Authors: Elizabeth S. Aby, Jason D. Eckmann, Jad Abimansour, David A. Katzka, Claire Beveridge, Joseph R. Triggs, Mohamad Dbouk, Tsion Abdi, Kevin O. Turner, Catiele Antunes, Justin Zhuo, Afrin N. Kamal, Parth Patel, C. Prakash Gyawali, Joshua A. Sloan
Categories: Original Articles, corticosteroid, dilation, dysphagia, endoscopy, esophagus, lichen planus
Source: Journal of Clinical Gastroenterology
To better understand the characteristics, treatment approaches, and outcomes of patients with esophageal lichen planus (ELP).
ELP is a rare, often unrecognized and misdiagnosed disorder. Data on this unique patient population are currently limited to small, single-center series.
A multicenter, retrospective descriptive study was conducted of adults diagnosed with ELP over a 5-year period, between January 1, 2015, and October 10, 2020, from 7 centers across the United States.
Seventy-eight patients (average age 65 y, 86% female, 90% Caucasian) were included. Over half had at least 1 extraesophageal manifestation. Esophageal strictures (54%) and abnormal mucosa (50%) were frequent endoscopic findings, with the proximal esophagus the most common site of stricture. Approximately 20% had normal endoscopic findings. Topical steroids (64%) and/or proton pump inhibitors (74%) dominated management; endoscopic response favored steroids (43% vs. 29% respectively). Almost half of the patients required switching treatment modalities during the study period. Adjunctive therapies varied significantly between centers.
Given its at times subtle clinical and endoscopic signs, a high index of suspicion and biopsy will improve ELP diagnosis, especially in those with extraesophageal manifestations. Effective therapies are lacking and vary significantly. Prospective investigations into optimal treatment regimens are necessary.
Key Words: lichen planus, esophagus, endoscopy, dilation, corticosteroid, dysphagia
Lichen planus (LP) is a chronic, idiopathic mucocutaneous inflammatory disease of squamous tissue that most frequently affects the skin, nails, hair, and mucous membranes.^1,2^ Cutaneous LP is estimated to affect 0.2% to 1% of adults worldwide, but esophageal lichen planus (ELP) is rare, with an estimated prevalence of 0.19%.^3,4^ Although patients with ELP may present with dysphagia or odynophagia, the majority are asymptomatic,^5,6^ and extraesophageal manifestations such as oral lesions may be the only clues to the diagnosis.^7,8^ These characteristics can make ELP challenging to diagnose.
Research to date has been based on small case series and single-center analyses. No clear treatment recommendations exist, given the lack of robust data on the topic. The aim of the current multicenter study was to better characterize the clinical presentation and management of ELP by analyzing a large data set acquired from multiple high-volume esophageal centers across the United States.
Adults (>18 y of age) carrying a diagnosis of LP with either an upper endoscopy or an encounter with a gastroenterology provider over a 5-year period (January 2015 to October 2020) were identified based on Current Procedural Terminology, ICD-9, and ICD-10 coding. Seven high-volume esophageal centers (University of Minnesota, Mayo Clinic Rochester, University of Pennsylvania, Johns Hopkins University, University of Oklahoma, Stanford University, and Washington University in St. Louis) contributed to this multicenter, retrospective descriptive study. Independent review of medical records based on an a priori data accrual plan identified patients with a diagnosis of ELP at each center. Deidentified patient data collected included clinical presentation, esophageal and extraesophageal manifestations, endoscopic findings, imaging findings, treatment modalities, and histologic and endoscopic outcomes. The institutional review board of each participating center reviewed and approved the study protocol.
Available data were analyzed in a descriptive manner to determine clinical presentation, management options utilized, and treatment outcomes in ELP patients. Continuous variables are expressed as medians and interquartile ranges or mean and SD, as appropriate. Categorical variables are summarized as counts and percentages. Data amalgamation and calculation of the above parameters were performed using Microsoft Excel. No imputation was made for missing data. A P-value ≤ 0.05 was considered statistically significant.
