Auditory event-related potentials and cognitive function of preterm children at five years of age

Abstract

Objective

In our previous study, auditory event-related potentials (AERPs) in preterm 1-year-old children had a positive deflection at 150–350 ms that correlated positively with their 2-year neurodevelopmental outcome. In a study of the same subjects at age 5, our aim was to assess AERPs and their relationship to neuropsychological test results.

Methods

Preterm small (SGA, n = 13), appropriate for gestational age (AGA, n = 15), and control (n = 13) children were assessed with an Easy paradigm presenting a large frequency change accompanied with occasional novel sounds, and a Challenging paradigm presenting small frequency and duration changes with a rapid rate. The preterm children underwent neurocognitive tests.

Results

Easy paradigm. The P1 response to frequency deviant was smaller and MMN larger in the preterm than in the control children. Challenging paradigm. The P1 response to standard, frequency, and duration deviants was smaller in the preterm than in the control children. The N2 response to frequency deviant was larger in the preterm than in the control children. AGA and SGA children had similar AERPs. The P1, N2, and MMN amplitudes correlated with verbal IQ and NEPSY language subtests.

Conclusions

Small P1 response(s) appears to be typical for preterm children.

Significance

Small P1 response in preterm children may suggest altered primary auditory processing.

Introduction

Preterm birth is a major risk factor for brain damage, inflicted by perinatal and postnatal mechanisms such as fetal inflammation, hypoxia, reperfusion injury, postnatal cerebrovascular insults, and nutritional deficiency (Amin, 2004, Folkerth, 2005, Lucas et al., 1998, Murphy et al., 2001, Raman et al., 2006, Vollmer et al., 2006). Volumetric magnetic resonance imaging (MRI) studies have shown that in preterm infants cerebral structural abnormalities are frequent. These include enlarged ventricles with reduced white matter volume, periventricular leukomalacia and cysts, myelination delay, thin cortical gray matter with poor gyration, and corpus callosum dysgenesia (Inder et al., 1999, Inder et al., 2003). The location and the magnitude of these cerebral abnormalities are related to adverse long-term neurodevelopment (Inder et al., 2003, Woodward et al., 2006). Incidences of both severe and moderate disabilities exceed 20% in very preterm children (Marlow et al., 2005, Mikkola et al., 2005). Even in those preterm children without handicaps, intelligence quotient (IQ) is below the normal average, and neuropsychological deficits such as attention, language, sensorineural, visuospatial, and memory disorders are common (Anderson and Doyle, 2003, Mikkola et al., 2005).

During structural and functional development of the auditory system, synaptogenesis occurs, synchronized with dendritic and axonal growth and myelination. The excess synaptic connections are removed up to adolescence (Huttenlocher and Dabholkar, 1997). Several auditory event-related potential (AERP) components have been used to evaluate auditory system development. In childhood, the AERPs are characterized by a large positive P1 at about 100 ms (Cunningham et al., 2000). The P1 decreases in latency and amplitude up to the age of 20 years, reaching the adult latency of approximately 50 ms (Ponton et al., 2000, Sharma et al., 1997). At age 5, the negative N1 response is not yet detectable (Ponton et al., 2002), whereas the negative peak N2 at approximately 250 ms is large and diminishes thereafter (Ceponiene et al., 2001).

Mismatch negativity (MMN) is a response to a perceived change in continuously repeated sounds (Näätänen et al., 1992). The response is present in infancy (Cheour et al., 1998, Kushnerenko et al., 2002) and in childhood (Ceponiene et al., 2001). MMN is obtained subtracting the response to the standard from that to the deviant (occasionally presented different) stimuli. MMN reflects the capacity to detect the difference between two sounds, a capacity important for language development (Näätänen et al., 1997). At age 5–7, MMN latency of 100–200 ms has been reported (Gomot et al., 2000). Responses to novel sounds reflect the child’s capacity to allocate attention. When a novel sound is presented, attention is automatically partially shifted towards that sound. This involuntary, automatic attention shift is suggested to elicit the P3a in AERP at around 300 ms (Alho et al., 1998, Escera et al., 1998). The P3a generators are thought to be located in the left auditory cortex and in the association regions temporoparietally and prefrontally on both hemispheres (Yago et al., 2003).

We have found (Fellman et al., 2004) that the AERPs of preterm infants differed from those of controls during the first year of life, and that of infants small for gestational age (SGA) differed from those of appropriate for gestational age (AGA). Instead of the MMN-like negativity peak observed in control children, a positive polarity response appeared in preterm children at age 12 months. This positivity, and several other AERP components recorded at 0, 6, and 12 months, correlated positively with the 2-year Bayley developmental index. Atypical AERP findings in preterm infants at term have also been reported by Kurtzberg et al., 1984, Therien et al., 2004. In older preterm children, MMN to syllables was smaller than in controls, and the absence of MMN at age 4 predicted naming difficulties at age 6 (Jansson-Verkasalo et al., 2004).

The aim of this study was to assess AERPs of preterm and control children at age 5, and to test the hypothesis that differing AERP components in preterm children correlate with neuropsychological test results. To address the question whether intrauterine growth restriction in preterm infants is associated with a poorer neurodevelopmental outcome than in those of normal birth weight, we recruited SGA and AGA preterm infants to this follow-up. We used the same Easy paradigm with an easily detectable frequency change as used in a previous study (Fellman et al., 2004). To address the automatic attention-allocation functions (Escera et al., 1998) we added a rare “novel” sound, which should elicit P3a responses to involuntary attention-allocation. Further, we used a Challenging paradigm with a duration deviant and a frequency deviant with only a slight change and a fast presentation rate to address change-detection capabilities and tolerance for rapid accumulation of sensory information.