PREVALENCE OF HEARING SCREENING FAILURES IN LOW-RISK CHILDHOOD CANCER SURVIVORS

Abstract

Background:

We sought to estimate the frequency of hearing screening failures in pediatric cancer survivors at low risk for hearing loss and evaluate the feasibility of administering screenings in this population.

Procedure:

Survivors in the St. Louis Children’s Hospital Late Effects Clinic were recruited. Eligibility included: 1) Diagnosis of a pediatric cancer treated without platinum chemotherapy or cranial radiation, 2) At least 6 months from completion of therapy, 3) Between the ages of 7–18 years, 4) Cognitively/behaviorally able to participate 5) English speaking. Behavioral hearing screenings from 1000–8000 Hz were performed by trained personnel using a calibrated audiometer. A failed screen was defined by a participant not responding to 2 or more of the 3 screening attempts for at least one frequency in at least one ear.

Results:

One hundred nine patients met eligibility criteria with 78 enrolled (71.5%). Diagnoses included leukemia (57.7%), sarcoma (11.5%), Wilms tumor (14.1%), lymphoma (12.8%), and other solid tumors (3.9%). The median age was 13.2 years ([Q1-Q3]: 9.6–15.4) and the median time from treatment completion was 3.7 years ([Q1-Q3]: 2.3–7.4). Eighteen patients (23%) failed the hearing screen (95% CI: 14–34%). No demographic or treatment-related variables were significantly correlated to screening failure. Six screen failures (33%) underwent formal audiology assessments with 3 demonstrating unilateral hearing loss: 2 conductive and 1 sensorineural.

Conclusions:

A significant fraction of pediatric cancer survivors at low risk for hearing loss failed hearing screening. Broader use of hearing screening should be considered.

Introduction

Survival rates for childhood cancer survivors (CCS) now exceeds 85%, due to advances in multi-modal therapy^1–4. However, more long-term chronic health conditions from their therapy accumulate, including hearing loss ^5,6. Hearing loss has been reported in up to 80% of CCS who received ototoxic therapies including platinum-based chemotherapy or radiation to the head ^7–13. Ototoxic monitoring following guidelines from the Children’s Oncology Group have been developed for children who received platinum-based cancer therapy or cranial radiation^10,14, but there are no recommendations for those who lack these risk factors¹⁵.

The possibility that others may be at risk for hearing loss was suggested in a report from the Childhood Cancer Survivor Study (CCSS)¹⁶. Self-reported data from 24,305 CCS who were at least 5 years post end of treatment, examined the frequency of significant hearing loss requiring a hearing aid or other intervention. Patients without prior exposure to platinum chemotherapy or cranial radiation, were still at higher risk of significant hearing loss compared to a sibling control group (HR 3.3, 1.9–5.6, <.0001). The authors concluded that CCS treated without known ototoxic therapy are still at risk.

The patients from this study fell into two major diagnostic categories: leukemia and solid tumors. Leukemia patients receive intrathecal chemotherapy, which is known to be neurotoxic. However, there have been no studies that have examined whether such therapy is ototoxic. Further, the use of aminoglycosides and loop-diuretics may be used during treatment, which can be ototoxic ¹⁷. Solid tumor patients often receive radiation to the site of their tumor, typically in the abdomen. A recent report demonstrated that radiation to parts of the body other than the head or neck was a risk factor for the development of late onset hearing loss in cisplatin treated patients¹⁸. In a separate report from the CCSS it was reported that patients were at a significantly increased risk of developing a chronic health condition subsequent to radiation therapy¹⁹. Thus, radiation may influence tissues outside the targeted area ^20,21.

Focused monitoring for late effects in CCS is essential for optimal medical management and to improve quality of life ^8,10,22. Unidentified childhood hearing loss can affect educational achievement, socioemotional development, quality of life, and employment as adults ^23–25. The American Academy of Audiology recommends annual hearing screening beginning in early childhood, with special focus on monitoring hearing of high-risk populations²⁶.

In the general population of children, the current literature demonstrates a rate of behavioral hearing screening failures for frequencies 1000–4000 Hz to be approximately 10% ^27,28. Survivors with school performance issues in our late effects program routinely undergo audiologic screening in the office, which has revealed 7 fails out of 24 screenings, or approximately 29%. Of this group, 5 patients received a full audiogram showing normal hearing. We therefore wished to assess the feasibility of performing hearing screening during a routine CCS clinic visit and acquire preliminary data on the prevalence of hearing screen failures in cancer survivors deemed to be at low risk for hearing loss.

