Brainstem Auditory Evoked Response in HIV-Infected Patients With and Without AIDS

Archives of Medical Research 33 (2002) 25–28

ORIGINAL ARTICLE

Brainstem Auditory Evoked Response in HIV-Infected Patients With and Without AIDS Laura Reyes-Contreras,a Andrés Silva-Rojas,a Antonio Ysunza-Rivera,b Graciela Jiménez-Ruíz,c,d Pedro Berruecos-Villalobosa and Gabriel Romo-Gutiérrezc a

Departamento de Audiología y Foniatría, Hospital General de México, Mexico City, Mexico b Departamento de Diagnósticos, Hospital Manuel Gea González, Mexico City, Mexico c Departamento de Infectología, Hospital General de México, Mexico City, Mexico d Departamento de Audiología, Hospital Central Sur PEMEX, Mexico City, Mexico Received for publication August 24, 2000; accepted June 4, 2001 (00/124).

Background. The purpose of this paper is to determine the type and frequency of brainstem auditory evoked potential abnormalities in HIV-infected patients with and without acquired immune deficiency syndrome (AIDS). Methods. Brainstem auditory evoked potentials were performed in 44 HIV-infected patients; 22 of these patients were asymptomatic carriers and 22 presented clinical manifestation of AIDS at the time of the study. Twenty healthy young adults were studied as controls. When brainstem auditory evoked potentials were performed, none of the HIVinfected patients had neurologic clinical findings. Results. The interwave intervals I–V showed significant differences (p 0.05) between both HIV-infected groups and the control group. There were non-significant differences between the two groups including HIV-infected patients with or without clinical features of AIDS. Conclusions. It is concluded that the HIV infection may produce subclinical pathologic changes in the cochlear nerve and brainstem, which can be detected by BAEP test. © 2002 IMSS. Published by Elsevier Science Inc. Key Words: Brainstem auditory evoked potentials, AIDS.

Introduction Human immune deficiency virus (HIV) enters the central nervous system (CNS) shortly after the onset of the infection. During the incubation period, the patient can be without neurologic symptoms for several years. However, some of these patients develop a progressive encephalopathy (1). Furthermore, 40–75% of these patients develop neurologic complications during the natural history of this illness. Postmortem autopsy findings of CNS involvement have been reported in 90% of the patients deceased as a consequence of acquired immunodeficiency syndrome (AIDS) (2). HIV neurotropism has been demonstrated at the DNA vi-

Address reprint requests to: Dr. Andrés Silva-Rojas, Netzahualpilli 760-1 (L1, M3), Colonia Ajusco Coyoacán, 04300 México, D.F., México. Tels.: (525) 631-3512; (52) 044-26-88-08-53; E-mail: andres14@ prodigy.net.mx

ral sequences in the cerebral tissue of patients with AIDS encephalopathy (3). The first pathologic changes in the CNS included a myelin loss at subcortical level (subcortical leukoencephalopathy), even without clinically evident neurologic manifestations (4). Histopathologic studies reveal local demyelinization areas due to direct infection of the glial and neurologic cells. At the brainstem level, a microglial aggregation may occur (5). The structures of the auditory pathway may be affected by pathologic changes secondary to the HIV infection. Therefore, neurophysiologic studies involving these structures such as brainstem auditory evoked potentials (BAEP) may be abnormal in early stages of HIV infection even in the absence of clinical manifestations (6). BAEP have come into widespread use for assessment of hearing status and assessment of the integrity of the pontine and midbrain auditory pathways. BAEP include five to eight peaks (or waves) positive to vertex. These waves are

0188-4409/02 $–see front matter. Copyright © 2002 IMSS. Published by Elsevier Science Inc. PII S0188-4409(01)0 0 3 4 2 - 3

