Approximately 1.4 million individuals suffer from traumatic brain injury (TBI) each year in the United States (Langlois, Rutland-Brown, & Thomas, 2006). TBI ranges in severity from mild (i.e., mild concussion also known as mild traumatic brain injury) to severe (i.e., an extended period of unconsciousness or amnesia after the injury). Even minor head injuries can cause a wide range of cognitive deficits (impaired attention, slowed information processing, impaired memory), expressive (word retrieval) and receptive (interpreting figurative language) language deficits, metacognitive deficits (self-monitoring, planning, organizing, and problem-solving), psychological deficits (personality changes including increased irritability, anxiety, depression, and social inappropriateness), and sensory deficits (auditory and visual).

By its very nature, head injury (HI) may cause damage to the central nervous system; therefore, the central auditory nervous system (CANS) must be considered in evaluating the patient with HI. Although it seems obvious, audiological evaluation of patients with HI often is overlooked. Head injury, like other disorders that affect the CANS such as strokes, tumors, and degenerative and developmental disorders, compromises the higher auditory system. Despite a unique onset and evolution of problems, HI bears many similarities to other CANS disorders.

Early identification of (central) auditory processing disorder ([C]APD), a disorder of the CANS, in patients with HI is crucial to improving central auditory processing as well as overall auditory function so that patients can benefit from interventions designed to improve cognitive and social functioning in the home, workplace, and school.

Causes of Head Injury

Among non-military personnel, the leading causes of HI/TBI are falls (especially among children under age 4 and adults age 75 and older), motor vehicle accidents, and sports injuries (Langlois et al., 2006). Blasts are a leading cause of TBI for active-duty military personnel in war zones (Defense and Veterans Brain Injury Center, 2005).

In many patients with head injury, both peripheral and central auditory systems are involved (Hall, Huangfu, Gennarelli, Dolinskas, Olson, & Berry, 1983). In fact, (C)APD is seen in more than 50% of adult and pediatric patients with TBI (Bergemalm & Lyxell, 2005; Flood, Dumas, & Haley, 2005). The number of individuals with possible TBI-induced (C)APD has increased as significant numbers of veterans (10%–20%) with head injuries have returned home from Iraq and Afghanistan (Defense & Veterans Brain Injury Center, Okie, 2005; Warden et al., 2005).

Assessment of the auditory and vestibular systems in cases of HI is warranted given the incidence of hearing loss, vertigo, tinnitus, and (C)APD associated with HI (Griffith, 1979; Wennmo & Svensen, 1989). Audiological evaluation is essential to ascertain the scope of system deficits and to maximize treatment effectiveness in rebuilding patients' lives (Musiek & Chermak, 2008). Although 68% of patients with TBI demonstrated abnormalities on one or more peripheral or central auditory measures (Bergemalm & Borg, 2001), the CANS is not always involved in HI. It is important, therefore, that the sensitivity of central audiologic procedures be determined on HI cases with known locus and degree of involvement.

Physiology of Comorbidity

(C)APD involves deficits in the neurobiological activity underlying perceptual (i.e., neural) processing of auditory stimuli (ASHA, 2005). Disruption of neural function in the CANS in HI is secondary to pathophysiologic factors such as immediate mechanical displacement and torquing of neural tissue, stretching and tearing of neurons and blood vessels, and later, edema, reduced circulation, demyelinization, and overall degeneration of affected neural substrate (Werner & Engelhard, 2007).

The brain is particularly vulnerable to shear and stress waves resulting from blasts, which can cause TBI due to diffuse axonal injuries occurring most often in the fronto-temporal areas, the internal capsule, deep grey matter (i.e., basal ganglia), upper brainstem, and corpus callosum (Taber, Warden, & Hurley, 2006). Contusions are often seen in the superficial grey matter of the (anterior, inferior, lateral) frontal and temporal lobes and subdural hematomas occur most frequently at the frontal and parietal convexities (Taber et al., 2006). With possible exception of the basal ganglia, all of these brain regions potentially involve auditory neural tissue.

Evaluating Patients with Head Injury

Radiology

Structural lesions associated with head trauma can directly impinge upon and potentially disrupt the CANS as do more focal disease processes such as strokes, tumors, and the plaque formation of multiple sclerosis (Meyers, Roberts, Bayless, Volkert, & Evitts, 2002). In some cases, the neural injuries caused by the primary injuries (i.e., contusions, etc.) and the secondary damage (i.e., ischemia, hypoxia, etc.) are detected using brain imaging technology. In many cases, however, imaging techniques and medical procedures fail to detect the neural damage (Kaipio et al., 2000; Musiek, Baran, & Shinn, 2004), suggesting that auditory testing of patients with minor head injuries should be conducted even in the absence of radiological evidence on brain imaging procedures.

