Introduction: As pediatric cochlear implant (CI) candidacy expands, children with greater degrees of residual hearing are receiving CIs. These nontraditional candidates have audiometric thresholds that meet adult manufacturer labeling but are better than current pediatric guidelines allow. The purpose of this study was to determine the impact of delayed cochlear implantation on speech perception in nontraditional pediatric CI recipients. Methods: Pediatric CI recipients with a history of progressive hearing loss and a preoperative 4-frequency pure-tone average of ≤75 dB HL at the time of implantation were considered for this retrospective study. Preoperative serial audiograms and word recognition scores were reviewed, and a method was created to establish a date when each individual ear 1st met nontraditional candidacy. The length of time between the date of candidacy and implantation was calculated and defined as the “delay time.” A multiple linear regression investigated delay time, age at surgery, surgery type (1st vs. 2nd side), and array type as predictive factors of maximum postoperative Consonant-Nucleus-Consonant (CNC) word scores. A one-way ANCOVA was performed comparing the postoperative CNC scores between subjects grouped by delay time. Results: A significant regression was found (F(4, 38) = 5.167, p = 0.002, R2 = 0.353). Both age at implantation (p = 0.023) and delay time (p = 0.002) predicted CNC word scores. Longer delay time was associated with poorer word recognition scores, while older age at implantation correlated with higher CNC word scores in this progressive hearing loss group. A significant difference was noted between subjects implanted with <1 year of delay and those with 3 or more years of delay (p = 0.003). All ears implanted within a year of candidacy achieved word recognition abilities that are generally accepted as above average (M = 84.91). Conclusion: CI candidacy for adults has evolved to allow for greater degrees of residual hearing, while audiometric guidelines for children have not changed since 2000. Our findings suggest that delay of cochlear implantation, even for children with significant levels of residual hearing, leads to poorer outcomes. Modified candidacy guidelines for children should be established to expedite referral to multidisciplinary CI teams and minimize delays in this population.

The current Food and Drug Administration (FDA) guidelines for pediatric cochlear implantation vary between manufacturer and patient age. Cochlear Americas’ device allows for implantation in children older than 24 months of age with severe-to-profound loss, and labeling for MED-EL and Advanced Bionics states that all children should have profound hearing loss to qualify for implantation. Acceptable aided word recognition scores for these indications range from <12 to <30% (Cochlear Corp., Cochlear Americas, Englewood, CO, USA; MED-EL GmbH, Innsbruck, Austria; Advanced Bionics Corp., Valencia, CA, USA). Adult criteria are more lenient. The least restrictive labeling is for potential candidates of electroacoustic stimulation or “Hybrid” devices from Cochlear Americas and MED-EL. They allow for thresholds ranging from normal to moderate in the low frequencies and severe to profound in high frequencies. These indications include for up to 60% word recognition in the ear to be implanted. The pediatric criteria are far more stringent than those of adults and exclude children with better than severe-to-profound audiometric thresholds who may still benefit from a cochlear implant (CI).

Traditional pediatric CI candidates meet FDA criteria set forth for children. Nontraditional pediatric candidates do not reach these guidelines but do meet adult criteria. Nontraditional candidates have increasingly been receiving implants off-label [Carlson et al., 2018]. In these cases, considerable focus is placed on aided speech perception and language development rather than audiometric thresholds when determining candidacy [Gantz et al., 2000; Dettman et al., 2004; Fitzpatrick et al., 2009; Teagle et al., 2019]. Studies have shown that nontraditional pediatric candidates benefit from cochlear implantation. Gratacap et al. [2015] examined word recognition in children with preoperative residual hearing. They found an average 26 percentage point increase in open-set word recognition after 1 year of CI use. Meredith et al. [2017] found a mean word recognition increase of 57 percentage points 1 month after activation in children with high-frequency hearing loss. Carlson et al. [2015] called for changes in pediatric candidacy criteria after finding a mean speech recognition improvement of 63 percentage points following cochlear implantation in children with less than severe hearing loss and speech perception skills exceeding 30%.

