Hydralazine-induced pericardial effusion

Abstract

Drug-induced lupus (DIL) is due to an autoimmune reaction to a drug with an estimated incidence of 15,000 to 30,000 cases every year in the US. Hydralazine is a well-known offender. Antinuclear antibody (ANA) is present in most cases, though four cases of ANA-negative DIL have been reported. In this report, we present another case of ANA-negative DIL secondary to hydralazine.

Drug-induced lupus (DIL) is a rare clinical syndrome due to an autoimmune response to a large group of medications. Numerous medications have been implicated since the first reported case in 1945,¹ and new medications continue to be added. Its association with hydralazine is well known. A positive antinuclear antibody (ANA) is the norm²; however, there are a few case reports of ANA-negative DIL.^3–6 In this report, we describe a case of pericardial effusion secondary to ANA-negative hydralazine-induced lupus.

CASE PRESENTATION

A 48-year-old man with known end-stage renal disease secondary to diabetic nephropathy and uncontrolled hypertension presented to the emergency department with the chief complaint of worsening dyspnea and lower-extremity edema for 2 days. He denied fever, joint pain, myalgias, and rash. His medications included aspirin 81 mg daily, pravastatin 40 mg daily, amlodipine 10 mg daily, carvedilol 25 mg twice daily, and hydralazine 50 mg three times a day. He had been on hemodialysis for 1 year with excellent outpatient hemodialysis compliance. In 2016, he was first noted to have a small pericardial effusion that had been stable, as demonstrated on multiple echocardiograms, the most recent of which was done 3 months earlier.

His blood pressure was 114/62 mm Hg; pulse, 79 beats/min; respiratory rate, 20 breaths/min; temperature, 97.5°F; and oxygen saturation, 100% in ambient air. The lungs were clear to auscultation. His jugular venous pressure was estimated to be 8 to 10 cm H₂O. His heart sounds were audible, with prominent S4 and no murmur. The liver was extended 4 to 5 cm below the right costal margin. There was 1+/4+ lower-extremity edema. The hemoglobin was 7 mg/dL (from a baseline of 9 mg/dL); blood urea nitrogen, 61 mg/dL; creatinine, 8.6 mg/dL; total bilirubin, 1.7 mg/dL; alkaline phosphatase, 278 mg/dL; albumin, 2.9 g/dL; total protein, 7.2 g/dL; serum iron, 59 μg/dL; percent saturation, 44%; ferritin, 2300 ng/mL; and thyroid-stimulating hormone, 1.24 IU/mL. White cell and platelet counts, serum electrolytes, and transaminases were normal, and troponin, serum markers for hepatitis B and hepatitis C infection, and interferon gamma release assay were negative. The erythrocyte sedimentation rate was significantly elevated at 140 mm/h. Antihistone antibody was positive, but ANA was negative. Other autoantibodies were not checked due to negative ANA. Chest x-ray showed an enlarged cardiac silhouette (Figure 1a). Transthoracic echocardiography revealed an ejection fraction of 65% and a large circumferential pericardial effusion without evidence of tamponade (Figure 1b).

Figure 1.

(a) Chest x-ray showing marked enlargement of the cardiac silhouette. (b) Echocardiography showing large pericardial effusion (arrow).

The patient underwent pericardial window with drainage of 1800 mL of bloody fluid; cultures and acid-fast bacilli stain of the fluid returned negative, ruling out infectious etiology. The patient did not have any symptoms suggesting a viral infection, and myocardial infarction was also ruled out. Uremic pericarditis was unlikely, because he was compliant with outpatient hemodialysis and had maintained adequate hemodialysis adequacy per current guidelines. The only possible explanation for the patient’s hemorrhagic pericardial effusion was hydralazine-induced lupus, especially with antihistone antibody positivity.

The patient had an uneventful postsurgical course and recovered well. Pericardial drainage gradually decreased over the next few days and chest tubes were subsequently removed. The patient had been on hydralazine 150  mg daily for many years. It was discontinued on admission, and he has had no recurrence of the effusion in the last 4 months.

DISCUSSION

Diagnosis of DIL is challenging and requires a high degree of clinical suspicion. Diagnosis of ANA-negative DIL is even more difficult due to its rarity. The diagnosis requires clinical vigilance, because more than 100 medications have been implicated. Sometimes the only clue is a temporal relationship between the offending drug and onset of symptoms. Once the diagnosis is established, the mainstay of treatment is prompt discontinuation of the offending medication and supportive care. A short course of corticosteroids has also been recommended in nonresolving cases.⁷

DIL is a drug-mediated immune reaction that leads to a clinical phenotype similar to idiopathic lupus. Up to 10% of lupus cases are drug induced, with an estimated incidence of 15,000 to 30,000 cases in the USA every year.^2,4 It usually presents after months or, as in this case, years of continuous exposure to an offending drug; however, it can also present after a few days or weeks, especially if rechallenged.² Of note, except for the large pericardial effusion, this patient had no other symptoms of systemic lupus.

Numerous medications have been implicated in DIL, with hydralazine (5%–8%) and procainamide (20%) having the highest risk.^8,9 Most cases have a positive ANA; in the literature, only four cases of ANA-negative DIL exist, with three cases from hydralazine^3–5 and one from lisinopril.⁶ Interestingly, all three cases of hydralazine-induced ANA-negative lupus presented with pericardial involvement.

The most frequent clinical presentations of DIL include arthralgia, arthritis, myalgia, fever, and weight loss; cutaneous involvement is less common compared to idiopathic lupus.^2,7 Life-threatening complications like pericardial tamponade^10,11 and vasculitis can also occur but are less common.¹² Borchers et al proposed the following diagnostic criteria: (1) continuing exposure to a specific drug, (2) at least one symptom compatible with lupus, (3) no history of lupus or symptoms suggesting of lupus prior to starting the drug, and (4) resolution of symptoms after discontinuation of the agent.² ANA positivity (>90%) is common, but a negative ANA should not preclude the diagnosis, particularly if a patient has other autoantibodies associated with lupus/DIL.² Associated risk factors include longer duration of therapy, higher cumulative dose, and slow acetylator status; some studies have suggested an association with certain human leukocyte antigens (HLA-DR2, HLA-DR3) and the C4B null complement allele.^2,9

This case demonstrates the diagnostic challenges of DIL. DIL should be suspected in patients with vague symptoms or features of vasculitis or pericardial effusion, especially if they are taking well-known high-risk medications.^8,9 A negative ANA should not exclude the diagnosis or cause a delay in discontinuation of the suspected drug.