Löwenstein-Buschke: Clinicopathologic Analysis of 78 Cases of Large and Giant Condyloma Acuminata of the Anus
Orhun CIG TASKIN1, Burcin PEHLIVANOGLU2, Michelle D. REID2, Theodore FRIEDMAN2, Michael LEE2, Talaat S. TADROS2, Sudeshna BANDYOPADHYAY3, Josephine AKINFOLARIN3, Ayse ARMUTLU1, Olca BASTURK4, Volkan ADSAY1
1Department of Pathology, Koç University Hospital, ISTANBUL, TURKEY
2Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USA
3Department of Pathology, Wayne State University, Detroit, MI, USA
4Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, NY, USA
Keywords: Condyloma acuminata, Giant, Anus, Buschke-Lowenstein, Squamous cell carcinoma
The nature and clinicopathologic associations of Löwenstein-Buschke disease are unclear.
Materials and Methods: 78 anal condylomatous lesions (≥2 cm) were analyzed. Cases were classified based on size as “medium-large”(2-5 cm,
n=59), “large” (5-10 cm, n=13) and “giant” (>10 cm, n=6).
Results: Patients were predominantly males (male/female=70/8). The mean age was 38 years (range:20-66). Two distinct lining types were
recognized: 1) Epidermal type, typically lacking overt koilocytotic change, with associated invasive carcinoma in 8%; 2) Mucosal type, often
manifesting koilocytotic change, with associated invasive carcinoma in 21%. Three types of high-grade dysplasia were discerned: 1) Basaloid,
8/9 showing high-grade dysplasia/carcinoma in-situ but non-invasive lesions; 2) Keratinizing, innocuous-appearing, but 5/6 was associated with
invasion; 3) Giant cell, showing scattered individual bizarre cells, with 3/5 showing invasive carcinoma. Overall, invasion was found in 14% of
the cases. The bulbous, broad-based destructive pattern characterizing verrucous carcinomas of the upper aerodigestive tract was not observed.
A statistically significant trend existed between the incidence of invasion and size: 8.5% for medium-large, 23% for large, and 50% for giant
(p=0.02). There was no discernable trend in the depth of invasion relative to condyloma size.
Conclusions: Our findings suggest that Löwenstein-Buschke lesions are mega versions of conventional condyloma. Being verrucoid, large and
minimally invasive, they can be conceptually regarded as a form of verrucous carcinoma, but they do not display the histologic characteristics of
verrucous carcinoma defined in the aerodigestive tract. They exhibit two types of linings: the mucosal type that often shows koilocytotic changes,
and the epidermal type that can be difficult to recognize in biopsies. These lesions may be associated with invasive carcinoma, albeit limited in
Anal condyloma acuminata refers to polypoid, cauliflowershaped
and pedunculated excrescences, histologically
characterized by hyperkeratosis, surface parakeratosis and
koilocytosis of the superficial cell layers (1,2). It is caused
by infection with the human papillomavirus (2). It is
frequently found in sexually active people, representing the
most common sexually transmitted disease in the United
Giant condyloma acuminatum (GCA) of the anus, also
called Giant Condyloma of Buschke and Löwenstein, was
initially thought to be a larger version of the conventional
condyloma with minimal biological aggressiveness. Later on, some authors emphasized that GCAs were distinguished
from traditional condylomas (3,4). Although they were
cytologically unremarkable, locally invasive features
with transformation to squamous cell carcinoma were
intermittently noted, characterizing GCA as a malignant
tumor (5). Therefore, GCAs have been categorized as a
distinct clinical entity, a “carcinoma-like condyloma,” with
the propensity to locally invade and recur, but without the
predilection to metastasize. It is classified as “verrucous
carcinoma of anus” by some.