Despite the evidence of the benefits of implantation, the timing of cochlear implantation in children with residual hearing proposes challenges. Records of children referred to this center indicate that referring clinicians are relying on unaided word recognition in combination with hearing aid verification to make decisions on whether to recommend a CI evaluation. Unfortunately, unaided word recognition does not correlate well with the aided word recognition scores used to determine candidacy [McRackan et al., 2018]. In addition, many of these children have hearing levels that can be amplified acoustically, but hearing aids may not provide the speech understanding predicted by verification methods, particularly in cases where high-frequency hearing thresholds reach 70 dB HL [Ching et al., 1998]. Reliance on unaided word recognition and audiograms alone may result in delay of candidacy evaluation. When children with residual hearing are evaluated and found to be CI candidates, parents often elect to wait until a child is no longer able to function with a hearing aid despite evidence and counseling to proceed with implantation.

Age at implantation has consistently been found to be a predictor of speech and language outcomes with CI [Dettman et al., 2007; Geers and Nicholas, 2007; Niparko et al., 2010; Ching et al., 2013; Leigh et al., 2013; Dettman et al., 2016; Ching et al., 2017; Karltorp et al., 2020]. Oftentimes, a delay in implantation is related to progressive hearing loss [Fitzpatrick et al., 2015] likely because these children have the opportunity to develop some speech understanding and derive benefit from their residual acoustic hearing. While age at implantation has historically been seen as one of the most influential variables in pediatric cochlear implantation, it may not be as predictive as the delay in implantation in cases of progressive hearing loss. When speech understanding deteriorates in a child with progressive hearing loss, they begin a period of significantly degraded listening. We hypothesize that the longer this period lasts, the more significantly a child’s speech understanding with a CI will be impacted.

The purpose of this study was to determine the impact of delayed implantation on speech perception in children who are considered nontraditional CI candidates. The secondary aim is to identify the length of delay that significantly impacts outcomes. We predicted that longer delays in implantation will be negatively correlated with postoperative word recognition. The intent of this work is to provide evidence that helps guide decision-making for referring clinicians and parents of children who may benefit from a CI.

A retrospective chart review was performed on all children who received a CI under the age of 18 to identify those who had a history of progressive hearing loss. Each ear was evaluated as a separate subject. Progressive loss was chosen as an inclusion criterion to enable identification of a time when a subject crossed from benefiting from hearing aids to having insufficient audibility and needing a CI. To be included, potential subjects needed to have had at least 1 year of CI use, postoperative Consonant-Nucleus-Consonant (CNC) word scores, lack of major inner ear malformation or cochlear nerve deficiency, typical development, and a preoperative 4-frequency pure-tone average (4FPTA; average threshold at 250, 500, 1,000, and 2,000 Hz) of ≤75 dB HL at the time of implant. There were 43 ears (28 children) that met these criteria. All had audiometric thresholds that were lower than allowed by current pediatric FDA guidelines at the time of implantation but would have met adult FDA criteria.

Demographic information, serial unaided hearing thresholds, and preoperative aided and unaided word recognition scores were documented. As different clinical protocols were followed dependent on subject age and clinic, 3 criteria were used to establish a date when each ear 1st reached nontraditional CI candidacy. The primary metric was aided word recognition. For children with serial aided CNC word scores, the candidacy date was set as the point when the percentage of words correct fell below 60% (n = 15) in order to meet the least stringent of adult criteria. If a subject did not have serial aided word recognition available, but did have a history of unaided word recognition scores, a candidacy date was determined using the formula suggested by Hoppe et al. [2015] which incorporates the maximum unaided word score (PBMax). Subjects were considered to have met candidacy when the PBMax score fell below the 4FPTA – 8 (n = 20). For younger children, the candidacy date was logged as the date when the referring audiologist 1st made the recommendation for evaluation by the implant team due to inadequate progress with spoken communication despite greater levels of residual hearing (n = 8). The length of time between the candidacy date and implantation was calculated in years and defined as the “delay time.”

