Prevalence of Esophageal Webs in Patients Undergoing Direct Laryngoscopy

ABSTRACT

Introduction

The prevalence of esophageal webs in the pharyngoesophageal segment (PES) is unknown, in part because webs produce inconsistent symptomatology and can be difficult to diagnose. This study aims to determine the prevalence of PES webs in patients undergoing endoscopic laryngeal surgery.

Methods

This retrospective cohort study included patients undergoing laryngoscopy for pathology unrelated to webs. Evaluation for the presence of a web was performed on standard/routine examination of the PES during direct laryngoscopy. Demographic and clinical factors were analyzed for associations with webs.

Results

Of 123 patients included in this study, 42 (34.1%) were found to have a PES web. A total of 22 webs (52.4%) were on the left, 16 (38.1%) were on the right, and 4 (9.5%) were bilateral. There was no difference in age (58.1 vs. 58.7, p = 0.864) or BMI (29.7 vs. 29.8, p = 0.900) between patients with and without PES webs. Webs were significantly more common in patients with a history of irradiation to the head and neck (70% vs. 31%, p = 0.031) with a RR of 2.26 (CI: 1.38–3.69). There was no association of webs with gender, race, history of gastroesophageal reflux disease, or other clinical factors. Only 33.3% of patients with a web had documented symptoms of dysphagia.

Conclusions

PES webs may be more prevalent than what is historically cited in the literature, and webs may be more common in patients with a history of irradiation to the head and neck.

Level of Evidence

4.

PES webs may be more common than what is traditionally cited in the literature. The prevalence of PES webs in this retrospective cohort study was found to be 34%. Webs may be more likely in patients with a history of irradiation to the head and neck, and only one‐third of patients with a PES web in our cohort had dysphagia symptoms.

1. Introduction

The pharyngoesophageal segment (PES) is comprised of the inferior pharyngeal constrictor, cricopharyngeus muscle, and cervical esophagus and maintains a basal tone responsible for preventing reflux, regurgitation of gastric contents, and aerophagia. Esophageal webs are thin, eccentric, mucosal and submucosal membranes of tissue that occur within the PES and may result in solid food dysphagia [1]. PES webs are historically associated with Plummer–Vinson syndrome [2, 3], though idiopathic PES webs can occur with no obvious etiology [4] and may be difficult to diagnose on flexible endoscopy [5] or fluoroscopy [6]. Although Mehdizadeh et al. demonstrated a high prevalence of webs in elderly cadavers [5], there have been no endoscopic studies to date regarding PES web prevalence in living patients, which is likely in part due to difficulty with visualization of the PES on flexible endoscopy due to its elastic collapse.

This study aims to identify the prevalence of PES webs in patients undergoing direct laryngoscopy for unrelated laryngeal pathology. In addition, we evaluate demographic and clinical factors associated with PES webs in this cohort.

2. Materials and Methods

This study is a retrospective review of patients who underwent laryngoscopic procedures in an operating room setting at the Medical College of Georgia at Augusta University from 2020 to 2022. All patients underwent surgery for pathology unrelated to webs. It is our routine practice to perform examination of the hypopharynx and PES during direct laryngoscopy, which allows for adequate distension of the PES using a rigid laryngoscope for examination for the presence of PES webs (Figure 1). Patients were excluded if they had prior endoscopic esophageal surgery or the known presence of an esophageal web from existing clinical information or fluoroscopic imaging. Institutional Review Board approval was obtained from the Medical College of Georgia at Augusta University (IRB #1073277‐23).

FIGURE 1.

Distension of the PES during direct laryngoscopy demonstrating various examples of webs (*) on the right, left, and bilateral in relationship to the cricopharyngeus muscle (+). Note that even bilateral webs are eccentric, as opposed to strictures, which are circumferential, usually concentric in nature, and occur more commonly in the esophagus distal to the PES.

Demographic data were collected, including age, gender, and ethnicity. Clinical factors obtained include BMI, history of irradiation to the head and neck, presence of symptoms of dysphagia or gastroesophageal reflux disease (GERD), and other past medical and surgical history. These data were collected as binary parameters from clinical documentation in the electronic medical record. Patients with documented symptoms of difficulty swallowing were considered to have dysphagia; globus/foreign body sensation was not included as a dysphagia symptom. The presence of GERD and other medical comorbidities were collected from the patient's past medical history and problem list of the electronic medical record. Patients with a documented history of surgery in the head and neck, including previous endoscopic laryngeal or airway surgery, tonsillectomy, sinus surgery, thyroidectomy, parathyroidectomy, tracheostomy, oncologic resection, spine surgery, and others, were considered to have a history of prior head and neck surgery.

