Antinuclear Antibody--Negative, Drug-Induced Lupus Caused by Lisinopril.

ABSTRACT: The clinical symptoms of drug-induced lupus (DIL) are similar to those of idiopathic systemic lupus erythematosus. The literature indicates that in patients with DIL, sera generally contain antinuclear antibodies (ANAs); however, ANA-negative DIL has been reported. The list of medications implicated as etiologic agents in DIL continues to grow. This list includes two different types of angiotensin-converting enzyme inhibitors--captopril and enalapril. We report the first case of DIL caused by lisinopril. Our patient had negative results on ANA testing and had histone antibodies (IgG anti--[H2A-H2B]-DNA) mirroring the disease course. Antibodies to the (H2A-H2B)-DNA complex are seen in more than 90% of patients with active DIL, excluding those with DIL due to hydralazine. Thus, it is important to recognize the clinical significance of IgG anti--(H2A-H2B)-DNA antibodies and that negative ANA test results do not preclude the diagnosis of DIL.

THE CLINICAL SYMPTOMS of drug-induced lupus (DIL) are similar to those of idiopathic systemic lupus erythematosus (SLE); however, its course is usually more benign. Clinical features of DIL most often include arthralgias, arthritis, myalgias, fevers, hematologic dyscrasias, and serositis. In DIL, the renal, pulmonary, visceral, and central nervous systems are usually spared. (1) In 95% to 100% of patients with DIL, serum is positive for antinuclear antibody (ANA), which most often has a homogenous pattern. (1) Of patients with active DIL, more than 90% have IgG anti-[(H2A-H2B)-DNA antibodies, with the exception of patients with DIL due to hydralazine. (2) The key to the diagnosis is the temporal relationship with the offending drug and the resolution of signs and symptoms over the weeks after the withdrawal of that particular agent. While ANA-negative DIL is rare, it has been described. (3) We report a case of ANA-negative DIL caused by a medication that has not previously been implicated as a possible etiolo gic agent for DIL.

CASE REPORT

A 44-year-old woman with no significant medical history was found to have a slightly elevated blood pressure on two separate occasions and was prescribed oral lisinopril 10 mg daily. Approximately 2 months later, the patient began to have fatigue, low-grade fevers, myalgias, left-sided pleurisy, and slight dyspnea on exertion. A chest radiograph showed mild cardiomegaly, and a 2-dimensional echocardiogram revealed a small to moderate-sized pericardial effusion without evidence of tamponade. The patient had negative results on ANA and rheumatoid factor testing, as well as a normal level of thyroid-stimulating hormone with a slightly elevated sedimentation rate (22 mm/hr; normal, 0-20 mm/hr). Renal function and serum albumin were also normal. The etiology of symptoms was thought to be viral.

The dyspnea on exertion, left-sided pleurisy, and pericardial effusion worsened to the point that pericardiocentesis was required 1 month later. A viral etiology was now believed to be unlikely given the amount of time elapsed, and prednisone (20 mg daily) was begun for immune-mediated pericarditis of uncertain etiology. Results of another ANA test were negative at that point. Symptoms resolved, and the prednisone was tapered over the next 2 months. Approximately 2 weeks after prednisone was stopped, the symptoms slowly returned. The pericardial effusion recurred, along with a small, left-sided pleural effusion, and prednisone 20 mg daily was resumed.

Over the next 3 months, the pericardial effusion recurred anytime the prednisone dose dropped below 15 mg/day, necessitating a pericardial window. A workup at this time again showed negative results on ANA testing, with increased antihistone antibodies of 1.5 (negative, <0.8; borderline, 0.8-1.2; positive, >1.2). The type of histone antibody measured was IgG anti-(H2A-H2B)-DNA. This was the first time that histone antibodies were measured, because of the previously negative ANA test results. Testing for other autoantibodies, including anti-dsDNA, anti-SS-A, anti--SS-B, anti-Smith, and anti-RNP, yielded negative results. The C3 and C4 levels were also normal.

The only medication that the patient had been taking throughout this time was lisinopril. The lisinopril was discontinued, and the prednisone was slowly tapered off over 3 months. The IgG anti-(H2A-H2B)-DNA level returned to negative (<0.8 ratio) 3 months after lisinopril was discontinued. At this writing, the patient has had no prednisone for 2 months and has had no recurrence of symptoms.

