Auditory Brainstem Response, Middle Latency Response, and Late ...

J Am Acad Audiol 13 : 367-382 (2002)

Auditory Brainstem Response, Middle Latency Response, and Late Cortical Evoked Potentials in Children with Learning Disabilities Suzanne C. Purdy* Andrea S. Kelly,' Merren G. Davies` Abstract Auditory evoked potentials (AEPs) and behavioral tests were used to evaluate auditory processing in 10 children aged 7 to 11 years who were diagnosed as learning disabled (LID) . AEPs included auditory brainstem responses (ABRs), middle latency responses (Iv1LRs), and late cortical responses (P1, N1, P2, P3) . Late cortical responses were recorded using an active listening oddball procedure . Auditory processing disorders were suspected in the LID children after a psychologist found phonologic processing and auditory memory problems . A control group of 10 age- and gendermatched children with no hearing or reported learning difficulties was also tested . Teacher ratings of classroom listening and SCAN Competing Words and Staggered Spondaic Word scores were poorer in the LD children . There were minor ABR latency differences between the two groups . Wave Na of the Iv1LR was later and Nb was smaller in the LID group . The main differences in cortical responses were that P1 was earlier and P3 was later and smaller in the LD group . Key Words : Auditory evoked response, auditory processing disorder, learning disability Abbreviations : ABR = auditory brainstem response ; APD = auditory processing disorder ; CAEP = cortical auditory evoked potential ; CAPD = central auditory processing disorder ; LD = learning disabled ; Iv1LR = middle latency response ; Iv1NIN = mismatch negativity ; WISC-R = Wechsler Intelligence Scale for Children-Revised Surnario Se utlizaron potenciales evocados auditivos (AEP) y pruebas conductuales para evaluar el procesamiento auditivo en 10 ninos con edades entre 7 y 11 anos, quienes fueron diagnosticados como portadores de trastornos del aprendizaje (LID) . Los AEP incluyeron respuestas auditivas de tallo cerebral (ABR), respuestas de latencia media (MLR) y respuestas corticales tardias (P1, N1, P2, P3) . Las respuestas corticales tardias fueron registradas utilizando un procedimiento inusual de escucha activa . Se sospecharon trastornos de procesamiento auditivo en los ninos LID despues de que un psicologo encontro problemas fonologicos y de memoria auditiva . Se evaluo tambien un grupo de control de 10 ninos ordenados por edad y genero, quienes no presentaban dificultades para escuchar o para aprender . La estimacion del maestro sobre la habilidad de escuchar en clase y los resultados de la prueba de Palabras SCAN de Competencia y la prueba SSW fueron peores en los n0os con LID . Existieron diferencias menores de latencia en el ABR entre los dos grupos . La onda Na en las IVILR fue mas tardia y la Nb fue mas pequena en el grupo LID . Las principales diferencias en las respuestas corticales fueron clue la P1 aparecia mas temprano y la P3 mas tarde y mas pequena en el grupo LID . Palabras Clave : Respuestas evocadas auditivas, trastornos de procesamiento auditivo, trastornos del aprendizaje Abreviaturas : ABR = respuesta auditiva del tallo cerebral ; APD = trastornos de procesamiento auditivo ; CAEP = potencial evocado auditivo cortical ; CAPD = trastornos central de procesamiento auditivo ; LID = discapacidad para el aprendizaje ; MLR = respuesta de latencia media ; IVIIV1N = negatividad desigual ; WISC = Escala Revisada Wechsler de Inteligencia para Ninos *National Acoustic Laboratories, Sydney, Australia: 'Discipline of Audiology, The University of Auckland, Auckland, New Zealand, 'Cochlear Corp ., Sydney, Australia

Reprint requests : Suzanne C . Purdy, National Acoustic Laboratories, 126 Greville Street, Chatswood, NSW 2067, Australia

