Long-term vancomycin use had low risk of ototoxicity.

BACKGROUND: Vancomycin is a commonly used antibiotic with potent activity against Gram-positive organisms, but prolonged use and high doses can lead to toxicity. While vancomycin-associated nephrotoxicity is widely reported, few cases of ototoxicity have been described. The objective of this study was to determine the prevalence of negative changes in audiograms in patients receiving long-term intravenous (IV) vancomycin and to identify high-risk patients who need audiogram monitoring.
METHODS: This was an IRB approved, cross-sectional study performed at an academic medical center from 1/2012-3/2019. Patients who were prescribed IV vancomycin for ≥ 14 days and had baseline and follow-up weekly audiometry were included. All data was extracted from the electronic medical record. The primary endpoint was worsening audiogram while on vancomycin. Descriptive and bivariate statistics were used to describe the patient population.
RESULTS: 424 patients were screened for inclusion; 92 received at least two audiograms while on vancomycin. Fifty-three percent of patients were men, the median (IQR) patient age was 44 (34-58) years, and 8% of patients had an estimated Cockcroft-Gault creatinine clearance ≤ 30 mL/min or received hemodialysis. The median (IQR) vancomycin exposure up until the last recorded audiogram was 30 (17-42) days. Vancomycin indications were: 53 (58%) bone and joint infections, 17 (18%) infective endocarditis, 10 (11%) bacteremia, 12 (13%) other infections. Seven (8%) patients experienced a worsening change in hearing from baseline, two (2%) of them suffered mild loss, two (2%) had mild to moderate loss, and three (3%) developed moderate-to-severe hearing loss. In bivariate analyses, no variables were found to be associated with a worsening change in audiogram, including baseline abnormal audiogram, age ≥ 40 years, elevated serum vancomycin levels, or vancomycin doses ≥ 4 grams/day.
CONCLUSIONS: The prevalence of negative changes in audiograms among patients receiving long-term intravenous vancomycin was low. The utility of routine audiogram testing in this population remains questionable except in high-risk patients; however, larger prospective studies with controls may be warranted to further explore the risk of ototoxicity.

Background

Vancomycin, one of the most widely used antibiotics in the United States, is a glycopeptide antibiotic that has been in clinical use since 1954. It is mainly used for the treatment of severe Gram-positive infections, such as methicillin-resistant Staphylococcus aureus (MRSA) [1-3]. While initial vancomycin utilization was limited due to formulation impurity issues and a low prevalence of invasive MRSA infections, recent usage has had a dramatic increase due to the commonality of community and healthcare-associated MRSA infections [4-5].

Increasing vancomycin dosages are often needed to account for elevated vancomycin minimal inhibitory concentrations (MICs) observed in MRSA isolates, which are thought to be related to treatment failure among patients with bacteremia and pneumonia, especially in obese patients [6-7]. Nephrotoxicity is a relatively common adverse effect associated with prolonged vancomycin use, with new reports suggesting a higher incidence of vancomycin-associated nephrotoxicity when larger doses are used to maintain therapeutic efficacy for trough-based goals [8].

Ototoxicity is a less frequently described adverse effect of vancomycin and has not been as readily demonstrated, however vestibular damage and/or cochlear damage associated with tinnitus and sensorineural hearing loss has been reported in humans after administration of vancomycin [9-14]. The use of other ototoxic drugs known to cause sensorineural hearing loss, such as aminoglycosides or loop diuretics, concomitantly with vancomycin makes it difficult to prove a causal association [15].

The 2009 therapeutic monitoring guidelines for vancomycin in adults do not recommend routine audiogram monitoring for ototoxicity [1], but rather to consider the ototoxic potential of the drug in patients receiving additional ototoxic medications [1,16]. Questions remain related to the true vancomycin ototoxicity risk potential, particularly in the setting of more aggressive vancomycin dosing and in the aging population seen in clinical practice. Monitoring ototoxicity with serial audiograms is costly and patient compliance may be a barrier to routine monitoring [17]. Given the perceived risks of vancomycin- associated ototoxicity and the commonality of invasive MRSA infections in our region, audiogram monitoring for patients receiving prolonged vancomycin for any indication was adapted by select infectious diseases physicians at our institution.

The objective of this study was to evaluate the prevalence of and identify risk factors for worsening audiogram changes in patients receiving long-term intravenous vancomycin. Additionally, we sought to determine the need for the serial audiogram monitoring from an antimicrobial stewardship prospective.

Methods

This was a cross-sectional study performed at an academic medical center from January 2012 and March 2019. A nested case control analysis was conducted on patients who developed changes in audiograms. Cases included patients who developed a worsening change in follow-up audiogram, controls had no change in follow-up audiogram. Patients were included only once in the study to preserve statistical independence. The study was approved by the University of Tennessee Graduate School of Medicine institutional review board, and a waiver of informed consent was obtained.

