Introduction
Chronic rhinosinusitis is a frequently seen condition in otorhinolaryngology out-patients, and it carries significant quality-of-life and healthcare cost implications.Reference Smith, Orlandi and Rudmik 1
Chronic rhinosinusitis is diagnosed using the European Position Paper on Rhinosinusitis and Nasal Polyps (‘EPOS’) criteria.Reference Fokkens, Lund, Mullol, Bachert, Alobid and Baroody 2 Specifically, chronic rhinosinusitis is defined as inflammation of the nose and paranasal sinuses characterised by two or more symptoms. Along with facial pressure or pain, and/or reduction or loss of smell, one of the symptoms should be nasal blockage, obstruction, congestion or nasal discharge (anterior or posterior drip).
An inflammatory condition, chronic rhinosinusitis involves disparate underlying pathophysiological mechanisms. Chronic rhinosinusitis is more akin to conditions of immune dysregulation such as asthma than to acute infective inflammation.
The European Position Paper on Rhinosinusitis and Nasal Polyps criteria divides the condition into chronic rhinosinusitis with nasal polyps and chronic rhinosinusitis without nasal polyps. There is further subdivision dependent on allergy status and co-morbid respiratory conditions, to allow optimal patient management. In regard to this, there are ongoing efforts to classify by histological and immunological features (e.g. eosinophilia and interleukin markers). In our practice, we have seen increasing numbers of patients on anti-tumour necrosis factor (TNF) therapy presenting to ENT with the symptoms of chronic rhinosinusitis rather than recurrent acute sinusitis.
The immunopathology of chronic rhinosinusitis in the Western population is well documented. Indeed, recent work has reported differences in the inflammatory process between Eastern and Western populations.Reference Kato 3 These differences, the role of immune dysregulation in chronic rhinosinusitis, and the observed increase in self-reported rates of rhinosinusitis symptoms whilst on anti-TNF therapy lead us to believe that there is an association between anti-TNF therapy and chronic rhinosinusitis, when compared against Europe-wide published data on chronic rhinosinusitis prevalence.
Materials and methods
The sample population was identified from a prospectively maintained database of patients receiving anti-tumour necrosis factor (TNF) treatment at the rheumatology department within our hospital.
The Global Allergy and Asthma European Network (‘GA2LEN’) questionnaire for chronic rhinosinusitis diagnosisReference Hastan, Fokkens, Bachert, Newson, Bislimovska and Bockelbrink 4 was added to questions regarding smoking history, atopy and chronology of nasal symptoms (Appendix 1). Whilst the lack of specificity in diagnosing chronic rhinosinusitis by questionnaire alone has been reported,Reference Bhattacharyya and Lee 5 the use of the Global Allergy and Asthma European Network questionnaire allows for direct comparison of our data with published chronic rhinosinusitis prevalence.
The sample anti-TNF therapy population was contacted by telephone and asked to participate using a standard wording of the survey. The calls were all made by the co-authors following randomisation of the prospective database. If no response was made to the phone call, the call was repeated at another time at least one week later. Medication history was recorded from the most recent rheumatology out-patient review and confirmed using online records.
Chronic rhinosinusitis was defined according to the European Position Paper on Rhinosinusitis and Nasal Polyps criteria; these were required to be present for longer than 12 weeks and to have occurred in the last 12 months. Patients were also asked if they had received a diagnosis of chronic rhinosinusitis made by a physician, with a positive answer to this question being considered diagnostic. Disease duration and self-reporting of symptoms were included to allow for comparison of our findings against those of the national equivalent study centres in the Global Allergy and Asthma European Network study.
Age, sex, smoking history and atopy (in the form of physician-diagnosed asthma, eczema or hay fever type symptoms) were calculated from the population data for responders. The prevalence of chronic rhinosinusitis, and 95 per cent confidence intervals (CIs), were calculated using the standard error derived from central limit theorem, 1.96. A two-by-two contingency table was constructed using the observed data, and the figures for our nearest geographical centre from the Global Allergy and Asthma European Network, in London. Chi-square and relative risk were calculated.
Results
Of 234 patients, 120 agreed to participate in the survey. Of these, 72 patients (60 per cent) were on anti-tumour necrosis factor (TNF) therapy for rheumatoid arthritis. The remaining 48 patients (40 per cent) had either ankylosing spondylitis or psoriatic arthritis.
The age, sex and smoking status of the anti-TNF therapy sample and anti-TNF therapy chronic rhinosinusitis sample were tabulated with the corresponding Global Allergy and Asthma European Network London data. Despite the difference in sample size, there was no significant difference between the samples in these criteria (Table I).
