RESUMO
OBJECTIVES: To determine the agreement between HNNE and TIMP at TCA for preterm infants born <32+0 weeks' gestation, and to evaluate their correlation to PDMS-2 at 12-month corrected age (CA). METHODS: Infants born between November 2013 to June 2022 who had both HNNE and TIMP performed at TCA of 37+0-41+6 weeks gestation, and motor outcome assessed using the PDMS-2 at 12-month old were enrolled. The HNNE and 12-month PDMS-2 findings were categorized as optimal vs sub-optimal. TIMP was categorized as typical vs atypical. Cohen's kappa was used to determine the agreement between HNNE and TIMP. Sensitivity analysis and Receiver Operating Characteristic (ROC) curves were used to evaluate the predictive values of HNNE and TIMP on motor outcome at CA of 12-months. RESULTS: HNNE and TIMP done on 125 infants at TCA do not show reliable agreement. HNNE demonstrated slight and fair agreement with the 12-month Total Motor Quotient (TMQ) and Fine Motor Quotient (FMQ) of the PDMS-2 respectively. TIMP at TCA demonstrated fair agreement with all sub-domains of motor function on PDMS-2 at 12-months. In comparison with TIMP, HNNE at TCA is more sensitive at predicting suboptimal total, gross and fine motor outcomes at 12-month CA with sensitivity of 68.4 %, 51.9 %, and 83.3 % vs 44.4 %, 31.8 % and 53.3 % respectively. Atypical TIMP at TCA is more specific for suboptimal total, gross and fine motor outcomes at 12-month CA with specificity of 90.3 %, 89 % and 90.5 % respectively. Neurobehavioral assessments at TCA using HNNE and TIMP were predictive of suboptimal fine motor quotient at CA of 12-months with AUC of 0.760 (p = 0.011) and 0.718 (p = 0.032) respectively. The difference in AUC between the 2 instruments of 0.042 was not statistically significant (p = 0.741). CONCLUSIONS: While the HNNE and TIMP done at TCA did not demonstrate significant agreement, suboptimal HNNE and atypical TIMP at TCA were predictive of suboptimal FMQ on PDMS-2 at 12-month CA.
