Muckle-Wells syndrome (MWS) is an inherited autoinflammatory disease characterized by fevers, rashes, arthralgia, conjunctivitis, and sensorineural hearing loss. In MWS, NLRP3 gene mutations are associated with excessive interleukin-1 release. The aims of this study were to determine the otologic characteristics of MWS, define trajectories of hearing loss, and explore the association with distinct NLRP3 genotypes.
A prospective observational cohort study of children and adults diagnosed as having MWS was conducted at a single center. NLRP3 gene mutations were determined. Patients underwent standardized clinical, laboratory, and otologic assessments, including pure tone audiometry, vestibular organ testing, and tinnitus evaluation. Trajectories of hearing loss were defined for each genotype. The genotype-specific risk of progression of hearing loss was determined.
A total of 33 patients ages 3–75 years who were members of 5 families with 4 different NLRP3 gene mutations were included. The majority of patients (67%) experienced bilateral sensorineural hearing loss. Even in cases of profound hearing loss vestibular reactivity remained normal. Fourteen adult patients reported nondebilitating tinnitus. Overall, hearing impairment progressed with age. Patients with the T348M mutation were at highest risk of rapid progression of sensorineural hearing loss.
Patients with MWS are at risk of developing progressive sensorineural hearing loss without vestibular involvement. Hearing impairment starts at high frequencies and can subsequently progress to profound hearing loss. Progression is age dependent. Patients with different NLRP3 mutations had distinctly different trajectories of hearing loss, suggesting a mutation-specific risk that should be considered when making treatment decisions.
Muckle-Wells syndrome (MWS) is an inherited autoinflammatory disease within the spectrum of cryopyrin-associated periodic syndromes (CAPS) (1). At the molecular level CAPS is characterized by mutations in the NLRP3 gene encoding the protein cryopyrin (2). The clinical phenotype of CAPS encompasses 3 syndromes: familial cold autoinflammatory syndrome (FCAS) (mildest), MWS (moderate), and neonatal-onset multisystem inflammatory disease (NOMID) (most severe), also known as chronic infantile neurologic, cutaneous, articular syndrome (3, 4).
Mutations in the NLRP3 gene on chromosome 1q44 in patients with MWS lead to altered function of the gene product, the NLRP3 protein, which is essential for the activation of intracellular caspase 1 and the processing of interleukin-1β (IL-1β) (5–7). The NLRP3 protein is part of the NLRP3 inflammasome. The activation of NLRP3 together with ASC and procaspase 1 leads to IL-1β release (8). Activation of the NLRP3 inflammasome induces cleavage of procaspase 1 to form the active caspase 1 that in turn cleaves proIL-1β to yield IL-1β, which is released from cells (9). Macrophages from patients with MWS show a constitutive increase of IL-1β (10, 11). Subsequently, excessive amounts of IL-1β, which occur periodically, lead to characteristic inflammatory features in patients with MWS. Patients with MWS commonly present with acute attacks of fever, rash, musculoskeletal symptoms, and conjunctivitis. Progressive sensorineural hearing loss and renal failure are the most devastating sequelae of MWS. IL-1 inhibition controls and possibly reverses sensorineural hearing loss in CAPS patients (12–18).
Sensorineural hearing loss in children and adults is etiologically heterogeneous with many known genetic and environmental causes (19–21). Genetic factors account for half of all cases of congenital profound or progressive hearing loss. These can be classified by the mode of inheritance or the presence of characteristic clinical features that may result in the diagnosis of a specific form of syndromic deafness. The identification of >120 independent genes for deafness has provided profound new insights into the physiology of hearing and the pathology of hearing loss (22). The majority of identified gene products are essential in the transduction process in the cochlea.
Even in the first description of the syndrome by Muckle and Wells in 1962 histopathologic findings in the temporal bones of 2 patients were addressed; these included a complete degeneration of the organ of Corti in the inner ear (1). However, knowledge about the pathophysiology of hearing loss in MWS remains limited. No prospective standardized audiologic assessments in MWS patients have been reported, and the association of distinct NLRP3 gene mutations and trajectories of hearing loss have not been explored. This study aimed to determine the genetic, clinical, laboratory, and otologic characteristics of patients with MWS, to define trajectories of hearing loss, and to explore the association of hearing loss with distinct NLRP3 genotypes.
