Clinical significance of electrically evoked auditory brainstem response

Electrically evoked auditory brainstem responses (EABR) were recorded in 31 postlingually deafened adults, who had recently received cochlear implants (mini-system, Cochlear Ltd). The wave consisted of three distinct positive peaks labeled P1, P2, and P3 with latency of 1.35 (+/- 0.14), 2.17 (+/- 0.18) and 4.08 (+/- 0.31) ms, respectively. The P3 threshold (EABR-T) and slope (EABR-S) were 0.9 (+/- 0.47) mA and 0.6 (+/- 0.28) muv/mA, respectively. The relationships between the EABR parameters (EABR T and -S of the P3 wave) and age, duration of deafness, promontory test and subjective response (T and C-level) were investigated. The scattergram showed a strong negative linear relationship between EABR-S and subjective T-level. This finding suggests that EABR-S is good measure of postoperative perception.     

Speech-evoked cognitive P300 potentials in cochlear implant recipients

The cognitive P300 evoked potential was elicited by speech stimuli in successful cochlear implant recipients, and the resulting P300 morphology was remarkably similar to that of normal-hearing individuals. The P300 was elicited by the synthesized speech pair/da/and/di/ presented using an oddball paradigm to nine "good" Nucleus cochlear implant users and nine age-matched normal-hearing subjects (34-81 yr old). There were no significant differences in P300 amplitude and latency between the two groups. Moreover, the N1 and P2 potentials occurred at similar latencies in the two groups, although the N1 amplitude was significantly smaller in the cochlear implant users. The P300 was absent in one "poor" cochlear implant user. The results suggest that the P300 may serve as a useful tool for evaluating the cognitive aspects of auditory processing in cochlear implant recipients, and that it may aid in assessing the success of cochlear implantation.     

Response of the primary auditory cortex to electrical stimulation of the auditory nerve in the congenitally deaf white cat

Neural activity plays an important role in the development and maintenance of sensory pathways. However, while there is considerable experience using cochlear implants in both congenitally deaf adults and children, little is known of the effects of a hearing loss on the development of the auditory cortex. In the present study, cortical evoked potentials, field potentials, and multi- and single-unit activity evoked by electrical stimulation of the auditory nerve were used to study the functional organisation of the auditory cortex in the adult congenitally deaf white cat. The absence of click-evoked auditory brainstem responses during the first weeks of life demonstrated that these animals had no auditory experience. Under barbiturate anaesthesia, cortical potentials could be recorded from the contralateral auditory cortex in response to bipolar electrical stimulation of the cochlea in spite of total auditory deprivation. Threshold, morphology and latency of the evoked potentials varied with the location of the recording electrode, with response latency varying from 10 to 20 ms. There was evidence of threshold shifts with site of the cochlear stimulation in accordance with the known cochleotopic organisation of AI. Thresholds also varied with the configuration of the stimulating electrodes in accordance with changes previously observed in normal hearing animals. Single-unit recordings exhibited properties similar to the evoked potentials. Increasing stimulus intensity resulted in an increase in spike rate and a decrease in latency to a minimum of approximately 8 ms, consistent with latencies recorded in AI of previously normal animals (Raggio and Schreiner, 1994). Single-unit thresholds also varied with the configuration of the stimulating electrodes. Strongly driven responses were followed by a suppression of spontaneous activity. Even at saturation intensities the degree of synchronisation was less than observed when recording from auditory brainstem nuclei. Taken together, in these auditory deprived animals basic response properties of the auditory cortex of the congenitally deaf white cat appear similar to those reported in normal hearing animals in response to electrical stimulation of the auditory nerve. In addition, it seems that the auditory cortex retains at least some rudimentary level of cochleotopic organisation.     