Table I Age, sex, smoking and atopy by patient sample
IQR = interquartile range; GA2LEN = Global Allergy and Asthma European Network; N/A = not applicable; TNF = tumour necrosis factor; CRS = chronic rhinosinusitis
The prevalence of chronic rhinosinusitis was: 20 per cent (95 per cent CI = 12.9–27.2) in the anti-TNF therapy population as a whole, 22 per cent (95 per cent CI = 14.0–33.3) in the rheumatoid arthritis sample, and 17 per cent (95 per cent CI = 10.0–33) in the ankylosing spondylitis or psoriatic arthritis sample. Anti-TNF treatment was associated with an elevated prevalence of chronic rhinosinusitis in our population. The probability of difference, when compared with Global Allergy and Asthma European Network (London) data using chi-square for a two-by-two contingency table, was 0.0005. The relative risk was 2 (95 per cent CI = 1.37–2.94). There was significance also when only the rheumatoid arthritis population was considered (p = 0.0009; relative risk = 2.23 (95 per cent CI = 1.42–3.51)). There was no significance when only the ankylosing spondylitis or psoriatic arthritis population was considered (Table II).
Table II Chronic rhinosinusitis prevalence and statistical results by patient sample
CI = confidence interval; GA2LEN = Global Allergy and Asthma European Network; N/A = not applicable; TNF = tumour necrosis factor; CRS = chronic rhinosinusitis
The rates of individual symptoms in each group are shown in Table III. These data reflect the increased prevalence of chronic rhinosinusitis. The patients who received a physician-made diagnosis of chronic rhinosinusitis had positive responses for the other criteria too.
Table III Chronic rhinosinusitis symptom prevalence by patient sample
Data represent percentages. CRS = chronic rhinosinusitis; GA2LEN = Global Allergy and Asthma European Network; N/A = not applicable; TNF = tumour necrosis factor
The medication subgroups and the frequency of users in the anti-TNF therapy chronic rhinosinusitis sample and anti-TNF therapy non-chronic rhinosinusitis sample are shown in Tables IV and V. The anti-TNF therapy chronic rhinosinusitis sample received a total of 140 prescribed medications (5.8 per patient) other than anti-TNF therapy in the 6 months prior to data collection. Ten patients (42 per cent) remained on concurrent disease-modifying anti-rheumatic drugs, whilst 14 patients (58 per cent) remained on steroid treatment. The most common medications were prescribed for improving bone density (bisphosphonates or calcium supplements) (n = 18, 75 per cent) and for management of gastroesophageal reflux disease (n = 17, 71 per cent). There was no record of medication history for one non-chronic rhinosinusitis patient. The non-chronic rhinosinusitis sample received 609 prescribed medications (6.4 per patient). All patients receiving disease-modifying anti-rheumatic drugs, prednisolone or thyroid supplements were on single agents. There was no significant difference in the rates of medications prescribed between the chronic rhinosinusitis and non-chronic rhinosinusitis groups. For the remaining medications, patients were often on multiple agents for the same disease process or from the same drug class. There was no difference in the rate of prescriptions for any individual medicine.
Table IV Numbers of drug prescriptions by patient sample: part one*
* Statistical test data shown for those groups with single prescriptions per patient (disease-modifying anti-rheumatic drugs, prednisolone and thyroid supplementation). † n = 24; ‡ n = 95. DMARD = disease-modifying anti-rheumatic drugs; NSAID = non-steroidal anti-inflammatory drugs; TNF = tumour necrosis factor; CRS = chronic rhinosinusitis
Table V Numbers of drug prescriptions by patient sample: part two
*Other drugs included warfarin, anti-histamines and anti-angina medications. † n = 24; ‡ n = 95. PPI = proton pump inhibitor; TNF = tumour necrosis factor; CRS = chronic rhinosinusitis
Discussion
The results indicate an increase in the prevalence of chronic rhinosinusitis (all subtypes) in a population treated with anti-tumour necrosis factor (TNF) therapy, despite the known benefits to some subtypes of chronic rhinosinusitis.Reference Karosi, Csomor and Sziklai 6 The difference in prevalence is significant when compared with published data from the nearest geographical centre. These findings are in contrast to recent studies that implicate TNF in the pathogenesis of some chronic rhinosinusitis subtypes,Reference Karosi, Csomor and Sziklai 6 and raise the possibility of a chronic rhinosinusitis subgroup (with or without polyposis) in which TNF is protective against the condition.
