- Email: [email protected]
Neural conduction impairment in the auditory brainstem and the prevalence in term babies in neonatal intensive care unit
Clinical Neurophysiology 126 (2015) 1446–1452
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Neural conduction impairment in the auditory brainstem and the prevalence in term babies in neonatal intensive care unit Ze D. Jiang ⇑ Division of Neonatology, Children’s Hospital, Fudan University, Shanghai, China
a r t i c l e
i n f o
Article history: Accepted 15 October 2014 Available online 30 October 2014 Keywords: Brainstem auditory function Brainstem impairment Neural conduction Neonatal brain damage Term babies
h i g h l i g h t s Term babies in neonatal intensive care unit (NICU) are at high risk of brain damage and neurological
impairment. The present study found that neural conduction in the auditory brainstem was impaired in NICU term
babies, which occurred in one-third babies. NICU term babies are at risk of the impairment and the high rates in maximum length sequence brain-
stem auditory evoked response enhance early detection of the impairment.
a b s t r a c t Objective: To detect neural conduction abnormality in the auditory brainstem in term babies in the neonatal intensive care unit (NICU), determine prevalence of the abnormality, and assess if maximum length sequence (MLS) technique improves early detection of the abnormality. Methods: One hundred and six term babies were recruited, and studied by recording and analysing MLS brainstem auditory evoked response (BAER). Interpeak intervals were analysed in detail, which were then compared with those in normal term babies. Results: Wave V latency and I–V and III–V intervals in MLS BAER were increased in the NICU term babies at all click rates 91-910/s, particularly at 455 and 910/s (p < 0.05–0.001). No major abnormalities were found in wave I and III latencies and I–III interval. The abnormal increase in I–V and III–V intervals were seen in significantly more cases at 455 and 910/s in MLS BAER than at 21/s in conventional BAER (X2 = 10.92–13.88, all p < 0.01). As a whole, 38 (35.8%) of the NICU babies had abnormal III–V and/or I–V intervals in MLS BAER, which was significantly more than 13 (12.2%) in conventional BAER (X2 = 16.14, p < 0.01). Conclusion: There is neural conduction impairment in the auditory brainstem in NICU term babies, which occurs in one-third of these babies. Significance: Term babies in NICU are at risk of neural conduction impairment in the auditory brainstem. High click rates in MLS BAER enhance early detection of the impairment. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The neonatal brain is known to be at risk of damage as a result of various perinatal problems that directly or indirectly affect the immature brain. Studies using the brainstem auditory evoked response (BAER), an objective test to assess the functional integrity of the brainstem and auditory pathway, has shown that the neonatal ⇑ Address: Neonatal Unit, Department of Paediatrics, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. Tel.: +44 1865 221364; fax: +44 1865 221366. E-mail address: [email protected]
brainstem is susceptible to some unfavourable perinatal conditions, typically hypoxia–ischemia, or asphyxia, and hyperbilirubinemia (Hall III, 2007; Wilkinson and Jiang, 2006). Conventional BAER, i.e. the BAER obtained using conventional averaging techniques, has relative high false negative results, and cannot effectively detect some neuropathology (Majnemer and Rosenblatt, 1996, 2000; Wilkinson and Jiang, 2006). Since early 1980’s, the maximum length sequence (MLS) has been developed to study the auditory evoked responses to increase understanding of the functional integrity of the auditory system (Bell et al., 2006; Jiang, 2012; Jirsa, 2001; Lasky, 1997; Nagle and Musiek, 2009). More recently, the MLS technique has been
http://dx.doi.org/10.1016/j.clinph.2014.10.147 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
introduced to study the BAER in babies with perinatal problems. These studies have proven that the MLS technique enhances detection of some neuropathology in the neonatal brain, typically hypoxia–ischemia (Jiang, 2012; Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). Newborn babies, including term babies, who are admitted to the neonatal intensive care unit (NICU) have various problems or complications, some of which may damage the immature brain, typically hypoxia–ischemia, hyperbilirubinemia, and meningitis (Majnemer and Rosenblatt, 1996, 2000; Wilkinson and Jiang, 2006). Prompt detection of neonatal brain damage and understanding of prevalence of the damage is of great importance for early intervention and proper management to protect the immature brain and reduce any damage. Previous MLS BAER studies revealed that there is functional impairment of the brainstem in term babies after perinatal hypoxia–ischemia (Jiang et al., 2003, 2010). It is of interest to know if perinatal problems or conditions other than hypoxia–ischemia also have such impairment. We carried out a detailed study of MLS BAER in term babies who were admitted to the NICU due to perinatal problems or conditions other than hypoxia–ischemia. Our primary aims were to detect any abnormality in neural conduction in the auditory brainstem in NICU term babies and determine prevalence of the abnormality. This was mainly achieved by analyzing interpeak intervals, which reflect neural conduction in the auditory brainstem, in MLS BAER. Comparison of the prevalence of abnormal findings in MLS BAER was made with that in conventional BAER to determine if the MLS technique improves early detection of brainstem conduction abnormality in individual babies. 2. Patients and methods 2.1. Subjects The subjects’ data are summarized in Table 1. Since hypoxia– ischemia is a known high risk factor of brainstem auditory impairment in term babies, babies were excluded from the study entry if he or she had hypoxia–ischemia, including clinical signs of HI (hypotonia with reduced or no spontaneous movements, increased threshold for primitive reflexes, lethargy or comatose, absence or very weak suck and requirement of tube feeds, or seizures), umbilical cord blood pH < 7.10, and depressed Apgar score (less than 6 at 5 min), as defined in our previous studies (Jiang et al., 2003, 2010). Excluded were also those who exposed to potentially ototoxic drugs such as aminoglycosides (mainly amikacin), and diuretics
Table 1 Subjects’ data in NICU term babies and normal term babies. NICU Gender (n) Male/female Gestational age (weeks) Mean ± SD Range Postconceptional age (weeks) Mean ± SD Range Birthweight (g) Mean ± SD Range BAER threshold (dB nHL) Mean ± SD Range
Normal
47/59
19/26 *
39.6 ± 1.3 37–42
39.1 ± 1.5 37–42
39.7 ± 1.4 37–42
39.4 ± 1.3 37–42
3384 ± 603 2485–4640
3451 ± 471 2565–4539
13.1 ± 8.1+ 5–35
11.0 ± 5.4 0–20
+NICU babies who had a BAER threshold P40 dB nHL had been excluded from study entry and data analysis to avoid any significant effect of peripheral hearing problems on MLS BAER wave components. * P < 0.05 (t-test) for comparison between NICU and normal term babies.