Methods

Patient Recruitment:

Patients managed in the Division of Hematology and Oncology at St. Louis Children’s Hospital (SLCH) and affiliated clinics were recruited to participate during routine follow-up appointments from 6/14/2019 to 2/24/2020. Eligible patients were identified by reviewing the medical records prior to their scheduled visits. Eligibility criteria included: 1) Diagnosis of a cancer < 18 years of age, 2) completed all therapy for at least six months, 3) Age: ≥ 7 years old to ≤ 18 years, 4) English speaking, 5) Cognitively and behaviorally able to be evaluated by standard conventional hearing screening. Exclusion criteria included: 1) Prior exposure to treatments known to be associated with ototoxic hearing loss (cranial radiation, carboplatin/cisplatin, bone marrow transplant), 2) A diagnosis of a central nervous system tumor, 3) Family history/genetic predisposition to hearing loss 4) Pre-existing hearing loss prior to the diagnosis of cancer 5) Audiogram within one year of enrolment (to exclude those whose recent hearing status is known, without the need for screening).

Eligible patients were approached during routine outpatient visits to participate in the study. A signed informed consent and assent approved by the Human Research Protection Office’s Institutional Review Board was obtained. All participants enrolled on a voluntary basis, with a $10 gift card was provided as compensation. Designated members of the study team were formally trained and approved by the Department of Audiology at SLCH to use the screening equipment. A patient’s participation was complete upon the completion of the hearing screen. Those patients who failed the hearing screen were referred to the Department of Audiology for full diagnostic testing.

Hearing Screening:

Hearing screens were performed in exam rooms at SLCH or an affiliated clinic. Quality control of the environment was assured by measuring ambient noise in the patient exam rooms (Supplement 1). These noise levels were chosen based on the prevailing literature, as well as the screening recommendation from the American National Standards Institute²⁹.

Screeners utilized TDH-39 supra-aural headphones and a calibrated standard commercial Grason-Stadler GSI-18 audiometer. The child was seated facing away from the audiometer to ensure that he/she did not receive visual cues from the screener. Instructions for how to respond were given to the child prior to placing headphones on his/her ears.

The screening protocol included screening 1000 Hertz (Hz), 2000 Hz, 3000 Hz, 4000 Hz, 6000 Hz, and 8000 Hz. School screenings typically only test 1000–4000 Hz, but this study included 6000 and 8000 Hz as most ototoxic hearing loss is first observed in these high frequencies. The screener began testing at 1000 Hz and presented the tone beginning at 40 decibel (dB) HL to condition the patient to the task. If a response was obtained, the screener decreased the level to 25 dB HL and presented the tone again. The screener presented the tone at least two times at each frequency to verify the response. Criteria for a screen failure consisted of no hand raise response or only 1 response that could not be repeated at 20 dB HL at 2000–6000 Hz or 25 dB HL at 1000 Hz or 8000 Hz. Three attempts were made at each frequency and in both ears.

Hearing screen results were reported as pass or fail overall (and at each frequency in each ear). The location of screening, and the performance of a post hearing screen diagnostic hearing test were also recorded. Audiograms were extracted from the electronic chart if obtained within a year after the hearing screen. A standard procedure for the interpretation of audiometric data and determination of hearing loss was followed.

Demographic and medical information collected through review of the medical record included date of birth, date of diagnosis, date of completion of therapy, age, gender, race, ethnicity, diagnosis, exposure to furosemide or aminoglycosides, details of the patient’s cancer therapy including: all chemotherapy agents, number of intrathecal administrations, site of radiation, total dose of radiation, boost dose of radiation.

Study Design/Statistical Analysis:

This was a cross sectional study to acquire preliminary data to estimate of the prevalence of hearing screening failures in a pediatric cancer survivor population at low risk for therapy induced ototoxicity. We arbitrarily chose a screening failure rate of 20%, which was deemed to be a significant increase over the 10% rate documented among the general pediatric population in current literature. A sample size of 78 provided 80% power to detect an increase to 20% over the historically documented rate of 10% at a .05 significance level using a one-sided exact test for difference of one proportion.