The definition of “giant”, on the other hand, has been highly
variable, and controversy exists over the terminology,
size cut-off, histology, and degree of malignant potential of this lesion. The incidence of conventional malignant
change (carcinoma in-situ or invasive carcinoma) and
the relationship of size with malignant potential in GCAs
have not been thoroughly investigated. In this study, we
analyzed the largest series of large condylomas of the anal
region with the intent to determine the distinct histological
features including patterns of dysplasia and correlate them
with clinical and demographic data.
|Case Selection and Classification Criteria
78 resected anal condylomas that measured 2 cm or more
and were diagnosed between 1986 and 2007 in the authors’
institutional files were reviewed. Institutional review board
approval was obtained. Data on patient demographics
including age, gender, HIV status and clinical presentation
were extracted from the medical records, including the
All of the cases were procedures performed with the goal
of complete removal of the lesion. The cases were classified
according to their size, arbitrarily, as medium-large (2-5
cm) (n=59, 75%), large (5-10 cm) (n=13, 17%) and giant
(>10 cm) (n=6, 8%). The slides were reviewed to determine
the histologic features of the condyloma, the cell types
occurring in the lesion, including the presence of a granular
layer and keratotic layer, and the type and the grade of
For the purposes of this study, the high-grade squamous
intraepithelial lesions were regarded in two groups: 1)
High-grade dysplasia and 2) frank carcinoma in-situ, where
there is florid atypia and pattern that raises the concern for
invasive carcinoma, but no definitive invasion is identified
in the same region. The incidence of conventional dysplasia,
carcinoma in-situ, and invasive carcinoma were recorded
and correlated with the cell types and size of the lesion.
Statistical analysis was performed using IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version
25.0. Armonk, NY: IBM Corp. The normality of continuous
variables was investigated by the Shapiro-Wilk’s test.
Descriptive statistics were presented using mean and
standard deviation for normally distributed variables and
median (and minimum-maximum) for the non-normally
distributed variables. Non-parametric statistical methods
were used for values with a skewed distribution. The Mann-
Whitney U test was used for the comparison of two nonnormally
distributed groups. The Kruskal-Wallis test was
used or the comparison of three non-normally distributed
groups. The χ² test (Fisher Exact test where available) was
used for categorical variables and expressed as observation
counts (and percentages). Statistical signiﬁcance was
accepted when the two-sided p value was lower than 0.05.
The clinical findings of the patients with these tumors are
shown in Table I
. The vast majority of the patients were male
with a male to female ratio of 70/8. The male predominance
decreased with increasing size of the condyloma, with a
male/female ratio of 14/1 in the group with size 2-5 cm,
and 5.5/1 in the group with tumors measuring 5-10 cm, and
dropped down to 2/1 (p=0.08) in cases with tumors larger
than 10 cm. The overall mean age was 38 years (range: 20-
66). The giant (>10 cm) condylomas occurred in patients
nearly a decade older (mean age 46 vs. 38, p=0.09).
Click Here to Zoom
|Table I: Distribution of age, gender, dysplasia, invasion and HIV status by condyloma size.
Histologic subtypes: Two distinct subsets of condyloma
acuminata were recognized by the morphology of the lining
epithelium: epidermal and mucosal. 74% of the cases were
classified in one of these two categories. The characteristic
histomorphologic findings of these categories were as
1) Epidermal type (50%) recapitulated the characteristics
of the epidermis with a well-defined granular layer, distinct stratum corneum, and decreased glycogen in the cells. Half
of the cases (50%) were of this (epidermal) type. If taken
in isolation and examined at high-power, this type could
be indistinguishable from the normal epidermis. Highgrade
dysplasia was identified in 13% of cases and invasive
carcinoma in 8% from this epidermal subtype.
2) Mucosal type (50%): This type of condyloma has an
inconspicuous stratum corneum with prominent koilocytic
changes. High-grade dysplasia was identified in 2.5% and
invasive carcinoma was seen in 20.5% of this condyloma
subtype. Although the frequency of invasive carcinoma
appeared to be higher in the mucosal type cases, this did
not reach statistical significance (p=0.7).
Endoscopic appearance and histologic subtypes are seen in
Figure 1 and 2A, B.
Click Here to Zoom
|Figure 1: The endoscopic examination of anal condyloma reveals a polypoid, cauliflower-shaped lesion.
Click Here to Zoom
|Figure 2: The two distinct subsets of condyloma: A) Epidermal, harboring a distinct granular layer, stratum corneum, and decreased
glycogen in the cells. If observed in isolation at high-power, this type could be indistinguishable from normal epidermis (H&E; x200).
B) Mucosal, with an inconspicuous stratum corneum and prominent koilocytic changes (H&E; x200).
Patterns of dysplasia: All cases were reviewed for patterns
of dysplasia that were distinct from conventional. Three
distinct patterns of dysplasia were identified: basaloid,
dyskeratotic, and giant cell-type.