1447
(furosemide), Genetic and metabolic disorders associated with hearing loss. Any babies who had a BAER threshold P40 dB normal hearing level (nHL) had been excluded from study entry to avoid any significant effect of peripheral hearing problems on the measurements of MLS BAER wave components so as to analyse MLS BAER variables more reliably and accurately, the same as our previous MLS BAER studies (Jiang, 2012; Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). All our babies passed the neonatal hearing screening. Informed consent of parents was obtained for each baby before study entry. This study was approved by the Central Oxford Research Ethics Committee and the Ethics Committee of the Children’s Hospital of Fudan University. The study group (NICU babies) consisted of 106 babies born between 37 and 42 weeks (39.6 ± 1.3 weeks) of gestation, determined by the best estimate of last menstrual period, obstetrical record, and clinical examination. Birthweight ranged between 2485 and 4640 g (3284 ± 603 g). Forty-eight babies were recruited from the NICU of Children’s Hospital of Fudan University in Shanghai. The remaining 58 babies were recruited from the NICU of the John Radcliffe Hospital of Oxford University in Oxford. Of the 58 babies, 41, who had both conventional BAER recordings and MLS BAER recordings, were from those who were briefly reported before (Jiang et al., 2013). The NICU term infants had various perinatal problems or complications (e.g. hypotension, hypoglycaemia, meconium aspiration syndrome, sepsis, metabolic acidosis, pneumonia, and haemolytic or non-hemolytic hyperbilirubinemia). Some of the subjects had more than one perinatal condition. The control group (normal term babies) was 45 healthy term babies, 24 from the neonatal unit of Children’s Hospital of Fudan University in Shanghai and the remaining 21 from the neonatal unit of the John Radcliffe Hospital of Oxford University in Oxford. The gestational age ranged between 37 and 42 weeks (39.1 ± 1.4 weeks), which was marginally smaller, though statistically significant, than in the study group (Table 1). Their birth weights were between 2565 and 4539 g (3464 ± 473 g), which did not differ significantly from the study group (Table 1). None had any major perinatal conditions. At time of MLS BAER testing, monaural hearing thresholds in the controls were all 20 dB nHL or less. 2.2. Procedures of recording MLS BAER The procedures and instrumentation of the recording were the same in the two institutes (Children’s Hospital of Fudan University and the John Radcliffe Hospital of Oxford University). The Nicolet Spirit 2000 Portable Evoked Potential System (Nicolet Biomedical Inc. Madison, WI, USA) was used to record and analyze MLS BAER and conventional BAER. For all subjects, only the left ear was tested to keep consistency of recording conditions and reduce recording time, the same as in our previous MLS BAER studies (Jiang and Chen 2014; Jiang et al., 2003, 2005, 2010). All babies were studied between days 2 and 5 after birth. There were no significant differences in the postconceptional ages between the NICU babies (39.7 ± 1.4 weeks) and the normal control babies (39.4 ± 1.3 weeks). Three gold-plated disk electrodes were placed, respectively, at middle forehead (positive), ipsilateral earlobe (negative) and contralateral earlobe (ground). The impedance between any two electrodes was reduced to 5 kX or less, which were remained during the whole session of BAER recording. Sweep duration was 24 ms. Recording of the BAER started immediately after the infant fell asleep naturally, often after a feed without using any sedatives. The infant remained asleep throughout the recording session. Rarefaction clicks with a duration 0.1 ms were delivered monaurally through a TDH 39 headphone to the left ear. We first recorded conventional BAER with 21/s clicks as a basic repetition
1448
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
rate to compare the results with those of MLS BAER. Then, MLS BAER was recorded with clicks presented in the sequence of 91, 227, 455 and 910/s in the first run and a reverse sequence in the second run. All babies were tested at an intensity 60 dB nHL. For those with a BAER threshold, which was determined as the lowest intensity at which wave V can be recognized reliably in conventional BAER, >20 dB nHL (3 babies at 25 and 2 at 30 dB nHL), higher intensities (70 and 75 dB nHL, respectively) were also used. This allowed to collect MLS BAER data at the intensity 40 dB or slightly higher above the threshold of each subject, and thus minimize any effect of threshold elevation and slight peripheral hearing loss on MLS BAER components (Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). Two runs of MLS BAER recording were made for each stimulus condition to evaluate reproducibility of the recorded waveforms. Evoked brain responses to 1,500 trains of clicks were preamplified, bandpassed at 100–3000 Hz, and then averaged for each run of recording. During the signal averaging, amplitude artifact rejection was active to eliminate any on-line signals with amplitude exceeding ±25 lV. The tester kept monitoring the ongoing filtered EEG and the running averaged MLS BAER, and manually discontinued sampling whenever there were excessive muscle artefacts on the monitoring oscilloscope. As such, contamination of artefacts to the recorded MLS BAER waveforms was minimized. Sample MLS BAER recordings at different repetition rates of clicks in a normal term baby are shown in Fig. 1. 2.3. Data analysis Analyses of the BAER were conducted blinded to the medical history and clinical data of each baby. The latency of each MLS BAER major wave (I, III, and V) was measured, and the I–V, I–III and III–V interpeak intervals were then calculated, as previously described (Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). A SPSS package (SPSS, Chicago, IL) was used for statistical analyses. We had compared the MLS BAER data between the two institutes (Children’s Hospital of Fudan University and the John Radcliffe Hospital of Oxford University), and found that the results were basically same, without any significant differences. Therefore the data from the two institutes were combined together for detailed statistical analysis. The subject’s data in Table 1 were compared between the NICU babies and the normal babies. Categorical variable (gender) was analysed using
Pearson’s v2 statistic and Fisher’s exact test. Continuous variables (gestational age, postconceptional age, and birthweight, BAER threshold) were compared using Student’s t tests. A 2-tailed value of p < 0.05 was considered statistically significant. Measurements from two replicated MLS BAER recordings to each stimulus condition were averaged for each subject. Mean and standard deviation of each BAER variable at each stimulus condition were obtained for the study and control groups, respectively. Student’s t test was used for comparison between the study and control groups for each MLS BAER variable. The Chi-square test was used for comparison of the abnormal results in the NICU babies between MLS BAER and conventional BAER to examine whether MLS BAER improves the detectability of abnormality in the auditory brainstem. A 2-tailed value of p < 0.05 was considered statistically significant for these data analysis.
3. Results As presented in Table 1, there were no significant differences between the NICU and normal term babies in gender, postconceptional age at time of MLS BAER testing, birthweight and BAER threshold. Only gestational age in the NICU babies was marginally greater, though statistically significant, than in the normal babies. No significant difference was found in hearing levels (i.e. the dB above the thresholds of individual subjects), at which measurements of MLS BAER recordings were made and analysed, between the NICU babies (49.5 ± 6.4 dB nHL) and the normal term babies (49.3 ± 5.5 dB nHL). In both the NICU and normal term babies, major MLS BAER components (waves I, III and V) were all reliably identified at all click rates. No missing MLS BAER waves were found in the normal term babies. Only at the highest click rate 910/s, three NICU babies had less clear wave I, with 2 having less clear wave V. Nevertheless, the three babies had not been included in the study entry and data analysis. 3.1. Wave latencies Fig. 2 shows boxplots of the latencies of waves I (A), III (B) and V (C), respectively, at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in the NICU term babies and the normal term babies. In conventional BAER at 21/s, wave I latency in the NICU babies was similar to that in the normal babies (Fig. 2A). The latencies of waves III and V in the NICU babies were both longer that those in the normal babies (p < 0.05 and 0.01, Figs. 2B and 2C). In MLS BAER, wave I latency in the NICU babies was slightly shorter than those in the normal babies at all 91-910/s clicks, without any statistically significant difference (Fig. 2A). Wave III latency tended to be longer than those in the term babies at all rates, which did not reach statistical significance at 91 and 227/s but differed significantly at higher rates 455 and 910/s (p < 0.05 and 0.05, Fig. 2B). By comparison, wave V latency in the NICU babies was significantly longer than that in the normal babies at all click rates, particularly at the very high rates 455 and 910/s (p < 0.05–0.001, Fig. 2C). 3.2. Interpeak intervals
Fig. 1. Sample MLS BAER waveforms at different repetition rates of clicks, recorded from a normal term baby.
Fig. 3 shows boxplots of the I–V (A), I–III (B) and III–V (C) intervals, respectively, at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in the NICU babies and the normal babies. In conventional BAER at 21/s, all the three intervals in the NICU babies tended to be longer than those in the normal babies, with the I–V and III–V intervals differing significantly between
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
1449
Fig. 2. Boxplot (bold line across the box, median; box, 25th and 75th centile; extensions, the largest and smallest values) of the latencies of wave I (A), III (B), and V (C) at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in normal term babies (Normal) and NICU term babies (NICU). ⁄p < 0.05, ⁄⁄p < 0.01, and ⁄⁄⁄p < 0.001 for comparison between NICU and normal term babies.
