Trajectories of post-surgical pain in infants admitted to neonatal intensive care

Prolonged pain in premature neonates hospitalised in neonatal intensive care units: A scoping review

BACKGROUND

Exposure to repetitive pain during the neonatal period has been shown to have important short and long-term effects on the neurodevelopment of the premature neonate and can contribute to experienced prolonged pain. A uniform taxonomy of neonatal prolonged pain is still lacking to this day which contributes to suboptimal prolonged pain management in neonatal intensive care units. Accordingly, a scoping review exploring the state of knowledge about prolonged pain in preterm neonates hospitalised in the neonatal intensive care unit will contribute to the developing field of neonatal prolonged pain and provide recommendations for clinical prolonged pain management.

OBJECTIVE

To determine the scope, extent, and nature of the available literature on prolonged pain in premature neonates hospitalised in neonatal intensive care units.

DESIGN

Scoping review.

METHODS

An electronic search was conducted from inception to November 2023 in the databases of CINAHL, PubMed, Medline, Web of Science, GeryLit.org and Grey Source Index. Included studies discussed concepts related to neonatal prolonged pain such as definitions of prolonged pain, indicators of prolonged pain, contexts that result in prolonged pain, prolonged pain evaluation tools, consequences of prolonged pain and interventions for prolonged pain management.

RESULTS

Key concepts of neonatal prolonged pain were identified in the 86 included articles of this scoping review such as definitions (n = 26), indicators (n = 39), contexts (n = 49), scales (n = 56), consequences of prolonged pain (n = 30) and possible interventions for prolonged pain management (n = 22). Whilst a consensus on a definition has yet to be achieved, no proximate event was shown to cause prolonged pain and a time criterion was identified by authors as being relevant in defining prolonged pain. Interestingly, the context of hospitalisation was identified as being the most indicative of prolonged pain in premature neonates and should guide its evaluation and management, whilst only limited pain management interventions and consequences were discussed.

CONCLUSION

The findings of this scoping review contribute to the foundation of growing knowledge in neonatal prolonged pain and shed light on the ambiguity that currently exists on this topic in the scientific literature. This review summarises knowledge of key concepts necessary for a better understanding of prolonged pain and stresses the importance of considering contexts of hospitalisation for prolonged pain evaluation and management in neonatal intensive care units, with the objective of improving developmental outcomes of premature neonates.

TWEETABLE ABSTRACT

A scoping review reveals that the contexts of prolonged pain in premature neonates hospitalised in the neonatal intensive care unit are essential in guiding its evaluation and management.

Performance Evaluation of a Supervised Machine Learning Pain Classification Model Developed by Neonatal Nurses

BACKGROUND

Early-life pain is associated with adverse neurodevelopmental consequences; and current pain assessment practices are discontinuous, inconsistent, and highly dependent on nurses' availability. Furthermore, facial expressions in commonly used pain assessment tools are not associated with brain-based evidence of pain.

PURPOSE

To develop and validate a machine learning (ML) model to classify pain.

METHODS

In this retrospective validation study, using a human-centered design for Embedded Machine Learning Solutions approach and the Neonatal Facial Coding System (NFCS), 6 experienced neonatal intensive care unit (NICU) nurses labeled data from randomly assigned iCOPEvid (infant Classification Of Pain Expression video) sequences of 49 neonates undergoing heel lance. NFCS is the only observational pain assessment tool associated with brain-based evidence of pain. A standard 70% training and 30% testing split of the data was used to train and test several ML models. NICU nurses' interrater reliability was evaluated, and NICU nurses' area under the receiver operating characteristic curve (AUC) was compared with the ML models' AUC.

RESULTS

Nurses weighted mean interrater reliability was 68% (63%-79%) for NFCS tasks, 77.7% (74%-83%) for pain intensity, and 48.6% (15%-59%) for frame and 78.4% (64%-100%) for video pain classification, with AUC of 0.68. The best performing ML model had 97.7% precision, 98% accuracy, 98.5% recall, and AUC of 0.98.

IMPLICATIONS FOR PRACTICE AND RESEARCH

The pain classification ML model AUC far exceeded that of NICU nurses for identifying neonatal pain. These findings will inform the development of a continuous, unbiased, brain-based, nurse-in-the-loop Pain Recognition Automated Monitoring System (PRAMS) for neonates and infants.

What is the definition of acute episodic and chronic pain in critically ill neonates and infants? A global, four-stage consensus and validation study

OBJECTIVES

To define and validate types of pain in critically ill neonates and infants by researchers and clinicians working in the neonatal intensive care unit (NICU) and high dependency unit (HDU).

DESIGN

A qualitative descriptive mixed-methods design.

PROCEDURE/S

Each stage of the study was built on and confirmed the previous stages. Stage 1 was an expert panel to develop definitions; stage 2 was a different expert panel made up of neonatal clinicians to propose clinical characteristics associated with the definitions from stage 1; stage 3 was a focus group of neonatal clinicians to provide clinical case scenarios associated with each definition and clinical characteristics; and stage 4 was a survey administered to neonatal clinicians internationally to test the validity of the definitions using the clinical case scenarios.

RESULTS

In stage 1, the panel (n=10) developed consensus definitions for acute episodic pain and chronic pain in neonates and infants. In stage 2, a panel (n=8) established clinical characteristics that may be associated with each definition. In stage 3, a focus group (n=11) created clinical case scenarios of neonates and infants with acute episodic pain, chronic pain and no pain using the definitions and clinical characteristics. In stage 4, the survey (n=182) revealed that the definitions allowed an excellent level of discrimination between case scenarios that described neonates and infants with acute episodic pain and chronic pain (area under the receiver operating characteristic=0.87 and 0.89, respectively).

CONCLUSIONS

This four-stage study enabled the development of consensus-based and clinically valid definitions of acute episodic pain and chronic pain. There is a need to define and validate other pain types to inform a taxonomy of pain experienced by neonates and infants in the NICU and HDU.

Classification of Individual Pain Response Trajectories Following Medically Indicated Heel Lances in Preterm Infants During Their NICU Admission

OBJECTIVES

Infants born preterm are exposed to repeated painful procedures during neonatal intensive care unit admission. Particularly in preterm infants, trajectories of pain response are not well understood. The aim of this study was to classify pain response trajectories over 2 minute following medically indicated heel lances in preterm infants.

MATERIALS AND METHODS

This study used existing clinical trial data (NCT01561547) that evaluated the efficacy of kangaroo care and sucrose for infant pain control. Pain was measured using the Premature Infant Pain Profile at 30, 60, 90, and 120 seconds following a heel lance. Group-based trajectory modeling was used to classify pain response in this 2 minute period.

RESULTS

A total of 236 infants with median gestational age of 33 weeks contributed 610 procedures. A model with 5 trajectory classes best fit the data. Three trajectories were stable over time at different levels of intensity from low-mild to low-moderate pain. One trajectory reflected a linear reduction from high-moderate to low-moderate pain. The final trajectory showed variable moderate-high pain. At all times points, 3 classes were at least 1-point different from the overall sample mean pain score. Only 21 (9%) infants maintained the same class for all 3 procedures.

DISCUSSION

In this sample of preterm infants receiving pain relief, most pain trajectories reflected mild to low-moderate pain that was stable over 2 minute after heel lance initiation. Trajectories were not consistent over multiple procedures within infants, and an overall mean pain score for the sample may misrepresent subgroups of pain response.