26

Reyes-Contreras et al./ Archives of Medical Research 33 (2002) 25–28

labeled with Roman numerals. Peaks I–V are the main peaks of clinical interest. The neural generators of these peaks have been described extensively (7–10). It has generally been accepted that the acoustic nerves generate the I and II waves, the ipsilateral cochlear nucleus generates wave III, the superior olivar complex of both sides generates wave IV, and the lateral lemniscus and the contralateral inferior colliculus are the main structures responsible for the generation of wave V (11). The purpose of this paper is to evaluate BAEP of HIV infected patients with and without AIDS. Materials and Methods Twenty healthy young adults (medical students and their relatives) at the Hospital General in Mexico City (40 ears) were studied as controls. The ages of these subjects ranged from 18 to 51 years (mean: 32.2 years). There were 7 females and 13 males; all had normal audiologic studies including the following: 1) pure-tone behavioral audiometry (ANSI, 1969), and 2) normal tympanometry and acoustic reflexes (using 1 kHz, 2 kHz, and 4 kHz). All normal subjects had a negative history for neurologic disease. Fifty-one confirmed HIV-infected patients were studied. Forty-four patients were included in the study group; all showed normal hearing status as confirmed by audiometric and immitance studies. These studies were performed to assure that only normal hearing patients were included in this study. The patients were recruited from the Department of Infectious Diseases of the Hospital General in Mexico City from April 1, 1998 to November 30, 1998. These patients were divided into two groups. Twenty-two patients without symptoms, i.e., 17 males and 5 females with ages ranging from 24 to 51 years (mean: 32.2 years) were included in the first group. The second group included 22 patients with confirmed clinical manifestations of AIDS; ages ranged from 18 to 53 years (median: 33.8 years) Brainstem auditory evoked potentials were performed in all subjects in a sound isolation environment (ANSI 1969). The site of electrode placement was vertex (active) and each mastoid process (12). Electrode impedance was maintained at 5,000 ohms in all cases. For all recordings and measurements, the recommendations of the American Electroencephalography Society (12) were followed. The electrodes were led to a Cadwell 5200 multisensory system with a gain of 1  100,000. The filter bandwidth used for recordings was 150–3,000 Hz. The recording was triggered at stimulus onset and proceded for 10 msec. Two thousand click stimuli were used. Stimuli were 100 us square waves (rarefaction polarity) generated by the same instrument. Every condition of 2,000 trials was repeated separately at least twice, and both traces were superimposed to ascertain the reproducibility of the waves. Masking was used in the

non-test ear at 35 dB below the click intensity that was used in the test ear. Two stimulus rates for recording BAEP were used in this study. First, a stimulus rate of 11 Hz was used at an intensity of 80 dBHL (60 dBSL or better). This condition allowed the authors to obtain well-defined waves for measuring the absolute latencies of waves I, III, and V. Interwave intervals I–V, I–III, and III–V were obtained. Finally, the V/I amplitude ratio was determined. Then the stimulus rate was increased up to 77 Hz at the same intensity (80 dBHL); this protocol was used for evaluating wave V latency as a function of stimulus rate. Two-tail, two-way analysis of variance was used for analyzing dimensional variables (13) such as interwave intervals in the two groups of HIV-infected patients. Fisher exact test was used for analyzing binary variables (13) such as normality vs. abnormality of each parameter, i.e., interwave intervals, amplitude ratios, and repetition rate difference (RRD) in the two groups of HIV-infected patients. Mann-Whitney U test was used for comparing the median age (13) of the patients of the control group, as well as the two groups of HIV-infected patients. Results Mean age was not significantly different (p 0.05) in the three groups of patients. Table 1 shows reference values for interwave intervals, amplitude ratios, and RRD in the control group. Analysis of variance demonstrated significant differences (p 0.05) in the mean I–V interwave intervals between the control group, the group of HIV-seropositive patients, and the group of AIDS patients (see Table 2). However, the difference between the mean I–V interwave intervals of the group of HIV-seropositive patients and the group of AIDS patients was not significantly different (p 0.05). Moreover, when the group of HIV-infected patients and the group of AIDS patients were compared, Fisher exact test demonstrated that the number of patients with prolonged I–V interval in both groups was not significantly different (see Table 3). In the group of HIV seropositive patients, eight (36%) patients showed a prolonged I–V interval; in six of these pa-

Table 1. Normal values from normal group Parameter I–V I–III III–V AR V/I RRD

Values X  3 SD  3.69–4.32 X  3 SD  1.83–2.37 X  3 SD  1.61–2.18 1 0.70 msec

X  mean; SD  standard deviation; AR  amplitude ratio; RRD  repetition rate difference.

