Glass‐blowing, brass‐playing, and other laryngocele etiologies: Fact or fiction? A systematic literature review

Abstract

Objective

Laryngocele is a rare condition, which can be congenital or acquired, characterized by an abnormal dilatation of the laryngeal ventricle filled with air. A correlation between laryngocele formation and increased intralaryngeal pressure has been postulated, as much as a frequent coexistence with laryngeal carcinoma. Despite this, systematic research on its potential etiology is still limited up to date.

Methods

We carried out a systematic an electronic search on the MEDLINE, Embase, Web of Science, Cochrane Library, Scopus, and ClinicalTrials.gov databases for all articles related to laryngocele or laryngopyocele. The authors performed a PRISMA‐compliant systematic review across multiple databases including all studies published until February 7, 2023, focusing on laryngocele etiology. Case reports and articles in languages other than English were excluded. The authors recorded etiology, age of patients, male‐to‐female ratio, history of smoking, symptoms, management, and overall follow‐up.

Results

The majority of publications related to laryngocele are case reports and case series. We found a total of 10 articles correctly reporting etiologies, the majority of which included only a limited number of patients. Most case series explored the association of laryngocele and laryngeal cancer. Other reported etiologies were wind instruments' playing, cricohyoidoepiglottopexy's execution, amyloidosis, and chronic constipation. Some cases of congenital laryngocele were also reported.

Conclusions

Actual knowledge concerning laryngocele etiology appears—based on literature studies—inferential at best. The association between laryngocele and laryngeal cancer is, on the other hand, well‐established and must be taken into account when evaluating patients with this condition. Wider and prospective data collections should allow further insight into this condition and its pathogenesis.

Key Points

• Actual knowledge regarding laryngocele's etiology appears limited and restricted to a large number of case series and case reports. Its association with laryngeal cancer seems well established, while other etiologic causes (such as brass‐playing) should be explored further.

• To the author's knowledge, this is the first systematic review analyzing the etiology of laryngocele. It represents a pioneering attempt to assess the reliability of the etiological theories that have been long‐established until the present.

INTRODUCTION

In humans, the anterolateral wall of the larynx is characterized by a small recess, the ventricle of Morgagni: at the bottom of the ventricle, a cul‐de‐sac mucosal extension, known as the laryngeal saccule, takes origin. ¹ While the ventricle has no true functional role, its origin might be traced back to laryngeal air sacs, which are present in anthropoid apes with a phonatory and respiratory function. ² These large air sacs growing laterally to the larynx provide apes' characteristic nonlinear, high resonance vocalization and production of long‐lasting call sequence. ³ , ⁴

Starting from Australopithecus, air sacs have been lost during humans' evolutionary history in a process known as “laryngeal simplification,” creating a more stable voice source and, eventually, allowing the evolution of human language. ⁵

In humans, the presence of an air sac communicating with the larynx is pathological and is due to the abnormal dilatation of the laryngeal saccule. This condition is known as laryngocele, a term coined by Virchow in 1867. ⁶

The laryngocele is a benign, cystic dilatation of the laryngeal saccule containing air. ⁷ It communicates with the laryngeal lumen and is usually unilateral, though anecdotal bilateral cases have been reported. ⁸ The laryngocele may also contain secretory products of the mucosa; if the mucus becomes thick, its drainage can be impaired and may lead to a valve‐like mechanism admitting air under increased pressure. Furthermore, the sac can become acutely infected and filled with pus (laryngopyocele). ⁹

Laryngocele can be classified as internal, when it lies completely within the larynx, external, if it protrudes from the saccule throughout the thyrohyoid membrane and extends as a neck mass, or mixed when both presentations occur. ¹⁰

It is considered to be a rare entity; is more common in men after the 5th decade and in the Caucasian population. It is more commonly acquired than congenital. ¹¹

Different theories have been proposed about the development of laryngoceles, but still its etiology remains unclear. Congenital factors, such as a greater length of the saccule or congenital weakness of laryngeal tissues may play a role. ¹² Despite this, it has been postulated to arise in people with prolonged periods of continuous increased laryngeal pressure, such as wind instrument musicians, glass‐blowers, or patients with chronic bronchitis. ¹³ However, although respiratory strain may be encountered in many of the patients, it seems not to be required for the development of the laryngocele. ¹⁴