367

Journal of the American Academy of Audiology/Volume 13, Number 7, July/August 2002

any studies have used electrophysiologic measures to objectively investigate auditory processing . Unlike speech and other behavioral auditory processing tests, auditory evoked potentials can be recorded regardless of a child's developmental age or language, motivation, or attention level. Auditory processing from the level of the eighth nerve to the auditory cortex has been investigated using the auditory brainstem response (ABR), middle latency response (MLR), late cortical, P3, and mismatch negativity (MMN) evoked responses. ABR, MLR, and the P1-N1-P2 late cortical responses are "obligatory" responses that depend on the physical properties of the stimulus . P3 and MMN are "discriminative" responses recorded using an oddball stimulus paradigm in which an infrequent, deviant stimulus is interspersed randomly among a frequently occurring standard stimulus . P3 is also referred to as "P300" or "P3b" (Stapells, 2002) and is usually recorded using an active listening paradigm with the subject responding to the deviant stimulus, whereas MMN is a preattentive response that is usually recorded with the subject ignoring the stimuli (Schroger and Wolff, 1998). The American Speech-Language-Hearing Association (ASHA) Task Force on Central Auditory Processing (1996) concluded that electrophysiologic measures are useful for the diagnosis of central auditory processing disorders (CAPDs) but acknowledged that further research is needed to establish the clinical utility of middle and late evoked potentials . More recently, the Bruton Conference held at the Callier Center in Dallas (Jerger and Musiek, 2000 ; Chermak, 2001) produced the recommendation that a minimal test battery for the diagnosis of auditory processing disorders (APDs) in school-aged children should include ABR and MLR testing . The P3 event-related response was included in the list of optional procedures that are potentially useful for strengthening the diagnosis of APD. ABR is well understood and is used routinely for the detection of brainstem and eighth nerve lesions (ASHA, 1996). An abnormal ABR in a child thought to have APD can be indicative of neuropathology requiring medical intervention (e .g., Musiek et al, 1991). There are many reports in the literature dating back to the 1970s of ABR abnormalities in patients with sensory and motor neuropathies affecting the auditory system (e .g ., Satya-Murti et al, 1979). An abnormal ABR combined with intact otoacoustic emissions now leads to a diagnosis of auditory

M

368

neuropathy/dyssynchrony (Starr et al, 2000 ; Berlin, 1999). ABR abnormalities have been reported in children with learning problems (Sohmer and Student, 1978 ; Greenblatt et al, 1983 ; Jerger et al, 1987 ; Cunningham et al, 2001), speech and language disorders (Mason and Mellor, 1984 ; Milicic et al, 1998), and auditory processing deficits (Yencer, 1998). In contrast, Marosi and colleagues (1990) did not see ABR latency or amplitude differences in their study of primary schoolchildren with a learning disability. Thus, there is conflicting evidence for ABR abnormalities in school-aged children with APD who do not have auditory neuropathy or any evidence of auditory neural pathology. Some years ago, Jerger and his colleagues (1988) described MLR as potentially "the single most important auditory evoked response in terms of its ability to help us identify and understand CAPD ." MLR abnormalities have been found in children with learning or speech/language disabilities (Squires and Hecox, 1983 ; Mason and Mellor, 1984 ; Jerger and Jerger, 1985 ; Jerger et al, 1987, 1988 ; Fifer and SierraIrizarry, 1988 ; Arehole et al, 1995 ; Milicic et al, 1998). The MLR is also affected in adults with APD (Jerger et al, 1991 ; Marvel et al, 1992), cortical lesions (Ozdamar et al, 1982 ; ShehataDieler et al, 1991 ; Kraus and McGee, 1993 ; Musiek and Lee, 1997 ; Musiek et al, 1999 ; Setzen et al, 1999 ; Akkuzu et al, 2001), and multiple sclerosis (Jerger and Jerger, 1985 ; Stach and Hudson, 1990 ; Celebisoy et al, 1996). MLR appears to be a sensitive indicator of central nervous system pathology involving the auditory system. Many of the MLR investigations in children have been case studies, however; hence, there are still relatively few data on the nature of MLR abnormalities in children with APD. MLRs can be reliably recorded in young children and infants if appropriate stimulus and recording parameters are used (Tucker and Ruth, 1996). In younger children, the MLR typically consists of a broad, late Pa, followed by a negative trough, Nb (Musiek et al, 1988 ; Suzuki and Hirabayashi, 1987). Pb is typically absent . Pa and Nb latencies approximate adult values by 8 to 11 years (Suzuki and Hirabayashi, 1987). The change in waveform morphology with age has been attributed to differential maturation of multiple MLR generators . Kraus and McGee (1993) hypothesized that the earlier-maturing mesencephalic reticular formation dominates the younger child's MLR, and adult-like MLR morphology occurs with maturation of the thalamocortical pathways .