A total of 78 patients with ELP were identified across the 7 centers participating in the study. The baseline clinical characteristics of this study population are shown in Table 1. The majority of patients were female (85.9%) and white (89.7%), with a median age of 65 years. Extraesophageal manifestations were present in the majority of patients (55%). Oral lesions were the most common, followed by skin and genital involvement. Multiple sites of extraesophageal involvement were noted in 23%. Of 51 patients tested, 1 patient (1.9%) was found to have hepatitis C and was treated with direct-acting antiviral therapy.
Focal stricture was the most frequent endoscopic finding (53.8%), most often in the proximal esophagus (16/42, 39.0%) followed by distal esophagus (9/42, 22.0%); multifocal strictures were seen in 13/42 (31.0%). Mucosal sloughing or friability (50%) and narrow caliber esophagus (26.9%) were other common endoscopic findings. One patient was diagnosed with esophageal squamous cell cancer at the time of endoscopy. The esophagus appeared endoscopically normal in 20.5%. When barium esophagram was performed, esophageal luminal narrowing was noted in 84.1%. High-resolution esophageal manometry, performed in 10 patients, demonstrated esophagogastric junction outflow obstruction in 4 patients and was normal in 3 patients. None of the patients with esophagogastric junction outflow obstruction had distal esophageal strictures that could explain this manometric finding.
Proton pump inhibitors (PPIs) were the most frequent initial treatment modality (74.4%), either alone (17.9%) or in combination with additional therapies (56.4%) (Table 2). Overall, 50 patients (64.1%) received swallowed topical steroids (swallowed fluticasone or compounded budesonide), alone (17.9%) or in combination (46.2%). Across sites, these were administered 1 to 2 times daily, with total fluticasone doses ranging from 440 to 880 mcg, and total daily budesonide doses ranging from 2 to 6 mg. PPI combined with topical steroids was the single most common unique treatment combination (41.0%). As part of initial therapy, 6 patients (7.7%) received systemic steroids, 5 (6.4%) underwent intralesional steroid injections, and 4 (5.1%) were started on noncorticosteroid immunosuppressant medications. Endoscopic and histologic improvement was seen in 42.9% and 25.0% of patients treated with topical steroids, and 29.4% and 24.0% with PPI-based regimens, respectively. Responses to additional initial treatment modalities are shown in Table 2.
Ultimately, 32 patients (41.0%) underwent an adjustment in their ELP treatment regimen after initial treatment. The choice of this subsequent therapy was variable, with no consistent pattern between centers. Ten patients had a change to a different formulation of topical corticosteroid regimen, only 2 of which showed endoscopic or histologic improvement. Four patients were started on systemic corticosteroids after failing topical steroids, none of which were documented to have improvement. Twenty-three (29.5%) patients eventually required the use of a noncorticosteroid immunosuppressant medication, the majority of which (n=19) had extraesophageal organ system involvement. The immunosuppressant medications used included methotrexate, hydroxychloroquine, azathioprine, cyclosporine, and tacrolimus. Five of these 23 patients (21.7%) showed esophageal endoscopic or histologic improvement after initiating this therapy.
Fifty-seven patients (73.1%) underwent a total of 338 endoscopic dilations (median 4 dilations, interquartile 2 to 8) after the diagnosis of ELP. Twelve patients (15.4%) required 10 or more dilations over the course of the study. Adverse events were reported in 6 of 338 dilations (1.8%), 3 of which were perforations (0.9%); other adverse events consisted of severe chest pain (2 patients) and bleeding (1 patient). Flares of ELP symptoms were not documented after dilation.
In this multicenter retrospective descriptive study, we report the clinical manifestations, management modalities, and treatment outcomes in the largest reported cohort of ELP patients to date. Our findings confirm existing single-center data that ELP is most frequently diagnosed in middle-aged and elderly White women with variable clinical presentations.^9^ Oral or multisystem manifestations of ELP were noted in over half of the patients. Similar to previously published reports, strictures were frequently encountered in endoscopy and barium studies, however, ~20% had a normal endoscopy. In addition, high-resolution esophageal manometry did not have significant value in the evaluation of these patients. Finally, we report marked variation in management across the esophageal centers, with variable success in treatment outcomes.