Fig. 3. Boxplot (bold line across the box, median; box, 25th and 75th centile; extensions, the largest and smallest values) of the interpeak intervals of I–V (A), I–III (B), and III–V (C) at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in normal term babies (Normal) and NICU term babies (NICU). ⁄p < 0.05, ⁄⁄ p < 0.01, and ⁄⁄⁄p < 0.001 for comparison between NICU and normal term babies.
the two groups of babies (p < 0.01 and 0.05, Figs. 3A, and 3C). The III–V/I–III interval ration was similar in the two groups. In MLS BAER, the I–V interval in the NICU babies was significantly longer than in the normal babies at all click rates, and the difference between the two groups was generally increased with increasing click rate (p < 0.05–0.001, Fig. 3A). On the other hand, the I–III interpeak interval in the NICU babies was only slightly longer than in the normal babies, which did not differ significantly at any click rates (Fig. 3B). Similar to the I–V interval, the III–V interval was also significantly longer than in the normal babies
at all click rates, particularly at 455 and 910/s (p < 0.05–0.001, Fig. 3C). The III–V/I–III interval ration in the NICU babies was slightly increased at 91 and 227/s, but was significantly increased at 455 and 910/s (p < 0.001 and 0.001). 3.3. Abnormal rate of interpeak intervals To obtain the rate of neural conduction impairment in the auditory brainstem in the NICU babies we calculated the percentage rate of abnormal increase in the two abnormal MLS BAER intervals,
1450
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
Table 2 Numbers of NICU term babies with abnormal increase in I–V and III–V interpeak intervals at various click rates and statistical comparison of the abnormality in MLS BAER at each click rate (91-910/s) with the abnormality in conventional BAER at 21/s. BAER interval
Subjects
21/s
91/s
227/s
455/s
910/s
Normal criteria NICU abnormal cases (%) X2 p
5.40 11 (10.4%)
5.62 13 (12.3%) 0.19 >0.05
6.12 19 (17.9%) 2.27 >0.05
6.50 33 (31.1%) 13.88 <0.01
6.51 32 (31.1%) 12.87 <0.01
Normal criteria NICU abnormal cases (%) X2 p
2.49 7 (6.6%)
2.69 7 (6.6%) 0.00 >0.05
2.97 12 (11.3%) 1.45 >0.05
3.27 24 (22.6%) 10.92 <0.01
3.26 26 (25.2%) 13.26 <0.01
I–V (ms)
III–V (ms)
Normal criteria refer to the upper limit (mean + 2.5 SD) of a BAER variable in normal term babies. Values exceed the criteria are defined as abnormal.
i.e. III–V and I–V intervals, which reflect neural conduction in the auditory brainstem. As a common practice in BAER, any interval values that exceeded 2.5 standard deviations above the mean measurement in the normal term babies were considered to be abnormal. The results are summarized in Table 2. For the I–V interval, the numbers (and percentage rates) of babies with an abnormal increase at 91, 227, 455 and 910/s clicks were, respectively, 13 (12.3%), 19 (17.9%), 33 (31.1%) and 32 (30.2%). For the III–V interval, the numbers (and percentage rates) of babies with an abnormal increase at the four click rates were, respectively, 7 (6.6%), 12 (11.3%), 24 (22.6%) and 26 (24.5%). All of the babies who had an abnormal increase at 91 and/or 227/s also had an abnormal increase at 455 and/or 910/s. Most of the babies with an abnormal increase in the III–V interval also had an abnormal increase in the I–V interval. Conversely, most of the babies with an abnormal increase in the I–V interval also had an abnormal increase in the III–V interval. As a whole, 38 (35.8%) of the 106 NICU babies had and abnormal increase in the III–V and/or I–V intervals. The noticeable associated perinatal problems in these babies were hypoglycemia, hyperbilirubinemia, and sepsis. By comparison, the abnormal increase in conventional BAER at 21/s clicks was found in only 11 (10.4%) of the 106 NICU babies for the I–V interval, and 7 (6.6%) for the III–V interval. The total number of subjects with the abnormal increase was 13 (12.2%) at 21/s. All these babies also had an abnormal increase in MLS BAER. Comparison of the abnormal increase in I–V and III–V intervals in the NICU babies was made between MLS BAER and conventional BAER using the Chi-square test (Table 2). The abnormal rates for the I–V and III–V intervals at 91 and 227/s in MLS BAER did not differ significantly from those at 21/s in conventional BAER, although the abnormal increase at 227/s was seen in more than cases than at 21/s. At very high rates 455 and 910/s in MLS BAER, however, the abnormal increase in the I–V interval was seen in significantly more cases than at 21/s in conventional BAER (X2 = 13.88 and 12.86, p < 0.01 and 0.01, respectively). Similarly, the abnormal increase in the III–V interval at 455 and 910/s was seen in significantly more cases than at 21/s (X2 = 10.92 and 12.96, p < 0.01 and 0.01, respectively). The total rate of the abnormal increase was seen significantly higher in MLS BAER (35.8%) than in conventional BAER (12.2%, X2 = 16.14, p < 0.01). 4. Discussion The I–V interpeak interval is the major and most widely used BAER variable that reflects neural conduction in the auditory brainstem (Hall III, 2007; Jiang, 2012; Moore, 1987; Moore et al., 1995). It has been regarded as an index of brainstem conduction time. The present MLS BAER study in NICU term babies demonstrated a general increase in the wave V latency, and the I–V and III–V interpeak intervals. The most significant increase in these variables occurred mainly at very high click rates 455 and 910/s clicks. These results
indicate that neural conduction in the auditory brainstem is abnormal or impaired in term babies who receive intensive care, which is more evident following more stressful high-rate stimulation. As reflected by the rate of total abnormal increase in the I–V and III–V intervals, the neural conduction impairment was seen in 35.3% of the NICU term babies we studied. The I–III and III–V intervals are the earlier and later components of the I–V interval, and generally reflect neural conduction at the more peripheral (or caudal) and more central (or rostral) regions of the auditory brainstem, respectively (Jiang, 2012; Jiang et al., 2009). In our NICU term babies, while the III–V and I–V intervals were significantly increased, the I–III interval was only slightly increased at all click rates, without any statistical significance. Apparently, the significant increase in the I–V interval is mainly produced by the significant increase in the III–V interval and, to a less extent, by the slight increase in the I–III interval. The significant increase in the III–V interval also made a main contribution to the increase in wave V latency. Thus, the impaired neural conduction in the auditory brainstem in our NICU term babies occurs mainly at the more central regions of the brainstem, reflected by the increase in the III–V interval and, to a less degree, at the more peripheral regions, reflected by the insignificant increase in the I–III interval. For BAER studies in NICU term babies, there are previous reports in conventional BAER, mainly related to hypoxia–ischemia and hyperbilirubinemia (Akinpelu et al., 2013; Hecox et al., 1981; Newton, 2001; Streletz et al., 1986; Wilkinson and Jiang, 2006). So far, there are very few reports on MLS BAER in term babies, mainly in those who had perinatal hypoxia–ischemia (Jiang et al., 2003, 2010). Here, we compared the present findings in the NICU babies who had perinatal problems other than hypoxia–ischemia with those of the previous MLS BAER studies in term babies who had perinatal hypoxia–ischemia. During the first few days after birth, the babies after hypoxia–ischemia demonstrated a significant increase in the III–V and I–V intervals at all click rates 91-910/s click rates (Jiang et al., 2003, 2010). The I–III interval was also increased, which occurred mainly at the very high rates 455 and 910/s click rates. By comparison, the MLS BAER abnormalities in our NICU term babies were relatively less significant. The major increase in the I–V and III–V intervals occurred mainly at the very high click rates (455 and 910/s), and there was no significant increase in the I–III interval at any click rates. Such differences between the two groups of term babies suggest that that perinatal hypoxia–ischemia and other perinatal conditions exert somewhat differential effects on the neonatal auditory brainstem, and the brainstem impairment due to other perinatal problems is relatively less severe, compared with that due to hypoxia–ischemia. So far, the prevalence of brainstem impairment in term babies in the NICU remains unclear. The present study revealed that 35.3% of the NICU term babies had an abnormal increase in the
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