PATIENTS AND METHODS
A single-center observational study of consecutive patients diagnosed as having MWS between 2000 and 2008 was performed. Consecutive patients with MWS were eligible if they demonstrated clinical features of MWS and genetic evidence of an NLRP3 mutation. All MWS patients were followed up at a tertiary care center (departments of otorhinolaryngology and rheumatology at the University Hospital Tuebingen) according to standardized protocol. Informed consent for genetic testing for the NLRP3 mutations was obtained from all patients or from the legal guardian if the patients was a minor. Normal values from hearing assessments of age-matched controls were derived from large published data sets (23). Approval from the local ethics committee was obtained (Research Ethics Board no. 326/2007B01).
Demographics, family history, and NLRP3 mutation.
Demographic information was collected including sex, ethnicity, and age at diagnosis. Assessment of family history determined parental consanguinity and identified family members diagnosed as having autoinflammatory syndromes and possible MWS-related organ disease, such as hearing loss. NLRP3 gene mutations were determined as previously described (24).
Constitutional symptoms recorded included fever (pattern and duration) and fatigue. Organ-specific clinical characteristics recorded included headache, ocular symptoms, e.g., conjunctivitis, uveitis, papilledema, sensorineural hearing loss, oral ulcers, abdominal pain, arthralgia, arthritis and myalgia, and skin symptoms, including erythematous and cold-induced rash. In addition, proteinuria, hematuria, and renal failure requiring organ replacement were recorded.
Age-specific pure tone audiometry was performed using conventional, play, or behavioral methods. The audiologic examination included air and bone conduction thresholds for pure tone frequencies from 250–10,000 Hz and middle ear pressure testing by impedance audiometry. All assessments were performed using calibrated equipment in an International Organization for Standardization–certified soundproof room.
Pure tone thresholds were recorded at 500, 1,000, 2,000, and 4,000 Hz for both ears. The pure tone average was calculated according to the equation PTA = (HT 500 Hz + HT 1,000 Hz + HT 2,000 Hz + HT 4,000 Hz)/4, where PTA is the pure tone average and HT is the hearing threshold. The individual hearing levels were adjusted to the age-adequate hearing threshold to account for the potential bias of age-related hearing loss.
The vestibular organ was tested by videonystagmography. The test included calibration, observation of spontaneous or provoked (by head shake) nystagmus, positional nystagmus, and caloric stimulation. History of vertigo was also recorded.
All adult patients were asked to complete the 52-item tinnitus questionnaire (25). The questionnaire is validated for adult patients with chronic tinnitus and evaluates severity and effect of tinnitus. The 5 domains and their maximum scores include emotional distress (24 points), cognitive distress (16 points), intrusiveness (16 points), auditory perception difficulties (14 points), sleep disturbances (8 points), and somatic complaints (6 points). The maximum summative score is 84 points.
Treatment regimens and responses to IL-1 inhibition were documented. Patients received either anakinra (an IL-1 receptor antagonist) or canakinumab (a human monoclonal antibody targeted at IL-1β).
Assessment of trajectories of hearing loss.
Audiograms of all family members with an identical NLRP3 mutation were analyzed. Trajectories of hearing loss were constructed for each distinct NLRP3 mutation trajectory using the right ear pure tone average of all MWS patients at diagnosis. These were compared to the right ear pure tone averages from healthy controls.
Mutation-specific trajectories were compared with those of healthy controls by linear regression analyses using pure tone averages. Statistical analyses were performed with SAS software for Windows, version 9.2 (SAS Institute).
A total of 35 patients clinically diagnosed as having MWS and demonstrating evidence of an NLRP3 mutation were identified. Two patients were excluded: 1 patient had audiologic assessments at another center, which did not meet study protocol for technical standards, and 1 patient refused standard testing.
Thus, the study cohort consisted of 33 patients (16 men and 17 women). Demographic characteristics are shown in Table 1. Patients came from 5 different families (Figure 1). All patients were Caucasian. The median age at diagnosis was 35 years (range 3– 75 years). All 33 patients were found to be heterozygous carriers of mutations located on exon 3 of the NLRP3 gene. Four different NLRP3 gene mutations were found: 14 patients had the A439V mutation, 12 patients had the E311K mutation, 3 patients had the T348M mutation, and 4 patients had the V198M mutation (Table 1).