1) Basaloid: The basaloid phenotype demonstrated an
expansion of basal or parabasal-like cells towards the upper
layers of the mucosa with readily identifiable mitoses. This
type of dysplasia recapitulated what was seen in a category
of anal intraepithelial neoplasia with advanced progression
preferably designated as squamous cell carcinoma of basaloid
type (6). This type of dysplasia was identified in 12% of
the cases (n=10), nine of which were in the medium-large
group and one in the large group. Eight of the nine cases
that had isolated high-grade dysplasia were of the basaloid
type. Invasive carcinoma was present in three of the cases
with this type of dysplasia.
2) Giant-cell type of dysplasia was characterized by bizarre
large, highly pleomorphic nuclei, often with multinucleated
giant cells, and typically occurring in relatively mature, squamous mucosa as dispersed individual cells. This type of dysplasia was identified in 5% of the cases; two cases were
present in the medium-large group, and one case each was
present in the large and giant groups. One of the nine cases
with isolated high-grade dysplasia was of the giant cell type.
Three of the cases with this type of dysplasia had associated
invasive carcinoma, two of which also had concurrent
dyskeratotic dysplasia (see below, group 3).
3) The dyskeratotic form of dysplasia was characterized by
groups of or individual cells with inappropriate maturation
relative to its level within the epithelium. This dysmaturation
often manifested as zones of densely acidophilic
cells. Dyskeratotic single cells and overtly keratinizing nests
were seen. Dyskeratotic dysplasia was present in 6 cases
(8% of all cases): 3 cases from the medium-large group, one
from the very large group and two from the giant group.
This subtype was commonly seen in association with
conventional well-differentiated squamous cell carcinoma (5 cases) and one case of high-grade dysplasia. Patterns of
dysplasia are seen in Figure 3A-D.
Click Here to Zoom
|Figure 3: A-B) Basaloid type of dysplasia, showing expansion of basal or parabasal-like cells towards the upper layers of the mucosa
with readily identifiable mitoses. (H&E; x40&200). C) Giant-cell type dysplasia, characterized by large pleomorphic nuclei (H&E; x400).
D) Dyskeratotic dysplasia, characterized by keratinizing nests (H&E; x40).
Incidence of dysplasia, carcinoma in-situ and invasive
squamous cell carcinoma
Medium-large group (2-5 cm): This group included 59
cases (75% of all cases), 56 males and 3 females with a mean
age of 38 years (±11.8). Among these cases, high-grade
dysplasia was identified in 11 cases (18.6%), carcinoma insitu
in 4 cases (6.7%) and invasive squamous cell carcinoma
in 5 cases (8.5%). Sixteen patients (27.1%) were known to
be HIV positive.
Large group (5-10 cm): This group included 13 cases (17%
of all cases), 11 males and 2 females with a mean age of
37 years (±12.3). Among these cases, high-grade dysplasia
and carcinoma in-situ were not identified but invasive
squamous cell carcinoma was seen in 3 cases (23%). Three
patients (23%) were known to have a positive HIV status.
Giant group (>10 cm): This group included 6 cases (8%
of all cases), 4 males and 2 females with a mean age of 46
years (±11.3). Among these cases, high-grade dysplasia
and carcinoma in-situ were not identified but invasive
squamous cell carcinoma was seen in 3 cases (50%). One of
the patients was HIV positive (16.7%).
Overall incidence: The incidence of conventional highgrade
dysplasia in the absence of an associated squamous
cell carcinoma was 11.2% (n=9), all of which occurred
within the 2-5 cm category. Three of these cases were HIV
positive. The incidence of invasive carcinoma was 14%
(n=11). There was a statistically significant trend (p=0.02)
for increasing incidence of invasion with size. Seven (63% of
all invasive carcinomas) cases from the invasive carcinoma
category were microinvasive, penetrating <3 mm below
the basement membrane of the condyloma. Four cases of invasive squamous cell carcinoma penetrating greater
than 3 mm were distributed as follows: Medium-large: 1
case, Large: 2 cases, and Giant: 1 case. Invasive carcinoma
patterns are seen in Figure 4A-D.