Moreover, laryngoceles have been found in patients with laryngeal amyloidosis, laryngeal papilloma, laryngeal chronic inflammatory disease, and laryngeal cancer. ¹⁵ Symptoms may include cough, hoarseness, dyspnea, inspiratory stridor, dysphagia, and sore throat. ¹¹

Since Larrey first described laryngeal air sacs in man ¹⁶ in 1829, a vast number of cases have been reported and studied. At current, though, literature regarding laryngocele's etiology consists mainly of case reports, whilst cohort studies represent a minor part of publications available. In this context, information about etiology—especially the most picturesque cases such as brass‐playing and glass‐blowing patients—appears to have been passed on and overestimated when compared to their real‐world occurrence rates.

The aim of this systematic review is to analyze current information about laryngocele's etiology to understand how much of what we actually know has been consistently proved and what instead, has been handed down through times as an ancient story, and hence, little more than a myth.

METHODS

Information sources and search strategy

We conducted a Cochrane‐method systematic review between February 2023 and March 2024, according to PRISMA reporting guidelines. ¹⁷ We carried out systematic electronic searches for studies in English that reported original data regarding the etiology of laryngocele.

On February 7, 2023, we searched the MEDLINE, Embase, Web of Science, Cochrane Library, Scopus, and ClinicalTrials.gov databases for all articles related to laryngocele or laryngopyocele.

The complete search strategies and the number of items retrieved from each database are provided in Table 1.

TABLE 1.

Search strategy details and items retrieved from each consulted database.

Database	Search key	Date of Search	Items retrieved (n)
Pubmed	“laryngocele” [MeSH Terms] OR “laryngocele” [All Fields] OR “laryngoceles” [All Fields] OR “laryngocoele” [All Fields] OR “laryngocoeles” [All Fields] OR “laryngopyocele” [All Fields] OR “laryngopyoceles” [All Fields] OR “laryngocele” [MeSH Terms] OR “laryngocele” [All Fields] OR “laryngoceles” [All Fields] OR “laryngocoele” [All Fields] OR “laryngocoeles” [All Fields] OR “laryngopyocoele” [All Fields] Translations laryngocele: “laryngocele” [MeSH Terms] OR “laryngocele” [All Fields] OR “laryngoceles” [All Fields] OR “laryngocoele” [All Fields] OR “laryngocoeles” [All Fields] laryngopyocele: “laryngopyocele” [All Fields] OR “laryngopyoceles” [All Fields] laryngocoele: “laryngocele” [MeSH Terms] OR “laryngocele” [All Fields] OR “laryngoceles” [All Fields] OR “laryngocoele” [All Fields] OR “laryngocoeles” [All Fields]	February 7, 2023	606
Clinicaltrials.gov	laryngocele OR laryngopyocele OR laryngocoele OR laryngopyocoele	February 7, 2023	0
Cochrane Library	laryngocele OR laryngopyocele OR laryngocoele OR laryngopyocoele Febraury, the 7th, 2023 in Title Abstract Keyword ‐ (Word variations have been searched)	February 7, 2023	0
Scopus	TITLE‐ABS‐KEY (laryngocele OR laryngopyocele OR laryngocoele OR laryngopyocoele)	February 7, 2023	784
Web of Science	laryngocele OR laryngopyocele OR laryngocoele OR laryngopyocoele (all fields)	February 7, 2023	312
Embase	‘laryngocele’/exp OR laryngocele OR ‘laryngopyocele’/exp OR laryngopyocele OR laryngocoele OR laryngopyocoele	February 7, 2023	772
‐	‐	Total non‐unique items	2474
‐	‐	Unique items	914

Note:‐, no data

Inclusion and exclusion criteria

We included only original studies that clearly reported laryngoceles etiology. Studies focusing on other laryngeal malformations (i.e., saccular cyst) were excluded from the review. Only studies in the English language were included in the analysis. We excluded case reports, meta‐analyses, and systematic and narrative reviews. There were no restrictions based on the minimum study population or the publication date.

Selection process

Abstracts and full texts were reviewed in duplicate by different authors. At the abstract review stage, we included all studies that were deemed eligible by at least one reviewer. At the full‐text review stage, disagreements were resolved by achieving consensus among reviewers. Systematic review references were manually reviewed to identify any missing articles potentially eligible for inclusion.