Data were retrospectively gathered using chart review from the electronic medical record. Patients were included if they were ≥ 18 years old, prescribed vancomycin for ≥ 14 days, and had a baseline (within 3 days of initiating vancomycin therapy) and weekly follow-up audiograms. Patients were excluded if they had dementia or other cognitive dysfunction, a documented history of perforated tympanic membrane or the presence of hearing aids or cochlear implants, or if they did not have any follow-up audiograms.

Medical records were reviewed after patient identification and then standardized per protocol, and data collected included the following: patient age, race, sex, baseline serum creatinine and estimated creatinine clearance using the Cockroft-Gault equation, vancomycin dosing regimen used (i.e., dose, frequency, duration), all available vancomycin serum concentrations, infection type and vancomycin indication, microbiology and isolated organism MIC, and concomitant use of other ototoxic medications (i.e., aminoglycosides, furosemide and platinum-based chemotherapy).

The primary endpoint was worsening change in audiogram while on vancomycin. The procedure for audiogram monitoring in patients receiving vancomycin was not formally protocolized, but patients thought to be at risk for ototoxicity by the ordering provider typically received orders for weekly audiograms throughout the duration of vancomycin. Baseline and follow-up audiogram results were collected; certified audiologists interpreted audiograms. Findings were recorded for both ears ranked in order of increasing severity based on the sound intensity required for hearing as follows: normal, <25 dB; mild, 25 to 40 dB; moderate, 40 to 70 dB; severe, 70 to 90 dB; profound, >90 dB sensorineural hearing loss. Patients were considered to have a negative change in audiogram if there was a noted increase in the sound intensity required for hearing and worsening of more than 5 db from baseline.

Given the exploratory nature of this project, no formal sample size calculations were performed. Descriptive and bivariate analyses were utilized to describe the patient population. Categorical and continuous variables were compared using the Chi-Square, Fisher’s Exact, or Mann-Whitney U-tests. Classification and regression tree (CART) analyses were performed to identify dichotomous breakpoints in continuous variables associated with a negative change in audiogram. Rates of missing data for each key variable will be monitored throughout the study with a benchmark of <10% of individuals with missing data for a single key variable. Demographic or potentially confounding variables with missing data in excess of the benchmark will be removed and a suitable proxy variable will be chosen if possible. Patients without confirmed outcome status will be excluded from the analysis of that outcome.

Results

A list of adult patients who received vancomycin and at least one audiogram or audiogram order placed in the electronic health record was created in order to further screen patients for inclusion. Four-hundred and twenty-four patients were screened for inclusion; 92 (22%) received at least two audiograms while on vancomycin and were treated for at least 14 days. Baseline characteristics of the patient population are listed in Table 1. Most patients were men (53%) and Caucasian (95%); the median (IQR) patient age was 44 (37–61) years. Seven (8%) patients had an estimated creatinine clearance ≤ 30 mL/min using the Cockroft-Gault equation; 6 patients received hemodialysis.

Table 1

Baseline characteristics of patients based on change in audiogram during vancomycin therapy.

Variable	Total Population (n = 92)	No change in Audiogram (n = 85)	Negative Change in Audiogram (n = 7)	P-value†
Patient Demographics
Median (IQR) Age, years	44 (34–58)	43 (33–58)	43 (38–61)	0.57
Race, Caucasian	87 (95%)	81 (95%)	6 (86%)	0.33
Sex, Male	49 (53%)	46 (54%)	3 (43%)	0.7
Median (IQR) Body Mass Index, kg/m2	25.0 (20.5–31)	25.1 (20.3–30.6)	23.0 (20.0–37.7)	0.8
Estimated Creatinine clearance, at time of VAN
>100 mL/min	46 (50%)	42 (51%)	3 (43%)	1.0
80–99 mL/min	9 (10%)	9 (11%)	0	1.0
60–79 mL/min	15 (16%)	13 (15%)	2 (29%)	0.32
40–59 mL/min	12 (13%)	11 (13%)	1 (14%)	1.0
20–39 mL/min	3 (3%)	3 (4%)	0	1.0
< 20 mL/min	7 (8%)	6 (7%)	1 (14%)	0.44
Abnormal audiogram at baseline	51 (55%)	47 (56%)	4 (57%)	1.0
Vancomycin Characteristics
Number of audiograms, Median (IQR)	3 (2–5)	3 (2–5)	2 (2–6)	0.9
VAN mg/kg per dose, Median (IQR)	16.8 (14.5–19.2)	16.9 (14.6–19.4)	16.2 (13.6–17.9)	0.54
VAN concentrations, mcg/mL, Median (IQR)	17.6 (15.4–20.1)	17.6 (15.5–20.8)	18.6 (14.5–24.5)	0.83
VAN serum concentration ≥ 20 mcg/mL	68 (74%)	64 (75%)	4 (57%)	0.37
Daily VAN dose > 4 g/day	19 (21%)	17 (20%)	2 (29%)	0.63
VAN duration, days, Median (IQR)	42 (31–42)	42 (30–42)	42 (26–56)	0.82
Concomitant ototoxins	11 (12%)	11 (13%)	0	0.59
Infection Types
Osteoarticular	53 (58%)	47 (55%)	6 (86%)	0.12
Endocarditis	17 (18%)	17 (20%)	0	0.59
Bacteremia	10 (11%)	10 (12%)	0	1.0
Other	12 (13%)	11 (13%)	1 (4%)	1.0