Test Battery

CANS involvement can be revealed through sensitive electrophysiological and behavioral tests of central auditory function. After the status of the peripheral auditory system is determined and a comprehensive case history is obtained, a central auditory test battery that includes sensitive and specific tests should be conducted. Given the complexity and redundancies of the CANS, a test battery should be employed to assess different processes and various levels and loci of CANS function (ASHA, 2005). Because HI may impact the auditory system from periphery to cortex, the optimal test battery should examine the brainstem, sub-cortex, cortex, and corpus callosum. In practice, some compromise may be necessary to maintain test sessions as short as possible (and with frequent rest periods) due to fatigue and other confounds common in the HI population.

Non-auditory Confounds

The wide range of attention, memory, language, self-monitoring, and psychological deficits associated with TBI present potential confounds that can influence audiologic assessment. Audiologists must carefully select tests that have a low language level and present a low cognitive load to reveal CANS deficits rather than other non-auditory TBI-related deficits. Tests with documented sensitivity to CANS dysfunction that use non-verbal or simple verbal stimuli (such as gap detection, masking level differences [MLDs], dichotic tests, and pattern perception tests) and tests involving non-verbal response modes (e.g., gap detection, MLDs, etc.) may be more appropriate for patients with compromised cognition and/or language. To minimize confounds, it may be appropriate to modify test procedures (e.g., administer half of the list or half of the test items) or even repeat a test, if doing so is not contraindicated by fatigue or attention. Electrophysiological procedures may be particularly useful in evaluating patients with HI because minimal patient cooperation or participation is required. Audiologists should strive to administer the minimum number of tests necessary to reach a definitive diagnosis, given attention, concentration, and memory deficits and the potential for fatigue.

Peripheral Hearing Loss

Peripheral hearing loss affects an individual's performance on most central auditory tests. Evaluation of central auditory processing in the presence of peripheral hearing loss is possible, however, by employing certain testing strategies and administering tests less influenced by peripheral hearing loss. Audiologists should select tests and procedures that incorporate stimuli that are less influenced by hearing loss (e.g., dichotic digits, evoked potentials with broad-band stimuli such as clicks or tone-bursts), and administer central tests and procedures employing frequencies (i.e., typically low frequencies) where normal hearing sensitivity is present (e.g., pattern tests and evoked potentials). Audiologists also may take advantage of interpretive strategies when analyzing test outcomes. For example, the finding of asymmetrical central test results combined with symmetrical peripheral hearing sensitivity and speech recognition often indicates a neurological focus resulting from the HI.

Interpreting Test Results

Patient-referenced, intra-subject comparisons (e.g., laterality or interaural differences) can solidify interpretation of behavioral test results. On behavioral tests, strong laterality effects point to auditory involvement; bilaterally symmetrical results, however, should be interpreted cautiously. Finding a laterality effect on a central test in the presence of symmetrical peripheral function suggests central auditory dysfunction. By contrast, poorer performance on tests toward the end of a diagnostic session or poor response reliability suggests fatigue or reduced motivation or attention rather than a true manifestation of CANS dysfunction.

(C)APD and Head Injury

A number of studies have documented abnormal auditory brainstem response (ABR), especially in the acute phase of head injury (e.g., Abd Al-Hady et al., 1990; Greenberg et al., 1981; Hall, Huang-fu, & Gennarelli, 1982). Fewer studies have examined the performance of patients with HI on the later auditory evoked responses or on behavioral tests of central auditory function (Musiek et al., 2004). Nonetheless, these investigations clearly have implicated brainstem compromise in a number of patients with TBI, as well as compromise in the thalamocortical and other regions of the CANS and abnormal performance on monaural low-redundancy tests (such as time-compressed speech, speech in competition), dichotic speech tests, click fusion, and temporal pattern tests (Bergemalm & Borg, 2001; Lackner & Teuber, 1973; Musiek et al., 2004).  Central auditory test results from a patient with a head injury (i.e., left pontine contusion) resulting from a motor vehicle accident are shown in Figure 1 [PDF]. As expected in cases of low brainstem lesions, ipsilateral deficits were seen both in the ABR and on the filtered speech and dichotic listening behavioral tests, as was depressed binaural interaction performance.