Figure 4: A) Condyloma acuminata with focal microinvasion (H&E; x200). B) Condyloma acuminata with high-grade dysplasia and
focal microinvasion highlighted by a cuff of chronic inflammation (H&E; x200). C-D) Condyloma acuminata with invasive squamous
cell carcinoma [...] arising in keratinizing type of dysplasia (H&E; x100 & x200).
Since Buschke and Löwenstein’s description in 1925, it has
been well-established that GCAs harbor the potential for
malignant transformation (3,7–9). The anal condylomas
reported as “giant” in the literature, ranged from 1.5 cm to 30
cm in maximum dimension (7). Consequently, the question
arises: Are giant condylomas a morphologic variant of
verrucous carcinomas or are they distinct entities?(1) Some
believe that they are the same entity, using the terms GCA
and verrucous carcinoma interchangeably (4), whereas
others speculate that GCAs and verrucous carcinomas
represent distinct entities with divergent mechanisms for pathogenesis (10,11). There is some credence to the latter
viewpoint because it appears that verrucous carcinomas
usually do not arise from transformation of a pre-existing
condyloma, although exceptions have been reported (12).
Instead, it is a well-differentiated, cytologically bland, lowgrade
squamous cell carcinoma with broad, pushing borders
and local invasion (1). Conversely, GCAs share analogous
histologic features with the conventional condyloma
mentioned earlier. Kraus and Perez-Mesa proposed that
condyloma acuminatum, GCA, verrucous carcinoma,
and squamous cell carcinomas lie on the same pathologic
continuum (13). Clinically, there are a number of similar
treatment options for GCAs and verrucous carcinomas,
including local surgical excision, chemotherapy and
radiation, depending on the extent of disease (3,14,15).
This study presents the largest series-to-date subjecting
large and giant anal condylomas, their clinicopathologic
analysis and histological classification. The striking male
predominance (overall M/F: 70/8, with a tendency to
decrease with increasing size of the lesion) and young
age were remarkable features, along with the documented
positive HIV status in 25% of the patients. It is possible
that some of the remaining cases may also have been HIV
positive but the testing and other information was not
available to the authors.
Two distinct histological subsets of anal condylomas were
discerned in this study: epidermal and mucosal, however,
overlaps occurred in nearly 1/4 of the cases. The main
significance of this classification is the need of recognizing
different morphological aspects of these lesions, especially
in small and fragmented biopsies. Epidermal type, a form
of condyloma/dysplasia not recognized previously, is
virtually indistinguishable from normal epidermis. The
design of the epithelium, the texture of the cytoplasms,
the relationship of the cells with each other, and the layers
were characteristic of skin. In fact, in many areas, a granular
layer was also noted, completing the picture of epidermis.
Because of its close resemblance to normal epidermis, these
areas would have been almost impossible to recognize as
abnormal, let alone as a part of condyloma, if they were
taken in isolation on high-power examination. However,
they were lining florid condylomatous lesions, proving
their pathologic nature. We have noted this type of lesion in
the mucosa of oral cavity and it has been illustrated in some
publications. Recently, a group of cases were reported in the
esophagus under the heading of “esophageal epidermoid
metaplasia”, and are believed to be early precursor lesions
(16). Our study proves that there is indeed epidermallike
neoplastic transformation of mucosal sites. Therefore, this epidermal-type of condyloma elucidated in this study
is not only significant diagnostically as a subtle form of
dysplasia, but also important in terms of proving that
epidermal-like dysplasia exists as a concept. Moreover,
there were differences in the clinicopathologic associations
of the mucosal versus epidermal types of dysplasia. The
incidence of high-grade dysplasia was higher in epidermaltype
condyloma (13% vs. 2.5%), whereas the incidence of
invasive squamous cell carcinoma was higher in mucosaltype
condyloma (20.5% vs. 8%, respectively).
In this study, three distinct types of dysplasia were also
observed in large anal condylomas: basaloid, giant cell and
dyskeratotic types. Among those, dyskeratotic dysplasia was
seen in association with squamous cell carcinoma and its
presence in a biopsy should raise suspicion for an invasive
lesion. In fact, it may be better to regard it as “surface”
component of a keratinizing squamous cell carcinoma and
evaluate the case accordingly.