PICOS criteria

The Population, Intervention, Comparison, Outcomes, and Study (PICOS) framework for the review was defined as follows:

P: any patient with a diagnosis of laryngocele

I: no intervention available

C: no comparator available

O: etiology reported

S: all original study types, excluding case reports, meta‐analysis, and systematic and narrative reviews

Study quality assessment

All studies selected during the second review round underwent quality assessment. Case series and cohort studies were evaluated according to the “National Heart, Lung and Blood Institute Study Quality Assessment Tools” (NHI‐SQAT). Articles were categorized as “Good” if they met at least 80% of the NHI‐SQAT criteria, “Fair” if they met between 50% and the 80% of the criteria and “Poor” if the criteria met were less than 50%, using the methodology we consolidated in prior mixed‐level‐of‐evidence reviews. ¹⁸ , ¹⁹ , ²⁰ , ²¹ Levels of evidence were determined following the “Oxford Center for Evidence‐based Medicine” (OCEBM) guide. ²² All studies of good or fair quality according to the NHI‐SQAT rating tool were included in the final systematic review. Due to the qualitative nature of the data collected no meta‐analysis was initially planned or performed a posteriori.

Data extraction

For each study included, we recorded the following: study type, total number of patients, number of laryngoceles, etiology, and presence of an underlying malignancy. Other information such as: male to female ratio, mean age, type of laryngocele (i.e., internal, external, both), side, history of smoking, and symptoms were recorded where available.

Two authors extracted data and rated studies in duplicate, and disagreements were resolved by consensus.

RESULTS

Among the 914 unique research items initially identified, 533 published reports were selected for full‐text evaluation. No further report was identified for full‐text evaluation after reference checking. Ultimately, a total of 10 articles were retained for analysis, ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² as depicted in Figure 1.

FIGURE 1.

Articles' selection process.

Among the included studies, six articles were single‐center retrospective studies, two were observational cohort studies and two were case‐control studies. Their level of evidence, according to the OCEBM scale, was IV. According to the NHI‐SQAT, nine articles were rated as low‐risk‐of‐bias/good‐quality studies, and one article was rated as mild‐risk‐of‐bias/fair‐quality study. No significant biases towards the objectives of our systematic review were identified. Table 2 provides information on the study type, evidence level, and quality rating for all the included studies.

TABLE 2.

Type of study, evidence, and quality rating of reviewed articles.

Reference	Study type	OCEBM rating	Quality rating
Baker et al., 1982	RS	4	F
Close et al., 1987	OS	4	G
Cohen et al., 2017	CC	4	G
Lindell et al., 1978	RS	4	F
Macfie, 1966	OS	4	F
Micheau et al., 1978	OS	4	F
Naiboglu et al., 2010	RS	4	G
Pietrantoni et al., 1959	RS	4	G
Prasad et al., 2008	RS	4	F
Slonimsky et al., 2022	RS	4	G

Abbreviations: CC, case‐control study; OS, observational cohort study; RS, retrospective study.

A total of 4246 patients were involved in the analyses, among which, 410 laryngoceles were found. Male to female ratio and patients' average age were available in only 4 out of 10 articles. Incidence was shown to be higher in males, with an average age >50 in 3 out of 4 studies. The etiologies reported included 55 wind instrument players; 3 cases occurred after cricohyoidoepiglottopexy's execution; 2 patients with chronic cough; 1 patient with laryngeal amyloidosis and 1 case of congenital laryngocele. Association with laryngeal cancer was attested in 236 cases. Not all the studies accounted for symptoms (8 out of 10): hoarseness was the most referred symptom, present in 98 patients; 29 patients referred dysphagia, 19 dysphonia, and 11 dyspnea. In a total of 14 patients, the laryngocele presented as a neck swelling mass; 60 cases were totally asymptomatic.

In seven of the studies analyzed, diagnosis was made using X‐rays, CT scans, or both. None of the studies compared the sensitivity of these exams in identifying the laryngocele. One study ²⁷ compared the sensitivity of X‐rays during forced expiration with closed and open glottis, demonstrating the major diagnostic efficacy of the latter. One study ²⁸ evaluated the presence of laryngoceles in the laryngeal specimens after total laryngectomy. More frequently, treatment was proposed only for symptomatic patients, while asymptomatic were kept under follow‐up. Surgical treatment options ranged from endoscopic resection, trans‐cervical resection, and laryngectomy in patients with concomitant laryngeal cancer.