Evoked Potentials in Learning-Disabled Children/Purdy et al

In adults, Pa is largest at the vertex (Cz) and symmetrically distributed over the temporal lobes (Ozdamar and Kraus, 1983 ; Kraus and

McGee, 1988 ; Cacace et al, 1990) . A trend for broader and later MLR peaks at more lateral recording sites has been seen in adults (Deiber et al, 1988) . Mason and Mellor (1984) found that differences between MLR amplitudes in children with speech and language disorders and control group children were greater at hemispheric (T3, C3, C4, T4) electrode sites than at the vertex . The effects of neuropathology in adults may also be more evident when recordings are made over temporoparietal regions (Kileny et al, 1987) . A slight left ear Pa amplitude advantage has been reported in adults (Cacace et al, 1990), but the opposite has also been found (Deiber at al, 1988) . Kadoya and colleagues (1988) found that Pa amplitudes recorded over the hemisphere contralateral to the test ear in guinea pigs and adult humans were significantly larger than those recorded over the ipsilateral hemisphere . Ozdamar and Kraus (1983) also noted this contralateral advantage in their data . Marvel and colleagues (1992) found significant hemispheric asymmetry in MLR topographic brain maps in elderly listeners with suspected APD, whereas control subjects had symmetric MLR distributions . Musiek and his colleagues have suggested that MLR hemispheric and ear asymmetries are more reliable indicators of APD than absolute MLR latencies or amplitudes (Musiek et al, 1994 ; Cbermak and Musiek, 1997 ; Musiek et al, 1999) . Normal maturational changes and ear and recording electrode effects on the MLR need to be better characterized to establish normative data for MLR ear and hemisphere differences in children . The cortical P1-N1-P2 evoked potentials that occur within about 300 msec after stimulus onset in adults depend primarily on the physical properties of the stimulus . Discriminative cortical potentials elicited using an oddball stimulus paradigm result from either preconscious (e .g ., MMN) or conscious (e .g., P3b) perception of a change in the auditory stimulus and hence have been referred to as "processing-contingent potentials" (Stapells, 2002) . Both obligatory and discriminative potentials have been investigated as objective indices of central auditory function since they correlate well with

perception and discrimination of auditory stimuli (Hyde, 1997 ; Stapells, 2002) and are abnormal in individuals with brain lesions affecting auditory cortical regions (e .g ., Hood et al, 1994) . Cortical auditory evoked potential (CAEP) gen-

erators include primary auditory cortex, auditory association areas, frontal cortex, and subcortical regions (see reviews by Picton et al,

1999, and Stapells, 2002) . Although these responses are present in infants (Steinschneider et al, 1992), they undergo considerable maturational changes, and some cortical potentials may not be fully mature until close to adulthood (Ponton et al, 2002) . In infants and young children, CAEPs are dominated by P1, which becomes earlier and smaller as N1 and P2 begin to emerge in the waveform at about 8 to 10 years of age (Sharma et al, 1997 ; Ponton et al, 2002) . These maturational changes complicate the use of CAEP for diagnosis ofAPD since more extensive normative data are required than for the earlier-maturing evoked potentials . The scalp distribution of P1, N1, and P2 is normally symmetric with maximal amplitude near the vertex (Picton et al, 1999), but, as for MLR wave Pa, a contralateral hemisphere advantage (earlier latencies, greater amplitudes) has been reported in adults (Verkindt et al, 1995 ; Picton et al, 1999 ; Ponton et al, 2002) . The amplitude, latency, and scalp distribution of the discriminative cortical potential P3 depends on subject age as well as state of arousal and attention (Squires et al, 1975 ; Pearce et al, 1989 ; Johnstone et al, 1996 ; Oades et al, 1997 ; Stapells, 2002) .

