ARTICLE IN PRESS

Characteristics and Voice Outcomes of Ulcerative Laryngitis

,
D1X X
*,1VyVy N. YoungD2X1X , †D3X XJackie L. Gartner-SchmidtD4X X, ‡D5X XNecati EnverD6X X, §D7X XScott D. RothenbergerD,8X X and *D9X XClark A. RosenD10X X,
*San Francisco, California, yxPittsburgh, Pennsylvania, and z Istanbul, Turkey
Summary: Objectives. Ulcerative laryngitis (UL) is challenging in terms of treatment and patient counseling,
with few reports in the literature. This study describes UL patients and their clinical course including detailed
voice and stroboscopic outcomes after treatment which have not been described in previous literature.
Methods. Single-institution, retrospective review of 23 UL patients. Demographics, historical factors, disease
course, treatment, and outcomes are presented. Treatment results were compared to prior studies.
Results. Seventy four percent had inflammatory/infectious precipitating event. Average presenting Voice-Hand-
icap-Index-10 (VHI-10) was 25 (range: 638) and average final VHI-10 was 9 (range: 026). Ninty five percent
had improvement in VHI-10 (average decrease of 15). Only 50% had final VHI-10 within “normal” limits. Treat-
ment comprised reflux medications (85%), antibiotics (22%), antifungals (39%), antivirals (52%) steroids (52%),
and/or voice rest (65%). Average symptom duration before evaluation was 42 days; average follow-up was 6.8
months. Final laryngovideostroboscopy revealed no ulcers in 78%, but 65% had persistently decreased mucosal
wave vibration. Average time to ulcer resolution was 2.25 months but resolution or plateau of voice symptoms
occurred later, average 2.7 months. Multiple regression analyses revealed that younger age, shorter symptom
duration, and antireflux treatment were significant predictors of decrease in VHI-10 (P < 0.05).
Conclusions. Most patients have good voice outcomes following resolution of UL, although vocal fold muco-
sal wave abnormalities may persist. This study provides the most detailed report of UL, disease course and treat-
ment outcomes to date. Additionally, this study is also the first to suggest that earlier initiation of treatment may
improve voice outcome after UL.
Key Words: LaryngitisUlcerativeVoiceOutcomeTreatment.

TAGEDH1INTRODUCTIONTAGEDN Reports on the duration of the clinical course for UL are

Ulcerative laryngitis (UL) was first described in 2000. Since
then, there have been few reports about this clinical
entity14 with the most robust description being the multi-
institutional report based on author recollection by Simpson
et al in 2011.4 These authors suggested the following criteria
for the diagnosis of UL: preceding upper respiratory infec-
tion with cough, bilateral ulcerative lesions of the vocal
folds, prolonged course of ulcer resolution >6 weeks, and
lack of response to pharmacologic management.4 This was
the first attempt to provide specific diagnostic criteria for
this disease entity and was based on the 15 patients in this
study. Future replications of these diagnostic criteria remain
pending. In most reports, UL has been characterized as an
acute/subacute process that involves unilateral or bilateral
ulcerative lesion(s) of the midmembranous vocal folds.17
Accepted for publication July 25, 2018.
Presented at the American Broncho-Esophagological Association meeting. May
18-22, 2016, Chicago, IL.
Financial disclosures: none.
Conflict of interest: none.
Level of evidence: 4.
From the *University of California—San Francisco, Department of Otolaryngol-
ogy—Head and Neck Surgery, San Francisco, California; yUniversity of Pittsburgh
Voice Center, University of Pittsburgh School of Medicine, Department of Otolaryn-
gology, Pittsburgh, Pennsylvania; zIstanbul University Faculty of Medicine, Depart-
ment of Otolaryngology, Istanbul, Turkey; and the xCenter for Research on Health
Care Data Center, University of Pittsburgh School of Medicine, Division of General
Internal Medicine, Pittsburgh, Pennsylvania.
1
Present address: University of California  San Francisco, 2330 Post St., 5th floor,
San Francisco, CA.