Table 3 presents demographic and clinical information regarding the patients affected by laryngocele for all included studies.

TABLE 3.

Demographic and clinical information on the patients presenting laryngocele for all included studies.

Reference	Patients (n)	Laryngoceles (n)	Male: Female ratio (n:n)	Patients' mean age at diagnosis (years)	Etiology	Symptoms	Underlying malignancy (n)	Diagnosis	Treatment
Baker et al., 1982	174	6	5:1	47 (excluding the congenital case)	Amyloidosis (1); wind instrument playing (2); congenital (1); non‐reported (2)	Hoarseness (2); neck swelling (3); asymptomatic (1)	0	Laryngeal endoscopy and X‐rays	Observation (3); surgical excision (3)
Close et al., 1987	304	38	‐	‐	‐	Hoarseness (19); dysphagia (19); dyspnoea (9)	17	CT scan	Observation (21) Total laryngectomy for laryngeal cancer (11)
Cohen et al., 2017	29	9	15:14	60.5 (±11.2)	‐	Dysphonia (19); asymptomatic (5)	1	Laryngeal endoscopy, CT scan	Endoscopic resection (4), transcervical surgery (2), total laryngectomy (1), unknown (2)
Lindell et al., 1978	2068	87	‐	‐	‐	Hoarseness (75); dysphagia (13)	51	‐	‐
Macfie, 1966	94	52	‐	‐	Wind instrument playing	Asymptomatic	‐	X‐rays with open glottis strain	None
Micheau et al., 1978	546	104	‐	‐	‐	‐	104	Endoscopy, X‐rays	Laryngectomy
Naiboglu et al., 2010	62	3	‐	‐	Cricohyoidoepiglottopexy	Neck swelling (3); dyspnoea (2)	‐	Endoscopy and CT scan	Transcervical surgery
Pietrantoni et al., 1959	857	53	‐	‐	‐	‐	53	X‐rays (all) and histology (31)	‐
Prasad et al., 2008	5	5	5:0	52.8	Chronic cough (2); wind instrument playing (1)	Hoarseness (2); neck swelling (3)	1	X‐rays and CT scan	Observation (2), transcervical surgery (2), total laryngectomy (1)
Slonimsky et al., 2022	107	53	40:13	60 (±18)	‐	Neck swelling (5); dysphagia (2); dyspnoea (1)	9	CT scan	‐

Note: ‐, no data

DISCUSSION

Laryngocele is a rare entity, frequently asymptomatic or showing extremely mild symptoms. Given this, it is frequently diagnosed accidentally while performing neck radiological imaging or during laryngeal cancer assessment. As it follows, there could be many underdiagnosed cases and the actual incidence among the general population might be underestimated. Its association with continuous increased intralaryngeal pressure has been postulated, and some studies aimed to investigate its association with upper airways (UA) strain‐related work exposure and with laryngeal malignancies.

The medical evidence collected in this systematic review shows that we do not currently have enough data to confirm laryngocele etiological factors. The only ascertained point is the positive association with laryngeal cancer, with an unclear causative relationship. The incidence of this pathology in the general population, i.e., not exposed to hyperinflation of UA or to a chronic laryngeal pathology, should be more thoroughly assessed.

EPIDEMIOLOGY

Demographics were reported in only 4 out of 10 articles included in the review. The analysis revealed a consistent predominance of male patients (average M:F ratio 35:18). Average age was >50 years in three of the aforementioned articles; in one study, the mean age was 47.

Slonimsky et al., ³² after analyzing the radiological reports of their center and searching for laryngocele cases, suggested an incidence of laryngoceles in the adult population equal to 151 per 2.5 million people per year, way higher than previously reported in the literature. ³³

Laryngocele and cancer

Based on literature available until 2023, it noticeably seems that laryngocele has some sort of correlation with laryngeal cancer. Despite this, further studies should focus on their causal link and, more specifically, on “what comes first.” Some pathophysiological theories point out that probably the cancer, leading to the closing of the laryngeal ventricle, gives rise to an increased laryngeal pressure, which eventually brings to laryngocele's formation. That to be said, we cannot exclude a concomitance of common risks factors, that is, tobacco smoking, which leads to chronic bronchitis and cough, as well as being a well‐established cancerogenic agent.