Our findings establish that dominant esophageal strictures seen at endoscopy are often found in the upper-esophagus and mid-esophagus, with almost one third of patients having multifocal strictures.^9–11^ The proximal location of strictures is atypical of reflux-associated esophageal strictures and brings eosinophilic esophagitis into the differential diagnosis, particularly when only mucosal edema or rings are present without sloughing or a lacy appearance. Endoscopically identified strictures or narrow caliber esophageal lumen are under-recognized, as barium radiography demonstrated narrowing in >80% when performed (Fig. 1D). These findings are similar to previously published work that demonstrates that symptomatic esophageal narrowing identified by barium esophagram is common and under-recognized at endoscopy in other inflammatory esophageal conditions like eosinophilic esophagitis.^12^ Other common endoscopic findings included mucosal friability and sloughing (Figs. 1A–C), lacy white papules, desquamation, superficial pinpoint erosions, pseudomembranes, esophageal webs, and narrow caliber esophagus.^13^ One patient had squamous esophageal cancer, which has been described previously.^14^ This emphasizes the importance of considering endoscopic evaluation in lichen planus patients with esophageal symptoms.
FIGURE 1 Appearance of ELP endoscopically with friable mucosa in a narrow caliber esophagus (A); narrow caliber esophagus with mucosal friability and sloughing (B); esophageal mucosal edema with stricturing, narrow caliber, and a subtle lacy appearance (C); barium esophagram with multiple strictures (D), and histopathologically highlighting esophageal squamous mucosa with epithelial atrophy, lymphocytic infiltrate within the epithelium and lamina propria, Civatte bodies (red circles), and epithelial separation from the lamina propria (E).
A normal-appearing esophagus was encountered in 1 in 5 patients in our study, making esophageal biopsies essential when the ELP is suspected. Histologic findings reported include lichenoid lymphocytic infiltrate involving the superficial lamina propria and basal epithelium, parakeratosis, atrophic epithelium, lack of hypergranulosis, and variable thinning or acanthosis.^9^ Civatte bodies (anucleated necrotic basal cells) at the dermal-epidermal junction are characteristic of ELP (Fig. 1E).^15^ Similar histologic changes can be seen in lymphocytic esophagitis (LE), another rare, poorly understood disease of the esophagus with pathogenesis proposed to be related to allergic, medication-induced, hypersensitivity, or autoimmune reaction.^16^ Although there is an overlap in histologic findings between LE and ELP, ELP is classically characterized by a band-like lichenoid distribution of lymphocytes with dyskeratinocytes (Civatte bodies), whereas LE often shows a peripapillary lymphocytic distribution and is not associated with Civatte bodies.^16,17^ Endoscopic findings are highly variable and similar to ELP the esophagus can appear normal. Therefore, when a patient presents with symptoms suggestive of ELP, LE should be on the differential for both the endoscopist and pathologist interpreting biopsy specimens.
To the best of our knowledge, our study is the first to demonstrate marked variation even among esophageal experts in the management of ELP across multiple high-volume esophageal centers. This is likely a reflection of the lack of formal treatment guidelines for this rare disease. Several management modalities were utilized, including PPI, steroids (topical, systemic, and intralesional injection, all at varying doses and intervals), immunosuppressive medications, and endoscopic dilation. Initial PPI or topical steroid use frequently required a change to combination therapy, adjustment of steroid formulation, or addition of immunosuppressives, with variable histologic or endoscopic benefit. Importantly, although not evaluated with formal structured questionnaires to determine symptomatic benefit, topical steroids with or without PPI seemed to provide better endoscopic improvement compared with other options, including immunosuppressive agents (Table 2). However, even with changes in therapy, endoscopic/histologic improvement was seen overall in a minority of patients, illustrating the poor understanding of the disease, and underscoring the difficulty in treating patients with ELP.
PPIs were the most common medications prescribed for ELP across all sites. The reason behind this is unclear, given that the current understanding of ELP pathogenesis does not support acid reflux as a primary mechanistic driver of the disease. Potential explanations include the treatment of concurrent or prior patient symptoms (eg, heartburn), the proposed anti-inflammatory effect of PPIs, misinterpretation of endoscopic findings as reflux-related disease, concern that underlying acid reflux could potentiate or worsen ELP, and the frequency of PPI prescription by gastroenterologists for other esophageal disorders coupled with a perceived low risk of adverse effects with this therapy.^18–20^ However, it is worth noting that only a small proportion of patients (18%) were treated with PPI monotherapy, while the majority of PPI usage occurred in combination with additional, more disease-directed therapies, which suggests an understanding among practitioners that acid suppression alone is unlikely to adequately treat ELP.
