Abstract
Introduction: Every year, 15 million premature infants are born worldwide, and approximately 1 million die from complications of prematurity. Due to their physiological immaturity, many premature infants require specialized care in the NICU and are subjected to painful and stressful procedures that can increase RR, HR, BP, and metabolic rate, cause hypoglycemia, decrease SpO2, and alter facial expression (crying, sleep, and wakefulness). Intensivists have implemented humanization strategies to control pain during painful procedures, with pharmacological and nonpharmacological measures, including therapeutic touch, functional positioning, and nonnutritive sucking, among others. However, the evidence on these measures is limited, with low confidence, and it is unclear which is most effective. This study aimed to establish which non-pharmacological measures are most effective in controlling pain and stress in premature infants in the NICU. Methods: Systematic review with network meta-analysis of randomized clinical trials, published in English, Portuguese or Spanish, focusing on premature infants in the NICU, comparing pharmacological and non-pharmacological measures for pain and stress control. The search was carried out in the Medline/PubMed, Lilacs, Embase, Cochrane Library and PEDro databases, from April/2020 to April/2022, based on descriptors found in MeSH, defined by the acronym PICO(S): P (Population) premature; I (Intervention) non-pharmacological measures for pain and stress control; C (Comparison) sedative analgesia or no intervention; (Outcome) Decreased pain and stress (primary outcome), weight gain or improvement in behavioral parameters (secondary outcome), and S (study design) randomized clinical trials. The RoB2 and GRADE tools assessed the methodological quality and quality of evidence and strength of recommendation, respectively, of the included articles. Results: 210 studies were identified. After selection, 12 articles related to the primary outcome were included in the study. The measures used were: oral sucrose or glucose, heartbeat or music, odor/taste of breast milk, non-nutritive sucking with a pacifier, magnetic acupuncture, skin-to-skin contact, facilitated containment, sensory saturation, fentanyl, and proparacaine. The meta-analysis included 10 articles. The following were grouped: “measurements with sugars” (sucrose, glucose, and dextrose); “multiple therapies” sensory saturation (three or more measures together); and “control” (placebo, standard routine, and incubator). Multiple therapies were the combination of breast milk odor/taste, maternal heartbeat, and non-nutritive sucking with a pacifier. The most effective measure was “multiple therapies” (sensory saturation), followed by sugar in the pacifier and pacifier. The control presented the worst result for this outcome. Discussion and conclusions: Identifying effective non-pharmacological measures for controlling pain and stress in premature infants admitted to the NICU adds cost-effective therapeutic options, guides decision-making based on scientific evidence, and expands knowledge in the management of neonatal pain. The use of non-pharmacological measures has grown in the NICU. The first Brazilian recommendation for sensorimotor stimulation of newborns and infants in the NICU recommended sensory saturation (multimodal stimulation) to reduce pain and stress or improve the behavioral state of newborns with a high degree of certainty. Sugar is one of the most used measures to calm and reduce pain in newborns, including in the NICU. With moderate quality of evidence, it is strongly recommended that non-pharmacological measures be considered for the management of pain and stress in premature infants in the NICU, especially sensory saturation and sugars.