All statistical analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC). Demographic and clinical characteristics were summarized by descriptive statistics, i.e., median and interquartile range (IQR) for continuous variables, and count and percentage for categorical variables. The group differences between hearing screen pass or fail were examined by Kruskal-Wallis test for continuous variables and Fisher’s exact test (if cell count less than 5) or Chi-square test for categorical variables.

This study was reviewed and approved by the Protocol Review and Monitoring Committee and Institutional Review Board at Washington University School of Medicine.

Results

Participants included in this study were consented and enrolled between June 14, 2019 and February 24, 2020. Figure 1 outlines the accrual for the study. A total of 147 patients were identified as potentially eligible based on age and diagnosis. Of this number, 38 did not meet full eligibility criteria. The remaining 109 patients met eligibility criteria. Of this group, 3 patients or families were deemed unapproachable, 2 patients were clinically ill at time of consent, 18 patients were unavailable for testing due to time constraints, and 8 patients declined participation in the study. The remaining 78 eligible patients in the cohort (71.5%) were consented and enrolled in the study.

Figure 1:

Consort Diagram of Patient Recruitment: Summary of patient recruitment for study participation. Reasons patients failing to meet eligibility criteria are displayed. Reasons characterizing recruitment failures are also outlined.

Table 1 summarizes characteristics for the cohort in the study. Of the 78 enrolled participants, 41 were males (52.6%) and 37 were females (47.4%). The cohort was predominantly Caucasian (71/78, 91%). Most patients had a diagnosis of leukemia (45/78, 57.7%), with other diagnoses including 9 patients with sarcoma (11.1%), 11 patients with Wilms tumor (14.1%), 10 patients with lymphoma (12.8%) and 3 patients with other solid tumors (3.9%). There were 51 patients (65.4%) who received intrathecal chemotherapy and 8 patients (10.3%) who received radiation. There were 15 patients (19%) that received either furosemide or gentamicin as part of treatment.

TABLE 1.

Patient characteristics

	Number (%)
Total patients	78(100%)
Gender
Male	41 (52.6%)
Female	37 (47.4%)
Race
African-American	5 (6.4%)
Caucasian	71 (91%)
Hispanic	2 (2.6%)
Diagnosis
Leukemia	45 (57.7%)
Sarcoma	9 (11.1%)
Wilms Tumor	11 (14.1%)
Lymphoma	10 (12.8%)
Other solid tumor	3 (3.9%)
Time from diagnosis (years)
Mean	6.9
Median (Range)	6.2 (1.0–14.9)
Age at analysis (years)
Mean	12.8
Median (range)	13.2 (7/18)
Intrathecal Chemotherapy
Yes	51 (65.4%)
No	27 (34.6%)
Radiation
Yes	8(10.3%)
No	70 (89.7%)
End of treatment to hearing screen (years)
Mean	5
Median (range)	3.7 (0.5–13.0)

The median age for participants was 13.2 years ([Q1-Q3]: 9.6–15.4). The median time from diagnosis was 6.2 years ([Q1-Q3]: 4–9.2). Participants who failed had a longer time interval from diagnosis to screening, with a median of 7.8 years ([Q1-Q3]: 4–11.2) compared to a median of 5.7 years ([Q1-Q3]: 4–8.7) for participants who passed. The median time from end of treatment was 3.7 years ([Q1-Q3]: 2.3–7.4). Participants who failed were also further from the end of treatment, with a median of 6.4 years ([Q1-Q3]: 2.2–9.5) compared to 3 years ([Q1-Q3]: 2.3–7) for participants who passed.

Of the 78 patients who successfully completed the screening, 18 (23.1%) failed (95% CI: 14–34%, Table 2). Twenty-seven ears of these 18 patients (36 ears) failed the screen with the most prevalent frequencies failed consisting of 6000 Hz (17/27) and 8000 Hz (10/27), and 1000 Hz, (8/27). There was an even distribution of patients who failed bilaterally versus unilaterally. Despite recommendations for formal audiology evaluations, only 6 (33%) underwent evaluations with 3 demonstrating hearing loss. The association between participant characteristics and hearing screen failures is shown in Table 2. Gender, race, diagnosis, intrathecal chemotherapy, radiation treatment, exposure to other ototoxic agents (furosemide, gentamicin), age, and time from diagnosis and treatment were not significant predictors of a hearing screening failure.

TABLE 2.

