To address the current limitations related to infant pain assessment in the NICU, our international team of knowledge users and health/natural science/engineering/social science researchers have come together to build a machine learning algorithm that will learn how to discriminate invasive and non-invasive distress. Furthermore, to improve the use of current pain management practices, our team seeks to better understand the developmental mechanisms underlying skin-to-skin contact over time and factors that may influence its efficacy in mitigating pain responses in preterm infants. This is an ongoing naturalistic observational study.
Unmanaged pain in hospitalized infants has serious long-term complications. Existing infant pain assessment approaches demonstrate several key flaws (e.g., dependent on human cognitive capacity to simply combine data from multiple indicators into a total pain score, none have passed a critical discriminant validity test, bias introduced from human caregivers). Thus, the complexity of preterm pain assessment necessitates a machine learning approach. Our international team of knowledge users and health/natural science/engineering/social science researchers have come together to build a machine learning algorithm that will learn how to discriminate invasive and non-invasive distress. Furthermore, better understanding how skin-to-skin contact works in caregiver-infant dyads and factors that influence the effectiveness of this pain management strategy is a critical step in improving infant pain care in NICUs. Relatedly, the design and sample of our current study (acute pain paradigm while infant is either in skin-to-skin contact with the birthing parent or in the cot) allows us to not only test the influence of skin-to-skin contact vs. cot on preterm newborn pain responding, but also interrogate potential mechanisms underlying the effectiveness of skin-to-skin contact (i.e., cardiac regulation attunement between caregivers and their infants during the procedure; influence of birthing parent perceived stress given the particularly elevated stress levels of NICU parents). A sample of 400 preterm infants (300 from Mount Sinai Hospital and 100 from University College London Hospital \[UCLH\]) and their birthing parents (if available) will be followed during a routine painful procedure (heel lance). Pain indicators (facial grimacing \[behavioural indicators\], heart rate, respiration rate, oxygen saturation levels \[physiologic indicators\], brain electrical activity) during the painful procedure will be used to train the algorithm to discriminate between different types of distress (pain-related and non-pain related). Heart rate and respiration rate, as well as maternal-reported perceived stress levels, will be collected from the birthing parent to examine factors impacting the effectiveness of skin-to-skin contact.
Study Type
OBSERVATIONAL
Enrollment
400
Mount Sinai Hospital
Toronto, Ontario, Canada
RECRUITINGUniversity College London Hospital
London, No Province, United Kingdom
RECRUITINGBehavioural Correlate of Distress
To be analyzed using machine learning via bedside videography: Facial Grimacing using Neonatal Facial Coding System (NFCS-P subset; Bucsea et al., 2022, 10.1097/j.pain.0000000000002798)
Time frame: NFCS-P coded in 1-5 minute epochs, over 2 hour surrounding painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance)
Cortical Correlate of Distress
To be analyzed using machine learning via bedside monitoring: Continuous EEG data capture
Time frame: For 2 hours surrounding painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance)
Cardiac Correlates of Distress
To be analyzed using machine learning via bedside monitoring: Heart Rate, Heart Rate Variability
Time frame: Over 2 hours surrounding painful procedure (time locked to heel lance)
Oxygen Saturation Correlate of Distress
To be analyzed using machine learning via bedside monitoring: amount of oxygen-carrying hemoglobin in the blood relative to the amount of hemoglobin not carrying oxygen
Time frame: Over 2 hours surrounding painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance)
Respiration Rate Correlate of Distress
To be analyzed using machine learning via bedside monitoring: Respiratory patterns
Time frame: [Time Frame: Over 2 hours surrounding painful procedure (time locked to heel lance; approximately 1 hour before to 1 hour after heel lance)]
Birthing Parent Heart Rate and Respiration Rate
We will measure parent cardiac stress in order to control for the effect of the parent context on infant pain processing. We will collect ECG data with 3 leads and a thin belt for respiration from the CNS monitor.
Time frame: [Time Frame: Over 1.5 hours surrounding painful procedure (time locked to heel lance)]
Chart review of general health indicators and other NICU experiences (control as potential confounders), such as:
* Gestational age at birth, post menstrual age at study, and postnatal age * Birthweight * Current weight * Sex * Birth order * Type of delivery and any resuscitation required * Appearance, Pulse, Grimace, Activity, and Respiration (APGAR) scores at birth (1 and 5 minutes) * SNAP II scores * Ventilation requirements * Type and time of last feed * What indwelling devices are present during the study * Area or location of study * Diagnoses (active problems in the last 72h and any diagnosis that have now resolved \[\>72h\]) * Antenatal history * Brain imaging (cranial ultrasound) * Acute procedure history (type, per day, number of attempts) * Amount of skin to skin per day * Hours since last skin-breaking procedure \& number of skin-breaking procedures in the 48-hours preceding the heel lance * Total doses of sucrose \& sucrose at time of study * Drug classes at time of heel lance
Time frame: Extracted from participants' medical charts following study completion
Chart Review of Cumulative Painful Procedures
We will collect data on the cumulative exposure of painful and/or invasive procedures that the infant experienced since birth. A published tool will be used to calculate total pain burden during their NICU stay (Laudiano-Dray et al., 2020; 10.1097/j.pain.0000000000001814).
Time frame: Extracted from participants' medical charts following study completion
Parent Self-Report Questionnaires
Four questionnaires will be completed to understand the perspectives and experiences of birthing parents at time of study: 1. Participant Information Sheet (PIS) - quantifies socio-economic status, ethnic background, and familial characteristics of the sample. 2. Parental Stress Scale: NICU (PSS:NICU) - measures NICU specific stress 3. Experiences in Close Relationships Scale (ECR) - measures adult attachment patterns 4. Perceived Stress Scale (PSS) - measures general stress levels
Time frame: Completed at the beginning of the study while EEG electrodes are being applied to the baby
Detailed documentation of noxious and sham stimulation details: (Control for their influence on infant responses)
* Time of lance/sham * Stimulation site * Infant position * Exact monitoring measurements taken or failed * Sleep/wake states * Any cuts visible on the heels (including the other heel not lanced) * Maternal position * Degree of skin contact * Blood test required for the lance and volume * Whether heel was squeezed during lance and duration (if applicable)
Time frame: Documented following the study completion
Semi-Structured Interview
Health Professionals and Caregivers will be asked about their thoughts on using AI for infant pain assessment
Time frame: These interviews are occurring at the beginning of the study and will be qualitatively analyzed. They are not linked to infants whose data we are collecting primary outcomes.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.