III–V and/or I–V intervals. This rate of total abnormal increase implies that about one-third of the NICU term babies had neural conduction impairment in the auditory brainstem. We did not conduct detailed analysis of aetiology for the MLS BAER abnormality, because this is not the purpose of this study. However, we did notice the abnormality was mainly seen in the babies who had hypoglycemia, hyperbilirubinemia, and sepsis. These problems are known to be risk factors for neonatal brain damage (AdamsChapman and Stoll, 2006; Burns et al., 2008; Johnson and Bhutani, 2011; Mallard and Wang, 2012; Montassir et al., 2009; Shapiro, 2010; Tam et al., 2012; Yalnizoglu et al., 2007). A clinical implication of this high prevalence of brainstem conduction impairment is that some neuroprotective measures may be required for NICU term babies, particularly those with had hypoglycemia, sepsis and hyperbilirubinemia, to protect the neonatal brainstem and reduce any impairment to neural conduction in the brainstem. The rates of abnormal increase in the III–V and I–V intervals in our NICU babies were generally seen more at higher than at lower click rate. The abnormal rates at 91 and 227/s in MLS BAER were similar or slightly higher than at 21/s in conventional BAER. However, the abnormal rates at 455 and 910/s were much higher than those at the lower rates 91 and 227/s, and significantly higher than in conventional BAER. These findings provide further evidence that high-rate stimulation used in MLS BAER enhances the detection of brainstem auditory impairment in babies with perinatal problems (Jiang, 2012; Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2007, 2010; Wilkinson et al., 2007). In a previous short report, we briefly described MLS BAER in term babies who had various perinatal problems or conditions, and found that these babies were at risk of brainstem auditory dysfunction (Jiang et al., 2012). In the present study, we carried out a detailed analysis of both MLS BAER and conventional BAER in NICU term babies. In addition to detection of neural conduction impairment in the auditory brainstem, we also determined the prevalence of the impairment, and, in particular, as assessed if the MLS technique improves early detection of the impairment. We found that the BAER abnormalities that are indicative of brainstem conduction impairment were particular significant at high-rate stimulation in MLS BAER. The abnormal rates at 455 and 910/s in individual babies were significantly higher than those in conventional BAER at 21/s. As a whole, 35.8% of the NICU babies had abnormal increase in the III–V and/or I–V intervals in MLS BAER, which was significantly more than the 12.2% in conventional BAER. Thus, about one-third of NICU term babies have neural conduction impairment in the brainstem, suggesting that term babies in the NICU are at risk of brainstem conduction impairment. Compared with conventional BAER, MLS BAER, typically at high click rates (455 and 910/s), enhances the detection of brainstem conduction impairment for babies in the NICU. Over the last decade, the BAER recorded and processed using the MLS technique has been documented to enhance the detection of neuropathology that involves in the brainstem auditory pathway in neonatal audiology and neurology. MLS BAER recorded with high-rate stimulation (particularly 455/s) could become routine for clinical use (Jiang, 2012). Nevertheless, there is currently a lack of commercially available evoked potential systems that are equipped with the MLS technique. The development of such evoked potential systems in the near future is warranted. Acknowledgements Mrs. Dorathea Brosi did excellent work in recruiting of subjects and collecting of data. Drs. Rong Yin and Lili Ping made major contribution to recruiting subjects, collecting and analysing data. Professor Andrew Wilkinson is gratefully appreciated for his
1451
support of this project. The research was supported by the WellChild, and Bridging Support Fund and Medical Sciences Division of Oxford University, UK. The author has no potential conflicts of interest to be disclosed. References Adams-Chapman I, Stoll BJ. Neonatal infection and long-term neurodevelopmental outcome in the preterm infant. Curr Opin Infect Dis 2006;19:290–7. Akinpelu OV, Waissbluth S, Daniel SJ. Auditory risk of hyperbilirubinemia in term newborns: a systematic review. Int J Pediatr Otorhinolaryngol 2013;77:898–905. Bell SL, Smith DC, Allen R, Lutman ME. The auditory middle latency response, evoked using maximum length sequences and chirps, as an indicator of adequacy of anesthesia. Anesth. Analg. 2006;102:495–8. Burns CM, Rutherford MA, Boardman JP, Cowan FM. Patterns of cerebral injury and neurodevelopmental outcomes after symptomatic neonatal hypoglycemia. Pediatrics 2008;122:65–74. Hall III JW. ABR: pediatric clinical application and populations. In: Hall III JW, editor. New handbook of auditory evoked responses. Boston: Pearson Education; 2007. p. 313–65. Hecox K, Cone B, Blaw M. Brainstem auditory evoked response in the diagnosis of pediatric neurologic diseases. Neurology 1981;31:832–9. Jiang ZD. Maximum length sequence technique improves detection of neuropathology involving infant brainstem. In: Lawson PN, McCarthy EA, editors. Pediatric neurology. New York: Nova Science Publishers; 2012. p. 1–38. Jiang ZD, Chen C. Impaired neural conduction in the auditory brainstem of high-risk very preterm infants. Clin Neurophysiol 2014;125:1231–7. Jiang ZD, Brosi DM, Wang J, Xu X, Chen GQ, Shao XM, et al. Time course of brainstem pathophysiology during first month in term infants after perinatal asphyxia, revealed by MLS BAER latencies and intervals. Pediatr Res 2003;54:680–7. Jiang ZD, Brosi DM, Li ZH, Chen C, Wilkinson AR. Brainstem auditory function at term in preterm babies with and without perinatal complications. Pediatr Res 2005;58:1164–9. Jiang ZD, Xu X, Brosi DM, Shao XM, Wilkinson AR. Suboptimal function of the auditory brainstem in term neonates with transient low Apgar scores. Clin Neurophysiol 2007;118:1088–96. Jiang ZD, Brosi D, Wu YY, Wilkinson AR. Relative maturation of the peripheral and central regions of the auditory brainstem from preterm to term and the influence of preterm birth. Pediatr Res 2009;65:657–62. Jiang ZD, Brosi DM, Wilkinson AR. Differences in impaired brainstem conduction between neonatal chronic lung disease and perinatal asphyxia. Clin Neurophysiol 2010;121:725–33. Jiang ZD, Brosi DM, Yin R, Wilkinson AR. Term neonates receiving intensive care at high risk of brainstem auditory impairment. Arch Dis Child-Fetal 2012;97. F359-61. Jiang ZD, Zhou Y, Yin R, Wilkinson AR. Amplitude reduction in brainstem auditory response in term infants under neonatal intensive care. Clin Neurophysiol 2013;124:1470–6. Jirsa RE. Maximum length sequences-auditory brainstem responses from children with auditory processing disorders. J Am Acad Audiol 2001;12:155–64. Johnson L, Bhutani VK. The clinical syndrome of bilirubin-induced neurologic dysfunction. Semin Perinatol 2011;35:101–13. Lasky RE. Rate and adaptation effects on the auditory evoked brainstem response in human newborns and adults. Hear Res 1997;111:165–76. Majnemer A, Rosenblatt B. Evoked potentials as predictors of outcome in neonatal intensive care unit survivors: review of the literature. Pediatr Neurol 1996;14:189–95. Majnemer A, Rosenblatt B. Prediction of outcome at school age in neonatal intensive care unit graduates using neonatal neurologic tools. J Child Neurol 2000;15:645–51. Mallard C, Wang X. Infection-induced vulnerability of perinatal brain injury. Neurol Res Int 2012;2012:102153. Montassir H, Maegaki Y, Ogura K, Kurozawa Y, Nagata I, Kanzaki S, Ohno K. Associated factors in neonatal hypoglycemic brain injury. Brain Dev 2009;31:649–56. Moore JK. The human auditory brain stem as a generator of auditory evoked potentials. Hear Res 1987;29:33–43. Moore JK, Perazzo LM, Braun A. Time course of axonal myelination in the human brainstem auditory pathway. Hear Res 1995;87:21–31. Nagle S, Musiek FE. Morphological changes in the middle latency response using maximum length sequence stimuli. J Am Acad Audiol 2009;20:492–502. Newton V. Adverse perinatal conditions and the inner ear. Semin Neonatol 2001;6:543–51. Shapiro SM. Chronic bilirubin encephalopathy: diagnosis and outcome. Semin Fetal Neonatal Med 2010;15:157–63. Streletz LJ, Graziani LJ, Branca PA, Desai HJ, Travis SF, Mikaelian DO. Brainstem auditory evoked potentials in fullterm and preterm newborns with hyperbilirubinemia and hypoxemia. Neuropediatrics 1986;17:66–71. Tam EW, Haeusslein LA, Bonifacio SL, Glass HC, Rogers EE, Jeremy RJ, Barkovich AJ, Ferriero DM. Hypoglycemia is associated with increased risk for brain injury and adverse neurodevelopmental outcome in neonates at risk for encephalopathy. J Pediatr 2012;161:88–93.
1452
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
Wilkinson AR, Jiang ZD. Brainstem auditory evoked response in neonatal neurology. Semin Fet Neonatol Med 2006;11:444–51. Wilkinson AR, Brosi DM, Jiang ZD. Functional impairment of the brainstem in infants with bronchopulmonary dysplasia. Pediatrics 2007;120:362–71.
Yalnizoglu D, Haliloglu G, Turanli G, Cila A, Topcu M. Neurologic outcome in patients with MRI pattern of damage typical for neonatal hypoglycemia. Brain Dev 2007;29:285–92.
