Abstract
Background
Melkersson–Rosenthal syndrome (MRS) is often classified under the term orofacial granulomatosis (OFG). A part of OFG patients eventually develop Crohn’s disease (CD), but the relationship between MRS and CD is unknown.
Goals
To evaluate the long-term outcomes of MRS patients, with specific interest in bowel-related symptoms.
Study
This follow-up study consisted of adult patients with MRS – including the monosymptomatic form, cheilitis granulomatosa (CG) – who had participated in our earlier MRS study (diagnosed after 1995 in Helsinki University Hospital). A phone interview was conducted with 27 patients (77.1% of the patients from the earlier study) and included questions about orofacial symptoms, facial palsy, intestinal symptoms, concomitant illnesses, medications, possible food avoidances and family history. Stool samples were collected to measure faecal calprotectin, a surrogate marker for intestinal inflammation.
Results
The median follow-up time from symptom onset was 30 years. Three (11.1%) patients had developed inflammatory bowel disease (IBD); one CG patient developed CD, and two MRS patients with facial palsy developed ulcerative colitis. In addition, several other patients reported intestinal symptoms, but the examination of faecal calprotectin did not indicate intestinal inflammation.
Conclusions
There was a connection between MRS and IBD – not only CD but ulcerative colitis as well.
Melkersson–Rosenthal syndrome (MRS) is a chronic condition often classified as a subtype of orofacial granulomatosis. Oral manifestations are common in MRS. We found a connection between MRS and inflammatory bowel disease, not only Crohn’s disease but ulcerative colitis as well.
Key message
Introduction
Melkersson–Rosenthal syndrome (MRS) is a rare disorder of unknown aetiology and incidence [Citation1,Citation2]. The classic form of MRS consists of a triad of symptoms: recurrent facial palsy, recurrent orofacial oedema and fissured tongue (lingua plicata) [Citation3]. All three symptoms are present only in a minority of patients, and symptoms can also manifest separately [Citation3]. The most common form of MRS is the monosymptomatic form, cheilitis granulomatosa (CG), which presents with only facial or labial oedema. Histology of MRS may show non-caseating lymphoepithelioid granulomas [Citation1,Citation3,Citation4].
Orofacial granulomatosis (OFG) is a rare inflammatory disease of the mouth and orofacial area. Primary symptoms are oedema of the lips and face, and ulceration in the mouth [Citation5,Citation6]. Patients with OFG may develop a symptomatic intestinal disease like Crohn’s disease (CD), and patients with CD may develop oral lesions [Citation7,Citation8]. As such, clinicians have debated whether OFG is an oral manifestation of CD or a separate inflammatory disorder [Citation9–11]. Histologically characteristic are non-caseating granulomas [Citation5,Citation12,Citation13], similar to those found in MRS.
It is not always possible to distinguish MRS from OFG. MRS commonly involves oral manifestations such as buccal and labial swellings, as well as gingival, palatal and lingual manifestations [Citation3]. Histologically, MRS, OFG and oral lesions in CD are indistinguishable. OFG is often used as an umbrella term that encompasses MRS (including CG), to broadly describe the occurrence of granulomatous inflammation in the orofacial region without any recognized systemic condition [Citation2,Citation14]. However, there is considerable overlap in the classification of OFG and MRS/CG.
In an earlier study, we investigated the connection between OFG and CD, and explored the factors leading to bowel disease in patients presenting with orofacial symptoms. We found that 72% of patients with paediatric-onset OFG developed a symptomatic bowel disease, namely CD, during a median of 3.1 years follow-up [Citation6]. However, whether patients with MRS develop a symptomatic bowel disease in long-term follow-up has not been determined.
The aim of this study was to investigate the long-term health outcomes of patients with MRS, with particular interest in bowel-related symptoms.
Patients and methods
Patients
This follow-up study was conducted between May and July 2018. The study included adult patients with triad or oligosymptomatic forms of MRS or monosymptomatic CG, who had been treated and diagnosed at Helsinki University Hospital after 1995 and had participated in our earlier MRS study [Citation1]. An invitation to participate in this follow-up study was mailed to each patient and followed up with a phone call from the researcher (AH). Of the 35 patients contacted, 27 (77.1%) agreed to participate in the study, 23 of which participated in both the interview and laboratory testing and four in the interview only. Among the eight patients who did not participate in this follow-up study, one had died, five could not be reached, and two declined. Following the ethics committee guidelines, reasons for declining were not requested.