2.1. Statistics

Statistical parameters are presented as mean (standard deviation) for continuous variables and number (percentage) for categorical variables. Statistical differences between groups were analyzed by Student's unpaired t‐test for continuous variables and Fisher's exact test for categorical variables. The relative risk (RR) with confidence interval (CI) was calculated for clinical factors found to have a statistically significant association with PES webs. Analyses were carried out using GraphPad Prism 10, and a p < 0.05 was considered to be statistically significant.

3. Results

A total of 123 patients were included in the study with a mean age of 58.5 (15.7) years. Sixty‐four (52%) were female and 59 (48%) were male. Patient racial demographics included 80 (65.0%) Caucasian, 37 (30.1%) African American, 3 (2.4%) Asian, 2 (1.6%) Hispanic, and 1 (0.8%) multiracial.

In total, 42 (34.1%) patients were found to have the presence of a PES web. Twenty‐two webs (52.4%) were on the left, 16 (38.1%) were on the right, and 4 (9.5%) were bilateral (Table 1). There was no statistically significant difference in age between patients with and without PES webs (58.1 vs. 58.7, p = 0.864). When categorized by age, patients younger than 40, 40–49, 50–59, 60–69, and older than 70 all had web prevalences between 32% and 36% (Figure 2). BMI was not significantly different in patients with PES webs (29.6 vs. 29.8, p = 0.900). In this cohort, 40.6% of women were found to have a PES web compared to 27.1% of men, though this difference was not statistically significant (p = 0.131). There was no difference in web prevalence between Caucasian and African American patients (32.5% vs. 35.1%, p = 0.834); other ethnicities were excluded from this specific analysis due to low sample size. Patient demographics and their associations with webs are summarized and presented in Table 2.

TABLE 1.

Presence of PES webs in patient cohort.

Web	n (%)
None	81 (65.9%)
Left	22 (17.9%)
Right	16 (13.0%)
Bilateral	4 (3.3%)

FIGURE 2.

The prevalence of PES webs when categorized by age was similar among all groups (32%–36%).

TABLE 2.

Demographics of patients with and without PES webs, showing mean (standard deviation) for continuous variables and number (percentage) for categorical variables.

Variable	Web	No Web	p
Age	58.1 (16.4)	58.7 (15.4)	0.864
BMI	29.6 (7.2)	29.8 (6.6)	0.900
Gender			0.131
Male	16 (27.1%)	43 (72.9%)
Female	26 (40.6%)	38 (59.4%)
Ethnicity			0.834
White/Caucasian	26 (32.5%)	54 (67.5%)
Black/African American	13 (35.1%)	24 (64.9%)
Asian	1 (33.3%)	2 (66.7%)
Hispanic	1 (50%)	1 (50%)
Multiracial	1 (100%)	0 (0%)

Associations between PES webs and clinical factors are presented in Table 3. Patients with a history of prior irradiation to the head and neck were found to have a significantly higher prevalence of PES webs than patients without prior irradiation (70% vs. 31%, p = 0.031), with a RR of 2.26 (CI: 1.38–3.69). Webs were not more common in patients who reported a history of GERD (33.3% vs. 34.8%, p = 1.0) or prior head and neck surgery (28.9% vs. 42.6%, p = 0.170). On subgroup analysis based on the type of surgery, there was no increase in web prevalence in patients with a history of prior endoscopic laryngeal surgery, endoscopic airway surgery, tracheostomy, thyroidectomy, parathyroidectomy, or spine surgery when compared to patients with no prior head and neck surgery. However, all patients (n = 4) with a history of head and neck oncologic resection were found to have PES webs (p = 0.043) with a RR of 2.35 (CI: 1.69–3.28). Each of these patients also had a history of irradiation to the head and neck.

TABLE 3.

Association of clinical factors with the presence of a PES web in patient cohort, presented as number (percentage).