Address correspondence and reprint requests to VyVy N. Young, MD, Department
of Otolaryngology  Head and Neck Surgery, University of California  San Fran-
cisco, 2330 Post St., 5th floor, San Francisco, CA 94115. E-mail:
vyvy.young@ucsf.edu
Journal of Voice, Vol. &&, No. &&, pp. 17
0892-1997
© 2018 The Voice Foundation. Published by Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.jvoice.2018.07.021
variable. Voice outcomes have not been reported with any
detail previously. Treatment varies widely with the optimal
treatment regimen for UL remaining unclear.
The purpose of this study was to provide specific details
about the clinical course of UL patients including detailed
voice and stroboscopic outcomes. We sought to answer these
questions: (1) What is the patient perception of voice handi-
cap before and after treatment? (2) What is the average time
to resolution? (3) Is there an optimal treatment algorithm?
TAGEDH1MATERIALS AND METHODSTAGEDN
Approval for this study was obtained from the Institutional
Review Board of the University of Pittsburgh (PRO13030372).
A retrospective review of clinical data from patients diag-
nosed with UL treated at the University of Pittsburgh Voice
Center between 2006 and 2015 was performed. Informed
consent was obtained from all patients before data entry
into a clinical research database. The data were retrieved
from the database by a research assistant.
For purposes of this study, diagnosis of UL relied solely
on the presence of an ulcer (or eschar overlying an ulcer) in
the midmembranous portion of the true vocal fold(s) (repre-
sentative examples shown in Figures 1 and 2) in association
with acute or subacute (defined as present for
3 months)
hoarseness. Inclusion in the study required review of
recorded laryngovideostroboscopic examinations by a fellow-
ship-trained laryngologist to confirm the presence of visible
ulcer(s) or eschar overlying an ulcer. Additionally, only
patients with adequate follow-up until resolution of ulcers
and/or voice symptoms were included. Patients were followed
until either of the following endpoints was reached: (1)
 ARTICLE IN PRESS
2 Journal of Voice, Vol. &&, No. &&, 2018
FIGURE 1. Left vocal fold ulceration with underlying bilateral
vocal fold erythema.
FIGURE 2. Left vocal fold eschar overlying ulceration as well as
right vocal fold eschar.
subjective resolution of voice symptoms or (2) return of the
Voice Handicap Index-10 (VHI-10) to normal range (defined
as <11).89 “Final” VHI-10 was assessed at that time point.
Data collected included demographic information (age,
gender, occupation, smoking history, and other past medical
history such as asthma); history of the illness (eg, duration of
symptoms and precipitating event); patient presentation (eg,
primary complaints/symptoms, VHI-10, and laryngovideos-
troboscopy findings); type of treatment (eg, reflux medica-
tions, antivirals, antibiotics, antifungals, voice rest, oral
steroids, steroid injection(s), or other); and follow-up (eg,
serial VHI-10 scores and stroboscopic exam findings, dura-
tion of time off work, or persistent voice restrictions).
Regression analyses were conducted to identify predictors of
symptom improvement. Linear regression models were used to
identify variables associated with change in VHI-10 scores,
and logistic regression models were fit to determine factors
related to the return of VHI-10 to normal range (ie, final VHI-
10 < 11).8 Simple regression models were first employed to
identify unadjusted univariate relationships between outcomes
and explanatory variables. Subsequently, a multivariable
model was constructed for each outcome to identify multiple
significant predictors via forward selection of covariates. A
forward selection criterion of P < 0.10 for entry of covariates
into the models was employed, and final models would have
been pruned if any P values were greater than 0.05. Diagnos-
tics were performed to verify that assumptions of regression
were satisfied. For linear regression in particular, residual plots
showed no departures from the assumptions of normality, lin-
earity, or homoscedasticity. All statistical analyses were con-
ducted using Stata 14.2 (StataCorp. 2015. Stata Statistical
Software: Release 14. College Station, TX: StataCorp LP).
