Pulmonary disease induced by cardiovascular drugs

INTRODUCTION — A number of cardiovascular drugs have the potential to cause respiratory impairment, although diffuse parenchymal lung disease is quite rare (table 1A-B). Several different respiratory adverse effects have been identified: upper airway angioedema or hematoma, bronchoconstriction, cough, interstitial pneumonitis, organizing pneumonia, eosinophilic pneumonia, drug-induced lupus, acute respiratory distress syndrome, diffuse alveolar hemorrhage, pleuritis, pleural effusion, methemoglobinemia, and solitary lung mass.

This topic review will provide an overview of the lung diseases induced by various cardiovascular drugs. The clinical manifestations, diagnosis, and management of pulmonary toxicity due to amiodarone and an approach to the diagnosis of interstitial lung disease are discussed separately.

●(See "Amiodarone pulmonary toxicity".)

●(See "Approach to the adult with interstitial lung disease: Clinical evaluation".)

●(See "Approach to the adult with interstitial lung disease: Diagnostic testing".)

AMIODARONE — Pulmonary toxicity is a well-known adverse effect of the antiarrhythmic agent amiodarone. Several forms of pulmonary disease have been described, including interstitial pneumonitis, organizing pneumonia, acute respiratory distress syndrome (ARDS), diffuse alveolar hemorrhage (DAH), eosinophilic pneumonia, pulmonary nodules, solitary masses, and also (rarely) pleural effusion. The clinical presentation, pathogenesis, diagnosis, and treatment of amiodarone pulmonary toxicity are discussed separately. (See "Amiodarone pulmonary toxicity".)

ANGIOTENSIN-CONVERTING ENZYME INHIBITORS — Angiotensin-converting enzyme (ACE) inhibitors are associated with cough, angioedema, and, rarely, pneumonitis.

●Cough – All of the ACE inhibitors can induce a dry, persistent, and often nocturnal cough (in 5 to 20 percent of patients). The cough may develop within hours of the first dose or weeks to months later. It is more common in women, non-smokers, and persons of Chinese origin. The cough typically resolves one to four weeks after discontinuation of the ACE inhibitor but in a subgroup of coughers, resolution may take several months [1]. One important caveat is that cough may be a symptom of heart failure, and a thorough history and physical examination are needed to ascertain whether the cough is truly related to the ACE inhibitor therapy. (See "Major side effects of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers" and "Causes and epidemiology of subacute and chronic cough in adults".)

●Angioedema – Angioedema, particularly of the oropharynx and larynx, complicates ACE inhibitor therapy in approximately 0.1 to 0.7 percent of recipients. ACE inhibitor-induced angioedema is mediated by excess bradykinin, which builds up due to both lower ACE activity and reduced angiotensin II production. Since ACE inhibitor-associated angioedema typically causes swelling of the mouth, lips, tongue, larynx, pharynx, and subglottic tissues, upper airway compromise may be the presenting sign. Pruritus and urticaria are typically absent. Unlike classical hereditary angioedema, C1 inhibitor level and function and thus the C4 levels are normal in ACE inhibitor-associated angioedema. (See "An overview of angioedema: Pathogenesis and causes", section on 'ACE inhibitors' and "ACE inhibitor-induced angioedema", section on 'Pathophysiology'.)

●Interstitial pneumonitis – Captopril and perindopril have rarely been associated with the development of diffuse interstitial pneumonitis. An eosinophilic pneumonia was found in most cases, but an acute hypersensitivity pneumonitis with lymphocytic infiltration has been described [2-6]. The radiographic appearance is nonspecific, with bilateral patchy ground glass or consolidative opacities being the most frequent finding. The diagnosis is usually made on the basis of peripheral blood or bronchoalveolar lavage eosinophilia, exclusion of infection, and/or response to empiric drug discontinuation. Occasionally, a lung biopsy is needed. Therapy consists of drug withdrawal; occasionally systemic glucocorticoids have been used for patients with more severe respiratory impairment [5,6]. (See "Approach to the adult with interstitial lung disease: Diagnostic testing".)

ANGIOTENSIN RECEPTOR BLOCKERS — The incidence of cough appears not to be increased with use of the angiotensin receptor blockers (ARB). Angiotensin receptor blockers (ARB) are associated with a low rate of angioedema (0.1 to 0.2 percent in large trials). However, the cross-reactivity with angiotensin-converting enzyme (ACE) inhibitors is difficult to ascertain because of the phenomenon that patients with ACE inhibitor induced angioedema may continue to have episodes for weeks to months following discontinuation of the ACE inhibitor, which may overlap with initiation of the ARB. In a systematic review and meta-analysis of 11 randomized trials evaluating ARB in patients intolerant to ACE inhibitors, ARB had cough and angioedema incidences similar to placebo [7].

