604 - Are developmental screening evaluations at 3 months of age predictive of future diagnosis of cerebral palsy in former NICU infants?

Poster Number: 604
Publication Number: 604.310

Tara L. Johnson, Arkansas Children's Hospital, Little Rock, AR, United States; Namarta Kapil, University of Arkansas for Medical Sciences, North Little Rock, AR, United States; Sowmya Sivakumar, University of Arkansas for Medical Sciences College of Medicine, Little Rock, AR, United States; Beverly J. Spray, Arkansas Children's Hospital, North Little Rock, AR, United States; Bittu Majmudar-Sheth, Riley Hospital for Children at Indiana University Health, Fishers, IN, United States

Background: Infants in the NICU, such as those with prematurity and birth complications, have a greater risk of Cerebral Palsy (CP) than the general population. CP is generally diagnosed around two years of age. This prevents the majority of this population from harnessing the benefits of early medical intervention when their neuroplasticity is at its greatest. To remedy this, clinicians need reliable early clinical biomarkers to assist in early diagnosis of CP.

Objective: We evaluated the predictive validity of three developmental screening measures, (1) Hammersmith Infant Neurological Examination (HINE), (2) Prechtl’s General Movement Assessment (GMA), and (3) Capute Scales to predict CP in a high-risk population.

Design/Methods: We identified a cohort of infants who had an inpatient neurologic consultation in a level IV NICU from 5/2019-10/2020. Reasons for consultation included the following: < 37 week gestational age, grade 2-4 intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), and neonatal encephalopathy (HIE).
Patients were evaluated in clinic at 3 months of corrected age by a Pediatric Neurologist (BM), who performed HINE, GMA, and Capute Scales language (CLAMS) and visual-motor (CAT) evaluations. GMA was evaluated by three independent readers (BM, TJ, NK), and a consensus score was reported.
In 12/2021, we performed a nested case-control study on this cohort, defining cases as patients who had been diagnosed with CP and designating all other patients as controls. We compared demographics, risk factors, and early developmental evaluations between the groups. The sample was sufficiently powered to detect clinically significant differences in screening tool outcomes at a power of 80% and alpha < 0.05.

Results: Out of the 108 patients who were evaluated, 13 patients had CP, and 95 patients did not have CP. We compared demographic factors, known risk factors, and developmental screening outcomes between the two groups, with inference using two proportions (Table 1). Infants with prematurity, IVH, and PVL had an increased risk of CP. Infants with HIE had no increased risk of CP, likely due to protective effects of therapeutic hypothermia. Abnormal HINE, GMA, and CAT scores were predictive and highly specific for CP (Table 2).Conclusion(s): Given their high predictive value and specificity, GMA, HINE, and Capute Scales are good screening tests for CP and can play a pivotal role in early diagnosis of CP when implemented clinically during the early follow-up stages of these infants. Early diagnosis will lead to early medical interventions, which will lead to better quality of life.
Table 1. Comparison of demographics, risk factors, and developmental examination results between CP and control groups.There were no demographic differences in patients with and without CP. Patients with known risk factors were more likely to have CP than those who did not have those risk factors. Patients with poor outcomes on developmental screening examinations (HINE < 56, Abnormal GMA, CLAMS developmental quotient < 70, CAT developmental quotient < 70) were more likely to have CP than those who had good outcomes on developmental examinations.
Table 2. Comparison of developmental screening evaluations.
<img src=https://www.abstractscorecard.com/uploads/Tasks/upload/16020/FGOVBGGC-1177567-2-IMG(1).png width=440 hheight=119.96705107084 border=0 style=border-style: none;>The sensitivity and specificity in determining the CP diagnosis at 2 years of age was 33% and 96%, respectively, for abnormal GMA; 42% and 94% for HINE; 25% and 90% for CLAMS DQ; and 32% and 96% for CAT DQ. Positive predictive value and negative predictive value of these tests ranged from 23%-77%, and 78%-91% respectively. Overall, the screening tests have >90% specificity, which is a good marker for a screening test in this high-risk population.