Close et al. ²⁴ investigated the presence of laryngocele in a cohort of asymptomatic patients undergoing a CT scan of the neck, showing an incidence equal to 29%. Furthermore, they analyzed the anatomical slides of the patients undergoing total laryngectomy, showing the presence of a squamous cell carcinoma in the laryngeal ventricle site of the laryngocele in a total of 7 out of 11 patients. The authors suggested that laryngocele formation is secondary to increased laryngeal pressure from airway obstruction and other responses to the laryngeal cancer itself. ²⁴

Lindell et al. ²⁶ showed a 4.2% incidence of laryngocele in patients with cancer, either laryngeal or hypopharyngeal, theorizing again the possibility of development of the laryngocele secondary to the cancer; they did not find substantial evidence indicating that laryngocele should be a precursor of carcinoma.

Micheau et al. ²⁸ showed an 18% incidence of laryngocele in cases of laryngeal cancer, and particularly in 55 cases, the tumor developed along the epithelium of the laryngocele, accounting for 52 poorly differentiated squamous cell carcinomas, two adenocarcinomas and one lymphoepithelioma.

Pietrantoni et al., ³⁰ report based on X‐ray investigation, an incidence equal to 6.18% in laryngeal cancer. Despite this, they found the laryngeal ventricle involved by the tumor only in half (57%) of the laryngeal specimens examined, postulating a non‐strictly‐association between these two entities, and theorizing the role of other factors (i.e., anatomical predisposition, cough, and dysphonia) that are due to the presence of the tumor and may cause a rise of intralaryngeal pressure.

To the author's knowledge, no existing studies proposed a long‐term follow‐up of patients with laryngocele to observe the eventual development of cancer.

Other etiologies

One case‐control study ²⁵ included in the review proposed an association between saccular disorders and smoking or other chronic laryngeal pathologies: their analysis showed a higher prevalence of active smoking (55.2%) in patients with saccular disorders (saccular cysts and laryngoceles) compared with the control group, consisting in vocal fold cysts (19.7%); despite this they did not account the prevalence of smoking specifically in the subgroup of laryngoceles (9 over the total 29 cases analyzed).

Considering the data collected from this analysis, laryngocele seems to be more frequent in wind instrument players. A total of 55 wind instrumentalists among the 410 laryngoceles cases were found. The heterogeneity of data didn't allow for statistical analysis, but data are still significant to assess a positive correlation. What should be accounted for is that 52 cases out of 54 belong to a work carried out by Macfie et al. in 1966. ²⁷ They selected a sample of asymptomatic wind instrument players and confronted radiographic image results during forced expiration with open or closed glottis, showing that the exam performed during open glottis expiration has a better sensitivity for diagnosis. The inclusion criteria for this latter article, which revolved around a subpopulation of brass players, might have produced some sort of selection bias in the results, further amplified by the relatively surpassed diagnostic techniques employed. These have some significance in terms of diagnosis as well as in terms of etiology: increased intralaryngeal pressure exercised during open glottis strain (i.e., playing a wind instrument) could be mostly related to laryngocele's development compared to closed glottis strains (i.e., coughing). Based on this statement, we could theorize on the occurrence of laryngoceles in people with iatrogenic supraglottic muscle damage; for example, after cricohyoidepiglottopexy: the surgical damage brings to insufficient closure of the glottis, eventually leading to chronic open glottis strains. Naiboglu et al. ²⁹ reported an incidence of 4.8% of laryngoceles after cricohyoidoepiglottopexy, this data being concordant with other findings previously reported in the literature. ³⁴ , ³⁵

However, what emerges from this inceptive analysis is that the literature lacks recent studies focusing on layngocele's incidence in the modern wind instrumentalists population. Considering that glass‐blowing workers nowadays represent a very small population, and that wind instrument players avail of new sound emission techniques and, more in general, better education on the correct sound production (deep—improperly called “diaphragmatic”—breathing and no‐pressure mouthpiece use), the incidence of this pathology could be different from the past. Studies focusing on the real incidence of laryngoceles in the modern brass‐musician population should be carried out to clearly assess it as an etiological factor nowadays.