Word recognition testing was administered as part of routine preoperative candidacy evaluations and postoperative audiologic evaluations in the sound field using recorded word lists presented at 0° azimuth at 60 dB SPL in the study ear alone. Preoperative tests used varied based on what was developmentally appropriate and part of the standard of care at the time of testing. These data are presented with demographic information in Table 1. Some subjects were tested preoperatively using a recorded PBK list. In order to compare pre- and postoperative word recognition, PBK scores were 1st converted to estimated CNC scores using the formula cited in Teagle et al. [2019]. They were then included with the preoperative CNC scores for analysis. All postoperative scores were obtained using the CNC word list, and the reported scores represent the highest CNC word score achieved for each ear.

Table 1.

Participant information and averages

 Participant information and averages
 Participant information and averages

Statistical analysis was performed using the Statistical Package for the Social Sciences v26 (IBM Corp.). To investigate whether delay time impacts postoperative word recognition, a multiple linear regression was completed that measured maximum postoperative CNC word scores as the dependent variable and delay time, age at surgery, surgery type (1st vs. 2nd side), and array type (perimodiolar vs. lateral wall) as predictive factors. To examine the effect of different amounts of delay time on outcomes, subjects were stratified by ranges of delay: <1 year (0–0.99 years), 1–2 years (1.0–1.99 years), 2–3 years (2.0–2.99 years), and ≥3 years. A one-way ANCOVA was performed comparing the postoperative CNC scores between each of the groups, with age of implant as a covariate. A paired t test was completed on a subset of 26 subjects who had preoperative CNC word scores available (converted or estimated as noted above).

A total of 43 ears met inclusion criteria and were included in the analysis as separate subjects. Fifty-eight percent of subjects were implanted with a lateral wall electrode, while the remaining were implanted with a perimodiolar electrode. Sixty-five percent were 1st side devices and the remaining were 2nd sides. The average age of implantation was 8.4 years (SD = 3.62), and the average 4FPTA was 65.03 dB HL (SD = 8.27). The average delay time was 2.6 years (SD = 2.10). The mean postoperative CNC score was 75.4% words correct (SD = 12.81). Preoperatively aided word recognition scores ranged from 0 to 56% words correct. These data are presented in Table 1.

A multiple regression analysis was completed to address the 1st aim, determining whether delays in implantation impact speech perception outcomes in children meeting nontraditional candidacy criteria. The results indicated that delay time, age at implantation, 1st versus 2nd side surgery, and array type explained 35% of the variance in postoperative CNC word scores (R2 = 0.353, F(4, 38) = 5.176, p = 0.002). Coefficients are presented in Table 2. When controlling for the other independent variables, delay time was found to significantly predict CNC word scores (p = 0.002). Each year delay was predicted to result in a 3.29 point decrease in word recognition. Age at surgery was also found to be a significant factor when controlling for the other predictors (p = 0.023); however, older ages at implantation were associated with better outcomes. Array type (p = 0.132) and 1st or 2nd side surgery (p = 0.562) were not found to be significant factors. Notably, all children implanted with less than a year delay achieved word recognition scores of 74% or better. These are generally accepted as good scores [Lawson and Peterson, 2011]. Figure 1 demonstrates the relationship between surgical delay in years and highest CNC word score.

Table 2.

Results of the multiple linear regression

 Results of the multiple linear regression
 Results of the multiple linear regression
Fig. 1.

Greater delays in implantation are correlated with poorer word recognition outcomes (p = 0.002). The bands represent 95% confidence intervals. CNC, Consonant-Nucleus-Consonant.

Fig. 1.

Greater delays in implantation are correlated with poorer word recognition outcomes (p = 0.002). The bands represent 95% confidence intervals. CNC, Consonant-Nucleus-Consonant.

Close modal

As both age and delay time were found to be significant factors, age at implantation was added as a covariate, and a one-way ANCOVA was completed to compare outcomes when grouped by range of delay. There was a significant difference in CNC word scores between groups (F(3, 38) = 5.541, p = 0.003). Post hoc testing with a Bonferroni correction indicated that those implanted within 1 year of candidacy (M = 84.91, SD = 7.50) achieved significantly higher scores than those implanted 3 or more years after reaching candidacy (M = 68.43, SD = 12.53, p = 0.003). There were no other significant differences between groups. The results are illustrated in Figure 2.

Fig. 2.