ANTIPLATELET (NON-ASPIRIN), ANTICOAGULANT, AND THROMBOLYTIC MEDICATIONS — Use of anti-platelet medications (glycoprotein IIB/IIIA antagonists) [8-17], anticoagulants (vitamin K antagonists, direct thrombin inhibitor, factor Xa inhibitor) [18-22], and thrombolytic agents [23-25] in patients with coronary artery disease has been associated with bland diffuse alveolar hemorrhage (DAH). DAH has a similar presentation to pulmonary edema, so a high index of suspicion is needed, particularly when a patient with presumed pulmonary edema does not respond promptly to diuresis. The diagnosis and management of DAH are discussed separately. (See "The diffuse alveolar hemorrhage syndromes".)

Rarely, spontaneous pharyngeal and laryngeal hematomas associated with warfarin anticoagulation have caused airway obstruction [26-28].

A few cases have been reported of interstitial pneumonitis due to ticlopidine and clopidogrel [29-32].

ASPIRIN — Patients with aspirin-exacerbated respiratory disease (AERD), also known as Samter's triad (asthma, nasal polyposis, and acute bronchoconstriction secondary to aspirin ingestion), can experience acute bronchoconstriction after ingestion of aspirin. Some patients tolerate 81 mg of aspirin but develop symptoms at 162 or 325 mg. If patients experience dyspnea within three hours of starting or increasing aspirin dosage, AERD may be the cause. Most patients will experience a combination of nasal congestion, rhinorrhea, wheezing, and dyspnea. Additional symptoms may include facial flushing/erythema, laryngospasm, abdominal cramps, epigastric pain, and hypotension. Bronchoconstriction is typically reversible with an inhaled bronchodilator, which should be given promptly. Aspirin desensitization is an option for patients with AERD who require aspirin therapy. In addition, such protocolized desensitization has been shown to improve nasal-sinus and asthma symptoms in those with suboptimal control of their asthma [33]. When undertaking desensitization, premedication with a leukotriene-modifying agent such as montelukast is recommended. The protocol for aspirin desensitization is discussed separately. (See "Aspirin-exacerbated respiratory disease" and "Aspirin-exacerbated respiratory disease: NSAID challenge and desensitization", section on 'Challenge protocols and procedures'.)

BETA-ADRENERGIC RECEPTOR BLOCKERS — Beta-adrenergic receptor blockers ("beta-blockers") can exacerbate diseases of the airways (eg, chronic obstructive lung disease [COPD] and asthma) and the pulmonary vasculature (eg, portopulmonary hypertension). However, only rarely have they been associated with pleural or pulmonary parenchymal diseases, such as drug-induced lupus and interstitial pneumonitis.

Asthma and COPD — Beta-blockers are commonly used to treat hypertension, heart failure, and symptomatic coronary artery disease. However, beta-blocking medications that act on beta2 receptors can also cause bronchoconstriction. Since beta-adrenergic receptors of large and small airway smooth muscle are primarily the beta2 subtype, nonselective beta1/beta2-blockers (eg, propranolol) are more likely to cause bronchoconstriction in susceptible individuals [34]. In contrast, selective beta1-blockers (eg, atenolol, metoprolol) have a 20-fold greater affinity for beta1 adrenergic receptors than beta2 adrenergic receptors and, therefore, are less likely to induce bronchoconstriction.

Studies of selective beta1-blockers are reassuring regarding their safety in patients with COPD, including those with a reversible component. In a meta-analysis that examined the effect of cardioselective beta-blockers given as a single dose or for longer duration, the beta1-blockers did not reduce the forced expiratory volume in one second (FEV₁) or increase respiratory symptoms compared with placebo, and beta1-blocker treatment did not reduce the improvement in FEV₁ following inhaled beta2-agonists [35]. A subgroup analysis revealed no change in results for those participants with severe chronic airways obstruction or for those with a reversible obstructive component.

The benefits of using beta blockers, like any other drug, must be weighed on a case-by-case basis against the risk of side effects. Nonselective beta-blockers should be avoided in patients with asthma and used with caution in patients with an exacerbation of COPD. The use of selective beta1-blockers in patients who have COPD and a cardiovascular indication is discussed separately. (See "Management of the patient with COPD and cardiovascular disease", section on 'Treatment of CVD in patients with COPD' and "Arrhythmias in COPD", section on 'Multifocal atrial tachycardia' and "Primary pharmacologic therapy for heart failure with reduced ejection fraction", section on 'Beta blocker'.)