Usefulness of ictal and interictal 99mTc ethyl cysteinate dimer single photon emission computed tomography in patients with refractory partial epilepsy

OBJECTIVE: The aim was to investigate the feasibility of recording the electrical auditory brain stem response (EABR) evoked by electrical stimulation at the promontory (Prom-EABR) as a tool to assist selection of the ear for cochlear implantation in young children. STUDY DESIGN: The study group consisted of young children for whom the decision to proceed with implantation with the Nucleus mini 22-channel cochlear implant (Cochlear (UK) Ltd., London, UK) had already been made. SETTING: The Prom-EABR was recorded after the children had been anesthetised, but before the start of surgery. PATIENTS: A group of 25 children (11 boys and 14 girls), whose age at implantation ranged from 2 years 11 months to 6 years 8 months (mean age, 4 years 5 months), were investigated. INTERVENTION: Recordings of the Prom-EABR were used to determine which ear would receive the cochlear implant, providing there were no preexisting contraindications regarding selection of the ear. MAIN OUTCOME MEASURE: It has been suggested from earlier studies that the characteristics of the amplitude input/output (I/O) function of the EABR are related to neuronal survival. If the ear with the "better" I/O function is chosen for implantation, it might be expected that these children will perform better on average than those in whom the ear has been selected at random. RESULTS: Reliable recordings of the Prom-EABR were achieved in 40 ears (80%) of the 50 ears in the study. In 20 of the 25 children the technique was actively employed for selection of the ear for implantation. CONCLUSIONS: Recording of the Prom-EABR in the operating theater is a viable technique. Future analysis of long-term outcome measures of performance with the implant will confirm or dispute the benefit of ear selection using the Prom-EABR.     

Electrical stimulation of the auditory nerve. I. Correlation of physiological responses with cochlear status

The purpose of the present study was to evaluate evoked potential and single fibre responses to biphasic current pulses in animals with varying degrees of cochlear pathology, and to correlate any differences in the physiological response with status of the auditory nerve. Six cats, whose cochleae ranged from normal to a severe neural loss (< 5% spiral ganglion survival), were used. Morphology of the electrically evoked auditory brainstem response (EABR) was similar across all animals, although electrophonic responses were only observed from the normal animal. In animals with extensive neural pathology, EABR thresholds were elevated and response amplitudes throughout the dynamic range were moderately reduced. Analysis of single VIIIth nerve fibre responses were based on 207 neurons. Spontaneous discharge rates among fibres depended on hearing status, with the majority of fibres recorded from deafened animals exhibiting little or no spontaneous activity. Electrical stimulation produced a monotonic increase in discharge rate, and a systematic reduction in response latency and temporal jitter as a function of stimulus intensity for all fibres examined. Short-duration current pulses elicited a highly synchronous response (latency < 0.7 ms), with a less well synchronized response sometimes present (0.7-1.1 ms). There were, however, a number of significant differences between responses from normal and deafened cochleae. Electrophonic activity was only present in recordings from the normal animal, while mean threshold, dynamic range and latency of the direct electrical response varied with cochlear pathology. Differences in the ability of fibres to follow high stimulation rates were also observed; while neurons from the normal cochlea were capable of 100% entrainment at high rates (600-800 pulses per second (pps)), fibres recorded from deafened animals were often not capable of such entrainment at rates above 400 pps. Finally, a number of fibres in deafened animals showed evidence of 'bursting', in which responses rapidly alternated between high entrainment and periods of complete inactivity. This bursting pattern was presumably associated with degenerating auditory nerve fibres, since it was not recorded from the normal animal. The present study has shown that the pathological response of the cochlea following a sensorineural hearing loss can lead to a number of significant changes in the patterns of neural activity evoked via electrical stimulation. Knowledge of the extent of these changes have important implications for the clinical application of cochlear implants.     

Reduction in excitability of the auditory nerve following electrical stimulation at high stimulus rates. II. Comparison of fixed amplitude with amplitude modulated stimuli

