The effect of hemodialysis on hearing in chronic kidney disease patients – A systematic review

Search strategy

The search engines used to extract the data were PubMed, Science Direct, SCOPUS, Clinical key, EBSCO Medline, and Google Scholar. The electronic search was performed using different combinations of the terms, namely, hemodialysis, SNHL, sudden hearing loss, sudden deafness, chronic renal disease, CKD, and chronic renal failure. All the literature available from database inception till May 31, 2020, was included in the study. The selected articles were exported to Mendeley (Desktop version 1.19.4). These records were screened for duplication, by titles and abstracts, inclusion-exclusion criteria, and availability of full text. The entire process of screening and reviewing the full texts of eligible articles was done independently by two groups of three authors.

Inclusion criteria

This study attempted to investigate the association of SNHL and hemodialysis in CKD patients. The experimental studies, as well as observational studies such as case reports, cross-sectional studies, case–control studies, and cohort studies (prospective and retrospective), were included as the source of data. The inclusion and exclusion criteria were clearly defined. The articles needed to be in the English language. The study population was CKD patients. The intervention should be hemodialysis. Comparators included were CKD patients on other treatment modalities such as conservative treatment, peritoneal dialysis, and renal transplantation or age- and sex-matched healthy controls. The outcome observed in all the incorporated studies was the development of SNHL or SSNHL. Confirmed hearing loss with either pure tone audiometry (PTA), distortion product otoacoustic emissions (DPOAE), transitory evoked otoacoustic emission (TEOAE), or impedance audiometry was included in the study. The WHO classification or ASHA classification used to quantify the degree of hearing loss was reported in this study.^[14]

Exclusion criteria

Studies on CKD patients who had undergone hemodialysis but developed complications other than hearing loss were excluded from the study. Besides, patients with conductive hearing impairment or having known causes for hearing loss such as ototoxic drug intake and noise exposure were also eliminated from this review.

Quality assessment

The studies with keywords in the title and abstract were selected, full text retrieved, and critically appraised independently by two groups of three authors to assess for its relevance, validity, and potential risk of bias using the Critical Appraisal Skills Programme Checklist (CASP). Separate checklists were used for case-control and cohort studies.^[15]

Data extraction

In the included studies, several characteristics such as publication year, the country where the study was conducted, study design, the target population (age and gender distribution), type of treatment for Stage V CKD (hemodialysis, peritoneal dialysis, renal transplantation, or conservative treatment), duration of hemodialysis, duration of follow-up, and criteria to evaluate hearing levels were extracted. The results of the study were appraised for the quality of evidence and bias if any were reported.

Statistical analysis

Descriptive analysis was done and the characteristics of individual studies were compiled in a tabular form.

Search results

This systematic review was guided by the PRISMA framework [Figure 1]. A total of 1584 articles were retrieved and imported to Mendeley to be reviewed independently by two groups of three authors. One hundred and thirty-eight duplicates were removed. On screening for relevance with the titles and abstracts, 1265 articles were excluded (1195 articles were irrelevant, 49 articles were not in the English language, and 21 articles were not available in full text). A total of 181 full-text eligible articles were screened based on inclusion/ exclusion criteria and 145 articles were eliminated as the CKD patients had exposure to one or more factors that could also cause hearing loss such as radiation, noise, ototoxic drugs, Alport syndrome, or Fabry’s disease. Finally, 36 scientific articles were included for qualitative synthesis. The 36 articles extracted were from 1973 to 2020 and came from 16 different countries. The details are shown in [Table 1].

Close

Figure 1:

Preferred Reporting Items for Systematic Review and Meta-Analyses flowchart.

Export to PPT

Study population

The study population comprised 99,020 individuals (43,143 males and 55,877 females). Their ages ranged from 1 year to 80 years. Among them, 21,787 were CKD patients while 77,233 were healthy controls. Among the CKD patients, the mean duration of the disease was 5 years.^[9,10,16-35] The different treatment modalities for CKD reported were hemodialysis in 15,007 and peritoneal dialysis in 6443 while 195 were on conservative treatment and 142 participants had undergone renal transplantation.