Diagnostic modalities and treatment

Many studies proposed an X‐ray of the neck as the diagnostic exam, even though its poor positive predictive value is at present established. ²⁸ , ³⁰ When assessing this we have to consider that these studies are dated between the 60's and the 70's years of the XX century, thus CT scans were not already accessible. Macfie et al. ²⁷ demonstrated the major sensitivity of X‐rays executed during increased intraglottic pressure with the larynx open and relaxed (open glottis strain), over the performance during closed glottis strain.

When available, CT scans demonstrated to be the most sensitive radiological exam for the detection of laryngocele.

All the included studies were concord in proposing surgical treatment only in case of symptomatic laryngocele, unless laryngeal cancer was detected, with surgical options leading from endoscopic resection for the internal laryngoceles, to trans‐cervical resection or combined surgery for the external or mixed laryngoceles. ²⁴

Close et al. proposed an endoscopic follow‐up of at least a 2‐year period for asymptomatic laryngoceles, to detect evidence of cancer. In their sample, of the 21 patients with laryngocele and without evidence of laryngeal cancer, 18 performed a 36‐month follow‐up without developing any malignancies or laryngocele‐related symptoms.

Limitations

This review provides a general framework of the actual knowledge regarding laryngocele and its etiology. Despite this, several limitations must be acknowledged when interpreting the results of the included studies.

First of all, all studies were classified as Level IV evidence according to the OCEBM scale, being mostly retrospective or observational studies, and one case‐control study. As such, no causal inferences could be made. Secondly, etiology was proposed only in a, sometimes narrow, fraction of the sample taken into consideration, leading to an even weaker significance of the results.

Thirdly, there was substantial clinical heterogeneity between studies in variables such as cohort eligible criteria, patient characteristics, data collected, diagnostic criteria, and outcome definitions. This hindered direct comparisons and data analysis, significantly limiting the validity of the results.

Lastly, all the included articles, except one, were marred by some sort of population selection bias, focusing either on laryngeal cancer or on asymptomatic patients exposed to UA strain, and this seems to be the weakest point of the analysis. The literature lacks studies on laryngocele in asymptomatic patients not exposed to hyperinflation of the UA.

The great majority of articles regarding laryngocele available in literature until 2023 consist mainly of case reports and case series: we opted to exclude these articles from the review to pursue the highest level of evidence possible and provide readers with the best information available; these choice has drastically reduced the number of articles included in the review, but the authors believed that reducing exposure of unusual and mostly anecdotal cases could provide a better representation of the true etiologies.

CONCLUSIONS

In conclusion, what this analysis observed is that the information currently available in the literature about laryngocele's etiology is limited and weak. Actual knowledge concerning laryngocele etiology appears—based on literature studies—inferential at best. The association between laryngocele and laryngeal cancer is, on the other hand, well‐established and must be taken into account when evaluating patients with this condition. A noticeable correlation between laryngocele and open glottis strains, such as playing wind instruments, was found. Wider and prospective data collections should allow further insight into this condition and its pathogenesis.

Properly conducted standardized cohort studies, taking into consideration the general population, should be performed to understand its real incidence and clinical association in the present.

AUTHOR CONTRIBUTIONS

All authors contributed to the study's conception and design. Alberto Maria Saibene conceived the original idea for the study, performed the systematic electronic searches on databases and defined the exclusion criteria. Giorgia Pugliese and Leonardo Giudici performed the abstracts and full texts review of the articles. Giorgia Pugliese performed the study quality assessment and the data extraction. All authors contributed to the final version of this manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ETHICS STATEMENT

This study did not involve human participants, their data, or biological material. Therefore, it did not require institutional research ethics committee evaluation.

Pugliese G, Giudici L, Scotti A, et al. Glass‐blowing, brass‐playing, and other laryngocele etiologies: fact or fiction? A systematic literature review. World J Otorhinolaryngol Head Neck Surg. 2025;11:468‐476. 10.1002/wjo2.228

DATA AVAILABILITY STATEMENT

All data pertaining to this systematic review are available from the corresponding author upon reasonable request.