The overall incidence of invasive carcinoma in anal
condylomas in this study was 14% There was a statistically
significant trend (p=0.02) for increasing incidence of
invasion with size; however, there was no discernable
trend in depth of invasion relative to condyloma size.
More importantly, although half of the giant condylomas
displayed invasion, the majority was microscopic. This
may be highly pertinent to the biology of these lesions. In
essence, Löwenstein-Bushcke disease can be regarded as
a virally driven adenomatous (“papilloma”-type) tumor
of the anal squamous mucosa. As such, these can be
regarded as tumoral intraepithelial neoplasia. Similar to
most tumoral intraepithelial neoplasms, the neoplastic
cells in this disease somehow tend to grow in the surface,
eventually forming exophytic tumors without invading the
stroma. In other words, akin to the dichotomy that is well
known in urothelial neoplasms (17), there appears to be a
dichotomy in the HPV carcinogenesis in the anus as well.
In the urothelium, it has been demonstrated amply that the
papillary urothelial neoplasms represent a different pathway
of carcinogenesis than the “flat” carcinoma in-situ pathway,
in terms of molecular background, progression rate, and
clinical characteristics (18). The former has a much more
indolent, protracted clinical course.
Some authors regard Löwenstein-Bushcke as a verrucoustype
squamous carcinoma (19). Verrucous carcinoma is a
tumor that is well-recognized in the upper aerodigestive
tract. Similar to the tumoral intraepithelial neoplasms,
it continues to grow exophytically, before showing
conventional invasive carcinoma (20). However, verrucous
carcinoma appears to be something in between these two, a special type of invasive carcinoma rather than being a
purely pre-invasive neoplasm. Characteristically, it has
bulbous edges that represent broad-based pushing-type
infiltration, and that is why it can show destructive behavior,
even without showing regular type invasiveness. In fact, it
may be better to regard verrucous carcinoma as a form of
invasive carcinoma that has “blunt” invasion. In the case of
Löwenstein-Bushcke, the process has all the characteristics
of an ordinary condyloma acuminata, but a very large
one. In most cases, one does not get the impression of a
pushing-border invasion that is characteristic of verrucous
carcinoma. In fact, in the case of epidermal-type, they do
not even show much cytologic atypia. Of note, the fact
that most of the dysplastic changes occurred in mediumsize
cases may signify a different branching that takes place
during the advancement of these large condylomas, with
some cases acquiring conventional dysplastic changes.
On the other hand, the frequency of invasive carcinoma
increases by size. However, this is kind of expected, because
after all, the probability of changes that lead to invasion
increases with the number of neoplastic cells that are
present. In this study, no funds were available to conduct
specific HPV typing. It would be an interesting next step
to investigate the associations of different HPV types with
the histopathologic observations elucidated in this study.
Additionally, since clinical follow up data was limited, no
inferences could be made regarding the direct prognostic
associations of some of the histomorphologic observations.
In summary, it is advisable to further categorize anal
condylomas (≥2 cm) based on their size, due to the
increasing frequency of invasion in larger tumors.
Condylomas that are >10 cm seemed to occur in older
patients, suggesting a slower growing process. Our findings
suggest that Löwenstein-Buschke lesions are mega versions
of conventional condyloma. Being verrucoid, large and
minimally invasive, they can be regarded as a form of
verrucous carcinoma, but they do not display the histologic
characteristics of verrucous carcinoma as defined in the
aerodigestive tract. They exhibit two types of linings, the
mucosal-type that often shows koilocytic changes, and the
epidermal-type that can be difficult to recognize in biopsies.
Different types of high-grade dysplastic changes that occur
in these lesions (basaloid, keratinizing and giant cell) also
appear to have distinct behavioral characteristics. These
lesions may be associated with invasive carcinoma, albeit
limited in amount.
CONFLICT of INTEREST
The authors have no conflicts of interest or financial ties to
No funding was used.
The authors would like to thank Dr. Arzu Baygul for her
assistance with the statistics.
Concept: OT, Design: OT, MR, OB, Data collection or
processing: BP, TF, ML, TT, SB, Analysis or Interpretation:
BP, MR, TF, ML, TT, SB, JA, AA, OB, Literature search:
AA, Writing: OT, TF, VA , Approval: OB, VA.