TAGEDH1RESULTSTAGEDN
Overview and prior history
Fifty-four patients diagnosed with UL were initially identified
through the University of Pittsburgh Voice Center database
query. Following reviews of each patient's previously recorded
laryngovideostroboscopic examinations by a fellowship-
trained laryngologist, 37 patients were identified as having
either visible ulcer or eschar on one or both vocal folds. All
ulcers were located on the midmembranous portion of the true
vocal fold(s) and involved the medial vibratory margin; none
of these ulcers involved the vocal process or false vocal fold.
Among these 37 patients, six patients were lost to follow-up fol-
lowing initiation of treatment, five patients had only one visit
and never returned for any further evaluation or treatment
(also lost to follow-up), two patients had missing data (eg,
VHI-10 at presentation), and one patient was in the midst of
ongoing treatment with final outcomes unable to be assessed.
Therefore, 23 patients met the study inclusion criteria, includ-
ing 16 females and seven males. Fourteen patients (61%) had
bilateral vocal fold involvement. The average age was 50 years
(range: 2679, median 49). The average duration of symptoms
prior to evaluation in the University of Pittsburgh Voice Center
was 41.7 days (range: 4180, median 30) and average duration
of follow-up was 6.8 months (range: 136, median 3).
Fifty-two percent of these patients were in vocally
demanding professions (eg, teacher, physician, nurse, real-
tor, singer, sales person, radio announcer, etc). Current or
former smoking history was present in 39% of patients. The
onset of symptoms was sudden in 70% and gradual in 30%
of patients, with a preceding infectious/inflammatory event
identified in 74% (Table 1).
A majority (76%) of patients had treatment prior to pre-
sentation to the University of Pittsburgh Voice Center and
multidisciplinary evaluation including a fellowship-trained
laryngologist and a voice-specialized speech language
pathologist. These prior treatments included inhaler use
(52%), antibiotics (14%), oral steroids (32%), and/or voice
rest (23%)—with variable results.
Treatment
Treatment regimens following multidisciplinary evaluation
varied considerably, and many patients were treated with
 ARTICLE IN PRESS
VyVy N. Young, et al Characteristics and Voice Outcomes of Ulcerative Laryngitis 3

TABLE 1.
Demographic Overview of Ulcerative Laryngitis Patients in Current and Prior Studies
Rakel et al Hsiao Simpson et al Current Study
N 14 33 15 23
F:M 5:9 20:13 14:1 16:7
Bilateral:unilateral vocal fold involvement - 31:7* 15:0 14:9
Avg age, years 38.4 (2365) 49.5 (2676) 49 (3372) 50 (2679)
Avg initial VHI-10 - - - 25 (638)
Avg final VHI-10 - - - 9 (026)
Avg symptom duration before presentation, months - - 3.3 (210) 1.4 (0.16)
Avg follow-up, months - - - 6.77 (136)
Preceding infectious/inflammatory event 85% - 100%** 74%
Professional voice user 85% 33% - 52%
Current/former smoker 15% 12% 0% 39%
* This study described 38 instances of ulcerative laryngitis in 33 patients including several recurrences.
** This study required the presence of a prior upper respiratory infection for inclusion in the definition of UL.
Dash indicates data not available.Avg, average.

more than one modality. Antireflux medications were given voice therapy, and two were treated with intralesional injec-
in 85%, antivirals in 52%, antibiotics in 22%, antifungals in tion with dexamethasone (10 mg/mL concentration).
39%, and oral steroids in 52% of patients. Voice rest was
prescribed for 65% of these patients (Table 2).
