Morphine exposure and prematurity affect flash visual evoked potentials in preterm infants

The present study aimed to explore first the impact of perinatal risk factors on flash-VEP waves and morphology in a group of preterm infants studied at term equivalent age (TEA). Second, to correlate VEP morphology with neurological outcome at 2 years corrected age (CA).

Infants with a gestational age (GA) at birth <32 weeks, without major brain injury, were enrolled. Multivariate regression analyses were performed, and the models were run separately for each dependent variable N2, P2, N3 latencies and P2 amplitude. Logistic regression was applied to study N4 component (present/absent) and VEP morphology (regular/irregular). The predictors were GA, bronchopulmonary dysplasia (BPD), postmenstrual age at VEP registration, cumulative morphine and fentanyl dose, and painful procedures. Lastly, linear regression models were performed to assess the relation between the Bayley-III cognitive and motor scores at 2 years CA and VEP morphology, in relation to GA, BPD, painful procedures and cumulative morphine dose.

Eighty infants were enrolled. Morphine was the predictor of N2 (R² = 0.09, p = 0.006), P2 (R² = 0.11, p = 0.002), and N3 (R² = 0.13, p = 0.003) latencies. Younger GA was associated with lower amplitude (R² = 0.05, p = 0.029). None of the independent variables predicted the presence of N4 component, nor VEP morphology in the logistic analysis. VEP morphology was not associated with cognitive and motor scores at 2 years.

Morphine treatment and prematurity were risk factors for altered VEPs parameters at TEA. In our cohort VEP morphology did not predict neurological outcome.

Morphine administration should be evaluated according to potential risks and benefits, and dosage individually accustomed, according to pain and comfort scores, considering the possible risk for neurodevelopmental impairment.