A case of severe thrombocytopenia after the first exposure to rifampicin

Abstract

Severe immune thrombocytopenia is a rare side-effect of rifampicin (RFP) and can be life-threatening. Here, we report the case of a 74-year-old male with tuberculous pleurisy who developed severe thrombocytopenia after first exposure to RFP. Platelet count decreased to 1 × 10³/μL after 7 days of treatment with RFP, isoniazid, ethambutol, and pyrazinamide. After all the drugs were discontinued, the platelet count recovered. As thrombocytopenia did not occur after re-administration of drugs other than RFP, the patient was diagnosed with RFP-induced thrombocytopenia. Clinicians should be aware that RFP can induce acute and severe thrombocytopenia even without previous exposure to this drug.

List of abbreviations

DITP

drug-induced immune thrombocytopenia

EB

ethambutol

INH

isoniazid

IVIG

intravenous immunoglobulin

PAIgG

platelet-associated immunoglobulin G

PZA

pyrazinamide

RFP

rifampicin

1. Introduction

Tuberculosis requires treatment with multiple drugs with potential side-effects, including skin rash, loss of appetite, and liver toxicity. Although uncommon, the anti-tuberculosis drug rifampicin (RFP) may cause thrombocytopenia via immune reactions [1]. RFP-induced thrombocytopenia is more likely to occur in patients treated intermittently than in those treated continually [1,2]. In addition, thrombocytopenia is more severe in patients previously exposed to this drug [[3], [4], [5]]. Here, we report a case of acute and severe thrombocytopenia during continual treatment of RFP in a patient not exposed to RFP previously.

2. Case report

A 74-year-old man visited our hospital with dyspnea and a low-grade fever that had persisted for two weeks. He had no medical history of tuberculosis or use of anti-tuberculosis drugs. Chest radiography revealed a right pleural effusion (Fig. 1), and thoracentesis was performed. The pleural fluid was yellow, and laboratory findings were as follows: total cell count, >100 cells per high power field; lymphocytes, 79.2%; lactate dehydrogenase, 500 U/L; total protein, 4.9 g/dL; adenosine deaminase, 92.9 U/L. Furthermore, the interferon-gamma release assay results were positive. Thus, the patient was diagnosed with tuberculous pleurisy, and treatment with RFP (450 mg/day), isoniazid (INH, 300 mg/day), ethambutol (EB, 500 mg/day), and pyrazinamide (PZA, 1.2 g/day) was initiated.

Fig. 1.

Chest radiograph on admission showing pleural effusion in the right lung.

Seven days after treatment initiation, the patient visited our hospital because of oral bleeding. Physical examination revealed mild oral bleeding and multiple purpuric lesions in his legs. The platelet count had decreased from 217 × 10³/μL at baseline to 1 × 10³/μL. Other laboratory findings were as follows: white blood cell count, 5300 cells/μL; hemoglobin, 11.4 g/dL; prothrombin time, 10.9 s; fibrinogen, 513 mg/dL; D-dimer, 10.7 μg/mL; C-reactive protein, 3.51 mg/dL; and platelet-associated immunoglobulin G (PAIgG), 286 ng/10⁷ cells (normal value, less than 46 ng/10⁷ cells). Liver and renal functions were normal. Markers for hepatitis B and C virus and cytomegalovirus were negative. Thrombocytopenia due to the effect of anti-tuberculosis drugs was suspected, and the four drugs were discontinued. Although platelet transfusion of 20 units was done on days one and two after treatment discontinuation, platelet count increased only up to 16 × 10³/μL (Fig. 2). Since the bleeding subsided after the platelet transfusion and the platelet count gradually improved, we followed up until the side effects disappeared. The results of the drug-induced lymphocyte stimulation test were negative for the four drugs. On day 12 after discontinuation, levofloxacin was introduced, and EB, PZA, and INH (in this order) were reintroduced gradually. Since thrombocytopenia did not occur after restarting these drugs, he was diagnosed with rifampicin-induced thrombocytopenia. The patient was able to continue treatment since then, and the right pleural effusion disappeared.

Fig. 2.

The course of platelet count and treatment. EB: ethambutol, INH: isoniazid, LVFX: levofloxacin, PZA: pyrazinamide, RFP: rifampicin.

3. Discussion

Severe thrombocytopenia is a rare side effect of RFP [1] and can be life-threatening, especially in patients previously treated with this drug [[3], [4], [5]]. This study describes a case of acute and severe thrombocytopenia induced by RFP in a patient who had no previous exposure to RFP.

The two main mechanisms of drug-associated thrombocytopenia are the suppression of platelet production in the bone marrow, and increased platelet clearance in the peripheral blood via an immune response involving drug-dependent antibodies [6]. The second response, designated as drug-induced immune thrombocytopenia (DITP), can be severe and life-threatening [6]. RFP can cause DITP, and previous studies have indicated that RFP-dependent antibodies target the glycoprotein Ib/IX and/or IIb/IIIA complex on the platelet surface [[7], [8], [9]]. Anti-platelet antibodies are useful for diagnosing DITP. However, few specialized diagnostic centers have the infrastructure to quantify these antibodies. Although we did not quantify drug-dependent platelet antibodies, the acuteness of the course of the disease suggested an involvement of immune mechanism. In addition, high PAIgG levels and low platelet count despite platelet transfusion suggested platelet clearance.

Serum antibodies against RFP have been shown to increase after treatment discontinuation [2]. Continuous treatment with RFP is considered to lead to the neutralization of any antibodies formed and the continuous clearance of antigen–antibody complexes [10,11]. Therefore, the discontinuation of RFP might allow the antibody level to be built up, and the re-administration of RFP in this condition might cause more severe thrombocytopenia compared with that after the initial administration. In this respect, Mori et al. [3] reported that the platelet count decreased to 128 × 10³/μL two weeks after initiation of RFP and decreased further to 3 × 10³/μL the day after reintroduction. Mauricio et al. [4] reported that the platelet count decreased to 2 × 10³/μL six days after RFP reintroduction, although thrombocytopenia did not occur after administering the first dose. Conversely, our patient showed acute and severe thrombocytopenia despite not being previously exposed to RFP. There are some reports of severe thrombocytopenia occurring early, after initial exposure to RFP in patients with tuberculosis [[11], [12], [13], [14], [15]], the time to onset being 4–14 days, and the platelet count at the time of diagnosis 5–17 × 10³/μL. Thrombocytopenia in our patient was more severe than reported previously. The present case shows that RFP can lead to fairly severe thrombocytopenia even with first administration of the drug. Predisposing genetic or environmental factors for severe thrombocytopenia induced by RFP have not been identified yet, and hence we need to be careful with administration of RFP for all patients.

Generally the platelet count recovers within a week after discontinuation of exposure to the sensitizing agent [16]; however, severe cases may require other treatments, including corticosteroids and intravenous immunoglobulin (IVIG) therapy [6,17]. In our case, the bleeding stopped after platelet transfusion, and the platelet count gradually increased after discontinuation of the RFP. As in this case, several cases showed prompt improvement after discontinuation of RFP [[11], [12], [13], [14]]. Early detection of thrombocytopenia in patients being administered RFP, and its prompt discontinuation is important.

4. Conclusion

We have reported a case of severe thrombocytopenia after the first exposure to RFP. Clinicians should be aware that RFP can induce acute and severe thrombocytopenia even without previous exposure to this drug.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.