The failure of the questionnaire to accurately differentiate between recurrent acute sinusitis and chronic rhinosinusitis does raise the question of whether our findings are the result of confounding risk factors for recurrent acute sinusitis. Rheumatoid arthritis and disease-modifying anti-rheumatic drugs have been reported to be risk factors for infection.Reference Doran, Crowson, Pond, O'Fallon and Gabriel 7 The role of disease-modifying anti-rheumatic drugs and rheumatoid disease in rates of chronic rhinosinusitis has not been assessed. The chronic rhinosinusitis positive and negative patients in our sample demonstrated no significant difference in terms of prescription rates of disease-modifying anti-rheumatic drugs, steroids (which would normally be protective for chronic rhinosinusitis) or other non-rheumatoid related medication. In addition, anti-TNF therapy has been reported to be a general risk factor for infection,Reference Ding, Ledingham, Lugmani, Westlake, Hyrich and Lunt 8 specifically in sinusitis.Reference Maroon, Bond and Phelan 9
We have endeavoured to distinguish between recurrent acute sinusitis and chronic rhinosinusitis by careful wording of our verbal questionnaire. An unexpected incidental finding of using a telephone rather than postal survey was that 8 out of 24 patients (33 per cent) mentioned symptoms that occurred shortly after commencing anti-TNF therapy. A further 3 out of 120 patients (2.5 per cent) reported a short period of recurrent sinusitis following commencement of anti-TNF therapy, which settled spontaneously. Two more patients reported life-long sinusitis unchanged by anti-TNF therapy, and one patient felt that they had previously had chronic rhinosinusitis which had resolved entirely whilst on anti-TNF therapy. One of these patients had undergone endonasal sinus surgery with good therapeutic effects. These observations, whilst adding no statistical weight to our hypothesis, do add weight to the argument in favour of further investigation of chronic rhinosinusitis subtypes and the potential impact of TNF on these subtypes.
As rates of atopy in our sample anti-TNF therapy population were consistent with those in the Global Allergy and Asthma European Network population, it seems unlikely that our subgroup hypothesis proposals are based on the interaction of TNF with immunoglobulin E. Immunoglobulin A producing dendritic cells have been shown to be upregulated by TNF in mucosal-associated lymphoid tissue.Reference Tezuka, Abe, Iwata, Takeuchi, Ishikawa and Matsushita 10 This role of TNF in immunoglobulin A production, which has a major role in mucosal immune response, may represent the pathway through which anti-TNF therapy makes patients susceptible to chronic rhinosinusitis. In our original investigation plan, we had hoped to assess immunoglobulin levels and the percentage of eosinophils in blood. However, the retrospective nature of this investigation and the small sample population means that no meaningful conclusion could be drawn from these data.
The data also raise issues regarding patient information and consent for the administration of anti-TNF therapy. Perhaps rhinological assessment should be part of the initial planning for anti-TNF therapy. Such a service would seem impractical within the current cost restraints of our public health system. It is, however, prudent that doctors prescribing anti-TNF therapy are aware of the possibility of increased chronic rhinosinusitis prevalence and recurrent acute sinusitis. The availability of an otorhinolaryngology service for referral in cases of new or exacerbated symptoms is important to reduce the effects of chronic rhinosinusitis on quality of life.
-
• Chronic rhinosinusitis is common in ENT out-patients, with significant quality-of-life and healthcare cost implications
-
• Anti-tumour necrosis factor (TNF) treatment is associated with increased chronic rhinosinusitis prevalence
-
• Demographics, smoking, atopy and medications were comparable in chronic rhinosinusitis positive and negative patients
-
• No one medication was commonly prescribed to all chronic rhinosinusitis patients
-
• These results further highlight the complexity of immunity and immune dysregulation in chronic rhinosinusitis
-
• The results suggest that TNF and anti-TNF therapies have different roles depending on chronic rhinosinusitis subtype
In summary, we provide the first observational data relating to chronic rhinosinusitis prevalence in patients being treated with anti-TNF therapy. We show an increased prevalence, which suggests that there is a potential subtype of chronic rhinosinusitis in which TNF plays a protective role in symptoms and the disease process. Clinically, the immediate implication is one of awareness, whilst consideration must also be given to further research into the differentiation of the chronic rhinosinusitis subtypes.
Appendix 1. Telephone survey
Patient number:
Age:
Sex: M/F
Smoking status:
-
□ Current smoker □ Ex-smoker □ Non-smoker
Allergy status:
-
□ Asthma: Y/N If Y, onset?
-
□ Eczema: Y/N If Y, onset?
-
□ Hay fever: Y/N If Y, onset?
-
□ Other allergy: Y/N If Y, onset?
Has your nose been blocked for more than 12 weeks during the last 12 months? Y/N
Have you had pain or pressure around the forehead, nose or eyes for more than 12 weeks during the last 12 months? Y/N
Have you had discoloured nasal discharge (snot) or discoloured mucus in the throat for more than 12 weeks during the last 12 months? Y/N
Has your sense of smell been reduced or absent for more than 12 weeks in the last 12 months? Y/N
Has your doctor ever told you that you have chronic sinusitis? Y/N
If Y to any of the 5 sinusitis questions, have you had these symptoms all your life? Y/N
If N, when did they start?