DISCUSSION

Since DIL was first described about 50 years ago, more than 70 medications have been implicated as possible etiologic agents. (1) Included in this list are several antihypertensive medications, of which hydralazine is the best described and poses the most significant risk. Captopril and enalapril have been the only two angiotensin-converting enzyme inhibitors shown to induce lupus-like diseases. (1) However, Keung et al (4) reported that ramipril induced Evans' syndrome, an autoimmune destruction of red blood cells and platelets. The pathogenic mechanism of DIL and other druginduced autoimmune phenomena is unclear. Our patient is the first reported to have DIL due to lisinopril therapy.

Induction of ANAs by certain medications is common. In more than half the patients taking procainamide and slightly less than half taking hydralazine, ANAs will develop. However, a lupus-like disease occurs in only a small fraction of these patients. (5) Therefore, the presence of ANAs in no way implies disease. The converse, however, has generally been regarded to hold true for DIL; ie, that all patients with DIL will have a positive ANA test result. (6,7) Our patient had a negative result on ANA testing but had histone antibody that reflected the clinical course. It has also been described that some patients with DIL caused by quinidine or minocycline have serum that is ANA-negative. (3)

The type of histone antibody detected in our patient was IgG anti--(H2A-H2B)-DNA. This H2A-H2B antibody is seen in the majority of patients with SLE and in more than 90% of patients with active DIL, (2) with the exception of DIL caused by hydralazine, in which the antibody is directed against the Hi and H3-H4 complex. (8) In our patient, tests for other autoantibodies related to SLE all yielded negative results, the complement levels were normal, and the symptoms resolved after lisinopril was discontinued; these facts strongly argue against a diagnosis of idiopathic SLE. While the pericardial window could certainly explain resolution of the pericardial effusion, this would not explain the resolution of pleural effusion, fevers, myalgias, and fatigue.

With our ever-expanding armamentarium of medications, it is important to keep in mind the possibility of previously undescribed side effects, such as DIL. Our case stresses the importance of the occurrence of DIL in patients with a negative result on ANA testing. While this is rare, a negative ANA result does not preclude the diagnosis, particularly if the patient has IgG anti--(H2A-H2B)-DNA antibodies. In patients whose clinical symptoms fit DIL, the presence of IgG anti--(H2A-H2B)DNA antibodies can help identify the true etiology of the disease. In these cases, a simple discontinuance of a medication can alleviate all of the patient's symptoms.

References

(1.) Rich MW: Drug-induced lupus: the list of culprits grows. Postgrad Med 1996; 100:299-302,307-308

(2.) Burlingame RW, Rubin RL: Autoantibody to the nucleosome subunit (H2A-H2B)-DNA is an early and ubiquitous feature of lupus-like conditions. Mol Biol Rep 1996; 23:159-166

(3.) Rubin RL: Etiology and mechanisms of drug-induced lupus. Curr Opin Rheum, 1999; 11:357-363

(4.) Keung YK, Mallarino MC, Cobos E: Drug-induced Evans syndrome. Ann Intern Med 1998; 128:327

(5.) Yung RL, Richardson BC: Drug-induced lupus. Rheum Dis Clin North Am 1994; 20:61-86

(6.) Kelley NK, Harris ED Jr, Ruddy S, et al: Textbook of Rheumatology. Philadelphia, WB Saunders Co, 5th Ed, 1997, p 1019

(7.) Klippel JH, Dieppe PA: Rheumatology. London, CV Mosby Co, 2nd Ed, 1998, pp 7.36.4-5

(8.) Koopman WJ: Arthritis and Allied Conditions: A Textbook of Rheumatology. Baltimore, Williams & Wilkins Co, 13th Ed, 1997, pp 1335-1336

KEY POINTS

* While anti-nuclear antibodies (ANA) negative drug-induced lupus (DIL) is rare it has been described.

* Of patients with active DIL, more than 90% have IgG anti-[(H2A-H2B)-DNA] antibodies, with the exception of patients with DIL due to hydralazine.

* In our patient, tests for antibodies related to systemic lupus erythematosus all yielded negative results, the complement levels were normal, and the symptoms resolved after the lisinopril was discontinued; these facts strongly argue against a diagnosis of idiopathic lupus.

* A negative ANA result does not preclude the diagnosis of DIL, particularly if the patient has IgG anti-[(H2A-H2B)DNA] antibodies.

* Our patient is the first reported to have DIL due to lisinopril.