Portopulmonary hypertension — Beta-blockers are used prophylactically to prevent variceal hemorrhage in patients with cirrhosis complicated by portal hypertension. However, in patients with cirrhosis and portopulmonary hypertension, withdrawal of beta-blockers improved exercise capacity and decreased pulmonary vascular resistance, likely due to heart rate improvements causing increased cardiac output [36]. This indirect evidence suggests that beta-blockers should be used cautiously or avoided in those with symptomatic portopulmonary hypertension [37]. (See "Primary prevention of bleeding from esophageal varices in patients with cirrhosis", section on 'Preventive strategies' and "Portopulmonary hypertension" and "Prevention of recurrent bleeding from esophageal varices in patients with cirrhosis".)

Drug-induced lupus — The development of anti-nuclear antibodies is not uncommon with beta-blocker use, but the incidence of a symptomatic lupus syndrome attributed to a beta-blocker is distinctly less common (table 2) [38]. Beta-blocker induced lupus syndrome with pleuritis and pneumonitis has rarely been reported [39]. (See "Drug-induced lupus".)

Interstitial lung disease — Organizing pneumonia and eosinophilic pneumonia have rarely been reported with beta-blocking agents, such as sotalol and acebutolol [40-42]. As an example, migratory radiographic opacities were described in a patient on sotalol [40]. On lung biopsy, features of both organizing pneumonia and eosinophilic pneumonia were seen. Partial improvement was noted with addition of systemic glucocorticoids; complete recovery was seen only after sotalol was stopped. (See "Cryptogenic organizing pneumonia" and "Idiopathic acute eosinophilic pneumonia".)

HYDRALAZINE — Hydralazine, a vasodilating agent, is associated with drug-induced lupus (pleural and pericardial effusions), antineutrophil cytoplasmic antibody positive-pulmonary vasculitis, and diffuse alveolar hemorrhage [43]. (See "Drug-induced lupus", section on 'Causative drugs' and "Drug-induced lupus", section on 'Clinical spectrum of drug-induced lupus'.)

MINOXIDIL — Minoxidil is uncommonly used, mainly to treat recalcitrant hypertension. Fluid retention is a potential adverse effect of the drug. Minoxidil has been associated with the development of pericardial and pleural effusions, which may be exudative [44,45].

NITRATES — Overdoses of nitrate medications, such as intravenous infusions of nitroglycerin or nitroprusside, can cause methemoglobinemia. The clinical presentation may include dyspnea, respiratory depression, cyanosis, lethargy, altered consciousness, hypotension, and seizures. Measurement of oxygen saturation by a pulse oximeter may be inaccurate for assessing oxygen saturation as severe methemoglobinemia causes the SpO₂ to trend towards 85 percent and thus may either overestimate or underestimate the true arterial oxygen saturation (SaO₂) as measured by arterial blood gas analysis [46]. Methemoglobinemia is suspected when the arterial tension of oxygen (PaO₂) is normal despite clinical cyanosis. The diagnosis is based upon laboratory measurement of methemoglobin. (See "Methemoglobinemia", section on 'Evaluation and diagnosis (acquired/toxic)'.)

PACLITAXEL-ELUTING STENTS — Paclitaxel is an antineoplastic agent that can cause an interstitial pneumonitis when used to treat cancer. Paclitaxel has also been used to prevent restenosis after placement of coronary artery drug-eluting stents (DES), and case reports have described interstitial pneumonitis associated with these stents [47,48]. The onset of dyspnea, fever, and progressive respiratory insufficiency occurred within days of stent placement. Chest radiographs showed bilateral diffuse opacities. Due to the rarity of these events, the optimal management is not known. The three reported patients succumbed to progressive respiratory failure despite systemic glucocorticoid therapy [47,48]. Paclitaxel DES are no longer used frequently due to the better performance of newer generation DES. (See "Taxane-induced pulmonary toxicity" and "Intracoronary stents: Stent types", section on 'Early-generation drug-eluting stents' and "Periprocedural complications of percutaneous coronary intervention", section on 'Hypersensitivity reactions'.)

PROCAINAMIDE — The antiarrhythmic agent procainamide is associated with drug-induced lupus, antiphospholipid antibody syndrome, interstitial pneumonitis, and respiratory muscle weakness. A number of extrapulmonary adverse effects are also associated with use of procainamide, including nonspecific systemic symptoms, blood dyscrasias, and cardiac toxicity (see "Major side effects of class I antiarrhythmic drugs"). Use of procainamide is decreasing with availability of more effective antiarrhythmics such as amiodarone.