Prolonged delay demonstrates a trend toward worse speech perception performance. There was a significant difference in outcomes between the 0–1 and >3-year delay groups (p = 0.003). Error bars represent 95% confidence intervals, and the middle bands are means. CNC, Consonant-Nucleus-Consonant.

Fig. 2.

Prolonged delay demonstrates a trend toward worse speech perception performance. There was a significant difference in outcomes between the 0–1 and >3-year delay groups (p = 0.003). Error bars represent 95% confidence intervals, and the middle bands are means. CNC, Consonant-Nucleus-Consonant.

Close modal

Pre- and postoperative CNC word scores were compared using a paired t test. Postoperative scores (M = 74.46, SD = 13.34) were significantly higher than preoperative scores (M = 28.85, SD = 13.41) (t(25) = 12.54, p < 0.001). Postoperative word recognition scores exceeded preoperative scores in all cases. The results are noted in Figure 3.

Fig. 3.

Pre- and postoperative CNC word scores. There was a significant improvement when comparing preoperative CNC word scores obtained with a hearing aid and highest CNC word scores obtained with a cochlear implant. Error bars represent 95% confidence intervals, and the center bands are means. CNC, Consonant-Nucleus-Consonant.

Fig. 3.

Pre- and postoperative CNC word scores. There was a significant improvement when comparing preoperative CNC word scores obtained with a hearing aid and highest CNC word scores obtained with a cochlear implant. Error bars represent 95% confidence intervals, and the center bands are means. CNC, Consonant-Nucleus-Consonant.

Close modal

Recent research has shown that children with PTA poorer than 60 dB HL have a 75% chance of performing better with a CI than a HA [Leigh et al., 2016]. Despite the proven benefits of cochlear implantation in children with residual hearing and poor word recognition [(Carlson et al. 2015); Gratacap et al., 2015; Meredith et al., 2017], the current FDA guidelines are limited to children with, at best, a severe-to-profound hearing loss. With a significant improvement from pre- to postoperative word recognition, our results support previous work indicating benefits of implantation in nontraditional pediatric CI candidates.

The main goal of this study was to evaluate the impact of delaying implantation in children with better unaided thresholds than those cited under current pediatric FDA approvals. Our results demonstrated a relationship between delaying implantation in children who had reached adult-like candidacy and postoperative word recognition outcomes. Furthermore, delaying implantation 3 years or beyond was associated with poorer word recognition outcomes in nontraditional pediatric CI recipients. When considering clinical recommendations, all children implanted within a year of meeting nontraditional candidacy achieved word recognition skills that exceeded 74%, while the same could not be said for those who were delayed further.

The results of this study are similar to those of Gratacap et al. [2015] who found an association with the length of delay in implantation and open-set word recognition in children with preoperative residual hearing. In that study, delay was determined by calculating the date of onset of hearing loss rather than the date of candidacy. Half of the sample in that study had progressive or fluctuating hearing loss, and half would likely have met candidacy at diagnosis. The participants also had significant improvement in word recognition postoperatively. While the authors did not investigate a particular length of delay time that impacted outcomes most significantly, they stressed the importance of early implantation and limited deprivation in children who are nontraditional candidates. The present study supported those findings and established that 3 years of delay resulted in significantly poorer outcomes than more expedient implantation.

In a study surveying professionals, Fitzpatrick et al. [2009] found that clinicians were hesitant to refer children who may have greater levels of residual hearing because they felt that they were too young to truly ascertain how much benefit the children were receiving from their hearing aids. The onus for determining candidacy does not need to lie with a referring clinician when a CI evaluation includes a multidisciplinary team of professionals familiar with trends and best practices in pediatric cochlear implantation. Instead, clinicians should refer to a CI team for evaluation as soon as they suspect that hearing aids are not providing the input needed for spoken language development. When developmentally appropriate, aided word recognition should be evaluated routinely to validate speech understanding and determine if a referral is in order. Expanded pediatric guidelines for implantation may help inform referral decisions. This study suggests that early referral should occur when individual ear aided word recognition is below 60%.