Audiological tests used

It was observed that different researchers used varied audiological tests to detect and quantify the hearing loss. The most common being PTA used in 61.1% of the studies as a single assessment tool while it was combined with DPOAE in four studies,^{[22,24,33,36]} with TEOAE in three studies^[23,32,37] while all three hearing assessment tools; PTA, DPOAE, and TEOAE were used in one.^[38] In two studies, impedance audiometry was used along with PTA,^[20,30] while in one study, impedance audiometry was used with PTA and DPOAE.^[25] Speech audiometry was used along with PTA in one study.^[7] A single study incorporated Brainstem Evoked Response Audiometry (BERA) as a hearing assessment tool.^[38]

Study design

The varying study design was encountered in this systematic review with 13 cohort studies, 10 case-control studies, seven case reports, and six cross-sectional studies [Figure 2]. Among the cohort studies, there were 12 prospective studies and one retrospective study with a total of 96,040 participants. Cohort studies on this topic from 1977^[40] to 2018^[27] were obtained. There are conflicting results with six articles^{[24,27,35,37,40]} reporting hemodialysis as the cause for the development of SNHL while the remainder refuted it.^{[21,22,29,33,41,43]}

Close

Figure 2:

Distribution of the study design employed.

Export to PPT

The earliest case-control study on this topic was in 1992^[10] while the most recent was in 2020.^[16] The total sample from case-control studies was 1656. Case-control studies helped to compare the audiological results of the target population (patients on hemodialysis) with healthy, age- and gender-matched controls[10,16,20,23,25,28,44,45] or with CKD patients receiving conservative treatment.^[17,20,32] Six among them reported a positive association between hemodialysis and occurrence of SNHL,^{[16,23,28,32,44,45]} while four were unable to demonstrate a clear association of SNHL and hemodialysis.^{[10,17,20,25]}

Case reports constituted a total sample of 962. The earliest case report dates back to 1973^[47] while the latest was in 2017.^[6] The majority of these articles demonstrated a significant association between hemodialysis and SSNHL.^{[6,18,26,28,46,48]}

Cross-sectional studies formed 16.7% of the selected articles from 1996^[30] to 2020^[9] with a sample population of 363. Similar to case reports, cross-sectional studies linked hemodialysis to the occurrence of SSNHL.^{[9,30,34,38,49]}

Hemodialysis and hearing loss

High-frequency SNHL was reported in 89% of the studies while 11% did not specify the frequency affected. The time period of hemodialysis was mentioned in only 63.9% of studies with the mean duration of hemodialysis being 5.6 years. Although all studies reported the occurrence of SNHL, only 22% of studies reported an association of hearing loss with the duration of dialysis.[17,27,28,35,37,44,46] The follow-up period after hemodialysis varied from 3 months to 12 years. About 5.5% of participants reported SNHL after the very first session of hemodialysis.^[24,28] Pre-dialysis hearing assessment to detect any pre-existing hearing impairment was mentioned in only 13 (38%) of the 36 studies.^{[16,19,22,24,27,28,33,36,37,40,43,44,46]}

Recommendations of the study

All CRF patients need to undergo thorough pre-hemodialysis otologic and audiometric evaluation followed by repeated audiological assessments at regular intervals of hemodialysis to identify patients who may develop hearing impairment during treatment. This may prove to be the stepping stone in early detection as well as lead to a better understanding of the etiology of SSNHL in patients on hemodialysis. Further, a randomized controlled trial on this topic would help provide a stronger level of evidence to expand scientific knowledge and contribute to making conclusions about the effect of hemodialysis on hearing.

Limitations of the study

Inability to access some of the full-text articles might have affected this literature review.