●Drug-induced lupus – Procainamide can cause a syndrome of drug-induced lupus with protean clinical manifestations, including fever, arthralgia, rashes, myositis, vasculitis, serositis, and Raynaud phenomenon (table 2) [49,50]. Respiratory system involvement includes pleuritis (pleural effusions and pleurisy) and, less commonly, diffuse parenchymal lung disease. (See "Drug-induced lupus".)

Among patients with procainamide-induced lupus, pleuritis occurs in approximately half. While a high pleural fluid antinuclear antibody (ANA) may be seen in drug-induced lupus pleuritis, it lacks specificity and does not differentiate systemic lupus erythematosus (SLE) from drug-induced lupus. There are, however, a number of features that help to distinguish drug-induced lupus from SLE, including the general absence of renal and central nervous system disease with drug-induced lupus and the presence of anti-double stranded DNA antibodies and hypocomplementemia with SLE (table 2). The clinical manifestations and diagnosis of drug-induced lupus are discussed separately. (See "Drug-induced lupus", section on 'Diagnostic approach' and "Pleural fluid analysis in adults with a pleural effusion".)

One sensitive serologic indicator of drug-induced lupus is the presence of antibodies directed against the histone complex H2A-H2B [51]; the absence of this autoantibody essentially rules out drug-induced lupus. However, approximately 60 to 80 percent of patients with active spontaneous SLE also have this antibody. Thus, the presence of anti-histone antibodies does not discriminate between idiopathic SLE and drug-induced lupus. (See "Drug-induced lupus".)

●Antiphospholipid antibodies – Antiphospholipid antibodies are also associated with procainamide, although the antiphospholipid antibody syndrome due to procainamide is rare [52,53].

●Respiratory muscle weakness – Procainamide may rarely have adverse effects on respiratory muscles by several mechanisms [54]. First, it can competitively block the acetylcholine receptor, thereby impairing neuromuscular transmission and causing postoperative apnea and/or a myasthenia gravis-like syndrome. Hence, procainamide may induce a myasthenic crisis in patients with underlying autoimmune myasthenia gravis [55]. Second, myositis has also been reported in association with drug-induced lupus and can impair respiratory muscle function.

QUINIDINE — The antiarrhythmic agent quinidine has been associated with drug-induced lupus, respiratory muscle weakness, acute pneumonitis, and diffuse alveolar hemorrhage (DAH) in case reports [54,56,57]. The incidence of drug-induced lupus is much lower than with procainamide, which may be explained by quinidine's lack of an amine group [58]. Similar to procainamide, the pulmonary manifestations associated with quinidine-induced lupus are primarily related to pleuritis [58]. (See 'Drug-induced lupus' above.)

Quinidine can rarely cause a myasthenia gravis-like syndrome and has been shown to exacerbate autoimmune myasthenia gravis [54].

Two cases of DAH have been reported in patients with quinidine sulfate-induced thrombocytopenia [56,59].

STATINS — Various statins have been associated with interstitial lung disease (fibrotic, eosinophilic, nonspecific) in case reports [60-62]. Onset of lung disease was 1 week to 10 years after initiation of statin therapy. Radiographic findings included ground glass, consolidative, and reticular opacities. While a case control study of patients with idiopathic pulmonary fibrosis (IPF) did not find an increased risk of IPF associated with statin use [63], this observation does not exclude the possibility of a drug-induced pneumonitis due to a statin medication.

Conversely, a post hoc analysis of 624 patients randomized to placebo in three trials of pirfenidone in the treatment of IPF suggest that statins may have a beneficial effect on clinical outcomes in IPF [64]. Similarly, a potential beneficial effect on clinical outcomes has been found for patients taking a statin in trials of nintedanib for IPF [65].

A meta-analysis of randomized trials determining the clinical impact of statin therapy on patients with pulmonary hypertension secondary to lung diseases (group 3) suggested that statins might be safe and beneficial for patients with pulmonary arterial hypertension due to chronic lung diseases [66,67]. However, no prospective clinical trials have been performed to validate these findings.

TOCAINIDE — The anti-arrhythmic agent tocainide (no longer sold in the United States, limited availability elsewhere) has uncommonly been associated with an interstitial pneumonitis that develops after a few months of therapy [68-70]. Computed tomography (CT) typically shows septal thickening and consolidative opacities [69]. The lung disease may be characterized initially by a neutrophilic alveolitis with organizing pneumonia; irreversible fibrosis may develop with continuing inflammation.