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Neural conduction impairment in the auditory brainstem and the prevalence in term babies in neonatal intensive care unit Ze D. Jiang ⇑ Division of Neonatology, Children’s Hospital, Fudan University, Shanghai, China
a r t i c l e
i n f o
Article history: Accepted 15 October 2014 Available online 30 October 2014 Keywords: Brainstem auditory function Brainstem impairment Neural conduction Neonatal brain damage Term babies
h i g h l i g h t s Term babies in neonatal intensive care unit (NICU) are at high risk of brain damage and neurological
impairment. The present study found that neural conduction in the auditory brainstem was impaired in NICU term
babies, which occurred in one-third babies. NICU term babies are at risk of the impairment and the high rates in maximum length sequence brain-
stem auditory evoked response enhance early detection of the impairment.
a b s t r a c t Objective: To detect neural conduction abnormality in the auditory brainstem in term babies in the neonatal intensive care unit (NICU), determine prevalence of the abnormality, and assess if maximum length sequence (MLS) technique improves early detection of the abnormality. Methods: One hundred and six term babies were recruited, and studied by recording and analysing MLS brainstem auditory evoked response (BAER). Interpeak intervals were analysed in detail, which were then compared with those in normal term babies. Results: Wave V latency and I–V and III–V intervals in MLS BAER were increased in the NICU term babies at all click rates 91-910/s, particularly at 455 and 910/s (p < 0.05–0.001). No major abnormalities were found in wave I and III latencies and I–III interval. The abnormal increase in I–V and III–V intervals were seen in significantly more cases at 455 and 910/s in MLS BAER than at 21/s in conventional BAER (X2 = 10.92–13.88, all p < 0.01). As a whole, 38 (35.8%) of the NICU babies had abnormal III–V and/or I–V intervals in MLS BAER, which was significantly more than 13 (12.2%) in conventional BAER (X2 = 16.14, p < 0.01). Conclusion: There is neural conduction impairment in the auditory brainstem in NICU term babies, which occurs in one-third of these babies. Significance: Term babies in NICU are at risk of neural conduction impairment in the auditory brainstem. High click rates in MLS BAER enhance early detection of the impairment. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The neonatal brain is known to be at risk of damage as a result of various perinatal problems that directly or indirectly affect the immature brain. Studies using the brainstem auditory evoked response (BAER), an objective test to assess the functional integrity of the brainstem and auditory pathway, has shown that the neonatal ⇑ Address: Neonatal Unit, Department of Paediatrics, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. Tel.: +44 1865 221364; fax: +44 1865 221366. E-mail address: [email protected]
brainstem is susceptible to some unfavourable perinatal conditions, typically hypoxia–ischemia, or asphyxia, and hyperbilirubinemia (Hall III, 2007; Wilkinson and Jiang, 2006). Conventional BAER, i.e. the BAER obtained using conventional averaging techniques, has relative high false negative results, and cannot effectively detect some neuropathology (Majnemer and Rosenblatt, 1996, 2000; Wilkinson and Jiang, 2006). Since early 1980’s, the maximum length sequence (MLS) has been developed to study the auditory evoked responses to increase understanding of the functional integrity of the auditory system (Bell et al., 2006; Jiang, 2012; Jirsa, 2001; Lasky, 1997; Nagle and Musiek, 2009). More recently, the MLS technique has been
http://dx.doi.org/10.1016/j.clinph.2014.10.147 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
introduced to study the BAER in babies with perinatal problems. These studies have proven that the MLS technique enhances detection of some neuropathology in the neonatal brain, typically hypoxia–ischemia (Jiang, 2012; Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). Newborn babies, including term babies, who are admitted to the neonatal intensive care unit (NICU) have various problems or complications, some of which may damage the immature brain, typically hypoxia–ischemia, hyperbilirubinemia, and meningitis (Majnemer and Rosenblatt, 1996, 2000; Wilkinson and Jiang, 2006). Prompt detection of neonatal brain damage and understanding of prevalence of the damage is of great importance for early intervention and proper management to protect the immature brain and reduce any damage. Previous MLS BAER studies revealed that there is functional impairment of the brainstem in term babies after perinatal hypoxia–ischemia (Jiang et al., 2003, 2010). It is of interest to know if perinatal problems or conditions other than hypoxia–ischemia also have such impairment. We carried out a detailed study of MLS BAER in term babies who were admitted to the NICU due to perinatal problems or conditions other than hypoxia–ischemia. Our primary aims were to detect any abnormality in neural conduction in the auditory brainstem in NICU term babies and determine prevalence of the abnormality. This was mainly achieved by analyzing interpeak intervals, which reflect neural conduction in the auditory brainstem, in MLS BAER. Comparison of the prevalence of abnormal findings in MLS BAER was made with that in conventional BAER to determine if the MLS technique improves early detection of brainstem conduction abnormality in individual babies. 2. Patients and methods 2.1. Subjects The subjects’ data are summarized in Table 1. Since hypoxia– ischemia is a known high risk factor of brainstem auditory impairment in term babies, babies were excluded from the study entry if he or she had hypoxia–ischemia, including clinical signs of HI (hypotonia with reduced or no spontaneous movements, increased threshold for primitive reflexes, lethargy or comatose, absence or very weak suck and requirement of tube feeds, or seizures), umbilical cord blood pH < 7.10, and depressed Apgar score (less than 6 at 5 min), as defined in our previous studies (Jiang et al., 2003, 2010). Excluded were also those who exposed to potentially ototoxic drugs such as aminoglycosides (mainly amikacin), and diuretics
Table 1 Subjects’ data in NICU term babies and normal term babies. NICU Gender (n) Male/female Gestational age (weeks) Mean ± SD Range Postconceptional age (weeks) Mean ± SD Range Birthweight (g) Mean ± SD Range BAER threshold (dB nHL) Mean ± SD Range
Normal
47/59
19/26 *
39.6 ± 1.3 37–42
39.1 ± 1.5 37–42
39.7 ± 1.4 37–42
39.4 ± 1.3 37–42
3384 ± 603 2485–4640
3451 ± 471 2565–4539
13.1 ± 8.1+ 5–35
11.0 ± 5.4 0–20
+NICU babies who had a BAER threshold P40 dB nHL had been excluded from study entry and data analysis to avoid any significant effect of peripheral hearing problems on MLS BAER wave components. * P < 0.05 (t-test) for comparison between NICU and normal term babies.
