Background/Purpose:
     Hydoxychloroquine (HCQ) is widely used for the treatment of rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) with a side-effect profile including myopathy and cardiotoxicity. Acute HCQ poisoning has been reported to cause a prolonged QT interval, hypokalaemia and a prolonged QRS complex. RA itself is an independent risk factor for cardiovascular (CV) disease and a recent study has shown that patients are 40% more likely to develop atrial fibrillation. Our objective was to determine if alterations in cardiac conductivity with HCQ is appreciated and whether patients should undergo monitoring ECGs.

Methods:

       A retrospective analysis of our electronic medical records was undertaken to determine whether patients had had an ECG prior to, and at least 6 months after, starting HCQ and whether there had been a resultant change in the QT interval. Only resting standard 12-lead ECGs with both QT and QTc interval values measured were used. A past history of cardiac disease, arrhythmias or other medications know to prolong the QT interval were documented. The threshold for diagnosis of long QT syndrome was a QTc interval of 450 ms.

Results:

       1537 patients currently taking HCQ were identified. 102 patients were found to have had ECGs before, and at least 6 months after, starting HCQ therapy. Of these only 19 patients had suitable ECGs for analysis. This comprised 16 females and 3 males, with a mean age of 62.5 years (range 19-87). CV risk factors included hypertension (n=7), obesity (n=5), ischaemic heart disease (n=4), deep vein thromboses (n=3) and cerebral vascular accidents (n=3). Known diagnoses of arrhythmias included atrial fibrillation (n=2), supraventricular tachycardia (n=1) and an unspecified arrhythmia (n=1). The patients had been on HCQ therapy for a mean of 3.6 years (range 1.3-9.2) and were receiving either 200 mg (n=4) or 400 mg (n=15) per day. The initial ECGs had a mean QTc interval of 424 ms (range 377-584). The post HCQ ECGs had a mean QTc interval of 449 ms (range 387-620). The mean change in QTc was 25 ms (range 66-143). Overall 4 patients had a long QTc prior and 8 patients after initiation of HCQ therapy (table 1). 8 patients were also taking ≥1 medication known to prolong the QT interval.

Conclusion :

        The appreciation of potential drug-induced arrhythmias in Rheumatology patients to date has not been well described. Our analysis showed a trend in prolongation of the QTc following treatment with HCQ. As we examined only a surrogate outcome, namely a change in QTc interval, we were unable to determine whether this was clinically relevant. It is difficult to distinguish whether the ECG changes observed were due to HCQ or due to other factors. Nevertheless as only a small number of patients had ECGs we have highlighted the under-recognition of this particular problem. In order to truly determine HCQ as a culprit, a prospective study is required in this area.

Table 1