Chronic cough as a complication of treatment with statins: a case report

Introduction

Chronic persistent cough unresponsive to symptomatic treatment or persisting despite treatment of the underlying cause remains a common cause of referral for assessment by respiratory physicians. Chronic cough is said to be present when the duration of the cough exceeds 8 weeks and many causes have been implicated [Chung and Pavord, 2008]. Despite a thorough assessment, examination and use of relevant investigations, a cause may still not be identified thus constituting a diagnosis of idiopathic cough. However, the increased incidence of patients being positive to organ-specific autoantibodies suggests an association between idiopathic chronic cough and organ-specific autoimmunity [Birring et al. 2004].

Case report

We present the case of an 80-year-old gentleman with a past medical history of non-insulin-dependent diabetes mellitus for 10 years and hypercholesterolaemia for 4 years. He was recently diagnosed with ischaemic heart disease 2 years ago thus needing medical treatment as well as percutaneous angioplasty.

The patient was referred to the respiratory clinic at Mater Dei Hospital in Malta with a 2-month history of cough. He complained of a persistent cough 2 months after being started on statins only to resolve a few days after stopping this particular treatment. The patient did not complain of any throat clearing, dysphonia or altered taste. His other treatment included insulin, aspirin, isosorbide mononitrate and atenolol. He had not been prescribed an angiotensin-converting enzyme (ACE) inhibitor. He was initially prescribed simvastatin in June 2009 by his cardiologist. This symptom did not vary during the day nor did it disturb his sleep. He did recall that the cough came in bouts with increasing frequency as worsening from one to two bouts of coughing per day at symptom onset to five to six bouts per day after approximately 10 days from onset. Each bout lasted from 1 to 5 minutes. Furthermore, the onset of his symptoms was not associated nor preceded by other symptoms. Also, on direct questioning, he had never complained of similar symptoms in the past.

Of his own accord he omitted the simvastatin in view of the dry cough 2 months later. He continued taking his other treatment as advised. A decrease in frequency of symptoms occurred after stopping the statin with eventual complete resolution 8 days later.

In view of this observation, the patient was again reviewed by his cardiologist in September 2009 who advised switching over to fluvastatin. Two months into treatment, however, the same observation was made by the patient who again omitted his treatment. His symptoms again increased in frequency after onset only to resolve 1 week after omitting his treatment. In November 2009 he was reviewed by a gastroenterologist in view of the possibility of reflux associated cough. The patient was treated empirically with omeprazole and again switched over to simvastatin. It is important to note that there was no history of dysphonia, waterbrash, dyspepsia or worsening of cough in the supine position were reflux-associated symptoms would be expected to worsen. He was not obese with a body mass index of 22.3 g/m², was not exposed to heavy lifting and did not abuse alcohol, all being risk factors for nonacid reflux.

All other medications were continued and at no point was the patient treated with an ACE inhibitor or an angiotensin receptor antagonist. He had stopped and resumed his treatment for three times (twice with simvastatin and once with fluvastatin) with resolution and recurrence of his symptoms, respectively, before being referred to the respiratory clinic. Apart from this symptom he had no other ailments and led a very active lifestyle.

On examination he was found to have a clear chest with good air entry bilaterally. He was comfortable at rest with a respiratory rate of 14 breaths per minute and was neither clubbed nor cyanosed and there were no stigmata of cor pulmonale. The rest of his clinical examination was entirely normal. Chest X-ray and lung function testing were normal. In the first instance, the patient was advised to omit his simvastatin and keep a symptom diary. At follow up, results of erythrocyte sedimentation rate and serum C-reactive protein were both normal, at 6 mm first hour and <6 mg/l, respectively. The complete blood count, urea, electrolytes, calcium, creatinine and liver function testing were similarly within normal limits. Serum protein electrophoresis, antineutrophil cytoplasmic antibody (ANCA) and total extractable nuclear antibody levels were also normal. Rheumatoid factor IgM were just above the normal range (0–15 U/ml) at 19 U/ml. Fasting lipids were deranged, with a total cholesterol level of 7.30 mmol/l and a low-density lipoprotein level of 5.17 mmol/l. Triglycerides were normal.

A further attempt at introducing statin therapy was undertaken. The patient was advised to continue taking simvastatin at his usual dose until symptoms occur. He was then advised to change over to another formulation (vitamin C tablets), the nature of which was not revealed to the patient thus acting as a placebo drug to monitor symptoms. Symptoms recurred after 8 weeks of statin treatment, with resolution on stopping simvastatin and changing over to placebo.

