Diffuse Alveolar Hemorrhage Following Pirfenidone Initiation

Introduction

Diffuse alveolar hemorrhage (DAH) is bleeding into the alveolar space of the lungs, a condition that may be life-threatening. Several diseases, medications, and toxins can cause DAH. It generally presents as 1 of 3 characteristic patterns: vasculitis or capillaritis, bland pulmonary hemorrhage, or alveolar bleeding associated with another process or condition. Since DAH has numerous causes, the clinical presentation may reflect alveolar bleeding only or features of the underlying cause. Hemoptysis, diffuse alveolar infiltrates, anemia, and acute respiratory failure are often present although hemoptysis may be absent in up to one-third of patients at presentation. Reports of dyspnea, cough, and fever are common. Chest X-ray may demonstrate patchy or diffuse alveolar opacities while chest computed tomography (CT) can reveal areas of consolidation interspersed with areas of ground glass attenuation. Bronchoscopy is indicated in most patients to document DAH, rule out airway sources of bleeding, and exclude infection. Transbronchial biopsy is rarely necessary. When the diagnosis of DAH is confirmed, an underlying cause must be determined which may include obtaining serologic tests. Treatment of DAH may include systemic corticosteroids, supplemental oxygen, bronchodilators, and reversal of coagulopathy if present. ^1,2

Pirfenidone is a pyridone approved for the treatment of idiopathic pulmonary fibrosis (IPF). ³ During randomized, double-blind, placebo-controlled trials, pirfenidone significantly reduced decline in forced vital capacity (FVC), the total exhaled volume during a forceful and complete exhalation after a maximal inhalation. ^4,5 It is believed to exert its antifibrotic effect through scavenging reactive oxygen species and inhibiting profibrotic and inflammatory cytokines such as transforming growth factor-β. ⁶ The most commonly reported side effects are gastrointestinal including nausea, vomiting, dyspepsia, anorexia, and gastroesophageal reflux (GERD), and skin-related events consisting of rash and photosensitivity. ^{3 –5} Pirfenidone is known to have harmful effects on the liver; therefore, liver function tests including AST, ALT, and bilirubin should be monitored at baseline, monthly for 6 months, and every 3 months thereafter. ³ A postmarketing safety registry reflects side effects seen in published trials with gastrointestinal and dermatological concerns reported most commonly. ⁷ Pirfenidone should be administered with food to lessen gastrointestinal side effects and titrated to a target dose of 801 mg orally 3 times daily. ³

To our knowledge, there are no published reports of bleeding associated with pirfenidone therapy. The Food and Drug Administration (FDA) Adverse Events Reporting System reports 57 cases of hemoptysis and 3 cases of pulmonary alveolar hemorrhage with pirfenidone. ⁸ We present 3 cases of DAH in patients receiving pirfenidone for IPF.

Case 1

An 88-year-old female presented to the emergency department (ED) with increasing shortness of breath, cough, hemoptysis, and chest pain. Her medical history included IPF requiring 2 L/min supplemental oxygen, chronic obstructive pulmonary disease (COPD), hypertension, stroke, anxiety, arthritis, dementia, and depression. The patient reported no drug allergies and was a lifelong nonsmoker. Home medications included albuterol, amlodipine, ascorbic acid, aspirin, atorvastatin, benzonatate, cholecalciferol, guaifenesin, nitroglycerin, omeprazole, pirfenidone 801 mg 3 times daily, prednisone, promethazine, and senna.