Brainstem Potentials in HIV Patients With and Without AIDS Table 2. Interwave intervals in HIV-infected patients with AIDS and without symptoms

Group X Right ear intervals I–V I–III III–V Left ear intervals I–V I–III III–V

Without symptoms SD

Table 4. Normal and abnormal I/V ratio in the HIV infected with no symptoms group and the HIV-infected AIDS group

Amplitude ratio I/V AIDS X

4.15 msec 2.17 msec 1.98 msec

0.22 msec 0.16 msec 0.16 msec

4.21 msec 2.30 msec 1.98 msec

0.21 msec 0.17 msec 0.14 msec

4.19 msec 2.22 msec 1.97 msec

0.19 msec 0.16 msec 0.14 msec

4.30 msec 2.29 msec 2.01 msec

0.26 msec 0.20 msec 0.17 msec

tients the prolonged interval was bilateral, whereas two patients demonstrated a unilateral prolongation. In the same manner, 12 patients (55%) from the AIDS group showed prolonged I–V interwave intervals, while in six of these patients the prolonged intervals were bilateral (see Table 3). An abnormal V/I amplitude ratio was found in 10 (45%) HIV-seropositive patients; four patients showed this abnormality on both sides. In the group of patients with AIDS, an abnormal V/I amplitude ratio was found in 12 patients (55%); in three of these patients the abnormality was bilateral. A nonsignificant relationship was found between the number of patients with an abnormal I/V amplitude ratio in the group of HIV-seropositive as compared to AIDS patients (see Table 4). An abnormal RRD was found in two patients (9.09%) in the group of HIV-seropositive patients. This abnormality occurred only unilaterally. However, in the AIDS group an abnormal RRD was found in four (18.8%) patients. This abnormality occurred on one side only. Fisher exact test demonstrated that the number of patients with abnormal RRD was not significantly higher in the group of AIDS patients as compared to HIV-seropositive patients (p 0.05) (see Table 5). Finally, Fisher exact test demonstrated that the number of patients with at least one abnormality in BAEP parameters mentioned herein was not significantly higher in the group of AIDS patients as compared to HIV-seropositive patients (p 0.05) (see Table 6).

Table 3. Normal and abnormal I–V interwave interval in HIV-infected with no symptoms group and HIV-infected AIDS group

Normal Abnormal Total p 0.05.

a

HIV-infected without symptoms

AIDS group

Total

12 10a 22

10 12a 22

22 22 44

Group SD

X  mean; SD  standard deviation; msec  millisecond.

I–V interval

27

HIV-infected without symptoms

AIDS group

Total

14 8a 22

10 12a 22

24 20 44

Normal Abnormal Total p 0.05.

a

Discussion From the results of this study, it is evident that a great number of AIDS patients showed abnormal BAEP although they were free of neurologic signs and symptoms. Moreover, HIV-seropositive patients with no symptoms also showed significant BAEP abnormalities. Sixty eight percent of HIV-seropositive and AIDS patients studied in this paper demonstrated at least one abnormality in BAEP parameters. The most common abnormality found in this study was an abnormal V/I amplitude ratio, while the least frequent abnormality was an abnormal RRD. The number of patients with an abnormal BAEP parameter in the group of AIDS patients was not significantly higher as compared to number of HIV-seropositive patients with similar abnormalities. These findings may be related to the reported progressive encephalopathy, which may induce an abnormal neural conduction of the auditory pathway as demonstrated by BAEP in both groups of patients, i.e., with and without neurologic symptoms (1). This paper supports previous evidence that HIV infection may produce abnormal results in electrophysiologic tests such as BAEP even in asymptomatic HIV-seropositive patients (14,15). McArthur studied histopathologic CNS changes in early stage of HIV infection dementia (16) and was able to find some changes such as myelin pallor in the white matter. Furthermore, white matter pallor has been demonstrated in at least 50% of a group of patients with AIDS without clinical neurologic manifestations (1). It should be pointed out that the nonsignificant differences found in this paper, when the two groups including HIV-seropositive patients with or without AIDS symptoms were compared, do not support previous reports (14,15,17). It has been reported that the progression of the HIV infec-

Table 5. Normal and abnormal repetition rate difference in the HIV-infected with no symptoms group and the HIV-infected AIDS group

RRD Normal Abnormal Total p 0.05.

a

HIV-infected without symptoms

AIDS group

Total

20 2a 22

18 4a 22

38 6 44

28

Reyes-Contreras et al./ Archives of Medical Research 33 (2002) 25–28

Table 6. Normal and abnormal BAEP in the HIV-infected with no symptoms group and the HIV-infected AIDS group