Bivariate comparisons are between patients who did and did not have negative audiogram changes.

Abbreviations: VAN, vancomycin

A total of 351 audiograms were performed for 92 patients; the median (IQR) number of weekly audiograms while on vancomycin and time to first follow-up audiogram was 3 (2–5) and 7 (7–9) days, respectively. The median (IQR) vancomycin exposure up until the last recorded audiogram was 30 (17–42) days. Seventy-two percent of patients received at least 3 or more weekly audiograms while on vancomycin. The initial audiogram was performed while the patient was hospitalized, with subsequent audiograms performed in the outpatient setting for the majority of included patients.

Vancomycin indications were: 53 (58%) bone and joint infections, 17 (18%) infective endocarditis, 10 (11%) bacteremia, 12 (13%) other infections. The median prescribed vancomycin duration was 42 (28–42) days, and all patients had follow-up vancomycin concentrations; the median (IQR) vancomycin concentration was 17.6 (15.4–20.8) mcg/mL, baseline characteristics of patients based on change in audiogram during vancomycin therapy are listed in Table 1.

Seven patients (8%) experienced a worsening change in hearing from baseline audiogram, two patients (2%) experienced mild sensory neural loss, two patients (2%) had mild to moderate loss, and three patients (3%) developed moderate to severe loss. The median (IQR) time to negative audiogram change was 14 (14–28) days. Of the patients with worsening audiograms, four patients (57%) received concomitant antibiotics with vancomycin including cefepime, ertapenem and piperacillin-tazobactam. No patients received any concomitant ototoxins. Interestingly, no patients had vancomycin discontinued due to hearing loss, but rather for episodes of acute kidney injury (29%). The characteristics of patients with negative audiogram changes are listed in Table 2.

Table 2

Characteristics of patients with negative audiogram changes.

Pt. No.	Age, years	BMI, kg/m2	Estimated CrCl, mL/min	VAN mg/kg/ dose	Median VAN trough, mcg/mL	Indication	VAN duration, days	No. Audiograms	Baseline Audiogram Results	Follow-up Audiogram Results	Time to Negative Audiogram Change, days
1	56	33.0	71	16.75	18.33	Septic Arthritis	42	7.00	Mild HF HL bilaterally	Mild to sloping severe HL Bilaterally	28
2	59	21.0	8	14.12	20.38	Osteomyelitis	40	6.00	Normal Hearing Bilaterally	Mild to Mod HF hearing loss bilateraly	28
3	45	25.0	120	17.69	19.83	Brain abscess	48	5.00	Mild LF HL BL, Mod/Severe MidF HL BL	Increased HL MidF Left ear (still mod/severe)	35
4	41	32.0	71	18.90	14.55	Osteomyelitis	39	4.00	Mild sloping to at least severe HL bilaterally	Slightly decreased hearing bilaterally from baseline	14
5	64	54.9	50	12.06	37.02	Osteomyelitis	16	2.00	Mild HF HL bilateral ears	Mild to moderate-severe sensorineural HL bilaterally	14
6	40	21.0	124	17.23	14.20	Osteomyelitis	34	2.00	Normal Hearing Bilaterally	Mild conductive HL left ear	14
7	28	17.0	141.34	15.10	17.30	Osteomyelitis	40	2.00	Normal Hearing bilaterally	Mild HL at HF bilaterally	14

Abbreviations: Pt., patient; No., number; BMI, body mass index; CrCL, creatinine clearance; VAN, vancomycin; HL, hearing loss; LF, low frequency; HF, high frequency; Mid F, mid frequency; BL, bilateral

In bivariate analyses, no variables were found to be associated with a worsening change in audiogram, including a CART-derived age breakpoint of ≥ 65 years, elevated serum vancomycin levels, or vancomycin doses ≥ 4 grams/day (Table 3). No variables met criteria for inclusion in a multivariable regression model.