On admission, the patient’s vital signs were respiratory rate 26 breaths/min, temperature 37.1°C, heart rate 108 beats/min, blood pressure 152/77 mm Hg, and oxygen saturation 79% on 4 L/min oxygen via nasal cannula. Laboratory analysis revealed WBC 13.2 k/µL, Hgb 10.6 g/dL, Hct 32.3%, PLT 235 k/µL, PT 10.3 seconds, INR 1, d-dimer 1.43 mg/L, procalcitonin (PCT) 0.09 ng/mL, BUN 15 mg/dL, SCr 0.92 mg/dL, AST 22 U/L, ALT 32 U/L, and bilirubin 0.4 mg/dL. Blood cultures, urine culture, and respiratory pathogen screen were negative. Chest X-ray showed interstitial prominence with focal opacity in the lateral left midlung. CT of the chest demonstrated chronic changes of usual interstitial pneumonia (UIP) with superimposed patchy alveolar infiltrates.

The patient’s home medications were held except for amlodipine, atorvastatin, and omeprazole. She was given ipratropium/albuterol every 6 hours for COPD, methylprednisolone 40 mg intravenously (IV) daily followed by prednisone 40 mg orally daily for possible component of IPF exacerbation, and ceftriaxone 1 g IV every 24 hours plus azithromycin 250 mg orally daily for potential severe community-acquired pneumonia. During previous evaluation of her pulmonary fibrosis, the patient reported no known exposures to environmental toxins or medication therapies that may have led to pulmonary fibrosis. A serological evaluation revealed elevated C-reactive protein, erythrocyte sedimentation rate (ESR), and antinuclear antibodies which are nonspecific markers; however, all other tests were negative leading to the diagnosis of IPF.

During the patient’s hospitalization, a pulmonary consult was obtained, and CT chest findings were felt to be most consistent with alveolar hemorrhage. A bronchoscopy was not performed to confirm the diagnosis. Pirfenidone therapy was held upon admission and discontinued indefinitely as the attending physician thought it may be responsible for the DAH. The patient’s hemoptysis resolved over the course of her hospitalization. At her follow-up visit, she was initiated on nintedanib therapy without recurrence of hemoptysis.

Case 2

A 75-year-old male presented to the ED with complaints of blood-tinged sputum for a few days. During his initial course, hemoptysis had increased to a teaspoon amount. He also endorsed cough, weakness, fatigue, shivering, decreased appetite, and increase shortness of breath. His medical history included IPF requiring 4 L/min supplemental oxygen, hyperlipidemia, degenerative joint disease, diabetes mellitus type 2 with neuropathy, depression, GERD, stroke, hypertension, and obstructive sleep apnea (OSA). He had no medication allergies and reported smoking cessation about 50 years ago. His home medications included acetaminophen, aspirin, atorvastatin, escitalopram, gabapentin, glipizide, loratadine, losartan, metoprolol tartrate, multivitamin, omeprazole, pirfenidone 801 mg 3 times daily, and warfarin.

On admission, the patient’s vital signs were respiratory rate 18 breaths/min, temperature 36.6°C, heart rate 95 beats/min, blood pressure 90/55 mm Hg, and oxygen saturation 90% on 4 L/min oxygen via nasal cannula. Laboratory analysis revealed WBC 16 k/µL, Hgb 11.4 g/dL, Hct 34.3%, PLT 207 k/µL, PT 18.2 seconds, INR 1.7, PCT 0.09 ng/mL, BUN 19 mg/dL, SCr 1.06 mg/dL, AST 18 U/L, ALT 21 U/L, and bilirubin 0.9 mg/dL. Blood cultures, sputum culture, and respiratory pathogen screen were negative. Chest CT demonstrated diffuse subpleural reticular pattern consistent with known UIP as well as superimposed left-sided alveolar infiltrates.

The patient’s home medications were continued except for warfarin and aspirin which were held due to hemoptysis, and he was initiated on levofloxacin 750 mg IV every 24 hours for pneumonia. During previous workup of his pulmonary fibrosis, the patient reported no known exposures to environmental toxins or medication therapies that may have led to pulmonary fibrosis. Prior serological evaluation revealed elevated ESR; however, additional serological tests were negative leading to the diagnosis of IPF.