1447
(furosemide), Genetic and metabolic disorders associated with hearing loss. Any babies who had a BAER threshold P40 dB normal hearing level (nHL) had been excluded from study entry to avoid any significant effect of peripheral hearing problems on the measurements of MLS BAER wave components so as to analyse MLS BAER variables more reliably and accurately, the same as our previous MLS BAER studies (Jiang, 2012; Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). All our babies passed the neonatal hearing screening. Informed consent of parents was obtained for each baby before study entry. This study was approved by the Central Oxford Research Ethics Committee and the Ethics Committee of the Children’s Hospital of Fudan University. The study group (NICU babies) consisted of 106 babies born between 37 and 42 weeks (39.6 ± 1.3 weeks) of gestation, determined by the best estimate of last menstrual period, obstetrical record, and clinical examination. Birthweight ranged between 2485 and 4640 g (3284 ± 603 g). Forty-eight babies were recruited from the NICU of Children’s Hospital of Fudan University in Shanghai. The remaining 58 babies were recruited from the NICU of the John Radcliffe Hospital of Oxford University in Oxford. Of the 58 babies, 41, who had both conventional BAER recordings and MLS BAER recordings, were from those who were briefly reported before (Jiang et al., 2013). The NICU term infants had various perinatal problems or complications (e.g. hypotension, hypoglycaemia, meconium aspiration syndrome, sepsis, metabolic acidosis, pneumonia, and haemolytic or non-hemolytic hyperbilirubinemia). Some of the subjects had more than one perinatal condition. The control group (normal term babies) was 45 healthy term babies, 24 from the neonatal unit of Children’s Hospital of Fudan University in Shanghai and the remaining 21 from the neonatal unit of the John Radcliffe Hospital of Oxford University in Oxford. The gestational age ranged between 37 and 42 weeks (39.1 ± 1.4 weeks), which was marginally smaller, though statistically significant, than in the study group (Table 1). Their birth weights were between 2565 and 4539 g (3464 ± 473 g), which did not differ significantly from the study group (Table 1). None had any major perinatal conditions. At time of MLS BAER testing, monaural hearing thresholds in the controls were all 20 dB nHL or less. 2.2. Procedures of recording MLS BAER The procedures and instrumentation of the recording were the same in the two institutes (Children’s Hospital of Fudan University and the John Radcliffe Hospital of Oxford University). The Nicolet Spirit 2000 Portable Evoked Potential System (Nicolet Biomedical Inc. Madison, WI, USA) was used to record and analyze MLS BAER and conventional BAER. For all subjects, only the left ear was tested to keep consistency of recording conditions and reduce recording time, the same as in our previous MLS BAER studies (Jiang and Chen 2014; Jiang et al., 2003, 2005, 2010). All babies were studied between days 2 and 5 after birth. There were no significant differences in the postconceptional ages between the NICU babies (39.7 ± 1.4 weeks) and the normal control babies (39.4 ± 1.3 weeks). Three gold-plated disk electrodes were placed, respectively, at middle forehead (positive), ipsilateral earlobe (negative) and contralateral earlobe (ground). The impedance between any two electrodes was reduced to 5 kX or less, which were remained during the whole session of BAER recording. Sweep duration was 24 ms. Recording of the BAER started immediately after the infant fell asleep naturally, often after a feed without using any sedatives. The infant remained asleep throughout the recording session. Rarefaction clicks with a duration 0.1 ms were delivered monaurally through a TDH 39 headphone to the left ear. We first recorded conventional BAER with 21/s clicks as a basic repetition
1448
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
rate to compare the results with those of MLS BAER. Then, MLS BAER was recorded with clicks presented in the sequence of 91, 227, 455 and 910/s in the first run and a reverse sequence in the second run. All babies were tested at an intensity 60 dB nHL. For those with a BAER threshold, which was determined as the lowest intensity at which wave V can be recognized reliably in conventional BAER, >20 dB nHL (3 babies at 25 and 2 at 30 dB nHL), higher intensities (70 and 75 dB nHL, respectively) were also used. This allowed to collect MLS BAER data at the intensity 40 dB or slightly higher above the threshold of each subject, and thus minimize any effect of threshold elevation and slight peripheral hearing loss on MLS BAER components (Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). Two runs of MLS BAER recording were made for each stimulus condition to evaluate reproducibility of the recorded waveforms. Evoked brain responses to 1,500 trains of clicks were preamplified, bandpassed at 100–3000 Hz, and then averaged for each run of recording. During the signal averaging, amplitude artifact rejection was active to eliminate any on-line signals with amplitude exceeding ±25 lV. The tester kept monitoring the ongoing filtered EEG and the running averaged MLS BAER, and manually discontinued sampling whenever there were excessive muscle artefacts on the monitoring oscilloscope. As such, contamination of artefacts to the recorded MLS BAER waveforms was minimized. Sample MLS BAER recordings at different repetition rates of clicks in a normal term baby are shown in Fig. 1. 2.3. Data analysis Analyses of the BAER were conducted blinded to the medical history and clinical data of each baby. The latency of each MLS BAER major wave (I, III, and V) was measured, and the I–V, I–III and III–V interpeak intervals were then calculated, as previously described (Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2010; Wilkinson et al., 2007). A SPSS package (SPSS, Chicago, IL) was used for statistical analyses. We had compared the MLS BAER data between the two institutes (Children’s Hospital of Fudan University and the John Radcliffe Hospital of Oxford University), and found that the results were basically same, without any significant differences. Therefore the data from the two institutes were combined together for detailed statistical analysis. The subject’s data in Table 1 were compared between the NICU babies and the normal babies. Categorical variable (gender) was analysed using
Pearson’s v2 statistic and Fisher’s exact test. Continuous variables (gestational age, postconceptional age, and birthweight, BAER threshold) were compared using Student’s t tests. A 2-tailed value of p < 0.05 was considered statistically significant. Measurements from two replicated MLS BAER recordings to each stimulus condition were averaged for each subject. Mean and standard deviation of each BAER variable at each stimulus condition were obtained for the study and control groups, respectively. Student’s t test was used for comparison between the study and control groups for each MLS BAER variable. The Chi-square test was used for comparison of the abnormal results in the NICU babies between MLS BAER and conventional BAER to examine whether MLS BAER improves the detectability of abnormality in the auditory brainstem. A 2-tailed value of p < 0.05 was considered statistically significant for these data analysis.
3. Results As presented in Table 1, there were no significant differences between the NICU and normal term babies in gender, postconceptional age at time of MLS BAER testing, birthweight and BAER threshold. Only gestational age in the NICU babies was marginally greater, though statistically significant, than in the normal babies. No significant difference was found in hearing levels (i.e. the dB above the thresholds of individual subjects), at which measurements of MLS BAER recordings were made and analysed, between the NICU babies (49.5 ± 6.4 dB nHL) and the normal term babies (49.3 ± 5.5 dB nHL). In both the NICU and normal term babies, major MLS BAER components (waves I, III and V) were all reliably identified at all click rates. No missing MLS BAER waves were found in the normal term babies. Only at the highest click rate 910/s, three NICU babies had less clear wave I, with 2 having less clear wave V. Nevertheless, the three babies had not been included in the study entry and data analysis. 3.1. Wave latencies Fig. 2 shows boxplots of the latencies of waves I (A), III (B) and V (C), respectively, at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in the NICU term babies and the normal term babies. In conventional BAER at 21/s, wave I latency in the NICU babies was similar to that in the normal babies (Fig. 2A). The latencies of waves III and V in the NICU babies were both longer that those in the normal babies (p < 0.05 and 0.01, Figs. 2B and 2C). In MLS BAER, wave I latency in the NICU babies was slightly shorter than those in the normal babies at all 91-910/s clicks, without any statistically significant difference (Fig. 2A). Wave III latency tended to be longer than those in the term babies at all rates, which did not reach statistical significance at 91 and 227/s but differed significantly at higher rates 455 and 910/s (p < 0.05 and 0.05, Fig. 2B). By comparison, wave V latency in the NICU babies was significantly longer than that in the normal babies at all click rates, particularly at the very high rates 455 and 910/s (p < 0.05–0.001, Fig. 2C). 3.2. Interpeak intervals
Fig. 1. Sample MLS BAER waveforms at different repetition rates of clicks, recorded from a normal term baby.
Fig. 3 shows boxplots of the I–V (A), I–III (B) and III–V (C) intervals, respectively, at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in the NICU babies and the normal babies. In conventional BAER at 21/s, all the three intervals in the NICU babies tended to be longer than those in the normal babies, with the I–V and III–V intervals differing significantly between
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
1449
Fig. 2. Boxplot (bold line across the box, median; box, 25th and 75th centile; extensions, the largest and smallest values) of the latencies of wave I (A), III (B), and V (C) at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in normal term babies (Normal) and NICU term babies (NICU). ⁄p < 0.05, ⁄⁄p < 0.01, and ⁄⁄⁄p < 0.001 for comparison between NICU and normal term babies.
Fig. 3. Boxplot (bold line across the box, median; box, 25th and 75th centile; extensions, the largest and smallest values) of the interpeak intervals of I–V (A), I–III (B), and III–V (C) at different click rates in conventional BAER (21/s) and MLS BAER (91-910/s) in normal term babies (Normal) and NICU term babies (NICU). ⁄p < 0.05, ⁄⁄ p < 0.01, and ⁄⁄⁄p < 0.001 for comparison between NICU and normal term babies.