Table 3

Bivariate analyses of variables associated with negative audiogram changes.

Variable	Crude Odds Ratio(95% CI)	P-value
Vancomycin ≥ 4 grams/day	1.6 (0.3–8.8)	0.63
Vancomycin serum concentration ≥ 20 mcg/mL	0.4 (0.1–2.1)	0.4
Age ≥ 65 years	4.4 (0.5–38.2)	0.24
Concomitant ototoxins	0.9 (0.9–1.0)	0.6
Osteoarticular infections	5.8 (0.7–50.1)	0.12
Abnormal audiogram at baseline	1.1 (0.22–4.9)	1.0

Discussion

In this study, we found the overall prevalence of vancomycin-associated high frequency hearing loss to be low (8%). Most screened patients in our study had only a baseline audiogram and did not present for outpatient follow-up testing; only 22% of patients had a baseline audiogram within 3 days of vancomycin initiation with weekly follow-up audiograms. The majority of audiogram changes were mild to moderate in nature, and only 3% of our patient cohort developed moderate to severe hearing loss while on vancomycin; two of the three patients were receiving concomitant antibiotics including piperacillin tazobactam, and ertapenem, and none received another ototoxin. Ototoxicity occurred at different times of vancomycin exposures with as early as two weeks in two patients and as long as 4 or 5 weeks in others. No variables were found to be associated with a worsening change in audiograms in bivariate analyses. The results of this study prompted our antimicrobial stewardship program to stop routine ordering of serial audiograms for patients receiving vancomycin, which we expect to have significant patient satisfaction and cost savings implications. The median cost of an audiogram is more than $300 United States dollars [18].

Contrary to the low incidence of ototoxicity observed in our study, Forouzesh et al. reported a 12% rate of high-frequency hearing loss in patients receiving vancomycin [17]. The mean of the highest vancomycin trough level in their cohort was 19 mcg/mL, which did not differ between patients who developed ototoxicity or did not. The investigators found that patients older than 53 years old were more likely to develop ototoxicity, but no other study has been published since to confirm these findings [17, 19]. In our study, patients greater than 65 years old were found to have an overall higher proportion of negative audiogram changes, but this was not proven to be a significant finding possibly due to an underpowered event or due to variability in the population age distribution found in our cohort. James and colleagues observed reversible ototoxicity following vancomycin concentrations >40 mcg/mL and irreversible damage with concentrations >80 mcg/mL or the presence of pre-existing renal impairment [20]. We found no association between elevated vancomycin concentrations or high (i.e., more than 4 grams per day) vancomycin daily doses in our cohort, which is representative of current vancomycin dosing strategies [17,19]. We also did not find impaired renal function to be a risk factor for ototoxicity.

Vancomycin associated ototoxicity is a rare complication; it is mainly cochlear and rarely significant and is usually reversible except in few reported cases [3,21]. There are published reports of infants overdosed with vancomycin without ill effects [22], and even vancomycin exposure in utero did not cause ototoxicity in neonates [23]. While ototoxicity has been more frequently reported in patients with underlying hearing problems or who receive concurrent ototoxic medications, no clear association between ototoxicity and vancomycin serum levels has been established [15,24]. Despite the low risk of vancomycin ototoxicity, it is recommended to discontinue vancomycin in patients experiencing signs of ototoxicity including tinnitus, loss of balance or loss of hearing [16].

Audiogram monitoring in the in-patient setting has many limitations. This primarily requires patient cooperation that can be complicated by the use of other medications which may affect cognitive function, or could also be impacted by the patient’s underlying infection severity. Fifty-five percent of our patient population had an abnormal baseline audiogram, with noted improvement observed in some patients who received serial audiograms. Another limitation is that patients may have even a minor change of 2 decibels (dB) from baseline to follow-up audiogram and still meet the definition for a worsening audiogram. Campbell and colleagues reported that changes of > 5 dB are considered normal variability in an alert, oriented patient [25]. Alexander et al. performed large population-based cohort study evaluated the incidence of sudden sensorineural hearing loss in adults [26]. Results from that study suggested that the incidence of sudden sensorineural hearing loss (SSHL, unilateral loss of hearing occurring over 24 to 72 hours with 30 dB or more loss in at least 3 contiguous frequencies on pure-tone audiogram) was between 11 per 100,000 and 77 per 100,000 depending on the age group evaluated, with patients over 65 years being at highest risk. These findings add to the difficulty of the interpreting repeated audiograms and establishing plausibility of vancomycin-associated ototoxicity, especially in scenarios where no follow up audiograms performed after stopping vancomycin.