We have previously shown that acute electrical stimulation of the auditory nerve using charge-balanced biphasic current pulses presented continuously can lead to a prolonged decrement in auditory nerve excitability (Tykocinski et al., Hear. Res. 88 (1995), 124-142). This work also demonstrated a reduction in electrically evoked auditory brainstem response (EABR) amplitude decrement when using an otherwise equivalent pulse train with a 50% duty cycle. In the present study we have extended this work in order to compare the effects of electrical stimulation using both fixed amplitude electrical pulse trains and amplitude modulated (AM) pulse trains that more accurately model the dynamic stimulus paradigms used in cochlear implants. EABRs were recorded from guinea pigs following acute stimulation using AM trains of charge-balanced biphasic current pulses. The extent of stimulus-induced reductions in the EABR were compared with our previous results using either fixed amplitude continuous, or 50% duty cycle pulse trains operating at 0.34 microC/phase (2 mA, 170 micros/phase) at 400 or 1000 pulses/s (Tykocinski et al., Hear. Res. 88 (1995) 124-142). The AM pulse train, operating at the same rates, was based on a 1-s sequence of the most extensively activated electrode of a Nucleus Mini-22 cochlear implant using the SPEAK speech processing strategy exposed to 4-talker babble, and delivered the same total charge as the fixed amplitude 50% duty cycle pulse train. Two hours of continuous stimulation induced a significant, rate-dependent reduction in auditory nerve excitability, and showed only a slight post-stimulus recovery for monitoring periods of up to 6 hours. Following 2 or 4 h of stimulation using an otherwise equivalent pulse train with a 50% duty cycle or the AM pulse train, significantly less reduction in the EABR was observed, and recovery to pre-stimulus levels was generally rapid and complete. These differences in the extent of the recovery between the continuous waveform and both the 50% duty cycle and AM waveforms were statistically significant for both 400 and 1000 pulses/s stimuli. Consistent with our previous results, the stimulus changes observed using AM pulse trains were rate dependent, with higher rate stimuli evoking more extensive stimulus-induced changes. The present findings show that while stimulus-induced reductions in neural excitability are dependent on the extent of stimulus-induced neuronal activity, the use of an AM stimulus paradigm further reduces post-stimulus neural fatigue.     

Electrically evoked auditory brainstem response in peripherally myelin-deficient mice

The integrity of the myelin sheath is important for normal electrophysiological function and survival of neurons that make up the auditory nerve. It is hypothesized that myelin deficiency of the auditory nerve may change the electrophysiologic characteristics of the auditory system, especially the temporal properties. In this study, the electrically evoked auditory brainstem response (EABR) was systematically evaluated in TrJ and Po-DT-A mice. Both of these mice have a deficit of their peripheral myelin. Correlation between the EABR and degree of myelin deficiency was also evaluated. The EABR in both strains of poorly myelinated mice exhibited prolonged latency, decreased amplitude, elevated threshold of wave I evoked by short-duration stimuli (20 microseconds/phase). A 2-pulse stimulation paradigm was used to evaluate refractory properties. Myelin-deficient mice exhibited slower recovery from the refractory state than controls. Long-duration stimuli (4 ms/phase) were used to assess integration properties. Myelin-deficient mice demonstrated prolonged wave I latency and more gradual latency changes with current level. Myelin thickness showed a strong correlation with EABR threshold for short-duration stimulation (r = -0.784), maximum wave I latency (r = -0.778) and the time constant of the wave I latency-current level function (r = -0.736) for long-duration stimulation and normalized wave I recovery functions (r = -0.718). These findings suggest that EABR measurement may be a promising tool to assess the electrically stimulated properties of auditory neurons, particularly related to the status of myelin sheath.     

Prolonged P300 latency in attention deficit hyperactivity disorder predicts poor response to imipramine

P300 is a cognitive evoked potential that evaluates attention and information processing. This study uses auditory and visual P300 topography to develop a classification of attention deficit hyperactivity disorder (ADHD), and find predictors of treatment response. Of 45 ADHD children ages 6 to 15 treated with pemoline in a previous study, 25 were poor responders. Of these 25, 17 participated in an imipramine treatment protocol. Auditory and visual P300 testing was performed before and after treatment using 31 scalp electrodes. Good and poor responders to imipramine were clinically identical. Poor imipramine responders had longer auditory and visual P300 latencies than good responders. Treatment with imipramine decreased auditory P300 latencies and increased auditory P300 amplitudes. We have previously reported that ADHD patients with small right frontocentral auditory P300 amplitudes respond poorly to pemoline. Thus, P300 topography and latency classifies ADHD into three groups: group 1 with normal P300 topography, and good response to pemoline; group 2 with small right frontocentral auditory P300 amplitudes, poor response to pemoline, and good response to imipramine; and group 3 with long auditory and visual P300 latencies and small right frontocentral auditory P300 amplitudes, and poor response to pemoline and imipramine.     