During this hospitalization, a pulmonary consult was obtained. Due to persistent hemoptysis and increasing oxygen requirements, a bronchoscopy was performed on hospital day 2 which demonstrated evidence of alveolar hemorrhage. All bronchoscopic samples were sterile. Unfortunately, his respiratory status continued to worsen requiring transfer to the intensive care unit for bilevel positive airway pressure. On hospital day 4, a chest X-ray revealed increasing bilateral pulmonary consolidations and methylprednisolone 125 mg IV 3 times daily was initiated for possible inflammatory alveolitis and pirfenidone was discontinued. Despite extensive efforts, he did not improve, and his family opted to withdraw care. The patient passed away on hospital day 5.

Case 3

A 73-year-old male contacted his pulmonary case manager with complaints of increasing dyspnea with progressively bloodier sputum for 5 days. He denied any other symptoms. His medical history included IPF requiring 2 L/min supplemental oxygen, uncomplicated diabetes mellitus type 2, hypertension, and benign prostatic hypertrophy. He had no medication allergies and was a lifelong nonsmoker. His home medications included glipizide, lisinopril, multivitamin, and pirfenidone 801 mg 3 times daily.

On presentation to the ED, the patient’s vital signs were respiratory rate 26 breaths/min, temperature 37.3°C, heart rate 91 beats/min, blood pressure 101/59 mm Hg, and oxygen saturation 88% on 4 L/min oxygen via nasal cannula. Laboratory analysis revealed WBC 12.2 k/µL, Hgb 10.8 g/dL, Hct 32.1%, PLT 223 k/µL, PT 16.6 seconds, INR 1.3, PCT < 0.05 ng/mL, BUN 23 mg/dL, SCr 1.1 mg/dL, AST 35 U/L, ALT 51 U/L, and bilirubin 0.8 mg/dL. Microbiologic studies were negative. Compared to prior imaging, chest CT revealed unchanged findings of UIP with new-onset focal infiltrate.

The patient was admitted to an acute care bed, his home medications were continued, levofloxacin 750 mg IV every 24 hours was empirically initiated, and pulmonary consultation was requested. Bronchoscopic inspection and bronchoalveolar lavage were performed on hospital day 2 and revealed DAH; all bronchoscopic samples were sterile.

Following bronchoscopy, the patient’s clinical status worsened requiring transfer to a higher level of care for high flow oxygen at 60%. The patient’s pirfenidone was stopped and prednisone 60 mg orally daily was started for presumed drug-induced inflammatory alveolitis and DAH. Over the ensuing 10 days, his hemoptysis abated, supplemental oxygen needs returned to baseline, and prednisone was discontinued.

Discussion

Cellular proliferation, interstitial inflammation, and/or fibrosis within the alveolar wall characterizes interstitial lung disease (ILD). The main phenotype of ILD is interstitial fibrosis. Interstitial fibrosis is due to chronic pneumonitis, pulmonary sarcoidosis, underlying autoimmune disease, or idiopathic interstitial pneumonia (IIP) in most patients. IPF, a chronic, progressive, fibrotic ILD, is the most common IIP. ^9,10 The prevalence of IPF appears to be increasing and is 10 to 60 cases per 100 000 persons in the United States. ^11,12 Nonpharmacologic management strategies for IPF include supplemental oxygen to reduce exertional dyspnea and increase exercise tolerance, pulmonary rehabilitation to improve exercise capacity and health-related quality of life, and lung transplantation in highly selected candidates. ^9,10 Two pharmacologic agents, nintedanib and pirfenidone, are FDA-approved for use in IPF. ^4,5 The natural course of IPF is a gradual decline in lung function leading to respiratory failure and death. ^9,10 Among adults 65 years of age and older with IPF in the United States, the median survival is 3.8 years. ¹³ Complications and comorbidities associated with IPF include COPD, pulmonary hypertension, GERD, obesity, cardiovascular disease, lung cancer, venous thromboembolism, and OSA. ^9,10