Ototoxicity has important quality of life implications, and screening for ototoxicity is important in high-risk patients especially when receiving concomitant ototoxic medications. The use of rapid diagnostic techniques in clinical settings, such as polymerase chain reaction testing, can aid clinicians in promptly stopping vancomycin when not indicated and decrease unnecessary exposures [19].

Limitations to our study include that it was a small retrospective study with no randomization of the selected patients, and there were no follow up audiograms after completion of vancomycin treatment to assess reversibility in patients who developed ototoxicity. One other potential limitation was that audiograms were performed in different clinical settings (i.e., inpatient and outpatient), which may cause some variation in interpretation. The frequency of audiogram testing (i.e., weekly) was not protocolized or consistent among patients and may limit external validity, but we feel these are pragmatic data to reflect patient follow-up. Our study may not have been appropriately powered to detect risk factors for negative changes in audiograms, and did not assess vestibular toxicity. However, our data represent real-world practice of audiogram monitoring in what was perceived to be a high-risk population.

Conclusions

The prevalence of high-level ototoxicity changes of audiograms in patients receiving long-term intravenous vancomycin was low. The utility of routine audiogram testing in this population does not seem warranted. However, larger studies may be warranted to further explore the risk of ototoxicity in high-risk patients and reversibility.

Data analysis.

(SAV)

Click here for additional data file.

10 Oct 2019

PONE-D-19-25571

Long-term vancomycin use had low risk of ototoxicity

PLOS ONE

Dear Dr Shorman,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR:

In my opinion, authors should correct some missing data according to "Strobe Statement check list on cross-sectional studies": specifically:

point 9: Describe any efforts to address potential sources of bias;

point 10: Explain how the study size was arrived at;

point 11:Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and why ;

point 12: (a) Describe all statistical methods, including those used to control for confounding (b) Describe any methods used to examine subgroups and interactions (c) Explain how missing data were addressed (d) If applicable, describe analytical methods taking account of sampling strategy  (e) Describe any sensitivity analyses.

No conflicts between the reviews.

Please consider comments by reviewers.

We would appreciate receiving your revised manuscript by october 30th. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Martina Crivellari

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear Dr Martina Crivellari,

Re: PONE-D-19-25571 Long-term vancomycin use had low risk of ototoxicity

Thank you for the kind invitation to review the manuscript. This study addresses an important evidence gap on vancomycin related ototoxicity. Below are my comments to the authors as there are some areas that need to be addressed before the manuscript can be published.

General comments:

Language

There are some incomplete sentences and grammatical errors scattered within the manuscript. I would suggest for the authors to re-read their manuscript prior to the next submission and correct the necessary errors.

Main text:

Methods:

1) Given the inter-physician variability on usage of audiogram for monitoring of vancomycin related ototoxicity, are there general principles that the clinicians in your centre apply when selecting patients to undergo audiogram? It would be worthwhile to describe this in the methodology and also the uptake rate of using audiogram for patients among physicians in your institution.

2) What is the change in audiogram findings that is deemed as significant in the study? This needs to be defined in the methodology for clinical interpretability.

Results:

1) A significant proportion of patients were excluded as audiogram was not performed. Were the demographics of patients included representative or comparable with entire patient population on vancomycin?

2) p-value for table 1 should be reported as the authors mentioned that univariate analyses was performed in their methodology.

3) Results for the renal function of patients should be segregated by CKD stage or eGFR cut-offs for readers to have a better understanding on renal profiles of included patients.

4) For Table 2, it would be useful to know when the audiogram abnormalities were detected (during their course of vancomycin therapy) and also the specific concomitant ototoxic agents that the patients may be on. If the salient points of audiogram findings can be presented in a weekly basis, it would be able to add value in terms of points for discussion on the onset of vancomycin ototoxicity among your patient population. I am also interested to know if vancomycin therapy was stopped following detection of audiogram abnormalities.

5) For the bivariate analyses, how was the cut-off of 65 years old selected? I noted the study by Forouzesh et al (also cited by the authors) identified the age cut-off of 53 years old to be a risk factor for ototoxicity. The authors may wish to contemplate using that cut-off instead. I also did notice that there is some discrepancies in the age cut-off reported in Table 3 (age >=40 years old) and in the main text (age >65 years old). Do clarify which cutoff was used in the analyses.

6) It is intriguing that among patients who developed audiogram abnormalities, some get 2 audiograms while others get up to 7 audiograms. What I would like to understand is how the frequency of monitoring for audiogram was first derived and determined within your institution?

Discussion

1) The first 2 paragraphs of the discussion appear to be less relevant to the topic that this manuscript wishes to cover. Do consider removing it.

2) Page 10 writes that 424 patients with weekly audiogram orders were evaluated in the study due to perceived vancomycin ototoxicity. Do the authors mean that these patients reported some hearing deficits?