Variable	Web	No Web	p
Prior irradiation to head and neck			0.031*
Yes	7 (70%)	3 (30%)
No	35 (31.0%)	78 (69.0%)
History of GERD			1.000
Yes	19 (33.3%)	38 (66.7%)
No	23 (34.8%)	43 (65.2%)
Prior head and neck surgery			0.170
Yes	22 (28.9%)	54 (71.1%)
No	20 (42.6%)	27 (57.4%)

Note: Statistical significance denoted by * (α = 0.05).

Interestingly, only 33.3% of patients with webs had documented symptoms of dysphagia. The age of web patients with dysphagia was 61.9 (14.3) versus 56.3 (17.3) in web patients without dysphagia, which was not statistically significant (p = 0.152). Patients with webs were more likely to have dysphagia if they had prior head and neck irradiation, with a RR of 3.75 (CI: 1.90–7.40).

4. Discussion

This is the largest study to date regarding the endoscopic prevalence of PES webs, which we estimate to be approximately 34%. Mehdizadeh et al. suggested a higher prevalence of 68% in their study of elderly cadavers [5]. Though our data did not show a statistically significant association of PES webs with age, there was a large difference in mean age (58 in this study vs. 83 in the cadaver study) which may contribute to the difference in prevalence in these two studies. Additional confounders potentially include sample size, sampling error, or possible tissue effects during medical embalming.

The prevalence and clinical impact of PES webs is understudied. Inconsistency with nomenclature contributes to diagnostic confusion. We define webs as thin, discrete mucosal and submucosal tethered bands that occur in the PES and are eccentric in nature. Webs differ from redundant mucosal folds or plications, which are broad, thick, and non‐tethered. Strictures are circumferential, usually concentric in nature, and occur more commonly in the esophagus distal to the PES. Postradiation stenosis manifests as diffuse fibrosis to mucosa as well as musculature and represents a distinct entity from webs.

The paucity of literature regarding PES webs is also likely related to diagnostic challenges on endoscopy and fluoroscopy. Because of the basal tone of the cricopharyngeus muscle and the elastic collapse and mucosal redundancy of the PES, sufficient distension of the PES is imperative for adequate endoscopic visualization [5] and impacts the diagnostic sensitivity of endoscopy for webs. The use of a rigid laryngoscope in this study provides favorable exposure, simultaneously acting as a stent and endoscope to evaluate the PES. In contrast, flexible esophagoscopy relies on gaseous insufflation for esophageal distension and may be suboptimal for visualization of the PES and accurate diagnosis of webs, particularly if performed under general anesthesia.

Allen et al. and Ekberg and Nylander estimated the prevalence of PES webs on fluoroscopic imaging to be 12.5%–14% in patients with dysphagia and up to 7% in normal volunteers, and authors from both studies identified the importance of distension of the PES with a sizeable contrast bolus for accurate diagnosis [6, 7]. Radiologic evaluation of the PES on barium fluoroscopy is additionally limited by resolution and frame rate [5]. A barium swallow (esophagram) typically obtains images at 4 frames/s [8], although this can vary widely between institutions, while a modified barium swallow typically captures 15–30 frames/s [9]. Even with an adequate frame rate, PES webs are typically only visualized on a limited number of frames throughout an entire imaging sequence and may be missed without careful review during the passage of the contrast bolus through the PES [5, 6] (Figure 3). These diagnostic challenges with barium fluoroscopy may account for the lower prevalence in these imaging studies compared to our cohort.

FIGURE 3.

Web (*) demonstrated on the anterior esophageal wall during distension of the PES with contrast bolus on a barium esophagram.

Though some studies suggest a higher prevalence in women [4, 10], there was no statistically significant difference in web prevalence based on gender in our study. Similarly, there was no association of PES webs with any other demographic or clinical factors evaluated in this study apart from patients with prior radiation therapy to the head and neck and prior head and neck oncologic resection. Irradiated patients had a prevalence of PES webs of 70% in our cohort with a RR of 2.26 (CI: 1.38–3.69). A PES web was identified in all patients with a history of head and neck oncologic resection with a RR of 2.35 (CI: 1.69–3.28), though meaningful clinical correlation for this specific finding is limited by a small sample size of four and confounding, as each of these patients additionally had head and neck irradiation.