BAEP Normal Abnormal Total

HIV-infected without symptoms

AIDS group

Total

10 12a 22

4 18a 22

14 30 44

p 0.05.

a

tion increases the frequency of BAEP abnormalities (14, 15,17). A possible explanation for the absence of significant differences found in this study may be related to the high frequency of CNS involvement produced by HIV even in the absence of neurologic symptoms (2). However, the small number of patients studied herein does not allow drawing definite conclusions. The results of this study suggest that BAEP recordings in diagnosed HIV-seropositive patients may allow a more complete study of the disease, increasing prevention of further neurologic complications. Sequential BAEP recordings may also play a role for establishing the neurologic prognosis in HIV patients It may be concluded that BAEP can be a useful tool for detecting early neurologic dysfunction produced by the HIV infection even in asymptomatic stages.

References 1. Navia BA, Jordan BD, Price RW. The AIDS dementia complex. I. Clinical features. Ann Neurol 1986;19:517–524. 2. De la Monte SM, Mo DD, Schooley RT, Hirsh MS, Ridharson EP. Subacute encephalomyelitis of AIDS and its relation to HTLV-III infection. Neurology 1987;37:562–569. 3. Shaw GM, Harper ME, Hahn BH, Atkinson J. HTLV-III infection in brain of children and adults with AIDS encephalopathy. Science 1985;227:177–182.

4. Anders K, Guerra WF, Tomiyasu U, Rosen RJ, Telling S. The neuropathology of AIDS. UCLA experience and review. Am J Pathol 1986;124:537–558. 5. Petito CK. Review of central nervous system pathology in human immunodeficiency virus infection. Ann Neurol 1988;23(Suppl):54–57. 6. Jabbari B, Coats M, Salazar A, Martín A, Fernández FM, Johnson C, Fremer C. Longitudinal study of EEG and evoked potentials in neurologically asymptomatic HIV infected subjects. Electroencephalogr Clin Neurophysiol 1993;3:145–151 7. Moller AR, Jannetta P. Neural generators of brainstem evoked potentials: results from human intracranial recordings. Ann Otol Rhinol Laryngol 1981;90:591–596. 8. Hashimoto Y, Ishiyama Y. Brain-stem auditory-evoked potentials recorded directly from human brain-stem and thalamus. Brain 1981; 104:841–859. 9. Markand ON, Farlow MR. Brain-stem auditory evoked potential abnormalities with unilateral brain-stem lesions demonstrated by magnetic resonance imaging. Arch Neurol 1989;46:295–299. 10. Hughes JR, Fino JJ. A review of generators of brainstem auditory evoked potential: contribution of an experimental study. J Clin Neurophysiol 1985;2:355–381. 11. Silva A, Carrillo E, Ysunza A. Los generadores de los potenciales provocados auditivos de latencia corta. Revisión del tema. Anal Otorrinolaringol Mex 1997;42:200–207. 12. Coats A, Chiappa K, Picton T, Starr A, Stockard J. American Electroencephalographic Society guidelines for clinical studies: recommended standards for short-latency auditory evoked potentials. J Clin Neurophysiol 1984;1:32–40. 13. Feinstein A. Stochastic analysis. In: Feinstein A, editor. Clinical epidemiology: the architecture of clinical research. Philadelphia, PA, USA: WB Saunders Co.;1985. pp. 141–155. 14. Hausler R, Vibert D, Koralkink Y, Hirschel B. Neuro-otological manifestations in different stages of HIV infection. Acta Otoloaryngol (Stockh) 1991;481(Suppl):515–521. 15. Pierelli F, Soldati P, Zambardi P, D’Angostino D. Electrophysiological study (VEP, BAEP) in HIV-1 seropositive patients with and without AIDS. Acta Neurol Belg 1993;93:78–87. 16. McArthur JC, Becker PS, Parisi JE, Trapp B, Levinson M, Scull J. Neuropathological changes in early HIV-1 infection dementia. Ann Neurol 1989;26:681–884. 17. Bankaitis AE. The effect of click rate on the auditory brainstem response (ABR) in patients with varying degrees of HIV-infection: a pilot study. Ear Hear 1995;16:321–324.