3) Do the authors have any postulated reasons why age was not found to be a risk factor for vancomycin related ototoxicity?

4) The definition of changes in audiogram findings deemed to be significant should be described in the methodology instead of discussion.

5) Do the authors have any details on the baseline characteristics of their patients pertaining to dementia or other cognitive health problems which may confound their ability to comply with instructions for audiogram assessment?

Others minor comments

1) The keywords selected for the manuscript needs to be amended. Do use Mesh Terms where appropriate to allow wider reach of the article. I believe that the current keywords reflect the objective of the study instead of the keywords

Introduction:

2) Page 4 Line 87 to 90 – pls consider citing International journal of clinical pharmacy 40 (5), 977-981

3) There are some grammatical errors in the section for authors’ contributions. Kindly amend it.

4) There are some spelling errors in the manuscript. E.g. Page 8 Line 172

Reviewer #2: This is an original work which tries to find out whether vancomycin causes significant ototoxicity in long term. The patients size is somewhat sufficient and audiometric evaluations has been performed before and after or during the treatment. I think, this study merits to be published, because it provides a clear information about the safety of vancomycin in terms of ototoxicity.

I have an easy advice:

I recommend to use always ‘estimated glomerular filtration rate (eGFR)’ instead of ‘estimated creatinine clearance’. Probably, they used CKD-EPI equation for it. So it maybe it would be better to cite it (Ann Int Med 2009; 150(9):604-612).

In the abstract

Line 47, Line 153 and in Table 1.

Reviewer #3: The study presents the results of original research and the authors reported that the results have not been published elsewhere. The statistics analysis was performed with technical standard and is described in sufficient detail, although it is a simple analysis, it fulfills the purpose set by the authors. The conclusions presented are supported by the data; according to the objective of the study, the prevalence of negative changes in audiograms of the patients in their medical center was determined; however, the analysis was carried out with very small sample size. Since there are few reports that take into account the hearing damage associated to prolonged use of vancomycin, in an objective way, it can be considered that the data provided by this study could serve as a guide for other medical centers.

It is recommended that the authors review the document, to correct errors regarding the use of vancomycin concentration units, as well as the correct use of the abbreviation et al. There are errors in the writing of references, they are not uniform. Except for minor changes, it can be considered accepted, taking into account that the brief results could serve as a guide for other medical centers.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

14 Oct 2019

Academic Editor

Comment 1. Describe any efforts to address potential sources of bias

Response 1. We have addressed potential sources of bias of confounding in the discussion section of our manuscript (lines 268-278).

Comment 2. Explain how the study size was arrived at;

Response 2. We have added sample size information in the manuscript (lines 147-148)

“Given the exploratory nature of this project, no formal sample size calculations were performed.”

Comment 3. Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and why

Response 3. We have described our statistical methods in lines 147-157.

Comment 4. (a) Describe all statistical methods, including those used to control for confounding (b) Describe any methods used to examine subgroups and interactions (c) Explain how missing data were addressed (d) If applicable, describe analytical methods taking account of sampling strategy (e) Describe any sensitivity analyses.

Response 4. We have described our statistical methods in lines 147-157.

Reviewer #1

Comment 1. There are some incomplete sentences and grammatical errors scattered within the manuscript. I would suggest for the authors to re-read their manuscript prior to the next submission and correct the necessary errors.

Response 1. Thank you for this comment. We have critically reviewed the paper for grammar errors and misspellings. We believe we have addressed all of these issues.

Comment 2. Given the inter-physician variability on usage of audiogram for monitoring of vancomycin related ototoxicity, are there general principles that the clinicians in your centre apply when selecting patients to undergo audiogram? It would be worthwhile to describe this in the methodology and also the uptake rate of using audiogram for patients among physicians in your institution.

Response 2. We appreciate this insight. No formal principles were used to select patients under-going audiograms, outside of long-term vancomycin use, and this process was limited to the discretion of the ordering physician. We have clarified this throughout the manuscript (lines 101-104):

“Given the perceived risks of vancomycin- associated ototoxicity and the commonality of invasive MRSA infections in our region, audiogram monitoring for patients receiving prolonged vancomycin for any indication was adapted by select infectious diseases physicians at our institution.”

And (lines 136-139)

“The procedure for audiogram monitoring in patients receiving vancomycin was not formally protocolized, but patients thought to be at risk for ototoxicity by the ordering provider typically received orders for weekly audiograms throughout the duration of vancomycin.”

Comment 3. What is the change in audiogram findings that is deemed as significant in the study? This needs to be defined in the methodology for clinical interpretability.