Voice outcomes
Antireflux medications consisted of one of the class of pro-
Complete VHI-10 data were available for 22 out of the 23
ton pump inhibitors (often omeprazole 40 mg, once or twice
patients. The average VHI-10 score at presentation was 24.6
daily), with or without the addition of an H2 blocker (often
(range: 638, median 26) and the average VHI-10 score
ranitidine 300 mg nightly), along with reflux behavior modifi-
upon final follow-up was 9.4 (range: 026, median 10.5).
cation education (both verbal and written). These regimens
Overall, the average change in VHI-10 score was a decrease
were prescribed as an empiric trial of at least three months
of 15 (range: 033, median was ¡13; Tables 1 and 3).
duration. Antiviral medications included acyclovir 800 mg
The vast majority (95%) of all patients reported an
TID or valacyclovir 1 g daily or BID. Duration of antiviral
improvement in VHI-10 scores. However, a final VHI-10
treatment typically ranged from 7 to 14 days although one
score in the normal range (ie, <11)8 was reported in only
patient completed a 21-day course. Antibiotics were uncom-
50% (Tables 3 and 4). Although the remaining 50% had a
monly prescribed but included amoxicillin/clavulanate
persistently elevated VHI-10 score >11 suggesting a persis-
875 mg BID, amoxicillin 500 mg BID, or clindamycin
tent voice handicap, 64% of this group reported subjective
300 mg TID, typically over 1014 days. Approximately 40%
resolution in their voice symptoms (describing their voice as
of patients received fluconazole, 200 mg on first day and
either “normal” or “better,” or stating that they were
then 100 mg daily subsequently. The most common regimen
“happy” with their voice) (Table 4).
comprised a total of 14 days although both a 7- and 21-day
course were prescribed once. Half of the patients were treated
with prednisone (60 mg daily) for several days which was Categorical outcomes
then tapered over a total 1014 days. Voice rest was pre- The average time to resolution of ulcers as seen on laryngeal
scribed for 65% of the patients, but included both absolute endoscopy with stroboscopy was 2.25 months (range:
and relative voice rest. The duration of voice rest varied 0.256, median 1.75). However, the time to resolution or
from 7 to >30 days. Three patients additionally underwent plateau of voice symptoms was longer, with an average of

TABLE 2.
Treatment Distribution for Ulcerative Laryngitis in Current and Prior Studies
Rakel et al Hsiao Simpson et al Current Study
No. of pts 14 33 15 23
Antireflux meds - 0% 87% 85%
Antivirals - - - 52%
Antibiotics - 0% 13% 22%
Antifungals - 0% 93% 39%
Steroids—oral - 0% 27% 52%
Voice rest - - - 65%
Dash indicates data unavailable.
 ARTICLE IN PRESS
4 Journal of Voice, Vol. &&, No. &&, 2018

TABLE 3.
Outcomes of Ulcerative Laryngitis in Current and Prior Studies
Rakel et al Hsiao Simpson et al Current Study
Avg time to resolution of ulcers, months - 2.4 3.3 2.25
Persistent decreased vibration on - 75% 60% 70%
stroboscopy
Avg time to resolution of voice symp- - - - 2.7
toms, months
Needed to take time off work - - - 24%
Improvement in VHI-10 - - - 95%
Avg change in VHI-10, from presenta- - - - ¡12
tion to resolution of ulcers/symptoms
Final VHI-10 within normal range (<11) - - - 50%
Patients reporting subjective voice - - "All reported voice improved" 91%
improvement or resolution of
symptoms
Dash indicates data not available.

2.7 months (range: 0.756 months, median 2). The ulcera-
tions ultimately resolved in 78% of patients. Stroboscopic
examination at the patient's last office visit demonstrated
persistently decreased mucosal wave in 65% of patients,
even if no ulcers were visible (Table 3).
Regression analyses
Simple linear regression revealed significant unadjusted
relationships between age and change in VHI-10
(P = 0.035) as well as between antireflux treatment and
change in VHI-10 (P = 0.023). Multiple linear regression
identified younger age (P = 0.027), shorter symptom dura-
tion (P = 0.006) and the administration of antireflux treat-
ment (P = 0.049) as significant independent predictors of
decrease in VHI-10. Smoking status, gender, months of fol-
low up, and other treatments (antivirals, antibiotics, anti-
fungals, steroids, and voice rest) were not significant
predictors after adjusting for age, symptom duration, and
antireflux treatment. Simple logistic regression models

TABLE 4.