Association between participant characteristics and hearing screen fail

	Hearing Screen Results
Characteristics a	Pass (%)	Fail (%)	p b
No. of Patient	60 (76.9%)	18 (23.1%)
Categorical
Sex			0.8038
Male	32 (78.1%)	9 (21.9%)
Female	28 (75.7%)	9 (24.3%)
Race			0.7564
White/Caucasian	55 (77.5%)	16 (22.5%)
Hispanic or Latino	1 (50.0%)	1 (50.0%)
Black or African American	4 (80.0%)	1 (20.0%)
Diagnosis			0.6194
Leukemia	36 (80.0%)	9 (20.0%)
Sarcoma	7 (77.8%)	2 (22.2%)
Wilms Tumor	8 (72.7%)	3 (27.3%)
Lymphoma	6 (60.0%)	4 (40.0%)
Other Solid Tumor	3 (100.0%)	0 (0.0%)
Intrathecal Therapy			0.8963
Yes	39 (76.5%)	12 (23.5%)
No	21 (77.8%)	6 (22.2%)
Radiation			1.0000
Yes	6 (75.0%)	2 (25.0%)
No	54 (77.1%)	16 (22.8%)
Continuous
Age (years)	60, 12.5 (9.5–14.8)	18, 14.7 (12.4–16)	0.2045
Time from Diagnosis (years)	60, 5.7 (4–8.7)	18, 7.8 (4–11.2)	0.2701
Time from Treatment (years)	60, 3 (2.3–7)	18, 6.4 (2.2–9.5)	0.2356

^aFor categorical variables, (n (%)) is reported, excluding missing values. For continuous variables, “n, median (Q1-Q3)” is reported, where Q1 is 25^th percentile, and Q3 is 75^th percentile.

^bFisher’s exact test (if cell count less than 5) or Chi-square test for categorical variable; Kruskal-Wallis test for continuous variable.

Of the patients who ultimately underwent a comprehensive hearing test, all 6 failed the screen at least at 6000 and/or 8000 Hz in at least one ear. At follow up diagnostic testing, 3 of the 6 patients were confirmed to have hearing loss in the 6000–8000 Hz region. Of the remaining 12 screening failures, one patient was scheduled for a hearing test but has not yet been evaluated, 2 patients expressed interest in following up but preferred to do so closer to home, 3 patients declined pursuing a hearing test, and 6 patients did not specify their plans.

Table 3 displays the hearing screen results of the 6 patients who failed compared to their diagnostic audiologic results. For patient LRHL09, the audiogram showed a mild conductive hearing loss at 4000 Hz-8000 Hz in the right ear, with normal thresholds in the left ear. This patient had abnormal tympanograms bilaterally with no peak pressure, which is indicative of fluid present in the middle ear space. She ultimately received pressure equalization (PE).

TABLE 3.

Description of bilateral and unilateral screening failures

Patient ID	Age, in years, at screening	Diagnosis	Frequency (Hz) failed in right ear:	Frequency (Hz) failed in left ear:
Patient LRHL03	15	ALL	3000, 6000, 8000	2000
Patient LRHL08	16	Wilms tumor		6000
Patient LRHL09	8	ALL	1000, 2000, 3000 4000, 6000, 8000	1000, 2000, 8000
Patient LRHL16	17	AML	1000	1000
Patient LRHL21	15	ALL	8000	4000, 6000, 8000
Patient LRHL24	17	Lymphoma	8000
Patient LRHL35	8	ALL	6000	1000, 2000
Patient LRHL36	12	Wilms tumor	3000
Patient LRHL48	7	Sarcoma		6000
Patient LRHL54	13	ALL	6000	6000
Patient LRHL57	9	ALL	6000, 8000
Patient LRHL62	14	Lymphoma	6000, 8000	6000, 8000
Patient LRHL65	15	ALL	2000, 3000, 8000
Patient LRHL68	15	Lymphoma		1000 and 6000
Patient LRHL69	16	Lymphoma		6000
Patient LRHL73	16	ALL	1000, 6000	1000
Patient LRHL76	15	Sarcoma	6000	6000
Patient LRHL77	12	Wilms tumor		6000

ALL= Acute lymphoblastic leukemia, AML= Acute myelogenous leukemia

Patient LRHL62 failed the screen and received a diagnostic audiologic evaluation that showed a mild sensorineural hearing loss in her right ear from 6000–8000 Hz, with normal tympanograms bilaterally. This patient has a medical diagnosis of Noonan’s syndrome and previously had a hearing test in 2010 that showed normal hearing.