the two groups of babies (p < 0.01 and 0.05, Figs. 3A, and 3C). The III–V/I–III interval ration was similar in the two groups. In MLS BAER, the I–V interval in the NICU babies was significantly longer than in the normal babies at all click rates, and the difference between the two groups was generally increased with increasing click rate (p < 0.05–0.001, Fig. 3A). On the other hand, the I–III interpeak interval in the NICU babies was only slightly longer than in the normal babies, which did not differ significantly at any click rates (Fig. 3B). Similar to the I–V interval, the III–V interval was also significantly longer than in the normal babies
at all click rates, particularly at 455 and 910/s (p < 0.05–0.001, Fig. 3C). The III–V/I–III interval ration in the NICU babies was slightly increased at 91 and 227/s, but was significantly increased at 455 and 910/s (p < 0.001 and 0.001). 3.3. Abnormal rate of interpeak intervals To obtain the rate of neural conduction impairment in the auditory brainstem in the NICU babies we calculated the percentage rate of abnormal increase in the two abnormal MLS BAER intervals,
1450
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
Table 2 Numbers of NICU term babies with abnormal increase in I–V and III–V interpeak intervals at various click rates and statistical comparison of the abnormality in MLS BAER at each click rate (91-910/s) with the abnormality in conventional BAER at 21/s. BAER interval
Subjects
21/s
91/s
227/s
455/s
910/s
Normal criteria NICU abnormal cases (%) X2 p
5.40 11 (10.4%)
5.62 13 (12.3%) 0.19 >0.05
6.12 19 (17.9%) 2.27 >0.05
6.50 33 (31.1%) 13.88 <0.01
6.51 32 (31.1%) 12.87 <0.01
Normal criteria NICU abnormal cases (%) X2 p
2.49 7 (6.6%)
2.69 7 (6.6%) 0.00 >0.05
2.97 12 (11.3%) 1.45 >0.05
3.27 24 (22.6%) 10.92 <0.01
3.26 26 (25.2%) 13.26 <0.01
I–V (ms)
III–V (ms)
Normal criteria refer to the upper limit (mean + 2.5 SD) of a BAER variable in normal term babies. Values exceed the criteria are defined as abnormal.
i.e. III–V and I–V intervals, which reflect neural conduction in the auditory brainstem. As a common practice in BAER, any interval values that exceeded 2.5 standard deviations above the mean measurement in the normal term babies were considered to be abnormal. The results are summarized in Table 2. For the I–V interval, the numbers (and percentage rates) of babies with an abnormal increase at 91, 227, 455 and 910/s clicks were, respectively, 13 (12.3%), 19 (17.9%), 33 (31.1%) and 32 (30.2%). For the III–V interval, the numbers (and percentage rates) of babies with an abnormal increase at the four click rates were, respectively, 7 (6.6%), 12 (11.3%), 24 (22.6%) and 26 (24.5%). All of the babies who had an abnormal increase at 91 and/or 227/s also had an abnormal increase at 455 and/or 910/s. Most of the babies with an abnormal increase in the III–V interval also had an abnormal increase in the I–V interval. Conversely, most of the babies with an abnormal increase in the I–V interval also had an abnormal increase in the III–V interval. As a whole, 38 (35.8%) of the 106 NICU babies had and abnormal increase in the III–V and/or I–V intervals. The noticeable associated perinatal problems in these babies were hypoglycemia, hyperbilirubinemia, and sepsis. By comparison, the abnormal increase in conventional BAER at 21/s clicks was found in only 11 (10.4%) of the 106 NICU babies for the I–V interval, and 7 (6.6%) for the III–V interval. The total number of subjects with the abnormal increase was 13 (12.2%) at 21/s. All these babies also had an abnormal increase in MLS BAER. Comparison of the abnormal increase in I–V and III–V intervals in the NICU babies was made between MLS BAER and conventional BAER using the Chi-square test (Table 2). The abnormal rates for the I–V and III–V intervals at 91 and 227/s in MLS BAER did not differ significantly from those at 21/s in conventional BAER, although the abnormal increase at 227/s was seen in more than cases than at 21/s. At very high rates 455 and 910/s in MLS BAER, however, the abnormal increase in the I–V interval was seen in significantly more cases than at 21/s in conventional BAER (X2 = 13.88 and 12.86, p < 0.01 and 0.01, respectively). Similarly, the abnormal increase in the III–V interval at 455 and 910/s was seen in significantly more cases than at 21/s (X2 = 10.92 and 12.96, p < 0.01 and 0.01, respectively). The total rate of the abnormal increase was seen significantly higher in MLS BAER (35.8%) than in conventional BAER (12.2%, X2 = 16.14, p < 0.01). 4. Discussion The I–V interpeak interval is the major and most widely used BAER variable that reflects neural conduction in the auditory brainstem (Hall III, 2007; Jiang, 2012; Moore, 1987; Moore et al., 1995). It has been regarded as an index of brainstem conduction time. The present MLS BAER study in NICU term babies demonstrated a general increase in the wave V latency, and the I–V and III–V interpeak intervals. The most significant increase in these variables occurred mainly at very high click rates 455 and 910/s clicks. These results
indicate that neural conduction in the auditory brainstem is abnormal or impaired in term babies who receive intensive care, which is more evident following more stressful high-rate stimulation. As reflected by the rate of total abnormal increase in the I–V and III–V intervals, the neural conduction impairment was seen in 35.3% of the NICU term babies we studied. The I–III and III–V intervals are the earlier and later components of the I–V interval, and generally reflect neural conduction at the more peripheral (or caudal) and more central (or rostral) regions of the auditory brainstem, respectively (Jiang, 2012; Jiang et al., 2009). In our NICU term babies, while the III–V and I–V intervals were significantly increased, the I–III interval was only slightly increased at all click rates, without any statistical significance. Apparently, the significant increase in the I–V interval is mainly produced by the significant increase in the III–V interval and, to a less extent, by the slight increase in the I–III interval. The significant increase in the III–V interval also made a main contribution to the increase in wave V latency. Thus, the impaired neural conduction in the auditory brainstem in our NICU term babies occurs mainly at the more central regions of the brainstem, reflected by the increase in the III–V interval and, to a less degree, at the more peripheral regions, reflected by the insignificant increase in the I–III interval. For BAER studies in NICU term babies, there are previous reports in conventional BAER, mainly related to hypoxia–ischemia and hyperbilirubinemia (Akinpelu et al., 2013; Hecox et al., 1981; Newton, 2001; Streletz et al., 1986; Wilkinson and Jiang, 2006). So far, there are very few reports on MLS BAER in term babies, mainly in those who had perinatal hypoxia–ischemia (Jiang et al., 2003, 2010). Here, we compared the present findings in the NICU babies who had perinatal problems other than hypoxia–ischemia with those of the previous MLS BAER studies in term babies who had perinatal hypoxia–ischemia. During the first few days after birth, the babies after hypoxia–ischemia demonstrated a significant increase in the III–V and I–V intervals at all click rates 91-910/s click rates (Jiang et al., 2003, 2010). The I–III interval was also increased, which occurred mainly at the very high rates 455 and 910/s click rates. By comparison, the MLS BAER abnormalities in our NICU term babies were relatively less significant. The major increase in the I–V and III–V intervals occurred mainly at the very high click rates (455 and 910/s), and there was no significant increase in the I–III interval at any click rates. Such differences between the two groups of term babies suggest that that perinatal hypoxia–ischemia and other perinatal conditions exert somewhat differential effects on the neonatal auditory brainstem, and the brainstem impairment due to other perinatal problems is relatively less severe, compared with that due to hypoxia–ischemia. So far, the prevalence of brainstem impairment in term babies in the NICU remains unclear. The present study revealed that 35.3% of the NICU term babies had an abnormal increase in the
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