1) Bogomoletz WV, Potet F, Molas G. Condylomata acuminata,
giant condyloma acuminatum (Buschke-Loewenstein tumour)
and verrucous squamous carcinoma of the perianal and anorectal
region: A continuous precancerous spectrum? Histopathology.
2) Chang GJ, Welton ML. Human papillomavirus, condylomata
acuminata, and anal neoplasia. Clin Colon Rectal Surg.
3) Chu QD, Vezeridis MP, Libbey NP, Wanebo HJ. Giant condyloma
acuminatum (Buschke-Lowenstein tumor) of the anorectal
and perianal regions - Analysis of 42 cases. Dis Colon Rectum.
4) Grussendorf-Conen EI. Anogenital premalignant and malignant
tumors (Including Buschke- Lowenstein tumors). Clin Dermatol.
5) Alexander RM, Kaminsky DB. Giant condyloma acuminatum
(Buschke-Loewenstein tumor) of the anus: Case report and
review of the literature. Dis Colon Rectum. 1979;22:561-5.
6) Vincent-Salomon A, De La Rochefordière A, Salmon R, Validite
P, Zafrani B, Sastre-Garau X. Frequent association of human
papillomavirus 16 and 18 DNA with anal squamous cell and
basaloid carcinoma. Mod Pathol. 1996;9:614-20.
7) Trombetta LJ, Place RJ. Giant condyloma acuminatum of the
anorectum: Trends in epidemiology and management. Report
of a case and review of the literature. Dis Colon Rectum.
8) Prasad ML, Abcarian H. Malignant potential of perianal
condyloma acuminatum. Dis Colon Rectum. 1980;23:191-7.
9) Turkdogan P, Basturk O, Demir MA, Zeytinoglu A, Seyhan A.
Giant condyloma acuminatum - two cases with microinvasive
foci in one. Aegean Pathol J. 2004;1:62-5.
10) Zidar N, Langner C, Odar K, Hošnjak L, Kamarádová K, Daum O,
Pollheimer MJ, Košorok P, Poljak M. Anal verrucous carcinoma is
not related to infection with human papillomaviruses and should
be distinguished from giant condyloma (Buschke–Löwenstein
tumour). Histopathology. 2017;70:938-45.
11) Chan MP. Verruciform and condyloma-like squamous
proliferations in the Anogenital Region. Arch Pathol Lab Med.
12) Wawire J, Hansen K, Quddus MR. Verrucous carcinoma arising in
association with giant condyloma. Int J Surg Pathol 2019;27:407-8.
13) Kraus FT, Perez-Mesa C. Verrucous carcinoma. Clinical and
pathologic study of 105 cases involving oral cavity, larynx and
genitalia. Cancer. 1966;19:26-38.
14) Mistrangelo M, Mobiglia A, Cassoni P, Castellano I, Maass J,
Martina MC, Bellò M, Mussa A. Verrucous carcinoma of the anus
or Buschke-Lowenstein tumor of the anus: Staging and treatment.
Report of 3 cases. Suppl Tumori. 2005;4:S29-30.
15) Balik E, Eren T, Bugra D. A surgical approach to anogenital
Buschke Loewenstein Tumours (giant condyloma acuminata).
Acta Chir Belg. 2009;109:612-6.
16) Savant D, Zhang Q, Yang Z. Squamous neoplasia in the esophagus.
Arch Pathol Lab Med. 2020. (Epub ahead of print)
17) Shigehara K, Sasagawa T, Namiki M. Human papillomavirus
infection and pathogenesis in urothelial cells: A mini-review. J
Infect Chemother. 2014;20:741-7.
18) Van Batavia J, Yamany T, Molotkov A, Dan H, Mansukhani M,
Batourina E, Schneider K, Oyon D, Dunlop M, Wu XR, Cordon-
Cardo C, Mendelsohn C. Bladder cancers arise from distinct
urothelial sub-populations. Nat Cell Biol. 2014;16:982–91.
19) Virgilio E, Balducci G, Mercantini P, Tozzi F, Ziparo V, Ferri M.
Perianal giant condyloma acuminatum of Buschke-Loewenstein:
A carcinoma-like condyloma or a condyloma-like carcinoma?
ANZ J Surg. 2015;85:394-5.
20) Thompson LDR. Verrucous squamous cell carcinoma. Ear, Nose
Throat J. 2019;25:014556131987171. (Epub ahead of print)