Positron emission tomography of auditory sensation in deaf patients and patients with cochlear implants

The present study investigated the function of the auditory cortices in severely hearing-impaired or deaf patients and cochlear implant patients before and after auditory stimulation. Positron emission computed tomography (PET), which can detect brain activity by providing quantitative measurements of the metabolic rates of oxygen and glucose, was used. In patients with residual hearing, the activity of the auditory cortex measured by PET was almost normal. Among the totally deaf patients, the longer the duration of deafness, the lower the brain activity in the auditory cortex measured by PET. Patients who had been deaf for a long period showed remarkably reduced metabolic rates in the auditory cortices. However, following implantation of the cochlear device, the metabolic activity returned to near-normal levels. These findings suggest that activation of the speech comprehension mechanism of the higher brain system can be initiated by sound signals from the implant devices.     

Induction of endolymphatic hydrops in the guinea pig by perisaccular deposition of sepharose beads carrying and not carrying immune complexes

Click-evoked auditory brainstem responses were recorded in a prosimian primate, the bush baby (Otolemur garnettii), before and after depletion of serotonin (by systemic injection of para-chlorophenylalanine; pCPA) and up to 20 days after discontinuing pCPA injections (during the recovery of serotonin). Biphasic 100 micros clicks were presented at five repetition rates (13.2, 33.2, 53.2, 73.2, and 93.2 clicks/s; RATE) and sound pressure levels (SPL) were varied in 10 dB steps from 120-60 dB SPL peak equivalent. Absolute latencies of vertex-positive peaks I, III, IV, and V were measured from click onset. The latencies from each wave were statistically analyzed with a two-way analysis of variance using either RATE or SPL (but not both) and TIME AFTER pCPA as independent variables. Prior to pCPA, brainstem response latencies increased as a function of both decreasing SPL and increasing RATE. After pCPA, these normal increases in wave latency increased even more, particularly in response to high click rates. After pCPA was discontinued, measurements taken at weekly intervals indicated that latencies decreased after 1 week, increased to the highest values recorded after 2 weeks, and returned to normal after 3 weeks. These dynamic changes were interpreted to be the result of postsynaptic receptor up-regulation during the 10 days of continuous pCPA administration. These results suggest that serotonin plays an important role in sensory processing at the cellular level and, tonically, facilitates the auditory brainstem response to sound.     

Pediatric and psychiatric comorbidity. Part I: The future of consultation-liaison psychiatry

This study was conducted to investigate the influence of high-frequency aircraft noise on the function of the auditory system of school-age children. A total of 228 students attending a school near an airport (school A) and 151 students attending a school far from an airport (school B) were analyzed. Audiometry and brainstem auditory evoked potential (BAEP) detection were performed in all subjects to evaluate cochlear and retrocochlear function. The results of audiometry indicated that hearing ability was significantly worse in the children of school A, which was located under the flight paths. The values of pure tone average, high pure tone average, and threshold at 4 kHz were all higher in children who were frequently exposed to aircraft noise. There was no consistent difference in BAEP latencies between the two schools. These results indicate that central transmission is not affected in children who have been exposed to aircraft noise for several years. The results of the present study showed a significant association between aircraft noise exposure and prevalence of noise-induced hearing loss. Although damage to peripheral cochlear organs was confirmed in school-age children, involvement of the central auditory pathway could not be verified.     

Intensity discrimination and auditory brainstem responses in cochlear implant and normal-hearing listeners

Intensity-discrimination limens (IDLs) and auditory brainstem responses (ABRs) were measured as a function of stimulus intensity in 6 cochlear implant (CI) and 8 normal-hearing (NH) listeners. Pulse-train stimuli were delivered electrically to the auditory nerve in CI listeners and acoustically in NH listeners. In CI listeners, the IDLs expressed as Weber fractions decreased monotonically with increasing intensity. In NH listeners, a nonmonotonic IDL function showing a peak a midintensities was observed. ABR wave amplitudes increased regularly with intensity only in CI listeners. Results support the notion that the slight decrease in Weber's fractions with increasing sound intensity--generally referred to as "the near-miss to Weber's law"--is subtended by retrocochlear processes, whereas the increase in Weber's fractions at midlevels--known as "the severe departure from Weber's law"--originates in cochlear mechanisms.     