III–V and/or I–V intervals. This rate of total abnormal increase implies that about one-third of the NICU term babies had neural conduction impairment in the auditory brainstem. We did not conduct detailed analysis of aetiology for the MLS BAER abnormality, because this is not the purpose of this study. However, we did notice the abnormality was mainly seen in the babies who had hypoglycemia, hyperbilirubinemia, and sepsis. These problems are known to be risk factors for neonatal brain damage (AdamsChapman and Stoll, 2006; Burns et al., 2008; Johnson and Bhutani, 2011; Mallard and Wang, 2012; Montassir et al., 2009; Shapiro, 2010; Tam et al., 2012; Yalnizoglu et al., 2007). A clinical implication of this high prevalence of brainstem conduction impairment is that some neuroprotective measures may be required for NICU term babies, particularly those with had hypoglycemia, sepsis and hyperbilirubinemia, to protect the neonatal brainstem and reduce any impairment to neural conduction in the brainstem. The rates of abnormal increase in the III–V and I–V intervals in our NICU babies were generally seen more at higher than at lower click rate. The abnormal rates at 91 and 227/s in MLS BAER were similar or slightly higher than at 21/s in conventional BAER. However, the abnormal rates at 455 and 910/s were much higher than those at the lower rates 91 and 227/s, and significantly higher than in conventional BAER. These findings provide further evidence that high-rate stimulation used in MLS BAER enhances the detection of brainstem auditory impairment in babies with perinatal problems (Jiang, 2012; Jiang and Chen, 2014; Jiang et al., 2003, 2005, 2007, 2010; Wilkinson et al., 2007). In a previous short report, we briefly described MLS BAER in term babies who had various perinatal problems or conditions, and found that these babies were at risk of brainstem auditory dysfunction (Jiang et al., 2012). In the present study, we carried out a detailed analysis of both MLS BAER and conventional BAER in NICU term babies. In addition to detection of neural conduction impairment in the auditory brainstem, we also determined the prevalence of the impairment, and, in particular, as assessed if the MLS technique improves early detection of the impairment. We found that the BAER abnormalities that are indicative of brainstem conduction impairment were particular significant at high-rate stimulation in MLS BAER. The abnormal rates at 455 and 910/s in individual babies were significantly higher than those in conventional BAER at 21/s. As a whole, 35.8% of the NICU babies had abnormal increase in the III–V and/or I–V intervals in MLS BAER, which was significantly more than the 12.2% in conventional BAER. Thus, about one-third of NICU term babies have neural conduction impairment in the brainstem, suggesting that term babies in the NICU are at risk of brainstem conduction impairment. Compared with conventional BAER, MLS BAER, typically at high click rates (455 and 910/s), enhances the detection of brainstem conduction impairment for babies in the NICU. Over the last decade, the BAER recorded and processed using the MLS technique has been documented to enhance the detection of neuropathology that involves in the brainstem auditory pathway in neonatal audiology and neurology. MLS BAER recorded with high-rate stimulation (particularly 455/s) could become routine for clinical use (Jiang, 2012). Nevertheless, there is currently a lack of commercially available evoked potential systems that are equipped with the MLS technique. The development of such evoked potential systems in the near future is warranted. Acknowledgements Mrs. Dorathea Brosi did excellent work in recruiting of subjects and collecting of data. Drs. Rong Yin and Lili Ping made major contribution to recruiting subjects, collecting and analysing data. Professor Andrew Wilkinson is gratefully appreciated for his
1451
support of this project. The research was supported by the WellChild, and Bridging Support Fund and Medical Sciences Division of Oxford University, UK. The author has no potential conflicts of interest to be disclosed. References Adams-Chapman I, Stoll BJ. Neonatal infection and long-term neurodevelopmental outcome in the preterm infant. Curr Opin Infect Dis 2006;19:290–7. Akinpelu OV, Waissbluth S, Daniel SJ. Auditory risk of hyperbilirubinemia in term newborns: a systematic review. Int J Pediatr Otorhinolaryngol 2013;77:898–905. Bell SL, Smith DC, Allen R, Lutman ME. The auditory middle latency response, evoked using maximum length sequences and chirps, as an indicator of adequacy of anesthesia. Anesth. Analg. 2006;102:495–8. Burns CM, Rutherford MA, Boardman JP, Cowan FM. Patterns of cerebral injury and neurodevelopmental outcomes after symptomatic neonatal hypoglycemia. Pediatrics 2008;122:65–74. Hall III JW. ABR: pediatric clinical application and populations. In: Hall III JW, editor. New handbook of auditory evoked responses. Boston: Pearson Education; 2007. p. 313–65. Hecox K, Cone B, Blaw M. Brainstem auditory evoked response in the diagnosis of pediatric neurologic diseases. Neurology 1981;31:832–9. Jiang ZD. Maximum length sequence technique improves detection of neuropathology involving infant brainstem. In: Lawson PN, McCarthy EA, editors. Pediatric neurology. New York: Nova Science Publishers; 2012. p. 1–38. Jiang ZD, Chen C. Impaired neural conduction in the auditory brainstem of high-risk very preterm infants. Clin Neurophysiol 2014;125:1231–7. Jiang ZD, Brosi DM, Wang J, Xu X, Chen GQ, Shao XM, et al. Time course of brainstem pathophysiology during first month in term infants after perinatal asphyxia, revealed by MLS BAER latencies and intervals. Pediatr Res 2003;54:680–7. Jiang ZD, Brosi DM, Li ZH, Chen C, Wilkinson AR. Brainstem auditory function at term in preterm babies with and without perinatal complications. Pediatr Res 2005;58:1164–9. Jiang ZD, Xu X, Brosi DM, Shao XM, Wilkinson AR. Suboptimal function of the auditory brainstem in term neonates with transient low Apgar scores. Clin Neurophysiol 2007;118:1088–96. Jiang ZD, Brosi D, Wu YY, Wilkinson AR. Relative maturation of the peripheral and central regions of the auditory brainstem from preterm to term and the influence of preterm birth. Pediatr Res 2009;65:657–62. Jiang ZD, Brosi DM, Wilkinson AR. Differences in impaired brainstem conduction between neonatal chronic lung disease and perinatal asphyxia. Clin Neurophysiol 2010;121:725–33. Jiang ZD, Brosi DM, Yin R, Wilkinson AR. Term neonates receiving intensive care at high risk of brainstem auditory impairment. Arch Dis Child-Fetal 2012;97. F359-61. Jiang ZD, Zhou Y, Yin R, Wilkinson AR. Amplitude reduction in brainstem auditory response in term infants under neonatal intensive care. Clin Neurophysiol 2013;124:1470–6. Jirsa RE. Maximum length sequences-auditory brainstem responses from children with auditory processing disorders. J Am Acad Audiol 2001;12:155–64. Johnson L, Bhutani VK. The clinical syndrome of bilirubin-induced neurologic dysfunction. Semin Perinatol 2011;35:101–13. Lasky RE. Rate and adaptation effects on the auditory evoked brainstem response in human newborns and adults. Hear Res 1997;111:165–76. Majnemer A, Rosenblatt B. Evoked potentials as predictors of outcome in neonatal intensive care unit survivors: review of the literature. Pediatr Neurol 1996;14:189–95. Majnemer A, Rosenblatt B. Prediction of outcome at school age in neonatal intensive care unit graduates using neonatal neurologic tools. J Child Neurol 2000;15:645–51. Mallard C, Wang X. Infection-induced vulnerability of perinatal brain injury. Neurol Res Int 2012;2012:102153. Montassir H, Maegaki Y, Ogura K, Kurozawa Y, Nagata I, Kanzaki S, Ohno K. Associated factors in neonatal hypoglycemic brain injury. Brain Dev 2009;31:649–56. Moore JK. The human auditory brain stem as a generator of auditory evoked potentials. Hear Res 1987;29:33–43. Moore JK, Perazzo LM, Braun A. Time course of axonal myelination in the human brainstem auditory pathway. Hear Res 1995;87:21–31. Nagle S, Musiek FE. Morphological changes in the middle latency response using maximum length sequence stimuli. J Am Acad Audiol 2009;20:492–502. Newton V. Adverse perinatal conditions and the inner ear. Semin Neonatol 2001;6:543–51. Shapiro SM. Chronic bilirubin encephalopathy: diagnosis and outcome. Semin Fetal Neonatal Med 2010;15:157–63. Streletz LJ, Graziani LJ, Branca PA, Desai HJ, Travis SF, Mikaelian DO. Brainstem auditory evoked potentials in fullterm and preterm newborns with hyperbilirubinemia and hypoxemia. Neuropediatrics 1986;17:66–71. Tam EW, Haeusslein LA, Bonifacio SL, Glass HC, Rogers EE, Jeremy RJ, Barkovich AJ, Ferriero DM. Hypoglycemia is associated with increased risk for brain injury and adverse neurodevelopmental outcome in neonates at risk for encephalopathy. J Pediatr 2012;161:88–93.
1452
Z.D. Jiang / Clinical Neurophysiology 126 (2015) 1446–1452
Wilkinson AR, Jiang ZD. Brainstem auditory evoked response in neonatal neurology. Semin Fet Neonatol Med 2006;11:444–51. Wilkinson AR, Brosi DM, Jiang ZD. Functional impairment of the brainstem in infants with bronchopulmonary dysplasia. Pediatrics 2007;120:362–71.
Yalnizoglu D, Haliloglu G, Turanli G, Cila A, Topcu M. Neurologic outcome in patients with MRI pattern of damage typical for neonatal hypoglycemia. Brain Dev 2007;29:285–92.