Interview
The patients were interviewed over the telephone. The interview consisted of 42 questions regarding orofacial symptoms, facial palsy, intestinal symptoms, concomitant illnesses, medications, and possible food avoidances. Family history of MRS, CG, and inflammatory bowel disease (IBD) was inquired. Additional information about the studied condition was collected from the patient charts.
Laboratory tests
Faecal calprotectin was measured from stool samples in a routine clinical laboratory using a quantitative enzyme immunoassay (PhiCal Test, Calpro AS, Oslo, Norway). Values of <100 μg/g were considered normal [Citation15,Citation16]. Among the 23 patients that agreed to laboratory testing, 18 (66.7% of the study group) provided stool samples for analysis.
The study was approved by the Ethics Committee of the Helsinki University Hospital, and all patients signed an informed consent form.
Statistical analyses
Non-parametric Mann–Whitney and Spearman’s correlation tests were used to investigate the associations between variables. The groups were compared using an independent samples t-test or Mann–Whitney U test as appropriate. Statistical analyses were performed using GraphPad Prism 7 for Windows (GraphPad Software, Inc, San Diego, CA). Differences were considered significant when the p-value was <.05.
Results
The study group comprised 27 patients with a median age of 57.8 years (range 41.4–87.0). Patient characteristics are listed in . Fourteen (51.9%) patients had CG and 13 (48.1%) had triad or oligosymptomatic MRS (MRS t/o). The patients were 10–71 years old (median 24 years) when the MRS symptoms started, resulting in a median of 29.7 years of follow-up. The diagnosis was determined at a median of 38 years (range 23–73) age. Thus, the time interval between initial presentation of symptoms and diagnosis was more than 10 years in 11 patients; among these patients, it was over 20 years in four, and as much as 51 years in one. Concomitant illnesses, such as hypertension, hypothyroidism, hypercholesterolemia and type 2 diabetes, were reported in 18 (66.7%) of the 27 patients. Sixteen (59.3%) had continuous medications. Three (11.1%) patients had IBD.
Table 1. Patient characteristics.
The MRS symptoms experienced by the patients are listed in . Seventeen (63.0%) patients had experienced oedema of the lips at some point during the course of the disease, five (18.5%) of which were within the last year, and one (3.7%) within the last month. Labial oedema had occurred only once in five patients (29.4% of the patients with labial oedema). It was recurrent in 10 (58.8%) patients and continuous in two (11.8%) patients. One patient with persistent and disfiguring labial oedema underwent cheiloplasty of both upper and lower lip.
Table 2. Reported symptoms in patients with MRS t/o or CG within the course of the disease (ever) and within the last year before the study interview (last year).
During the course of the disease, nine (33.3%) patients had experienced facial oedema (other than labial), six (22.2%) of which within the last year and two (7.4%) within the last month. Facial oedema had occurred only once in one patient, but was recurrent in all other patients. There was no correlation between age and facial oedema (p = .8027). There was individual variation in the location of the oedema, with the area around the mouth most commonly reported (three patients). Six patients reported oedema in other body parts (ankles, three patients; hands, one patient; knee, one patient; abdomen, one patient; labia major, one patient; tongue, one patient; anus, one patient).
Of the 13 MRS patients with facial palsy (48.1% of the study group), one had experienced a palsy episode within the last year. Nine (69.2%) had experienced recurrent episodes, while a palsy had occurred only once in four (30.8%). The facial palsy had manifested on both sides of the face in separate episodes in six patients; four patients had experienced palsy on the left side only, and three on the right side only. None of the patients that were initially diagnosed with CG reported facial palsy in the follow-up questioning.
Nine (69.2%) of the MRS patients with facial palsy history had a fissured tongue. The complete symptom triad of recurrent facial palsy, orofacial oedema and fissured tongue, was found in four (14.8%) patients. Five (38.5%) had experienced oedema (facial or labial) or facial palsy within the last year, and one (7.7%) within the last month. Among the 14 CG patients, four (28.6%) had experienced oedema within the last year, and one (7.1%) within the last month.
Fifteen (55.6%) of all the 27 study patients reported some oral symptoms at some point during the course of their disease after diagnosis. Aphthous ulceration or blistering was the most common oral symptom, reported in nine patients. Three patients reported ulcers in the mouth. Three patients reported redness of the oral mucosa, and two reported extended nodules in the floor of the mouth (“staghorning”). Seven patients (25.9% of the study group) reported angular cheilitis (four within the last year and one within the last month). Some patients reported that the oral symptoms coincided with the oedema, while other patients had not noticed such an association.