Individual Voice Handicap Index-10 (VHI-10) Results for Patients with Ulcerative Laryngitis in the Current Study
Patient Initial VHI-10* Final VHI-10 Change in VHI-10 Patient’s Subjective Report of voice at Final Visit
1 30 0 ¡30
2 28 0 ¡28
3 28 0 ¡28
4 23 0 ¡23
5 12 0 ¡12
6 29 2 ¡27
7 38 5 ¡33
8 12 6 ¡6
9 6 6 0
10 30 8 ¡22
11 24 10 ¡14
12 22 11 ¡11 Voice "stable"
13 19 11 ¡8 "Happy" with voice
14 30 12 ¡18 No better but had been sick again
15 22 12 ¡10 "Happy" with voice
17 27 13 ¡14 "Happy" with voice
16 30 14 ¡16 "Normal"
18 23 15 ¡8 No better but had been sick again
19 25 16 ¡9 "Normal"
20 24 18 ¡6 Voice is "better"
21 29 22 ¡7 No better but had been sick again
22 31 26 ¡5 Voice is "better"
* VHI-10, Voice Handicap Index-10.
 ARTICLE IN PRESS
VyVy N. Young, et al Characteristics and Voice Outcomes of Ulcerative Laryngitis 5

TABLE 5.
Results of Linear Regression with VHI-10 Change as Outcome
Unadjusted (Simple Regression) Adjusted (Multiple Regression)
Covariate Coefficient (95% CI) P Value Coefficient (95% CI) P Value
Age 0.33 (0.02, 0.63) 0.035* 0.31 (0.04, 0.58) 0.027*
Smoker (current or Former) 4.19 (¡4.78, 13.17) 0.340
Female ¡1.23 (¡10.92, 8.46) 0.794
Symptom duration (days) 0.08 (¡0.01, 0.17) 0.077 0.10 (0.03, 0.17) 0.006*
Reflux treatment ¡12.78 (¡23.57, ¡1.98) 0.023* ¡9.19 (¡18.35, -0.04) 0.049*
Antiviral treatment 1.97 (¡6.67, 10.60) 0.640
Antibiotic treatment 0.39 (¡0.14, 0.91) 0.139
Antifungal treatment ¡3.68 (¡12.58, 5.05) 0.383
Steroid treatment ¡2.07 (¡10.70, 6.56) 0.623
Voice rest 4.36 (¡4.40, 13.11) 0.312
Follow up (months) 0.26 (¡0.18, 0.71) 0.232
*Significant at the a = 0.05 level.
Abbreviation: CI, confidence interval.

revealed that none of these factors were significant predic-
tors of whether final VHI-10 was within normal range (all
P > 0.10). As a result, no multivariable logistic regression
model was constructed (Tables 5 and 6).
TAGEDH1DISCUSSIONTAGEDN
Few studies of UL exist in the literature. Among these, this
is the first study to report detailed voice outcomes following
resolution of UL. Spiegel et al presented the first case report
of UL in 2000.1 Since that time, there have been few addi-
tional reports to further describe this “new” clinical entity.
The first description of a patient cohort with UL was pre-
sented by Rakel et al in 2002.2 More recently, an observa-
tional study of 33 patients3 and a multi-institutional review
of 15 patients were completed.4 To the best of our knowl-
edge, these studies, as well as three case reports,57 comprise
the current body of English-language literature regarding
UL. Building on the past description of UL from the litera-
ture, the current study presents the most complete clinical
picture and first detailed report of the patient perspective of
UL, particularly related to voice outcomes. The few previ-
ous reports of UL in the literature provided no information
about voice outcomes—from either an objective or subjec-
tive perspective—leaving clinicians and patients with no
guidance about what to expect over time in the course of
treatment of these patients. This lack of clarity has clouded
the question about what treatment to offer and/or what to
expect over time (with or without treatment), representing a
meaningful gap in knowledge and hampering quality
patient care.