At follow up at the respiratory clinic, the patient did not describe the occurrence of new symptoms. In view of his medical history and deranged lipid profile he was referred to the lipid clinic for consideration of treatment with nicotinic acid as well as being advised lifestyle and dietary modification.

Discussion

In this case report, the association between cough and treatment with simvastatin is discussed. A chronic cough is defined as persistent symptoms for 8 or more weeks and patients presenting with such symptoms should be referred to a pulmonologist for further evaluation.

In the presence of a normal chest X-ray causes of chronic cough to consider include asthma, postnasal drip, gastroesophageal reflux, drug-induced and idiopathic cough [Rosenow, 1995] with the first three remaining the most common found to account to up to 90% of diagnoses in a study of 88 patients [Mello et al. 1996]. There is increasing evidence showing an association organ-specific autoantibodies and idiopathic chronic cough [Birring et al. 2004]. The commonest drugs implicated in drug-induced cough are the ACE inhibitors were up 15% of patients being treated with ACE inhibitors may develop cough [Rosenow et al. 1992]. The impact of drugs on cough has also been extensively studied by the ‘Groupe d′Etudes de la Pathologie Pulmonaire Iatrogène’ (GEPPI) which is a clinical study group based in France that helps provide evidence-based guidance through their website Pneumotox [Foucher et al. 2010]. This is a valuable tool in providing guidance on drugs implicated in respiratory disease and respiratory symptoms [Camus et al. 2004; Dimopoulou et al. 2006].

Statins have been rarely associated as a cause of interstitial lung disease (ILD), however the evidence is limited to a few individual case reports. Fernandez and colleagues have shown that for every 10,000 reports of a statin-associated adverse event, approximately 1–40 reports were for statin-induced ILD [Fernandez et al. 2008]. This shows that statin-induced ILD is a newly recognized side effect of statin therapy which can present as a cough. Unfortunately, little is known about the mechanism of lung injury. With regards to nonacid reflux as a cause of cough, investigations to confirm reflux, such as oesophageal pH monitoring, were not conducted nor was the patient given a trial of gastrointestinal prokinetic treatment. However, the history given by the patient and the temporal association with statin use was very indicative that nonacid reflux was not high in the list of differential diagnosis.

The clinical course of interstitial changes may vary with some patients responding to drug cessation and glucocorticoids whilst in others cases the condition progresses slowly despite treatment [Walker et al. 2007]. The occurrence of ILD in patients taking statins suggests that several clinical presentations associated with mitochondrial toxicity may ultimately be reported with statins in susceptible individuals [Golomb and Evans, 2007]. Birring and colleagues suggest that the increased incidence of patients found positive to organ-specific autoantibodies means that an association exists between ILD and organ-specific autoimmunity [Birring et al. 2004].

Our patient developed a cough both on simvastatin and fluvastatin and therefore we conclude that this is indeed a class effect. To date there is only one report in the literature describing the association of cough with statins. Pascual Cruz and colleagues have shown that cough associated with treatment with statins showed a marked improvement after temporary interruption of the statin [Pascual Cruz et al. 2008]. Similarly, our patient was challenged with a statin and once cough developed he was changed over to placebo so as to eliminate any possible psychogenic associations with the cough. To date this is the second report in the literature discussing the association between statin and cough.

The authors hypothesize that statin therapy could lead to an increase in bronchial or cough reflex hypersensitivity thus resulting in an increased incidence of cough. This hypothesis is supported by the fact that the symptoms diminish when statin therapy is terminated, however such an interaction could be objectively assessed by direct bronchial provocation testing using histamine or metacholine. An underlying mechanism for this bronchial hypersensitivity could well be the HMG-CoA reductase inhibition of statins, however more research is needed so as to identify the link between the two mechanisms. A literature and online search did not reveal any available data from large phase IV clinical trials in statin therapy and the authors suggest that authorities with access to such data might be able to establish a possible link between cough and statin therapy.

Conclusion

Statin-induced cough has not been widely reported in the literature and little is known about the underling mechanisms. The presence of autoantibodies is unlikely to be significant in this case and, as such, the hypothesis that an association between autoimmune lung disease and statin treatment exists is not plausible at present. Other unknown genetic factors could possibly be associated with this symptom and further work is needed to confirm the link between statins and cough and unravel the underlying mechanisms. Since statins are very widely used throughout the world, knowing about the possibility of this side effect could have huge implications when treating our patients.