Aim of Investigation: Pain from invasive or noninvasive procedures, mechanical ventilation, or painful medical and surgical conditions is commonplace in neonatal intensive care units (NICUs). While prevention and treatment of neonatal pain seem essential, an adequate analgesic approach cannot be implemented without relevant and timely pain assessments. Data on neonatal pain assessment practices are scarce, with undefined best practices or clinical benefits. We aimed to describe pain assessment practices in 243 NICUs from 18 European countries and to examine the NICU and patient characteristics influencing pain assessments at the bedside.
Methods: Demographic data, modes of respiration, use of sedation, analgesia, or neuromuscular blockers, frequency and types of pain assessments were recorded for all newborns during the first 28 days of NICU admission. Multivariable models tested the associations between the performance of pain assessments and center and neonatal factors.
Results: Among 6648 neonates enrolled, highest level of ventilation during the study period classified patients into tracheal ventilation (TV, n=2138 [32%]), non-invasive ventilation (NIV, n=1493 [23%]), and spontaneous ventilation groups (SV, n=3017 [45%]). Pain assessments were performed in 1250 (58%), 672 (45%), and 916 (30%) of these groups respectively (p<0.001). Using data from 78,742 patient-days, we found that 2,838 (43%) neonates received 4.3 (5.2) pain assessments per neonate and per day (median (IQR): 2.4 (1-5)), whereas 3810 (57%) neonates did not receive any pain assessments. Pain assessments occurred on every day of the NICU stay in 461/2138 (22%) TV patients, 236/1493 (16%) NIV patients, and 393/3016 (13%) SV patients (p<0.001).Many different pain assessment methods were used; the EDIN scale was used most frequently (42.3% among those who had at least one pain assessment). We analysed 33,625 patient-days in the TV group to test for associations between pain assessment and the use of opioids, sedatives-hypnotics, or general anaesthetics (O-SH-GA). The rates of pain assessments on patient-days with and without O-SH-GA use were, respectively, 57% vs. 43% while receiving mechanical ventilation, and 60% vs. 34% while not receiving mechanical ventilation (both p<0.001). Multivariable analyses showed that NICU-based guidelines, nursing leadership, and increased surgical admissions promoted the use of routine pain assessments (p<0.001). More pain assessments were performed in newborns below 32-weeks gestational age, those with decreased severity of illness, those already intubated at admission, those requiring mechanical or non-invasive ventilation, or surgery, or use of O-SH-GA.
Conclusion: Even though pain is considered the 5th vital sign, only 43% of NICU neonates received bedside pain assessments. Clinical practice variability and low rates of pain assessments in NICUS may reflect weaknesses in the current paradigm used for neonatal pain assessments, their subjectivity, lack of inter-rater reliability, and other long-standing concerns. Results suggest that training to improve the rate of pain assessment in NICUs will enhance pain management in NICUs.
Trial Registration: ClinicalTrials.gov #NCT01694745