Psychophysical/Behavioral Test Findings

Psychophysical central auditory tests have been shown to reveal CANS involvement in TBI. A number of behavioral measures have been used, although dichotic listening has been most frequently examined in patients with HI. In fact, Richardson, Springer, Varney, Struchen, and Roberts (1984) recommended routine use of dichotic testing in evaluation for closed head trauma. Bergemalm and Borg (2001) found that 24% of patients with TBI failed a distorted (interrupted) speech test and 4% of patients failed a phase audiometry (interaural time difference) test at seven to 11 years after their injury. Meyers et al. (2002) reported 100% specificity and 60% sensitivity for mildly brain-injured patients and 80% sensitivity for more severely brain-injured patients using a dichotic speech test (Dichotic Word Listening Test). Poor dichotic listening performance, specifically left-ear deficits, was found to correlate with posterior corpus callosum damage in pediatric patients with HI (Benavidez et al., 1999). The degree of ear asymmetry in dichotic listening has been shown to vary in direct relationship to the severity of HI (Levin et al., 1989).

Electrophysiological Procedures

Although a number of electrophysiological procedures are relevant to diagnosing head injury, the following is an overview of three primary procedures: ABR, the middle latency response (MLR), and the late potentials (N1, P2, P3). As is true for psychophysical procedures, these electrophysiological procedures are sensitive to compromise of the auditory neural substrate, which may or may not be involved in all cases of HI.

ABR. The ABR has a long history of use in patients with HI. Because the ABR is generated by the auditory nerve and brainstem tracts, this anatomical region is best assessed by this procedure. Three major studies have converged on an approximate 50% true positive or hit rate for ABR in primarily mild HI (Gaetz & Bernstein 2001; Bergemalm & Borg, 2001; Rowe & Carlson, 1980), and a few other studies have reported a slightly poorer hit rate (Schoenhuber & Gentilini, 1986; McClelland et al.,1994). The ABR diagnostic index presenting the highest sensitivity is the I–V interwave interval, which reflects primarily the central conduction time (Bergemalm & Borg, 2001). Although high-repetition-rate ABR has not always provided a diagnostic advantage in other populations, it appears to be a worthwhile procedure in patients with HI (Soustiel et al., 1995). If so, then maximum length sequences (MLS) techniques, which permit high rates of stimulation using the ABR, should be investigated.

MLR. The MLR generators are located in the thalmocortical auditory tracts of the brain. These generators are rostral to the ABR generators; therefore, the MLR is an important supplement to the ABR in evaluating HI. The MLR can be recorded concurrently with the ABR, providing an efficient diagnostic advantage in locating the site(s) of involvement. Although the MLR has not been investigated as often as the ABR with patients with HI, studies have documented significant differences in Na, Pa amplitude and latency between controls and patients with minimal HI (Drake et al., 1996; Soustiel et al., 1995). Totally absent MLRs were reported in 30% (six of 20) of patients with HI. More research on the MLR in HI is indicated, especially given the efficiency in measuring the MLR concurrently with the ABR. 

Late Evoked Potentials (N1, P2, P3). The N1 and P2 late potentials are generated by the primary auditory cortex and nearby regions. The P300 is likely generated in a number of areas, including the primary auditory areas. The N1, P2 track record for utility in HI evaluations is somewhat mixed. Drake and colleagues (1996) reported the N1 to be delayed significantly for patients with mild HI compared to controls. Jones et al. (2000) found abnormal N1, P2 responses in almost 90% of patients with HI who had been comatose. Conversely, Harris and Hall (1990) and Segalowitz et al. (2001) reported no significant difference in N1 and P2 responses from patients with HI relative to control groups.

By contrast, more consistent findings have been reported for the P300 in patients with HI. The P300 can be evoked by different sensory stimulation, allowing for multimodal comparisons and information regarding the extent of involvement across modalities (Lew et al., 2004). Moreover, P300 may be one of the most sensitive evoked potentials in cases of HI. For example, in one of the most striking and cited studies, Segalowitz et al. (2001) compared college students who previously had suffered minimal HI with a control group of students who had no history of HI. The HI group demonstrated significantly smaller P300s for four different "oddball” paradigm tasks. Other studies also have shown reduced amplitudes and delayed latencies for P300 in patients with mild HI compared to control groups (Segalowitz et al. 1995; Solbakk et al., 2000).

More Accurate Diagnoses

Head injury involving TBI can result in significant central auditory deficits even in the absence of radiologic evidence and in the absence of any obvious deficits within the peripheral auditory system (Musiek et al., 2004). Psychophysical tests and electrophysiological procedures with the high degree of sensitivity and specificity needed to diagnose (C)APD can be used to assess the CANS, independent of the specific underlying disorder (e.g., multiple sclerosis, aphasia, tumor, or HI/TBI). Audiologists must select tests and procedures commensurate with the patient's overall profile and abilities to minimize potential confounds. Consideration of the contaminating effects of possible confounds, such as deficits in attention, working memory, concentration, comprehension, and fatigue, among others, is particularly important in patients with HI who can be more difficult to test (Musiek & Chermak, 2008).