The previous studies to date have suggested a tendency
toward female preponderance1,35,7; occurrence in middle
age24,67; and a common history of preceding infectious or
inflammatory event.2,4,6,7 These findings were confirmed in
the current study as well. The prevalence of professional

TABLE 6.
Results of Logistic Regression with Final VHI-10 Normal as Outcome
Unadjusted (Simple Regression)
Covariate Odds Ratio (95% CI) P Value
Age 0.97 (0.91, 1.04) 0.490
Smoker (current or former) 0.36 (0.06, 2.16) 0.262
Female 1.00 (0.15, 6.53) 1.000
Symptom duration (days) 0.97 (0.93, 1.01) 0.149
Reflux treatment 0.40 (0.03, 5.25) 0.485
Antiviral treatment 1.00 (0.19, 5.36) 1.000
Antibiotic treatment 5.71 (0.52, 62.66) 0.154
Antifungal treatment 2.22 (0.37, 13.18) 0.379
Steroid treatment 1.00 (0.19, 5.36) 1.000
Voice rest 0.45 (0.08, 2.67) 0.379
Follow-up (months) 0.84 (0.67, 1.07) 0.164
Abbreviation: CI, confidence interval.
 ARTICLE IN PRESS
6 Journal of Voice, Vol. &&, No. &&, 2018
voice users has been variable, ranging from modest levels
(33% in Hsiao,3 50% in current study) to a substantial
majority (85% in Rakel et al).2 Two of the three case reports
also occurred in professional voice users.1,5 The accumula-
tion of studies to date have provided a generalized picture
of characteristics including demographics related to UL.
There have been conflicting reports related to vocal fold lat-
erality of the ulcerative lesions. Simpson et al4 included the
presence of bilateral ulcerations as a diagnostic criteria for
UL; however, Hsiao's study,3 as well as the current study,
report unilateral ulcerations. Therefore, we believe the pres-
ence of a visible ulceration should raise suspicion for and
may be adequate to diagnosis UL, even if only present uni-
laterally. Additionally, further research will be required to
determine whether characteristics such as laterality, ulcer
size, or ulcer location (eg, anterior vs posterior vs mid vocal
fold, vocal process, superior surface of vocal fold, etc) car-
ries a significant clinical impact, in terms of either symptoms
or outcomes.
Little information from prior studies is available regarding
specifics of treatment for UL. In Simpson et al,4 antireflux
and antifungal medications appeared to be prescribed most
commonly (87% and 93%, respectively). In the current study,
antireflux medications were administered at a similar rate
(85%) but antifungals were prescribed to a much lesser degree
(39%). Other treatments, including antibiotics, antivirals,
oral steroids, and voice rest, were administered to varying
degrees in this study; limited or no information exists regard-
ing the frequency of these treatments in prior studies. This
broad range of treatment results in an inability to make spe-
cific conclusions about treatment or, more importantly, to
suggest a specific treatment algorithm based on prior experi-
ence. In this study, multiple regression analyses demonstrated
that reflux treatment may be predictive of a favorable change
in VHI-10, but it is acknowledged that 85% of patients in this
study received this treatment and thus this finding perhaps
should be viewed cautiously. Whether an early initiation of
this type of treatment may impact outcomes is as yet unclear.
Timing of the initiation of this medication has not been eval-
uated in this or any previous study of UL and may represent
a key point for future study. Until this impact is better eluci-
dated, the risks versus benefits of the clinical utility of an
early empiric antireflux regimen must be carefully weighed in
each individual patient's situation.
It is fascinating that Hsiao's study3 which included only
nonspecific medications as needed (such as acetaminophen,
mucolytics, antitussives, etc.) without any specific treatment
for the UL resulted in comparable findings in terms of time
to resolution of ulcers as well as frequency of persistently
decreased vocal fold vibration on stroboscopy. The implica-
tions of these similar results brings into question the nature
and need for a treatment protocol in these patients; clearly,
detailed study of treatment options in UL is needed to fur-
ther clarify this very important question.