Both obligatory and discriminative CAEPs have been investigated in children and adults thought to have APD. Researchers have found a variety of Pl-Nl-P2 and P3 abnormalities, including increased absolute and interwave latencies (Jirsa and Clontz, 1990 ; Arehole, 1995 ; Tonnquist-Uhlen, 1996a, 1996b; Bruneau et al, 1999 ; Seri et al, 1999), reduced N1 amplitude (Bruneau et al, 1999 ; Seri et al, 1999 ; Cunningham et al, 2001 ; Wioland et al, 2001), reduced P3 amplitude (Jirsa and Clontz, 1990), increased (Bernal et al, 2000) or decreased (Tonnquist-Uhlen, 1996b) P2 and N2 amplitudes, and increased hemispheric asymmetry (Mason and Mellor, 1984 ; Jerger et al, 1991). A number of studies have shown reduced MMN amplitudes (and sometimes increased latencies) in adults and children with speech/language, reading, or learning difficulties (Kraus et al, 1993, 1996 ; Korpilahti and Lang, 1994 ; Schulte-Korne et al, 1998, 1999, 2001 ; Baldeweg et al, 1999 ; Bradlow et al, 1999). Further research is required, however, before MMN can be regarded as a clinical tool for APD assessment owing to the small amplitude and high variability of the response (Picton et al, 2000 ; Dalebout and Fox, 2001 ; McGee et al, 2001). 369

Journal of the American Academy of Audiology/Volume 13, Number 7, July/August 2002

In the present study, ABR, MLR, P1-N1-P2, and P3 responses were recorded in children with a learning disability whose phonologic processing and auditory memory problems led to a suspicion of APD. Electrophysiologic responses, scores on behavioral tests of APD, and teacher assessments of classroom listening were compared with the results for a control group of children to determine which tests distinguished the two groups of children and the pattern of electrophysiologic and behavioral results for the children with a learning disability (LD) . METHOD Subjects Ten children (6 boys, 4 girls) aged 7 to 11 years (mean = 9.3 years, SD = 1.5 years) were recruited with the assistance of the University ofAuckland's Learning Assessment Centre . The children had been assessed by a psychologist and had been diagnosed as having an LD and a possible APD. Children with no significant attention deficit whose Weschler Intelligence Scale for Children-Revised (WISC-R) scores were within normal limits were invited to participate in the research . The psychologist made a diagnosis of possible APD if the children showed difficulties with auditory blended phoneme analysis and one or more of the following tasks: (1) auditory immediate memory and/or attention, (2) auditory and visual immediate memory and/or attention, (3) auditory discrimination of speech phonemes, and (4) auditory long-term memory. These auditory skills were assessed using the WISC-R digit span and arithmetic subtests, the Learning Efficiency Test (LET) of auditory and visual immediate memory and auditory long-term memory (Webster, 1981), the Detroit Tests of Learning Aptitude-2 (DTLA-2) Following Oral Directions subtest (Hammill, 1985), the revised Lindamood Auditory Conceptualization (LAC) tests of speech phoneme auditory discrimination and auditory blended phoneme analysis (Lindamood and Lindamood, 1979), and the Selective Reminding Test (Buschke and Fuld, 1974). The control group consisted of 10 children recruited via friends and colleagues who were matched for age and gender with the experimental group (6 boys, 4 girls ; mean age = 9 .2 years, SD =1 .6 years) whose parents reported no history of learning or hearing difficulties . Children in both groups had normal pure-tone thresholds (