Medical therapy was frequently combined with endoscopic dilation, often repeated. The need for endoscopic dilation did not correlate with any specific medication use. The adverse event rate with endoscopic dilation (1.8%) was higher than that reported in restrictive esophageal disorders such as eosinophilic esophagitis.^21^ Importantly, the risk of perforation in this patient population seems to be higher than the overall general risk of dilation for other esophageal stricturing diseases.^22^ Although not appreciated in our cohort of patients, endoscopists should be aware of Koebner phenomenon, where trauma from dilation may potentially flare ELP.^11^
An association has been suggested between Hepatitis C and LP.^23^ In our cohort, 1 patient was found to have hepatitis C and was treated with direct-acting antiviral therapy. Many patients were tested, however, 27 were not. Given the association between hepatitis C and LP, and because hepatitis C screening is recommended by the US Preventive Services Task Force for all patients between 18 and 79 years old ^24^ we suggest all patients found to have ELP should undergo screening with anti-HCV antibody testing followed by polymerase chain reaction testing for HCV RNA if antibody testing is positive.
On the basis of the findings of this descriptive series, we propose the following evaluation and management algorithm for individuals with suspected ELP (Fig. 2). Our study is limited by small patient numbers despite multicenter data collection, and by the retrospective study design that is unable to determine disease prevalence. Treatment outcome assessment is confounded by the concurrent use of multiple management modalities, lack of histologic follow-up, and lack of a formal symptom questionnaire or structured protocol assessing outcomes. Nevertheless, this multicenter data set from large academic centers provide important insights into disease characteristics and management, which can provide a basis for the formulation of a diagnostic approach to ELP. Esophageal involvement should be suspected not just in patients with esophageal symptoms, but also in LP patients with oral lesions or multisystem presentation. Endoscopic findings may be subtle, but findings of proximal or multifocal esophageal strictures, atypical inflammatory changes, or a lacy/reticular pattern to the mucosa should raise suspicion for ELP. Histopathologic evidence of ELP may be present even without esophageal endoscopic findings.^25^
FIGURE 2 Proposed treatment algorithm for the evaluation and management of a patient with suspected esophageal lichen planus. ELP indicates esophageal lichen planus. †Careful dilation is indicated for focal strictures or narrow caliber esophagus. *All treatment selections should be based on a patient-centered risk and benefit discussion along with management of other areas of lichen planus involvement.
Swallowed topical steroids (with or without PPI) should be considered first-line therapy, with careful endoscopic dilation when strictures are encountered. Endoscopic follow-up should be considered because of the risk of esophageal squamous cancer.^14^ Second-line agents include alternate formulations of topical steroids, systemic corticosteroids, and immunosuppressive medications. Prospective registry-based studies are needed to better understand the value of each therapeutic modality and to estimate the risk of long-term complications.
Elizabeth S. Aby, Email: abyxx002@umn.edu.
Jason D. Eckmann, Email: eckma044@umn.edu.
Jad Abimansour, Email: abimansour.jad@mayo.edu.
David A. Katzka, Email: dak2178@cumc.columbia.edu.
Claire Beveridge, Email: beveric@ccf.org.
Joseph R. Triggs, Email: Jrtriggs@gmail.com.
Mohamad Dbouk, Email: dbouk@wustl.edu.
Tsion Abdi, Email: tabdi1@jhmi.edu.
Kevin O. Turner, Email: turn0585@umn.edu.
Catiele Antunes, Email: catiele.antunes@yale.edu.
Justin Zhuo, Email: jzhuo92@gmail.com.
Afrin N. Kamal, Email: kamala@stanford.edu.
Parth Patel, Email: parth.a.patel@wustl.edu.
C. Prakash Gyawali, Email: cprakash@wustl.edu.
Joshua A. Sloan, Email: sloan151@umn.edu.