Approximately 10% to 20% of IPF patients experience an acute exacerbation (AE) each year. ^9,14 An AE of IPF (AE-IPF) is defined as an acute, clinically significant respiratory deterioration characterized by new bilateral ground glass opacification or consolidation not otherwise explained by cardiac failure or fluid overload. ¹⁵ There are no proven, effective therapies for AE-IPF and management primarily includes supportive care focused on palliation of symptoms and correction of hypoxemia with supplemental oxygen. ^10,15 Despite a paucity of evidence, systemic corticosteroids are commonly used due to anecdotal reports of effectiveness and the high in-hospital mortality rate associated with AEs, which can exceed 50%. ^10,14,15 Up to 46% of IPF deaths are preceded by an AE, and the median survival following AE is 3 to 4 months. ^14,15 In one retrospective study of 52 patients with an AE-IPF, diffuse alveolar damage (DAD) and alveolar hemorrhage were found in 78.8% and 28.8% of patients, respectively. ¹⁶ Because DAD can lead to alveolar hemorrhage, it could be interpreted that AE-IPF may cause DAH. We did not feel our patients’ radiographic findings were consistent with the definition of AE-IPF; however, without biopsies, we cannot definitively say our patients did not have DAD.

The first patient presented with shortness of breath, cough, hemoptysis, tachypnea, and hypoxia. Although a confirmative bronchoscopy was not performed, the chest CT findings were felt to be consistent with DAH. She had undergone a previous workup for connective tissue, autoimmune, and vasculitic diseases which was negative. An echocardiogram was completed due to chest pain and slightly elevated troponin at admission; no mitral stenosis was found, and the left ventricular ejection fraction was 40% to 45%. Laboratory analysis was normal outside of leukocytosis, anemia, hypoalbuminemia, and elevated d-dimer. Infectious workup was negative including multiple PCT measurements, but a sputum culture was unable to be obtained. The patient was on aspirin 81 mg orally daily prior to admission, which was held during hospitalization, but resumed upon discharge. Pirfenidone was initiated with an appropriate upward titration in dose approximately 7 weeks prior to admission. It was held during her hospital stay and discontinued at discharge. Several interventions were initiated during her hospitalization including systemic corticosteroids, antibiotics, and discontinuation of aspirin making it difficult to determine which measure led to resolution of hemoptysis. At her follow-up visit, no further bleeding was reported, and she was initiated on nintedanib approximately 2 months following her hospitalization. Of note, she did present to the ED with bloody nasal drainage and blood-tinged sputum about 2 months after nintedanib initiation. She has subsequently been on nintedanib therapy for over a year with no further hospitalization or reports of bleeding. According to the Naranjo method, the likelihood of pirfenidone-induced DAH is possible (Table 1). ¹⁷

OPEN IN VIEWER

Table 1.

Naranjo Method for Estimating the Probability of Adverse Drug Reactions (ADRs). ¹⁶

Assessment Questions	Yes	No	Don’t Know	Case #1	Case #2	Case #3
Are there previous conclusive reports of ADR?	+1	0	0	0	0	0
Did ADR occur after drug was administered?	+2	−1	0	+2	+2	+2
Did ADR improve when drug was discontinued?	+1	0	0	+1	0	+1
Did ADR reappear when drug was readministered?	+2	−1	0	0	0	0
Are there alternative causes for the ADR?	−1	+2	0	−1	−1	−1
Did reaction reappear when placebo was given?	−1	+1	0	0	0	0
Was the drug detected in toxic concentrations?	+1	0	0	0	0	0
Was the reaction more severe when dose was increased, or less severe when dose was decreased?	+1	0	0	0	0	0
Did the patient have similar reaction to previous exposure?	+1	0	0	0	0	0
Was the ADR confirmed by any objective evidence?	+1	0	0	+1	+1	+1
Total Score				3	2	3

Abbreviation: ADRs, adverse drug reactions.