Table 1. Baseline characteristics of and NLRP3 gene mutations in a large cohort of patients with Muckle-Wells syndrome (n = 33)
Age at diagnosis, median (range) years
No. of men/women
No. (%) Caucasian
No. (%) with NLRP3 mutation
All 33 patients had active MWS at diagnosis. Constitutional symptoms were common: severe fatigue was seen in all patients. The most common MWS-related organ system features were musculoskeletal and ocular manifestations: arthralgia was present in 30 patients (91%), arthritis in 18 patients (55%), and myalgia in 9 patients (27%). Ocular symptoms were seen in 28 patients (85%), including conjunctivitis in 24 patients (73%) and uveitis in 11 patients (33%). Papilledema was not found in this cohort. Hearing loss was present in 22 of the 33 patients (67%). Recurrent episodes of headaches were reported in 22 patients (67%) and oral ulcers in 13 patients (39%). Erythematous rash, independent of cold exposure, was described by 26 patients (79%), and cold-induced rash was described by 13 patients (39%). Febrile episodes lasting 3–5 days were reported by 5 patients (15%). The maximum temperature per episode ranged between 39°C and 40°C. Twelve patients (36%) reported episodic abdominal pain. Clinical data are summarized in Table 2.
Table 2. Clinical symptoms of Muckle-Wells syndrome patients with confirmed NLRP3 mutation and otologic assessment
Systemic features, no. (%)
Organ manifestations, no. (%)
Sensorineural hearing loss
Results of hearing assessment.
All patients had normal tympanic membrane mobility and middle ear pressure as measured by impedance audiometry. Twenty-two of the 33 patients (67%) presented with sensorineural hearing loss of different ranges compared to an age-matched population with normal hearing. Symmetric bilateral hearing thresholds were identified in 29 patients (88%). In the other 4 patients, minimal asymmetry of the hearing threshold in the higher frequencies was found. One female patient (age 19 years) has been deaf in the left ear since birth. Baseline audiometric data are summarized in Table 3.
Table 3. Hearing thresholds in 33 patients at time of Muckle-Wells syndrome diagnosis and in age-matched normal controls*
Age at diagnosis, years
PTA in controls, dB
Muckle-Wells syndrome patients
Hearing threshold at the highest tested frequency, dB
Pure tone average (PTA) was determined by using pure tone thresholds recorded at frequencies of 500, 1,000, 2,000, and 4,000 Hz. Hearing thresholds of >130 dB were coded if the maximum loudness of the audiometer (130 dB) was not perceived by the patient (i.e., nonmeasurable threshold, anacusis). Age-matched normal control hearing assessment values were derived from large published data sets.
Patient was found to have been deaf in the left ear since birth.
Pure tone hearing thresholds uniformly worsened at higher test frequencies. The slope of the threshold varied among patients, starting with an isolated pitch above 4 kHz in younger patients. Such isolated high-frequency hearing loss was not always noticed by patients. Due to the methodologic limitation of the pure tone average calculation, which only averages hearing thresholds assessed at frequencies of 500, 1,000, 2,000, and 4,000 Hz, such hearing impairment cannot be determined by pure tone average comparison. Therefore, characteristic hearing loss in MWS appears better illustrated by measuring the hearing threshold in decibels at the highest frequency tested by audiometry, as shown in Table 3.
None of the 15 patients tested complained of vertigo. Neither spontaneous nor provoked nystagmus was present. Caloric stimulation revealed normal nystagmus response even in patients with profound hearing loss.
Of the 17 patients who were asked to complete the tinnitus questionnaire, 14 responded. Seven (50%) reported that chronic tinnitus had been present for years, but none reported a significant effect of the tinnitus on everyday life.
Results of treatment.
Both IL-1 inhibition treatment regimens led to stability of hearing results in the majority of patients. Hearing improved in 3 patients treated with canakinumab compared to 2 patients receiving anakinra. All 5 patients carried the E311K NLRP3 mutation. In the anakinra-treated group, 1 patient's health deteriorated (Figure 2).
Trajectories of hearing loss.
Overall, MWS patients had a significantly steeper slope of hearing loss progression as compared with healthy controls (P < 0.05) (Figure 3). Each NLRP3 mutation was associated with a distinct trajectory of hearing loss. In patients with the V198M or A439V mutation, progression of hearing loss was similar to that in healthy controls. In contrast, the trajectories of hearing loss for patients with the E311K and T348 mutations had significantly steeper slopes (P < 0.001), reflecting rapidly progressive sensorineural hearing loss (Figure 3).
This is the first study to determine the spectrum of audiologic and neuro-otologic features in a large cohort of MWS patients and to explore the association of hearing loss with the NLRP3 mutation subtype. At diagnosis, sensorineural hearing loss was present in 67% of patients. Trajectories of hearing loss were plotted and demonstrated that the autoinflammatory disease overall does not have the same effect on hearing in all patients, even after adjustment for age. In some patients, it led to accelerated physiologic hearing loss; some patients lost their hearing rapidly at a young age. This study demonstrated that patients with different NLRP3 mutations had distinctly different trajectories of hearing loss, suggesting a mutation-specific risk that should be considered when making treatment decisions.