Of the 13 MRS patients with facial palsy, eight (61.5%) had received peroral cortisone treatment for facial palsy at least once. Of the 14 patients with CG, six (42.9%) had received peroral cortisone, and one (7.1%) patient each of the following treatments for oedema: cortisone injections, laser treatment, topical tacrolimuse (calcineurin inhibitor) and surgery.
Bowel symptoms
Three patients (11.1% of the study group) had been diagnosed with IBD; two MRS t/o patients with ulcerative colitis, and one CG patient with CD. In addition, three patients had a diagnosis of a functional condition of irritable bowel syndrome; two CG patients and one MRS t/o patient. In another six patients (4 CG, 2 MRS t/o), the causes of their intestinal symptoms had been under clinical evaluation, but with negative results. The intestinal symptoms included abdominal pain (n = 4), diarrhoea (n = 3), abdominal distension (n = 2), constipation (n = 2) and vomiting (n = 2). During the last month, nine (33.3%) patients had suffered from diarrhoea; one case was explained by acute gastroenteritis, and another by lactose intolerance.
The patient with CD was diagnosed with IBD nearly 20 years ago, at the age of 27, after which she developed CG at the age of 35. She had suffered from recurrent facial oedema located around the mouth, as well as swelling of both lips, which had become persistent after recurrences. She had also experienced oral ulceration and tags accompanied with angular cheilitis. One patient with MRS t/o was diagnosed with ulcerative colitis one week before the study interview, at the age of 55. The MRS symptoms had started more than 10 years ago, including recurrent swelling of the lower lip, one episode of facial palsy and a fissured tongue. The other MRS t/o patient with ulcerative colitis was diagnosed with IBD over 15 years ago, at the age of 32. He had experienced recurrent facial palsy with the first episodes before the presentation of IBD. He did not have a fissured tongue, but had experienced oral lesions, including aphthous ulceration, oedema and erythema of the oral mucosa, and angular cheilitis, as well as migraine, other headaches and tinnitus.
Stool samples were collected from 18 patients to determine the presence of intestinal inflammation, using faecal calprotectin as a surrogate marker. Faecal calprotectin levels ranged from <5 to 217μg/g (median 23μg/g). Faecal calprotectin levels were normal (<100μg/g) in 15 (83.3%) patients and elevated in three (16.7%) patients. Two of the patients with elevated calprotectin had IBD. One patient with no gut symptoms had faecal calprotectin above the normal range (143μg/g) and was referred for a control measurement (result not known).
Age was not correlated with bowel-related symptoms (p = .4175; median age for patients with and without bowel-related symptoms was 61.9 years and 55.3 years, respectively). Patients with facial palsy did not have more bowel-related diagnosis compared to CG patients (p = .9219, Spearman’s r = 0.01981).
There were two siblings with MRS t/o in the study group. None of the other patients reported first-degree family members with MRS. Two patients (7.4%) reported a family history of CD (one MRS t/o patient had a sibling with CD and another MRS t/o patient had a father with CD). None of the patients reported family members with ulcerative colitis.
Diet
Twenty (74.1%) patients reported avoiding some food items in their diet, most commonly because of heartburn or dyspepsia (reported by 12 patients). Ten (37.0%) patients avoided food items due to intestinal symptoms (four patients avoided lactose, and one avoided milk protein). Eight (29.6%) patients avoided food items that they connected to oral or facial symptoms. Tomato, mustard, seasoning, ketchup, chocolate, kiwi, bell pepper, chili, food additives, and fruits and berries were mentioned to cause or exacerbate labial oedema (n = 4), ulceration in the angle of the mouth (n = 1) or aphthous ulceration in the mouth (n = 2). One patient reported nitrite-containing foods to cause facial oedema.