Response 3. We have included the interpretive audiogram results that coincide with any significant changes as interpreted by certified, hospital-based audiologists (lines 139-145):

“Baseline and follow-up audiogram results were collected; certified audiologists interpreted audiograms. Findings were recorded for both ears ranked in order of increasing severity based on the sound intensity required for hearing as follows: normal, <25 dB; mild, 25 to 40 dB; moderate, 40 to 70 dB; severe, 70 to 90 dB; profound, >90 dB sensorineural hearing loss. Patients were considered to have a negative change in audiogram if there was a noted increase in the sound intensity required for hearing and worsening of more than 5 db from baseline.”

Comment 4. A significant proportion of patients were excluded as audiogram was not performed. Were the demographics of patients included representative or comparable with entire patient population on vancomycin?

Response 4. We did collect sufficient demographic data on patients who did not receive at least 14 days of vancomycin, or who did not receive at least two follow-up audiograms. However, we have no reason to suspect the population of patients included in this study are not representative of the general population of patients receiving vancomycin. We unfortunately do not have IRB approval to access data on patients who were excluded from this study.

Comment 5. p-value for table 1 should be reported as the authors mentioned that univariate analyses was performed in their methodology.

Response 5. We have included P-values for the data included in Table 1.

Comment 6. Results for the renal function of patients should be segregated by CKD stage or eGFR cut-offs for readers to have a better understanding on renal profiles of included patients.

Response 6. Thank you for this comment. We measured renal function using the Cockcroft-Gault equation, and this has been updated throughout the manuscript when creatinine clearance is discussed. We have also segregated patient renal function into different cut-offs as suggested, in Table 1.

Comment 7. For Table 2, it would be useful to know when the audiogram abnormalities were detected (during their course of vancomycin therapy) and also the specific concomitant ototoxic agents that the patients may be on. If the salient points of audiogram findings can be presented in a weekly basis, it would be able to add value in terms of points for discussion on the onset of vancomycin ototoxicity among your patient population. I am also interested to know if vancomycin therapy was stopped following detection of audiogram abnormalities.

Response 7. Thank you for this comment. None of the patients who developed a negative change in audiogram were on a concomitant ototoxin (line 189-190, Table 2), and no patients discontinued vancomycin secondary to hearing loss (line 190-191). In regards to timing of negative audiogram change, we have added this information in the results section of the manuscript (line 187) and in Table 2.

Comment 8. For the bivariate analyses, how was the cut-off of 65 years old selected? I noted the study by Forouzesh et al (also cited by the authors) identified the age cut-off of 53 years old to be a risk factor for ototoxicity. The authors may wish to contemplate using that cut-off instead. I also did notice that there is some discrepancies in the age cut-off reported in Table 3 (age >=40 years old) and in the main text (age >65 years old). Do clarify which cutoff was used in the analyses.

Response 8. Thank you for this observation. The cut-off of 65 years was derived from a classification and regression tree (CART) analyses, as described in our methods (lines 150-152):

“Classification and regression tree (CART) analyses were performed to identify dichotomous breakpoints in continuous variables associated with a negative change in audiogram.”

The breakpoint of 65-years is based only on our population distribution, as was 53-years determined by Forouzesh and colleagues. We feel using 65-years as a cut-off is a better representation of our own data as opposed to Forouzesh and colleagues. We have also updated Table 3 to state 65-years instead of 40 years. Thank you for identifying this oversight.

Comment 9. It is intriguing that among patients who developed audiogram abnormalities, some get 2 audiograms while others get up to 7 audiograms. What I would like to understand is how the frequency of monitoring for audiogram was first derived and determined within your institution?

Response 9. Thank you for this comment. While there is no standardized process in place, the general approach was to obtain weekly audiograms while the patient was on vancomycin. This was included in the manuscript in lines 136-139:

“The procedure for audiogram monitoring at our institution was not formally protocolized, but patients thought to be at risk for ototoxicity by the ordering provider typically received orders for weekly audiograms throughout the duration of vancomycin.”

And (lines 273-275)

“The frequency of audiogram testing (i.e., weekly) was not protocolized or consistent among patients and may limit external validity, but we feel these are pragmatic data to reflect patient follow-up.”

Comment 10. The first 2 paragraphs of the discussion appear to be less relevant to the topic that this manuscript wishes to cover. Do consider removing it.

Response 10. Thank you for this feedback. We have removed these paragraphs from the discussion.

Comment 11. Page 10 writes that 424 patients with weekly audiogram orders were evaluated in the study due to perceived vancomycin ototoxicity. Do the authors mean that these patients reported some hearing deficits?

Response 11. We originally generated a screening list based on the following criteria: age > 18 years, receipt of vancomycin order, and receipt of at least one audiogram or an audiogram order placed in the electronic health record. We then proceeded to screen the 424 patients to assess if they met all inclusion criteria.