Patients often inquire about the expected duration of the
ulcers and/or symptoms, but this clinically important ques-
tion remains poorly understood. Only two studies have
described the average time to resolution of the ulcerative
lesions thus far.3,4 Rakel et al broadly stated that “eventually
all ulcerations healed” and that “patients may be affected for
up to 1 year” but specific details regarding time were not pro-
vided.2 More recent studies have suggested a shorter interval
until resolution of the ulcers (average 2.4 months3 to 3.3
months4 - comparable to the average of 2.25 months
observed in this study). That the time to resolution of ulcers
may be markedly shorter than anecdotally reported (ie, only
a few months, rather than an entire year) is clinically relevant
and potentially impactful on treatment decision-making, as
expectant management alone for a full year is likely exces-
sively long, particularly in the presence of severe symptom-
atology. Clinicians should counsel patients about a more
realistic time course for expected recovery, both of their vocal
fold ulcers as well as of their voice symptoms. This informa-
tion may have important implications on patient's functional-
ity (eg, time off work) and/or treatment planning (eg, earlier
surgical intervention such as steroid injection). In fact, if
patients do not demonstrate reasonable in the first 23
months of treatment, then more aggressive intervention may
need to be more strongly considered.
Multiple studies (including the present study) have indi-
cated a strong tendency for persistence of decreased vibra-
tion of mucosal wave on stroboscopic examinations over
time, occurring in 6075% of patients.34 The findings of
the current study confirm that the average time to resolution
of the patient's voice symptoms was longer than physical
ulcer resolution. Perhaps, the durable voice symptoms
reflect the decrease in mucosal wave that persists. The
potential for prolonged symptoms even after resolution of
the ulcer(s) is a critically important educational point for
both patients and otolaryngologists. The occurrence of
voice changes with an acute upper respiratory infection is
not uncommon; however it is generally expected that these
types of symptoms would be short-lived. Unfortunately in
the case of UL, these symptoms have been demonstrated
now in multiple studies to linger over a prolonged period of
time. Appropriate counseling is required to guide expecta-
tions for progression of the disease course over time.
Recurrence of UL has been previously reported by both
Simpson et al and Hsiao in their cohorts of patients. There-
fore voice clinicians are encouraged to remain vigilant
about the potential risk of recurrence, particularly in light
of an additional infectious or inflammatory event.
Most patients with UL appear to show improvement in
their voice symptoms over time but this has been at best
vaguely alluded to in prior reports, leading to a lack of clar-
ity in how clinicians should counsel these patients. This is
the first study to report a detailed comparative characteriza-
tion of patient's demographics, symptoms, laryngeal charac-
terization, and voice handicap scores. In our study, 95% of
patients demonstrated improvement in their VHI-10 scores.
Only one patient reported no change in their VHI-10 score;
however that one patient initially presented with a VHI-10
score of 6 which is within the range for normal. This low
score at initial presentation suggests that the patient did not
 ARTICLE IN PRESS
VyVy N. Young, et al Characteristics and Voice Outcomes of Ulcerative Laryngitis
consider themselves to have a high degree of vocal handi-
cap; additionally, because this patient started with such a
low degree of voice handicap, there would not have been
significant room for improvement following any type of
treatment. Considering only those patients who initially pre-
sented with a voice handicap (as evidenced by an abnormal
VHI-10 score), 100% of those patients reported an improve-
ment in their VHI-10 score following treatment. This high
rate of improvement is encouraging and can be reassuring
to patients, especially early on in their course when their
voice symptoms are most bothersome.
However, only 50% of patients had final VHI-10 within
normal limits. While this finding may initially seem disheart-
ening, the majority of these patients (64%) reported a subjec-
tive satisfaction with their voice. Thus patient's expectations
should be optimistic but tempered. It is interesting to note
that three out of the four patients who reported that their
voice was not subjectively much improved had a subsequent
upper respiratory illness. Since these patients did not return
for additional follow-up, it is unknown whether their VHI-10
could have changed further over time. It is possible that their
VHI-10 score could have continued to decrease overtime,
possibly even to a categorically “normal” range and thus the
currently reported 50% rate of abnormal final VHI-10 may
not accurately reflect the final voice status of these patients.