Discussion
The diagnosis of MRS is difficult, especially when the classic triad of symptoms – facial palsy, orofacial oedema, and fissured tongue – is not present. In many cases, there can be long intervals of months or years without symptoms, and when symptoms are present they may occur one at a time. The diagnosis is clinical and based on patient history and clinical features. Histological evidence of granulomas is not mandatory for the diagnosis. The granulomatous changes confirm the diagnosis, but they are not always present in biopsies, and their absence does not exclude the diagnosis of MRS [Citation3]. Hence, the need for a biopsy is not clear, but if they are taken, they should be sampled during acute oedema to provide the highest probability of diagnostic findings [Citation1]. In some cases, neuroimaging examinations may be necessary to exclude other diseases (e.g. neurosarcoidosis). Because of the challenges in diagnosing MRS, this disease may remain undiagnosed for years or even decades. However, increasing knowledge of this mysterious condition can lead to diagnosis within a shorter timeframe, although because of the nature of the condition, many times the diagnosis can only be established after several years of follow-up. It is important to consider MRS, and question patients about other symptoms, including possible oedema episodes from facial palsy patients, especially those experiencing recurrent palsy. Additionally, the mouth should be carefully examined for fissured tongue. Unfortunately, there is no specific treatment, and the management thus far is mostly empirical.
Over half of the patients with MRS reported some oral symptoms during their follow-up. A similar proportion of adult patients with paediatric-onset CD also reported oral symptoms [Citation17]. In patients with MRS, aphthous ulceration and blistering were the most common oral symptoms. One-quarter of the patients reported suffering from angular cheilitis. In patients with OFG, angular cheilitis was found in half of the patients, and in patients with paediatric CD, it was the most frequently seen oral lesion, present in 15% of the patients [Citation6,Citation17]. However, angular cheilitis is not specific to CD or OFG, and it is also observed in otherwise healthy individuals. Other oral symptoms in patients with MRS were sporadic. Some of the patients had noticed that oral symptoms coincided with facial oedema, while others did not. Interestingly, over half of the patients who reported oral lesions had noticed that certain food substances caused or exacerbated the symptoms, and therefore avoided those substances in their diet. None of the patients reported food allergies that were formally diagnosed with a food challenge.
A part of patients with OFG are known to develop CD. A relationship between MRS and CD has also been proposed [Citation8], but seems controversial [Citation18]. In our cohort, one patient (3.7%) with CG had been diagnosed with CD. Additionally, two MRS t/o patients (7.4% of all patients, 15.4% of patients with MRS) had ulcerative colitis. Thus, in total, 11.1% were diagnosed with IBD, which greatly exceeds the current frequency in the Finnish population (0.86%; Kolho KL, personal communication). This finding suggests a link between MRS and IBD. To the best of our knowledge, there are no existing reports of a connection between MRS and ulcerative colitis.
The patient with CD first presented with IBD, after which she developed CG. However, in both patients with ulcerative colitis, the symptoms of MRS, including facial palsy, presented before the symptoms of the gut disease. Several patients had experienced intestinal symptoms that required clinical examination, but no definite diagnosis was determined. We screened these patients with faecal calprotectin for the presence of inflammation in the gut, but found only one non-IBD patient with a marginally elevated value. Extensive screening for CD has been deemed unhelpful in the absence of gastrointestinal symptoms [Citation18]. This is in line with our finding of negative screening results for the presence of intestinal inflammation. However, further studies with preferably larger cohorts are warranted. Physicians treating MRS should be aware of the likely association between MRS and IBD, and inform the patients to seek medical advice when experiencing long-lasting gastrointestinal symptoms.
The nomenclature and classification concerning the spectrum of MRS and OFG are ambiguous. OFG may be used as an umbrella term to encompass granulomatous conditions in the orofacial area, including MRS t/o and CG. While CG is considered to be a monosymptomatic or incomplete form of MRS, CG cannot be clinically or histologically distinguished from OFG; therefore, these two terms are often used to refer to the same condition. Precise classification is not always even possible, because the disease may be presented without all symptoms. As an example of the difficulty of classification of these conditions, two of our patients diagnosed with CG had fissured tongues, hence their actual diagnosis should have been MRS t/o. Notably, facial palsy is not considered to be a feature of OFG.
Strengths and limitations
Considering how rare MRS is in the general population, we were able to include a decent amount of patients in our study. To the best of our knowledge, this study has the longest follow-up period (median of 30 years after the presentation of symptoms) for patients with MRS.
The oral manifestations and other symptoms were self-reported by the patients, which raises a possible recall bias. Reporting can be affected by the variation among individuals in terms of their ability to recognize and assess the degree of symptoms. However, a proper clinical examination of symptoms is difficult to arrange due to long symptom-free intervals. It is possible that performing the study as a phone interview increased the likelihood of consent. An additional advantage of the phone interview was that we could ask the patients to clarify the responses if needed. Furthermore, we were able to reach patients who had moved away, as well as older patients who otherwise may have difficulties filling in questionnaires or attending study visits.