Cardiac Troponin T to evaluate myocardial protection via intermittent cold blood or continuous warm blood cardioplegia in coronary artery bypass grafting

The effects of a speech-coding strategy of cochlear implant (CI) on cortical activity were evaluated using positron emission tomography. The CIs used in the present study were those of a 22-channel system using the Multipeak speech-coding strategy (MPEAK) and the spectral peak strategy (SPEAK). On comparing the 2 groups, it was found that the speech-tracking performance was significantly higher in the SPEAK group than in the MPEAK group. Regional cerebral blood flow (rCBF) was measured during the silent resting, noise stimulus and speech stimulus periods. The increase in rCBF was localized mainly in the primary auditory area during the noise stimulus period. The increase in rCBF in the auditory association area during the speech stimulus period was stronger in the SPEAK group than in the MPEAK group. This finding suggests that the SPEAK strategy activates more speech processing neuronal networks in the auditory association area than the MPEAK strategy.     

Effects of prenatal exposure to PCBs on the neurological function of children: a neuropsychological and neurophysiological study

To determine the long-term neurotoxicity of prenatal exposure to polychlorinated biphenyls (PCBs), 54 children--27 'Yu-Cheng' ('oil disease') children and 27 controls--were administered a battery of tests, including the WISC-R, auditory event-related potentials (P300), pattern visual evoked potentials (P-VEPs) and somatosensory evoked potentials (SSEPs). Full-scale IQ scores on the WISC-R were lower for the Yu-Cheng group than for the control group. Mean P300 latencies were significantly longer, and P300 amplitude significantly more reduced, in the Yu-Cheng group than in the control group at Cz and Pz. There were no significant difference in peak latencies and amplitudes between the two groups for P-VEPs and SSEPs. These findings suggest that prenatal exposure to PCBs tends to affect high cortical function rather than the sensory pathway in the developing brain.     

Brainstem auditory evoked potentials during brainstem ischemia and reperfusion in gerbils

To evaluate the reversibility of neural function in the brainstem following ischemia, we investigated the effect of transient brainstem ischemia on the brainstem auditory evoked potential in gerbils. Brainstem ischemia was produced by bilateral extracranial occlusion of vertebral arteries. Local cerebral blood flow was measured by quantitative autoradiography after 5 min of ischemia and was reduced to less than 3 ml/100 g per min in the pons and lower midbrain, indicating severe and reproducible brainstem ischemia. During brainstem ischemia, brainstem auditory evoked potential waveforms disappeared completely. After a brief ischemic insult (5 min), all four brainstem auditory evoked potential components recovered to normal. After longer ischemic insults (10-30 min), brainstem auditory evoked potential components never recovered to normal. Microtubule-associated protein 2 immunoreactivity revealed differential vulnerability of the acoustic relay nuclei in the brainstem. Neurons in the lateral lemniscus were most vulnerable, followed in order by neurons in the trapezoid body, the superior olive and the cochlear nucleus. We also demonstrated a close relationship between the reversibility of ischemia-induced changes on brainstem auditory evoked potential and ischemic lesions of these relay nuclei. These data may be useful for evaluating the therapeutic window of thrombolytic therapy during acute vertebrobasilar occlusion.     

Analysis of evoked otoacoustic emissions in patients with high frequency cochlear hearing loss

Evoked otoacoustic emissions (EOAE) and auditory brainstem evoked response (ABR) were tested in patients with high frequency cochlear hearing loss and subjects with normal hearing. Results showed that: 1. In the normal group, the main component of EOAE varied between 0.5-5kHz. 2. There was a close correlation between the pure-tone audiogram figure and EOAE spectrum. 3. The I/O function curves of EOAE displayed non-linear characteristics and a trend of saturation with higher stimulus levels in the normal group. In the patient group the growth rates of EOAE and ABR were remarkably higher than those in the normal group with high stimulus levels. It suggests that the abnormal EOAE are probably relevant to the recruitment phenomenon in cochlear hearing loss.     

Subjective assessment of allergy relief following group hypnosis and self-hypnosis: a preliminary study