MWS patients presented with distinct otologic features. Sensorineural hearing loss remains the characteristic manifestation of MWS; it was found in two-thirds of patients at the time of diagnosis. In the initial stage of disease, hearing was affected at higher test frequencies, as previously reported (16). Similarly, age-related hearing loss, presumed to be primarily associated with arteriosclerosis, classic vascular risk factors, and environmental factors, typically affects sensitivity to higher frequencies first (frequencies above ∼2 kHz) (26, 27). Immune-mediated hearing loss is a leading cause of reversible sensorineural hearing loss. It is reported to develop in one-third of patients with systemic autoimmune diseases. Different subtypes are recognized, including acute sensorineural hearing loss, rapidly progressive bilateral sensorineural hearing loss, and Ménière disease (28).
Goldbach-Mansky et al and Ahmadi et al suggested that the underlying pathology of hearing loss in CAPS may be a disruption of the blood–brain barrier and subsequent “leakage” of inflamed microvessels (12, 29). This phenomenon can be visualized by gadolinium enhancement in the cochlea in patients with NOMID, when examined by high-resolution magnetic resonance imaging sequencing (12). It is interesting that the inflammatory, microvascular process selectively targets hearing and spares vestibular function in the inner ear even in MWS patients with severe inflammation and profound hearing loss. Associated nondebilitating tinnitus was reported by 50% of the MWS patients tested. Inflammatory hearing loss is not unique to MWS: Karmody et al reported sensorineural hearing loss in 22 of 38 patients with inflammatory bowel disease, with very limited improvement following immunosuppressive therapy (30). In contrast, MWS patients appear to demonstrate significant improvement or even complete reversibility of sensorineural hearing loss when IL-1 is inhibited (15, 16, 17, 31).
MWS patients with different NLRP3 mutations were found to have distinct hearing loss trajectories. This is particularly interesting because the V198M mutation is known to have broad phenotype variability, including that some carriers are asymptomatic. Dode et al reported low incidence of hearing loss in a family with the V198M mutation (32).
Patients with the A439V mutation appeared to have accelerated age-related hearing loss. MWS patients positive for the newly identified E311K mutation demonstrated a linear progression of hearing loss over time, as shown in Figure 3. Most importantly, patients with the T348M mutation demonstrated severe sensorineural hearing loss at a young age. Similarly, in the study by Dode and colleagues all 8 patients with MWS (from 4 different families), who were positive for the T348M mutation, were found to be deaf. Patients ranged in age from 9 to 57 years (32). These data suggest that hearing loss in MWS is closely associated with the NLRP3 mutation subtype. The results of this study suggest the presence of high, intermediate, and low-risk mutations for progressive hearing loss in MWS.
There were several limitations to this study. The study cohort was small in size, which was not unexpected given that MWS is a rare disease. However, this is the largest MWS cohort reported to date that has been systematically studied for neuro-otologic manifestations at diagnosis. Genetic mutations were identified in all patients, and neuro-otologic manifestations were prospectively determined in an unbiased manner in an attempt to address this challenge. The interdisciplinary clinical care setting at our center and the associated data collection process enhance the robustness of the data reported in this study and reduce the risk of bias. However, it remains a possibility that patients with potentially high-risk NLRP3 mutations, such as T348M, but a mild phenotype, including normal hearing and no other significant organ disease, may have been missed, since those patients would not necessarily seek medical attention. This would lead to an overestimation of the effect of the mutation subtype. However, the risk of not capturing the mild or asymptomatic end of the spectrum would apply to all mutations reported in this study.
This study determined the spectrum of otologic features in MWS patients at the time of diagnosis. Sensorineural hearing loss was the leading symptom, present in 67% of patients. Nondebilitating tinnitus was reported by 50% of patients tested. No patient had any evidence of impaired vestibular function in the inner ear. Distinct mutation-associated trajectories of hearing loss were identified, suggesting a genetically defined effect on hearing, even after adjustment for age. These data are consistent with patient information derived from published reports, but should be prospectively validated in other collaborative studies, since a possible mutation-specific risk would affect treatment decisions.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Benseler had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Kuemmerle-Deschner, Koitschev, Ummenhofer, Hansmann, Plontke, Koitschev, Koetter, Angermair, Benseler.