A fair number of patients consented to the stool screening, allowing us to explore possible intestinal inflammation. Notably, this is the first study to systematically evaluate the possibility of underlying intestinal inflammation using faecal calprotectin as a surrogate marker in patients with MRS.
Conclusion
Long-term follow-up of three decades of patients with MRS confirmed the typical course of the disease – recurrent and chronic – as described in the literature. One-third of our patients had experienced MRS-related symptoms within the last year, and many patients had suffered from the disease for years or decades. However, a number of our patients were currently asymptomatic, even after difficult recurrences of the disease previously.
Oral manifestations do not appear to increase during the course of the disease, although over half of the patients reported oral symptoms at some point of the disease. Allergens in food may play a role in oral and labial symptoms of MRS as self-reported by the patients, but further studies on this topic are needed. In our study, there was a connection between MRS and IBD, not only Crohn’s disease but ulcerative colitis as well.
Acknowledgements
We are grateful to RN Anne Nikkonen for her help with the study.
Disclosure statement
No potential conflict of interest was reported by the authors.
Additional information
Funding
References
- Kanerva M, Moilanen K, Virolainen S, et al. Melkersson-Rosenthal syndrome. Otolaryngol Head Neck Surg. 2008;138:246–251.
- Wehl G, Rauchenzauner M. A systematic review of the literature of the three related disease entities cheilitis granulomatosa, orofacial granulomatosis and Melkersson-Rosenthal syndrome. Curr Pediatr Rev. 2018;14(3):196–203.
- Greene RM, Rogers RS 3rd. Melkersson-Rosenthal syndrome: a review of 36 patients. J Am Acad Dermatol. 1989;21:1263–1270.
- Hornstein OP, Stosiek N, Schonberger A, et al. Classification and scope of clinical variations of Melkersson-Rosenthal syndrome. Z Hautkr. 1987;62:1453–1466. 1471-5.
- Wiesenfeld D, Ferguson MM, Mitchell DN, et al. Oro-facial granulomatosis-a clinical and pathological analysis. Q J Med. 1985;54:101–113.
- Haaramo A, Alapulli H, Aine L, et al. Detailed follow-up study of pediatric orofacial granulomatosis patients. J Pediatr Gastroenterol Nutr. 2017;65:388–393.
- Basu MK, Asquith P, Thompson RA, et al. Proceedings: oral lesions in patients with Crohn's disease. Gut 1974;15:346.
- Tyldesley WR. Oral Crohn's disease and related conditions. Br J Oral Surg. 1979;17:1–9.
- Challacombe SJ. Oro-facial granulomatosis and oral Crohns disease: are they specific diseases and do they predict systemic Crohns disease? Oral Dis. 2008;3:127–129.
- Sanderson J, Nunes C, Escudier M, et al. Oro-facial granulomatosis: Crohn's disease or a new inflammatory bowel disease? Inflamm Bowel Dis. 2005;11:840–846.
- Tilakaratne WM, Freysdottir J, Fortune F. Orofacial granulomatosis: review on aetiology and pathogenesis. J Oral Pathol Med. 2008;37:191–195.
- Grave B, McCullough M, Wiesenfeld D. Orofacial granulomatosis-a 20-year review. Oral Dis. 2009;15:46–51.
- Freysdottir J, Zhang S, Tilakaratne WM, et al. Oral biopsies from patients with orofacial granulomatosis with histology resembling Crohn's disease have a prominent Th1 environment. Inflamm Bowel Dis. 2007;13:439–445.
- Miest R, Bruce A, Rogers RS. 3rd,. Orofacial granulomatosis. Clin Dermatol. 2016;34:505–513.
- Kolho KL, Raivio T, Lindahl H, et al. Fecal calprotectin remains high during glucocorticoid therapy in children with inflammatory bowel disease. Scand J Gastroenterol. 2006;41:720–725.
- Hamalainen A, Sipponen T, Kolho KL. Infliximab in pediatric inflammatory bowel disease rapidly decreases fecal calprotectin levels. World J Gastroenterol. 2011;17:5166–5171.
- Haaramo A, Alapulli H, Aine L, et al. Oral and otorhinolaryngological findings in adults who were diagnosed with pediatric onset Crohn's disease: a controlled study. J Clin Gastroenterol. 2018. DOI:https://doi.org/10.1097/MCG.0000000000001074
- Worsaae N, Christensen KC, Schiodt M, et al. Crohn's disease in the oral cavity. Ugeskr Laeger. 1980;142:2701–2704.