Patient LRHL65’s audiologic evaluation showed a mild to moderate CHL at 4000–8000 Hz in the right ear. This patient had an abnormal tympanogram in the right ear with no peak pressure. A repeat audiogram is planned in conjunction with an otolaryngology evaluation to address the conductive hearing loss.

Discussion

The current study is one of the first reports of hearing screening in cancer survivors at low risk for hearing loss. Such patients are not routinely screened for hearing loss based on long-term follow-up guidelines from the Children’s Oncology Group ³⁰. However, the findings of this study bring these recommendations into question. Our findings support previous investigations of low risk pediatric cancer survivors suggesting a proportion of these patients may experience significant hearing loss as they age¹⁶. Given the observed screening failure rate of 23%, these patients may warrant further evaluations.

The failure rate of this study population is substantially higher than what has been reported in the general population based on failure rates reported in the literature; however, our testing included the frequencies 6–8 kHz, which are not typically included in hearing screenings which may account for the higher failure rate we observed. We have successfully demonstrated that performing hearing screenings on patients in our survivorship clinic was feasible, and we have established quality measures to ensure that the staff were trained, and the rooms utilized for testing were appropriate without excessive background noise.

While this population has not been routinely screened for hearing loss, many of these patients are at known risk for cognitive impairment, including deficits in attention, processing speed, and memory due to their anti-cancer therapy ^31–33. As sensory inputs impact cognitive development and function, a hearing loss can further exacerbate difficulties or delays. Recent studies have shown that neurocognitive deficits often manifest as academic difficulties in childhood cancer survivors who were treated both with and without CNS-directed therapies ^34–36.

Cognitive or behavioral problems in adolescence for CCS have been shown to significantly decrease their likelihood of higher education attainment and graduation from college ³⁷. Hearing loss was not addressed in these studies; however, any degree of loss can put children at risk for impaired academic advancement and unemployment as adults ³³. Early identification of hearing loss and intervention with appropriate amplification can moderate the effects on outcomes in language and academic performance ³⁸.

It is difficult to postulate possible explanations for the increased rate of failed hearing screenings in this low-risk population given that we were unable to obtain a diagnostic audiologic evaluation on most of the patients. The low percentage of screening failures who proceeded to formal audiologic testing limits our ability to speculate on potential mechanisms for hearing loss. Analysis of the variables collected in this report failed to identify significant risk factors. Future research should investigate larger cohorts to identify risk factors that were not discovered in this report or underlying genetic factors that may result in hearing loss.

One patient with sensorineural hearing loss had the diagnosis of Noonan Syndrome. Most studies examining the nature of hearing loss in patients with Noonan’s syndrome indicate that temporary conductive hearing loss due to otitis media with effusion is the most common cause ³⁹. While sensorineural hearing loss has also been described in Noonan Syndrome and may still be underreported it is not included as a criterion for diagnosis of Noonan Syndrome⁴⁰. This patient had previous hearing testing showing normal hearing and has not complained of any difficulties hearing.

Conductive hearing loss was observed in two of the six formally tested patients, so we suspect that this represents some fraction of the patients with hearing loss. These patients were also older in age: 8 years old and 15 years old. Conductive hearing loss is likely to occur more often in younger children ages 0–4 years old, with prevalence estimates for otitis media with effusion between 10–17% for this age group⁴¹. In children between 6–8 years old, the likelihood decreases to 3–4%, implying that older children are less likely to present with a conductive hearing loss from otitis media with effusion⁴². The identification of such losses and the positive impact of interventions such as PE tubes suggests that CCS may be at risk for conductive hearing loss and the identification and management of such losses could have a substantial impact on these patients.

Noise induced hearing loss could certainly account for others, however, noise induced hearing loss typically manifests as a “notch” in the thresholds, which was not reflected in the screens. Formal audiology testing would be necessary to clearly characterize the phenotype^43,44. Regardless of the etiology, hearing preservation should include strategies to protect from further loss due to noise.

The low participation rate in formal testing also limits our ability to establish the true incidence of hearing loss and prevents us from establishing clear recommendations for who should undergo formal testing. We did demonstrate that screening was feasible within the confines of a late effects clinic and that the percent of failures warrants further investigation. Difficulties in formally testing hearing screen failures have been also observed in primary care settings, where only 25% (7/28) of children ages 3–19 years old who were referred for a hearing test returned for audiologic evaluation ²⁷. In another study performing hearing screenings in a school environment, the return rate for children in grades 1–7 who failed the screening at 20 dB HL was as low as 17.5% (24/137)²⁸. Thus, the rate of non-compliance observed in this study is not unique to CCS but its impact given this population’s risk for cognitive impairment may be more significant.

