Long-Term Auditory Follow-Up of Preterm Infants after Neonatal Hearing Screening

In this preliminary study the long-term follow-up results at four and eight years of age of NICU graduates after neonatal established HL showed that the initial ABR at the PTA of three months was accurate in predicting the final stage of all types of HL.

In NICU graduates with SNHL the initial ABR and not the VRA was significantly accurate in predicting the severity of HL at four and eight years of age.

ANSD was set as no response at all above a stimulus level of >90dB while OAE’s and/or cochlear microphonics are present. Therefore no level of HL could be established by definition at three months PTA from ABR in children with ANSD. Also in this ANSD group VRA results showed unequal dB levels with play-audiometry at four and eight years of age. In 53% of these children VRA levels were >10dB better than play-audiometry. This means that hearing levels established by VRA were not predictive for hearing levels at four and eight years follow-up.

None of the children with CHL had congenital malformations in the auditory pathway. Nearly all cases recovered from a probably temporary CHL. In one case, CHL was mild and persistent without deterioration due to a chronic middle ear infection.

These findings are of importance in the process of parental guiding. In the SNHL group ABR findings at three months PTA supports future hearing type and level findings while in children with ANSD uncertainty about the level of HL is long lasting.

To our best knowledge this is the first study that takes the PTA of the preterm infant at the time of the initial diagnostic ABR into account as a more physiological milestone for the premature newborn when investigating long-term follow-up results. Due to the recent introduction of neonatal hearing screening, follow-up studies are limited and mostly refer to a (near) term population of newborns. The follow-up period in these studies ranges from six months to six years of age [5,6,10-15]. This study is unique in its insight into data after a lengthy follow-up period, i.e. four and eight years. These follow-up data were related to the PTA of three months which is in terms of pathophysiology more suitable to very preterm newborns. The diagnostic ABR is used as the gold standard in establishing HL after birth [4]. From literature it is known that in premature infants the ongoing process of myelination of the auditory pathway is delayed or alterated [5,6,16-18]. Eventually the maturational delay catches up at the postconceptional age of 40-42 weeks, which is normally the full-term age of healthy newborns [16-18]. Coenraad, et al., showed a prolonged I-V interval at ABR, combined with normal ABR thresholds, which declined with an increased postconceptional age referring to a maturational delay in the auditory pathway [4]. The study of Sleifer, et al., confirms these findings with an inverse correlation between gestational age and ABR inter peak latencies I-III, I-V and III-V with gestational age [5]. Jiang, et al., also found that the maturation in central regions of the brainstem is faster compared to the peripheral regions in preterm infants [6]. Other studies also showed that in preterm infants, <28 weeks of gestation, there is an improvement in HL up until the age of 40 weeks post-term, suggestive of maturational problems of the auditory pathway [17]. The study of Vohr, et al., supports this data [18]. They observed that even in the absence of brain injury there are alterations in maturation and vulnerability of the brain as a consequence of premature birth leading to structural changes within the brain. For these reasons follow-up data in this study were related to the PTA of three months which seems to be proper in the SNHL (very) preterm newborns.

The play-audiometry results at four and eight years of age reflect the most objective stage of HL in the child. The less severe hearing level thresholds at VRA in comparison with play-audiometry levels in the ANSD group may raise questions. The subjective character of the interpretation of the VRA test might possibly bean explanation for the different results between VRA and play-audiometry hearing levels. Another explanation could be a more progressive character or a more heterogeneous course of ANSD in NICU graduates. Robertson, et al., concluded earlier in 2002 that NICU graduates with severe respiratory failure frequently developed late-onset progressive ANSD without a clear explanation for this phenomenon [19]. One can also argue that the more positive results of VRA are partly consequence of the altered or delayed maturation of the brain [5,6]. This study does not support an ongoing deterioration reflected by stable play-audiometry levels between four and eight years of age in children with ANSD. Gardner-Berry, et al., followed twelve infants with ANSD until the age of two years. They found that a reliable VRA was difficult to obtain and resulted in a time delay in further audiological management [15].

Limitation of this study is the limited number of included newborns. We realize there was no sufficient statistical power to detect a lack of difference. Though, the initial results were encouraging for extending this study.

We conclude that the preliminary findings in this study suggests that in NICU graduates ABR of the better ear at three months PTA reliably predicts the type and severity of HL at four and eight years of age in children with SNHL. In the group of children with ANSD VRA levels were lower and not predictive for the severity of HL at four and eight years of age with play-audiometry. These initial results warrant more future research in a larger study group to substantiate reliable predictions with greater precision of the severity of HL in relation to the initial findings at the young age in very preterm NICU graduates to support early intervention and adequately rehabilitation.