) for all measured outcomes between the music and nonmusic groups in all studies to be 0.8268 (95% CI: 0.68–0.97). Although this analysis demonstrates a large and significant effect size, methods of meta-analysis of the review were not described, nor is combining effects across outcomes a typical approach to meta-analysis. Because it is unclear how the authors arrived at their conclusions and effect size estimates and the combined treatment effects for each outcome were not presented, it is difficult to draw any valid conclusions from the findings.Only four of the seven studies (57%) included in this review article revealed adverse events; none revealed any increase in the number of adverse events in the intervention versus the control arm. In a narrative review, Heird (13) examined the safety of the use of PUFAs in neonates. Theoretical concerns would be lower birthweights, increased risk of oxidant damage (ie, necrotizing enterocolitis, bronchopulmonary dysplasia, and retrolental fibroplasia), altered eicosanoid metabolism, and possible gene transcription defects. However, there were no significant differences in the risk of adverse clinical outcomes between supplemented and control groups in data derived from two systematic reviews and a large multicenter trial involving a total of 4,555 infants. (5)(6)(7)Vickers et al (9) reported no adverse effects of touch or massage in their systematic review on massage. Kelmanson and Adulas (10) and Guzzetta et al (11) did not comment on adverse events of massage in their preterm subjects. A systematic review on the safety of massage in children revealed that, overall, massage therapy is safe. (14) The majority of adverse events were mild, including local pain or swelling, mild fevers, and skin rashes. The few moderate or severe adverse events reported involve people untrained in massage. Newborn abdominal massage was associated with seven cases of intestinal volvulus in preterm infants who received massage from allied health personnel; abdominal massage in premature newborns by untrained providers should be avoided until investigated further.Neal and Lindeke (15) review guidelines, controversies, and cautions on the use of music in neonates and suggest that auditory stimulation should be limited to infants older than 28 weeks’ gestation, with precautions addressing sound levels. There was no comment on the occurrence of adverse events in the studies in Standley’s (12) systematic review.The authors thank the invaluable assistance of Connie Winther and Soleil Surette in the literature search for this article.