Incomplete follow-up by cancer survivors is not unique to audiology. A report from the CCSS indicated that only 42% of survivors had attended a risk-based appointment in the past 2 years ¹⁵. One study reviewing 400 medical charts of survivors of pediatric cancer showed that while 90% of the cohort was at risk for at least one late effect from treatment, only 40% of the patients had been seen for follow-up care within a year of their last recommended appointment ⁴⁵. Age and years from completion of therapy are significant predictors of non-compliance in CCS seen at late effects clinics, as they are less likely to maintain follow-up the further they are from the end of treatment ^45–47.

Other factors associated with reduced follow-up for CCS include socioeconomic status, race, access to medical insurance, distance from hospital, and lack of secondary cancer events ^48,49. Additionally, parents of CCS are often psychologically vulnerable for years after their child’s treatment has ended, which can impact their responsiveness to secondary diagnoses ^50,51. The low follow-up rate remains a challenge for all of survivorship care.

Strengths and Limitations:

Enrolling patients based on availability is a potentially biased approach that could lead to patient selection. However, the distribution of diagnoses seen in the study population is reflective of what is seen at this institution. The homogenous nature of participants with the overwhelming majority being Caucasian (Table 1) is a limitation that is also reflective of the patient population served by SLCH. Eligibility criteria for this study was open enough to allow a large number of patients to be approached based on the clinic schedule with a reasonably high recruitment rate (>70% of eligible patients). When determining eligible patients, the study team avoided bias by only reviewing the specific eligibility criteria, which did not include school performance difficulties or other possibly relevant issues.

Screenings were conducted by three personnel, leading to the possibility that there were inconsistencies between screeners. Of the 78 total screenings conducted, 44 (56%) of participants were screened by screener #1, 34 (43.5%) participants were screened by screener #2, and 2 (2.5%) participants were screened by screener #3. There were no noted differences in the methods employed by each screener as each followed the same protocol. Further, the number of screening failures attributed to each screener was fairly equal, with 10/18 (55%) of the screening fails performed by screener #1 and 8/18 (44%) of the screening fails performed by screener #2. Thus, our outcomes did not appear to be linked to the examiner.

A prospective trial including screening prior to the start of therapy would be useful than this cross-sectional trial. It would also be more accurate to perform diagnostic testing on the same day as the hearing screens. Follow-up tests were performed between 1–6 months after screening. We cannot rule out changes in hearing between the screens and the diagnostic testing. This would most likely impact conductive overlays of hearing.

Conclusion

We have demonstrated that hearing screening is feasible in a pediatric survivorship clinic. A significant fraction of pediatric cancer survivors at low risk of hearing loss failed hearing screening. Broader use of hearing screening may identify a group of survivors who would benefit from hearing interventions that could help with neurocognitive and educational performance as well as quality of life.

TABLE 4.

Audiogram information compared to screening failures

Patient ID	Age at screening (Years)	Diagnosis	Freq failed (kHz) Right ear	Freq failed (kHz) Left ear	Audiogram Right Ear	Audiogram Left Ear
Patient LRHL09	8	ALL	1, 2, 3, 4, 6, 8	1, 2, 8	Mild CHL at 4 kHz-8 kHz	WNL
Patient LRHL35	8	ALL	6	1, 2	WNL	WNL
Patient LRHL48	7	Sarcoma	None	6	WNL	WNL
Patient LRHL62	14	Lymphoma	6, 8	6, 8	WNL	Mild SNHL from 6 kHz-8 kHz
Patient LRHL65	15	ALL	2, 3, 8	None	Mild CHL at 4 kHz-8 kHz	WNL
Patient LRHL 73	16	ALL	1, 6	1	WNL	WNL

ALL= Acute Lymphoblastic Leukemia, CHL=Conductive Hearing Loss, SNHL= Sensorineural Hearing loss, WNL= Within Normal Limits

Abbreviations

CCS

Childhood Cancer Survivors

CCSS

Childhood Cancer Survivor Study

Hz

hertz

dB

decibel

HL

hearing level

IQR

interquartile range

PE

pressure equalization