Interaction between Viagra (sildenafil) and HIV-1 combination therapy

The Lancet- Volume 353, Number 9169 12 June 1999 

We report a death that may have been caused by this interaction.

A 47-year-old man was admitted after 2 h of severe central chest pain that occurred while he was at rest. He had no history of chest pain. In 1995, he had a diagnosis of HIV-1 infection and had been taking highly active retroviral therapy (HAART) in the form of ritonavir 400 mg twice daily and saquinavir 400 mg twice daily for over 1 year. The therapy led to suppression of plasma HIV-1 viral load to below detectable levels and his CD4 cell count was 580x106/L.

The patient had erectile dysfunction for which no physical cause was identified and had been prescribed sildenafil 25 mg 12 weeks before presentation. He had taken it eight times with no untoward effects. He took a 25 mg dose 1 h before the onset of chest pain. There was no history of prescribed or illicit nitrate use. He smoked 30 cigarettes daily and had been a smoker for 30 years. There was no history of hypertension or diabetes mellitus and no family history of ischaemic heart disease. His most recent total cholesterol was 56 mmol/L and triglycerides 40 mmol/L. On examination he was clammy and in pain. His pulse was 92 and he was normotensive. Cardiovascular and respiratory examinations were normal. There was no clinical evidence of left-ventricular failure.

An electrocardiogram (ECG) showed an extensive anterolateral myocardial infarction, with 2-4 mm ST segment elevation in leads V1-V6, I and aVL. A chest radiograph was initially normal. Urea and electrolytes were normal. Total cholesterol was 71 mmol/L, triglycerides 17 mmol/L, and random glucose was 60 mmol/L. There were no contraindications to thrombolysis. Recombinant tissue plasminogen activator 100 mg was administered 145 min after the onset of chest pain. There were no clinical or ECG indications of reperfusion.

Initially he remained haemodynamically stable but after 24 h became hypotensive and developed clinical and radiographic evidence of pulmonary oedema. Echocardiography indicated an akinetic left-ventricular anterior wall, septum, apex, and lateral wall, with a subjective ejection fraction of 20-30%. There was no evidence of acute valvular regurgitation or of chamber rupture. Inotropic support, diuretic therapy, and non-invasive continuous positive-airway-pressure ventilatory support produced an initial improvement in his condition.

Infectious diseases and respiratory opinions were investigated. There were few clinical or radiographical findings to support a diagnosis of HIV-1-related pneumonia, but he was too ill to undergo bronchoscopy. He was, therefore, treated empirically with co-trimoxazole (septrin) and azithromycin. After a brief improvement, he became hypotensive and increasingly hypoxaemic. He was intubated and ventilated but had a cardiac arrest which led to his death.

This death has been reported via the "purple card scheme" to the Committee on Safety of Medicines, to Pfizer, and to the US Food and Drug Administration. We believe that caution should be exercised when treating erectile dysfunction with sildenafil in patients receiving protease inhibitor-based HAART until more data are available.

*Mark C S Hall, S Ahmad
*Department of Cardiology, and Centre for Sexual Health, Royal Bolton Hospital, Farnworth, Bolton, BL4 0JR, UK