While younger age at implant is known to result in better outcomes in children [Dettman et al., 2007; Nicholas and Geers, 2007; Niparko et al., 2010; Ching et al., 2013; Leigh et al., 2013; Dettman et al., 2016; Ching et al., 2017; Karltorp et al., 2020], this study found better word recognition in children implanted at older ages. This may be due to the use of progressive hearing loss as an inclusion criterion. Older children may have had more exposure to spoken language prior to the progression of hearing loss and hence had more top-down skills to draw upon. While the reasons for this correlation were not investigated in the current study, this finding does acknowledge that age at implant may not always correlate to length of candidacy.

The present study was limited by its retrospective nature and relatively small sample size. Our findings indicated that delays between candidacy and CI surgery were associated with poorer postoperative word recognition scores, but did not examine the impact surgical delay may have on long-term speech and language development. In addition, the subjects in this study had progressive hearing loss and we were not able to account for the rate of hearing loss progression among participants. The rate and age at which progression occurred may be a factor that impacts speech perception and should be investigated further.

While CI candidacy for adults has evolved, audiometric candidacy guidelines for children have remained unchanged since 2000. Research supports the benefits of early implantation in children with severe-to-profound hearing loss, and evidence supporting cochlear implantation in children with greater levels of residual hearing continues to emerge. Our findings suggest that delay of 3 years or more between reaching candidacy and receiving a CI in pediatric candidates with preoperative residual hearing is associated with poorer outcomes. Pediatric CI candidacy guidelines should be amended to provide audiologists and physicians the information they need to ensure timely referral and provide children with the access to hearing they need to reach their full potential.

This study was approved by the biomedical institutional review board at the University of North Carolina at Chapel Hill under study No. 07-1655. Parents or guardians provided written informed consent, and subjects over the age of 7 provided assent for inclusion.

The authors have no conflicts of interest to declare.

Lisa Park receives research grant support from MED-EL Corporation.

Lisa Park: study design; data acquisition, analysis, and interpretation; drafting and revising; final approval; accountability. Elizabeth Perkins: study design; data interpretation; drafting and revising; final approval; accountability. Jennifer Woodard: data acquisition and interpretation; revising; final approval; accountability. Kevin Brown: study design; data interpretation; revising; final approval; accountability.