The pathogenesis of PES webs is poorly understood. Mehdizadeh et al. suggested that webs may result from progressive tethering of the cricopharyngeus muscle to the lateral and anterior walls of the PES from repetitive trauma from cumulative lifetime swallows and may be part of the aging process [5]. Interestingly, our data show a similar prevalence across adult age groups, suggesting that some webs may be congenital [11] or develop during childhood; a comparison to the prevalence of PES webs in the pediatric population is unfortunately unavailable in the existing literature and is also limited by previously discussed diagnostic challenges.

We postulate that PES webs are likely heterogeneous in etiology and can also form from inflammation or desquamation of PES mucosa with subsequent tethering during re‐epithelialization. Prior reported cases have been described in diseases of epithelial desquamation or diseases associated with mucositis, including epidermolysis bullosa [12, 13, 14], Stevens–Johnson syndrome [15], celiac disease [16, 17], graft versus host disease [18], and inlet patches (ectopic acid‐secreting gastric mucosa) in the cervical esophagus [19, 20]. Similarly, we hypothesize that radiation‐induced mucositis in the PES can serve as the inciting mucosal injury that predisposes for the formation of a web.

Symptomatology associated with webs is also poorly understood. Webs are more common in patients with dysphagia [6, 7], and treatment of PES webs with dilation can be effective in symptomatic patients [10, 21]. However, webs are also found in asymptomatic patients [7]. We suspect that the size and arc length of a web impact the distensibility and maximal cross‐sectional area of the PES, which then affects the degree of anatomic obstruction of bolus passage and associated symptomatology. Penetrance of symptoms of dysphagia in patients with documented PES webs in our study was only 33%. Post hoc analysis showed that webs were more likely to be associated with dysphagia in patients with prior irradiation to the head and neck, with a RR of 3.75 (CI: 1.90–7.40). This finding suggests another potential explanation for incomplete penetrance of dysphagia symptoms in web patients. Specifically, patients with normal pharyngeal strength and contractility may be able to overcome mild obstruction from a web with minimal or no symptoms. However, symptoms may be unmasked in patients with reduced pharyngeal contractility and reduced bolus lubrication due to decreased saliva production, both of which are common after irradiation to the head and neck [22].

The high prevalence of PES webs from this study suggests that many patients with webs remain undiagnosed on routine workup of dysphagia with flexible endoscopy and barium fluoroscopy. Though the role of empiric esophageal dilation in patients with a negative workup is controversial [23, 24, 25, 26, 27], Jacob et al. found that mucosal tears after empiric dilation were predictive of symptomatic improvement [28]. We postulate that some patients who improve with empiric dilation may have undiagnosed underlying PES webs that are lysed or ruptured during dilation of the PES, resulting in a mucosal tear and subsequent improvement in symptoms.

Limitations to this study include its retrospective nature and potential bias from sampling error, as all patients included in our cohort were treated by a tertiary academic laryngology practice and underwent surgery for laryngeal pathology, which may skew the measured prevalence in our cohort when compared to the overall population. In addition, patients with a known history of a PES web from prior fluoroscopic swallow studies were excluded from our cohort, which may result in an underestimate of the true prevalence in the general population. We planned to examine the association of webs with autoimmune and rheumatologic diseases, but our cohort did not yield sufficient numbers of patients with these diseases to power a meaningful analysis. Correlation between the presence of a PES web and preoperative hematologic labs to evaluate a potential association with iron‐deficiency anemia would be interesting, as the impact of iron‐deficiency anemia on the pathogenesis of webs is poorly understood [29, 30]. Hematologic bloodwork is not routinely collected for preoperative evaluation at our institution, and many patients are referred from outside our institution, so unfortunately, these data were not available in our cohort.

5. Conclusions

This is the largest study to date regarding the endoscopic prevalence of PES webs, which was approximately 34% in our cohort. PES webs may be more common than what is traditionally cited in the literature. Our study further suggests that webs may be more likely in patients with a history of irradiation to the head and neck. Symptomatic penetrance of webs is variable, as only one‐third of patients with a PES web in our cohort had dysphagia symptoms.

Conflicts of Interest

The authors declare no conflicts of interest.

Yang Z. M., Stockton S. M., Compton E. C., Butler J. J., Johnson C. M., and Postma G. N., “Prevalence of Esophageal Webs in Patients Undergoing Direct Laryngoscopy,” The Laryngoscope 135, no. 10 (2025): 3594–3599, 10.1002/lary.32282.