Naranjo et al developed this method to determine if an adverse drug event was caused by a drug or other factors. It classifies the reaction as doubtful < 0, possible 1-4, probable 5-8, or definite > 9. Bolded numbers apply to our patients’ cases.

The second patient presented with shortness of breath, cough, and hemoptysis which were investigated with bronchoscopy where DAH was confirmed by progressive reddening of serial aliquots of lavage fluid. He had undergone a previous workup in the outpatient setting for connective tissue, autoimmune, and vasculitic diseases which was negative. During this admission, antiglomerular basement membrane antibodies and anti-neutrophil cytoplasmic antibody (ANCA) were normal. The patient was on aspirin and warfarin at the time of admission, although his INR was subtherapeutic at admission and for 6 consecutive weeks prior to admit. He was given phytonadione 1 mg orally on hospital day 2. No echocardiogram was obtained during this admission, but a previous exam was normal. Laboratory analysis was normal outside of hyperglycemia, leukocytosis, anemia, and elevated PT/INR. Infectious workup was negative including PCT, but it should be noted that some of the cultures were obtained after initiating antibiotic therapy. Pirfenidone was initiated with an appropriate upward titration in dose approximately 4 months prior to admission. The patient was not initiated on systemic corticosteroid therapy until hospital day 4 at which point his pirfenidone was also discontinued. Given that the patient expired the next day, it is impossible to know if these interventions may have improved his hemoptysis. According to the Naranjo method, the likelihood of pirfenidone-induced DAH is possible (Table 1). ¹⁷

The third patient presented with increasing dyspnea and progressive increase in bloody sputum, which were investigated with bronchoscopy that revealed DAH through progressive reddening of serial aliquots of lavage fluid. He had a previous workup for connective tissue, autoimmune, and vasculitic diseases which was negative. During this hospitalization, his laboratory analysis and microbiologic workup were unremarkable. Cultures from the bronchoalveolar lavage were sterile although they were taken after initiation of antibiotics. Pirfenidone was initiated with an appropriate upward titration in dose 19 days prior to admission. Following the bronchoscopy, he clinically deteriorated with increasing oxygen requirements prompting initiation of systemic corticosteroid therapy and discontinuation of pirfenidone. Over the next 10 days, the patient’s hemoptysis resolved, his oxygen requirements returned to baseline, and he was discharged home. Given the fact that several interventions were initiated during his hospital course, it is difficult to determine which measure led to resolution of hemoptysis. At his follow-up visit, no further bleeding was reported, and he was initiated on nintedanib. He has been on nintedanib therapy for over a year with no further hospitalizations or reports of hemoptysis. According to the Naranjo method, the likelihood of pirfenidone-induced DAH is possible (Table 1). ¹⁷

Conclusion

To our knowledge, DAH has not been attributed to pirfenidone. The clinical presentation of all patients was consistent with DAH and confirmed by bronchoscopy in 2 of 3 patients. Upon discovery of DAH, appropriate treatment was initiated including systemic corticosteroids and discontinuation of pirfenidone. Two of 3 patients recovered with these measures; however, it is difficult to determine which interventions led to improvement and 1 case was complicated by concomitant discontinuation of aspirin. Both patients were initiated on nintedanib at their follow-up visits. One patient experienced an episode of blood-tinged sputum related to epistaxis after initiation of nintedanib, but this resolved, and she continues to tolerate nintedanib. Considering these cases, further investigation is warranted into this potential reaction.

With the increasing incidence of IPF and the aging population, the use of pirfenidone will likely rise. Because pirfenidone has been shown to decrease disease progression, patients may remain on therapy for longer periods of time. Increased and prolonged usage provides a greater responsibility for health-care professionals to carefully monitor for and report potential side effects to the medical community. It is vital that prescribers and patients are aware of the current safety concerns regarding this agent.