In most cases, symptoms remit with discontinuation of therapy. However, progressive respiratory failure has been described in patients with advanced fibrosis despite discontinuation of tocainide [68]. Systemic glucocorticoids may hasten recovery in more severe disease.

SUMMARY AND RECOMMENDATIONS

●Overview – Several cardiovascular drugs can cause symptoms like dyspnea, cough, and radiographic abnormalities. However, diffuse parenchymal lung disease linked to these drugs is rare. Identified respiratory side effects include: upper airway angioedema or hematoma, bronchoconstriction, cough, interstitial pneumonitis, organizing pneumonia, eosinophilic pneumonia, drug-induced lupus, acute respiratory distress syndrome, diffuse alveolar hemorrhage, pleuritis, pleural effusion, and solitary lung mass (table 1A-B). (See 'Introduction' above.)

●Angiotensin-converting enzyme inhibitors – Angiotensin-converting enzyme (ACE) inhibitors are associated with cough, angioedema, and, rarely, pneumonitis. The cough is typically nonproductive, persistent, and often nocturnal (in 5 to 20 percent of patients) and frequently requires cessation of therapy. (See 'Angiotensin-converting enzyme inhibitors' above and "Major side effects of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers".)

●Antiplatelet agents, anticoagulants, and thrombolytics – Treatment of coronary artery disease with anti-platelet medications (glycoprotein IIB/IIIA antagonist), anticoagulants (vitamin K antagonists, direct thrombin inhibitor, factor Xa inhibitor), and thrombolytic agents has been associated with bland diffuse alveolar hemorrhage (DAH). Rare cases of upper airway obstruction due to retropharyngeal or glottic hematoma have been reported. Clopidogrel and ticlopidine have infrequently been associated with interstitial pneumonitis. (See 'Antiplatelet (non-aspirin), anticoagulant, and thrombolytic medications' above.)

●Aspirin – Aspirin causes acute bronchoconstriction in patients with aspirin-exacerbated respiratory disease (triad of asthma, nasal polyps, acute bronchoconstriction due to aspirin). The dose of aspirin that provokes a reaction varies among patients; some patients are able to tolerate 81 mg, but develop symptoms at 162 or 325 mg. Aspirin desensitization, a protocol that may also help improve asthma symptoms, is an option for patients with AERD who require aspirin therapy. (See 'Aspirin' above and "Aspirin-exacerbated respiratory disease" and "Aspirin-exacerbated respiratory disease: NSAID challenge and desensitization".)

●Beta-adrenergic receptor blockers – Nonselective beta1/beta2 blockers (eg, propranolol) can cause bronchoconstriction in susceptible individuals, but this effect is substantially less likely to occur with selective beta1 blockers (eg, atenolol, metoprolol). (See 'Beta-adrenergic receptor blockers' above and "Management of the patient with COPD and cardiovascular disease".)

●Cardiovascular drugs associated with interstitial pneumonias – Eosinophilic pneumonitis has rarely been found in patients taking ACE inhibitors, beta-blocker medications, and statins. Organizing pneumonia has been reported in association with amiodarone, beta-blocker medications, and tocainide. Statins, paclitaxel-eluting stents, and tocainide have been associated with pneumonitis in case reports. (See 'Amiodarone' above and 'Angiotensin-converting enzyme inhibitors' above and 'Beta-adrenergic receptor blockers' above and 'Statins' above and 'Tocainide' above.)

●Drug-induced lupus – Drug-induced lupus has been reported with beta-blocker medications, hydralazine, procainamide, and quinidine. The diagnosis of drug-induced lupus is based on the combination of clinical manifestations (eg, pleuropericarditis), serologic evaluation, and response to discontinuation of the implicated medication. Features that help to differentiate drug-induced lupus from systemic lupus erythematosus (SLE) are listed in the table (table 2). The presence of anti-histone antibodies in the absence of other autoantibodies (eg, anti-double stranded DNA, anti-ribonucleoprotein, anti-Smith) strongly favors drug-induced lupus. (See 'Drug-induced lupus' above and 'Hydralazine' above and 'Procainamide' above and 'Quinidine' above and "Drug-induced lupus".)

Although respiratory muscle weakness may arise from drug-induced lupus (table 2), procainamide and quinidine can also cause respiratory muscle weakness by unmasking or exacerbating underlying myasthenia gravis. (See 'Procainamide' above and 'Quinidine' above.)