We have added the following sentence for clarity (lines 160-162):

“A list of adult patients who received vancomycin and at least one audiogram or audiogram order placed in the electronic health record was created in order to further screen patients for inclusion.”

Comment 12. Do the authors have any postulated reasons why age was not found to be a risk factor for vancomycin related ototoxicity?

Response 12. We hypothesize that our study may have been underpowered to detect other variables associated with ototoxicity when long-term vancomycin is used. We have included this rationale in the discussion, lines 221-225:

“In our study, patients greater than 65 years old were found to have an overall higher proportion of negative audiogram changes, but this was not proven to be a significant finding possibly due to an underpowered event or due to variability in the population age distribution found in our cohort.”

Comment 13. The definition of changes in audiogram findings deemed to be significant should be described in the methodology instead of discussion.

Response 13. We have included these data in the methods section of the manuscript (lines 140-145). Our comment in the discussion was only to highlight that a small decibel difference in would lead potentially lead significant changes in audiogram interpretation, and is an inherent limitation of our findings.

Comment 14. Do the authors have any details on the baseline characteristics of their patients pertaining to dementia or other cognitive health problems which may confound their ability to comply with instructions for audiogram assessment?

Response 14. We do not have specific data on dementia diagnoses or baseline impaired cognition in from electronic health record data. However, our audiologists screen and comment on any potential confounding factors that may skew the results of the audiogram test. None of the patients included would have received an audiogram if they were deemed to not have the capacity to reliably comply to testing, based on institutional standards. We have therefore added language that patients with dementia or cognitive disorders were excluded from the study (lines 123-125):

“Patients were excluded if they had dementia or other cognitive dysfunction, a documented history of perforated tympanic membrane or the presence of hearing aids or cochlear implants, or if they did not have any follow-up audiograms.”

Comment 15. The keywords selected for the manuscript needs to be amended. Do use Mesh Terms where appropriate to allow wider reach of the article. I believe that the current keywords reflect the objective of the study instead of the keywords

Response 15. We have modified the article keywords to MeSH Terms: vancomycin, hearing loss, drug toxicity, antimicrobial stewardship, methicillin-resistant Staphylococcus aureus

Comment 16. Page 4 Line 87 to 90 – pls consider citing International journal of clinical pharmacy 40 (5), 977-981

Response 16. Thank you for this feedback. We have incorporated this publication from Seng and colleagues to support our statement related to vancomycin ototoxicity.

Comment 17. There are some grammatical errors in the section for authors’ contributions. Kindly amend it.

Response 17. We have modified the grammatical errors for the Author’s Contributions (lines 13-17):

“CH established the electronic case report form, performed data collection, and assisted in manuscript preparation. MPV performed the statistical analyses and assisted in manuscript preparation. BW assisted in establishing the electronic case report form and performed data collection. MAS designed the study and wrote the manuscript.”

Comment 18. There are some spelling errors in the manuscript. E.g. Page 8 Line 172

Response 18. We have thoroughly reviewed the manuscript in the context of spelling and grammar. Thank you for this comment.

Reviewer #2

Comment 1. I have an easy advice:

I recommend to use always ‘estimated glomerular filtration rate (eGFR)’ instead of ‘estimated creatinine clearance’. Probably, they used CKD-EPI equation for it. So it maybe it would be better to cite it (Ann Int Med 2009; 150(9):604-612).

In the abstract

Line 47, Line 153 and in Table 1.

Response 1. Thank you for this comment. The actual equation used to assess renal function was the Cockcroft-Gault equation, and not eGFR. We have corrected this error throughout the manuscript and in Table 1. Furthermore, we have stratified the creatinine clearance by range instead of reporting the median (IQR), as suggested by Reviewer #1.

Reviewer #3

Comment 1. It is recommended that the authors review the document, to correct errors regarding the use of vancomycin concentration units, as well as the correct use of the abbreviation et al.

Response 1. Thank you for these comments. To stay consistent with vancomycin unit reporting, all units have been modified to micrograms per milliliter (mcg/mL). Additionally, we have made appropriate modifications regarding the correct use of et al., where applicable.

Comment 2. There are errors in the writing of references, they are not uniform.

Response 2. We have modified the reference formats per journal instructions.

Submitted filename: response to reviewers.docx

Click here for additional data file.

17 Oct 2019

Long-term vancomycin use had low risk of ototoxicity

PONE-D-19-25571R1

Dear Dr. Shorman,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Martina Crivellari

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

All comments have been addressed.

28 Oct 2019

PONE-D-19-25571R1

Long-term vancomycin use had low risk of ototoxicity

Dear Dr. Shorman:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Martina Crivellari

Academic Editor

PLOS ONE