Distortion Product Otoacoustic Emissions (DPOEs) are otoacoustic emissions evoked by two pure equilevel tones (f1, f2) called primaries and are believed to provide frequency-specific information regarding cochlear function. We recorded DPOEs at 2f1-f2 frequency with a constant frequency ration (f2/f1 = 1.22) in 8 normal hearing subjects (16 ears, mean age 28 +/- 1.5) to establish the characteristics of these emissions in the adult population. DPOEs were measured at the following F2 frequencies and respective fp geometric mean frequencies: 696/632, 1001/904, 1501/1360, 2002/1809, 3003/2714, 4004/3626, 5005/4531 e 6006/5435 Hz. Detailed testing included the recording of DPOE "audiograms" and input-output functions depicting the relationship of the amplitudes of DPOE to primary-tone levels ranging from 25 to 70 dB SPL in 5 dB steps. The present findings are in good agreement with investigations based on evoked otoacoustic emissions published by other researchers. The average DPOE "audiograms" demonstrated a low-frequency maximum at 1501 Hz (f2)/1360 (fp) and a high-frequency peak at 5005 Hz (f2)/4531 (fp). The two maximum regions were separated by a minimum around 3003 Hz (f2)/2714 (fp). This study confirms the feasibility of DPOE measurements among adults and provide a normal baseline for this age group. DPOEs could be useful, in association with evoked otoacoustic emissions and with auditory brainstem responses, in obtaining a precise evaluation of the peripheral auditory system.     

The effects of cochlear hearing loss, age and sex on the auditory brainstem response

This study analyzed the effects of auditory impairment, age and sex on the auditory brainstem response (ABR) wave latencies. ABR wave I, wave V and I-V interval measures were extracted from the clinical records of 201 patients with cochlear hearing loss. Females had consistently earlier wave V latencies and shorter I-V intervals than males. No age effects were observed. Degree of impairment had a systematic effect on ABR wave latencies and I-V intervals. Wave I displayed latency extension with increasing levels of high-frequency hearing loss, whilst for wave V increases in latency were dependent upon both degree and slope of the hearing loss. Present results suggest that many of the previously reported sex differences and variable interactions seen for the ABR can be accounted for by differences in the underlying distribution of audiogram shapes within and between study populations. Different audiometric configurations were found to produce consistent differential effects on both wave I and wave V latency and thus influence the I-V interval. This study underlines the need to develop a more detailed model of impairment effects if correction factors are to be employed more effectively in ABR testing for retrocochlear pathology.     

Cochlear implantation: a review of the literature and the Nijmegen results

The field of cochlear implantation is developing rapidly. In subjects with bilateral profound deafness who gain no benefit from conventional hearing aids the aim of cochlear implantation is to provide a means for them to receive auditory sensations. Throughout the world, most cochlear implant centres are still continuing their research efforts to improve the results with this technique. Although it is still difficult to predict how an individual will perform with a cochlear implant, the success of cochlear implantation can no longer be denied. In this paper, we review some recent papers and reports, and the results of the various Nijmegen cochlear implant studies. Data about subject selection, examinations, surgery and the outcome are discussed. Our results were in good agreement with those of other authors. It can be concluded once again that cochlear implantation is an effective treatment for postlingually deaf adults and children, and for prelingually (congenital or acquired) deaf children with profound bilateral sensorineural deafness.     

Aplasia and hypoplasia of the vestibulocochlear nerve: diagnosis with MR imaging

PURPOSE: To introduce aplasia or hypoplasia of the vestibulocochlear nerve (VCN) as a possible cause of hearing loss and to identify the magnetic resonance (MR) imaging characteristics of this entity. MATERIALS AND METHODS: In seven patients with congenital deafness or unexplained sensorineural hearing loss, MR imaging enabled diagnosis of aplasia or hypoplasia of the VCN. Axial (0.7-mm) three-dimensional Fourier transformation-constructive interference in steady state (3DFT-CISS) images and parasagittal reconstruction images perpendicular on the course of the VCN were obtained. Twenty normal inner ears were also studied; their findings were compared with those of the patients. RESULTS: The facial nerve and inferior and superior vestibular and cochlear branches of the VCN were identified on the MR images in the 20 normal inner ears. Aplasia of the VCN was detected in two patients with normal labyrinths but with a severe stenosis of the internal auditory canal. A common VCN with absence of the cochlear branch was found bilaterally in two patients with a congenital malformation of the labyrinth. A common VCN with absence or hypoplasia of the cochlear branch was found in three patients with normal internal auditory canals and labyrinths. CONCLUSION: Submillimetric gradient-echo images (eg, 3DFT-CISS) should always be used to exclude aplasia or hypoplasia of the cochlear branch of the VCN in all cochlear implant candidates and patients with congenital deafness. This entity, which can occur with or without associated labyrinthine malformation, should be confirmed in two planes.