1.
Carlson
ML
,
Sladen
DP
,
Gurgel
RK
,
Tombers
NM
,
Lohse
CM
,
Driscoll
CL
.
Survey of the American Neurotology Society on cochlear implantation: part 1, candidacy assessment and expanding indications
.
Otol Neurotol
.
2018
;
39
(
1
):
e12
9
.
2.
Carlson
ML
,
Sladen
DP
,
Haynes
DS
,
Driscoll
CL
,
DeJong
MD
,
Erickson
HC
,
.
Evidence for the expansion of pediatric cochlear implant candidacy
.
Otol Neurotol
.
2015
;
36
(
1
):
43
50
.
3.
Ching
TYC
,
Dillon
H
,
Button
L
,
Seeto
M
,
Van Buynder
P
,
Marnane
V
,
.
Age at intervention for permanent hearing loss and 5-year language outcomes
.
Pediatrics
.
2017
;
140
(
3
):
e20164274
.
4.
Ching
TY
,
Dillon
H
,
Byrne
D
.
Speech recognition of hearing-impaired listeners: predictions from audibility and the limited role of high-frequency amplification
.
J Acoust Soc Am
.
1998
;
103
(
2
):
1128
40
.
5.
Ching
TY
,
Dillon
H
,
Marnane
V
,
Hou
S
,
Day
J
,
Seeto
M
,
.
Outcomes of early- and late-identified children at 3 years of age: findings from a prospective population-based study
.
Ear Hear
.
2013
;
34
(
5
):
535
52
.
6.
Dettman
SJ
,
D’Costa
WA
,
Dowell
RC
,
Winton
EJ
,
Hill
KL
,
Williams
SS
.
Cochlear implants for children with significant residual hearing
.
Arch Otolaryngol Head Neck Surg
.
2004
;
130
(
5
):
612
8
.
7.
Dettman
SJ
,
Dowell
RC
,
Choo
D
,
Arnott
W
,
Abrahams
Y
,
Davis
A
,
.
Long-term communication outcomes for children receiving cochlear implants younger than 12 months: a multicenter study
.
Otol Neurotol
.
2016
;
37
(
2
):
e82
95
.
8.
Dettman
SJ
,
Pinder
D
,
Briggs
RJ
,
Dowell
RC
,
Leigh
JR
.
Communication development in children who receive the cochlear implant younger than 12 months: risks versus benefits
.
Ear Hear
.
2007
;
28
(
2 Suppl l
):
11S
8S
.
9.
Fitzpatrick
EM
,
Ham
J
,
Whittingham
J
.
Pediatric cochlear implantation: why do children receive implants late
.
Ear Hear
.
2015
;
36
(
6
):
688
94
.
10.
Fitzpatrick
E
,
Olds
J
,
Durieux-Smith
A
,
Mccrae
R
,
Schramm
D
,
Gaboury
I
.
Pediatric cochlear implantation: how much hearing is too much
.
Int J Audiol
.
2009
;
48
(
2
):
91
7
.
11.
Gantz
BJ
,
Rubinstein
JT
,
Tyler
RS
,
Teagle
HF
,
Cohen
NL
,
Waltzman
SB
,
.
Long-term results of cochlear implants in children with residual hearing
.
Ann Otol Rhinol Laryngol Suppl
.
2000
;
185
:
33
.
12.
Gratacap
M
,
Thierry
B
,
Rouillon
I
,
Marlin
S
,
Garabedian
N
,
Loundon
N
.
Pediatric cochlear implantation in residual hearing candidates
.
Ann Otol Rhinol Laryngol
.
2015
;
124
(
6
):
443
51
.
13.
Hoppe
U
,
Hast
A
,
Hocke
T
.
Audiometry-based screening procedure for cochlear implant candidacy
.
Otol Neurotol
.
2015
;
36
(
6
):
1001
5
.
14.
Karltorp
E
,
Eklöf
M
,
Östlund
E
,
Asp
F
,
Tideholm
B
,
Löfkvist
U
.
Cochlear implants before 9 months of age led to more natural spoken language development without increased surgical risks
.
Acta Paediatr
.
2020
;
109
(
2
):
332
41
.
15.
Lawson
G
,
Peterson
M
.
Speech audiometry
.
San Diego, CA
:
Plural Publishing Inc
;
2011
.
16.
Leigh
JR
,
Dettman
SJ
,
Dowell
RC
.
Evidence-based guidelines for recommending cochlear implantation for young children: audiological criteria and optimizing age at implantation
.
Int J Audiol
.
2016
;
55
(
Suppl 2
):
S9
S18
.
17.
Leigh
J
,
Dettman
S
,
Dowell
R
,
Briggs
R
.
Communication development in children who receive a cochlear implant by 12 months of age
.
Otol Neurotol
.
2013
;
34
(
3
):
443
50
.
18.
Meredith
MA
,
Rubinstein
JT
,
Sie
KCY
,
Norton
SJ
.
Cochlear implantation in children with postlingual progressive steeply sloping high-frequency hearing loss
.
J Am Acad Audiol
.
2017
;
28
(
10
):
913
9
.
19.
McRackan
TR
,
Fabie
JE
,
Burton
JA
,
Munawar
S
,
Holcomb
MA
,
Dubno
JR
.
Earphone and aided word recognition differences in cochlear implant candidates
.
Otol Neurotol
.
2018
;
39
(
7
):
e543
9
.
20.
Nicholas
JG
,
Geers
AE
.
Will they catch up? The role of age at cochlear implantation in the spoken language development of children with severe to profound hearing loss
.
J Speech Lang Hear Res
.
2007
;
50
(
4
):
1048
62
.
21.
Niparko
JK
,
Tobey
EA
,
Thal
DJ
,
Eisenberg
LS
,
Wang
NY
,
Quittner
AL
,
.
Spoken language development in children following cochlear implantation
.
JAMA
.
2010
;
303
(
15
):
1498
506
.
22.
Teagle
HFB
,
Park
LR
,
Brown
KD
,
Zdanski
C
,
Pillsbury
HC
.
Pediatric cochlear implantation: a quarter century in review
.
Cochlear Implants Int
.
2019
;
20
(
6
):
288
98
.