Simple logistic regression did not identify any factors which
were predictive of a normal final VHI-10. However in multi-
ple linear regression analyses, younger age, shorter duration
of symptoms, and the administration of antireflux treatment
were associated with improved VHI-10. These findings
should be confirmed in a larger cohort.
A particularly interesting finding in this study's multiple
linear regression analyses is that shorter duration of symp-
toms was associated with better ultimate voice outcomes.
Put another way, these results suggest that early diagnosis
could be crucial. Although this study, like the few others
that have preceded it, could not delineate which treatment
was most effective, this finding does suggest that starting
some form of treatment as soon as possible may be essen-
tial. Thus, clinicians may need to take a more aggressive
approach to treatment of patients with UL and start treat-
ment early. Also, it may be important to improve primary
care physician's awareness of this disease entity to recog-
nize symptoms early (eg, uncharacteristically severe voice
or cough symptoms with upper respiratory infection URI),
facilitate early referral, and thus expedite prompt initiation
of treatment. Other treatments should be studied further to
7
determine their degree of empiric benefit. The development
of an evidence-based treatment algorithm remains a goal
for this patient population.
TAGEDH1CONCLUSIONSTAGEDN
Most patients with UL improve, but some patients may
experience prolonged voice symptoms, even after resolution
of the vocal fold ulcer(s). Furthermore, stroboscopic exami-
nation in these patients demonstrates a persistent decrease
in mucosal wave, despite improvement in voice symptoms.
Almost all patients had a decrease in VHI-10, and most of
them reported subjective resolution of voice symptoms.
This study represents the most complete comparative
description of UL to date, including treatment outcome and
data, filling a critical gap in the literature, as no prior study
has provided detailed voice outcomes, either subjective or
objective. UL continues to vary widely and treatment for
this entity may even yield results comparable to observation
alone. Younger patients with short duration of symptoms
treated with antireflux medications had significantly greater
VHI-10 improvement; findings should be investigated fur-
ther, especially in relation to reflux treatment. It is impor-
tant to recognize this disease process and initiate treatment
early. Prolonged symptoms following an upper respiratory
tract infection should raise concern for UL and warrants
early laryngeal evaluation. A large-scale, prospective, and
multi-institutional study is needed to better determine an
optimal treatment regimen for patients diagnosed with UL.
TAGEDH1REFERENCESTAGEDN
1. Spiegel JR, Sataloff RT, Hawkshaw M. Prolonged ulcerative laryngitis.
ENT J. 2000;79:342.
2. Rakel B, Spiegel JR, Sataloff RT. Prolonged ulcerative laryngitis. J
Voice. 2002;16:433–438.
3. Hsiao TY. Prolonged ulcerative laryngitis: a new entity. J Voice.
2011;25:230–235.
4. Simpson CB, Sulica L, Postma GN, et al. Idiopathic ulcerative laryngi-
tis. Laryngoscope. 2011;121:1023–1026.
5. Toland BL, DeFatta RA, Sataloff RT. Prolonged ulcerative laryngitis in
an 18-year-old voice major. Ear Nose Throat J. 2013;92:E24.
6. Sinclair CF, Sulica L. Idiopathic ulcerative laryngitis causing midmem-
branous vocal fold granuloma. Laryngoscope. 2013;123:458–459.
7. Beaver ME, Stasney CR, Rodriguez M. Acute ulcerative laryngitis. Ear
Nose Throat J. 2005 May;84:268.
8. Rosen CA, Lee AS, Osborne J, et al. Development and validation of the
voice handicap index-10. Laryngoscope. 2004;114:1549–1556.
9. Arffa RE, Krishna P, Gartner-Schmidt J, et al. Normative values for the
Voice Handicap Index-10. J Voice. 2012;26:462–465.