Foot ischaemia and toe skin necrosis associated with interleukin-2 infusion in an HIV-infected patient

INTRODUCTION

The introduction of highly active antiretroviral therapy (HAART) has resulted in tremendous improvements in morbidity and mortality in HIV-infected patients. The initiation of HAART generally leads to a rapid reduction in HIV-1 RNA plasma levels. However, around 15% of patients experience a ‘discordant response’, whereby the HIV-1 RNA plasma level is below the limit of detection but the CD4+ cell count response is blunted. Another pattern of discordant response is characterized by a sustained CD4+ cell count response, despite persistent viraemia and to an increase in peripheral CD4+ cell counts. Discordant responses are independently associated with an increased risk of mortality. ^1,2 Intermittent interleukin-2 (IL-2) administration to HIV-infected patients leads to CD4(+) T-cell expansions. IL-2 leads to peripheral expansion of a population of long-lived CD4CD45ROCD25 cells that express high levels of foxP3 but exert weak suppressive function and play a role in the maintenance of a state of low turnover and sustained expansion of the CD4 T-cell pool. ³

IL-2 was effective in increasing CD4 cell counts without increasing viral load in HIV-positive patients with a baseline CD4 cell count of more than 200 CD4 cells per cubic millimeter. ⁴ Katlama et al. ⁵ reported that administration of IL-2 4.5 × 10⁶ IU subcutaneously twice daily for five days every six weeks produces significant and sustained increase in CD4 cell counts in HAART-treated patients with persistent CD4 cell counts <200 × 10⁶ cells/L. Recently, several studies confirmed that the addition of IL-2 to HAART can significantly expand CD4 T-cell counts in moderately advanced human immunodeficiency virus infection, without the loss of virologic control. ^6,7 Therefore, subcutaneous IL-2 has been conclusively shown to induce significant increases in CD4 cell counts in HIV-infected patients, in particular when given concomitantly with HAART. In Spain, the use of IL-2 in HIV-infected patients is for compassionate use. ⁸ The most common adverse effects associated with IL-2 in HIV patients have been mild constitutional symptoms and local erythema at the injection site. ⁹

We present a case of foot ischaemia and toe skin necrosis associated with IL-2 infusion in an HIV-infected patient. To our knowledge, there has been no previous mention of this association in the literature.

CASE REPORT

A 39-year-old white man who had been HIV positive for 13 years presented to the emergency room with symptoms of a severe foot ischaemia and toe skin necrosis. He had no history of drug allergy and illicit drugs abuse. The patient quit smoking in 1995. HIV infection discovered in 1995 was treated with zidovudine monotherapy. Two years later, the patient changed his antiretroviral therapy (ART) to stavudine and didanosine. Doses were in accordance with manufacturers' recommendations. After three years of therapy, his plasma HIV RNA viral load was 2247 copies/mL. In 2000, treatment including zidovudine, lamivudine and efavirenz was started with a low CD4 T-cell count (<100 cells/mm³) and plasma HIV RNA viral load under 5000 copies/mL. In 2001, he was commenced on abacavir, indinavir/ritonavir and efavirenz with similar CD4 T-cell count and plasma HIV RNA viral load. After two years, his HAART was changed to tenofovir, lopinavir/ritonavir and lamivudine. For this scheme there were no genotypic resistances. The patient adherence on ART was >95% since the commencement of this new HAART scheme; this was confirmed by outpatient pharmacy clinic's reports. Despite this treatment, his CD4 T-cell count was under 100 cells/mm³. In order to increase CD4 T-cell count, IL-2 (4.5 MIU twice a day, subcutaneously) for five days every two months was added to the treatment. After 10 courses of IL-2, the patient had responded appropriately to IL-2 with a CD4 T-cell count of >300 cells/mm³. We decided to continue IL-2 therapy for maintaining CD4 T-cell count increase. Two days after 12 courses of IL-2 was initiated, the patient noted a sudden pain and tenderness in left foot. He presented to our hospital with a severe left toe foot ischaemia. No abdominal pain, jaundice, vomiting, ulcers or fever were present. The HIV RNA level was persistently undetectable during IL-2 therapy. He had no history of substance abuse and denied the use of any other drugs or herbal remedies. He had no known general risk factors for thrombosis other than HIV infection. Physical examination showed a cyanotic bilateral violaceous discolouration of the left foot toe with a necrotic ulceration in the external foot side. He had cold left foot with normal peripheral pulses. Any other thromboembolic disorder was observed. The results of laboratory tests were normal. Serum lipid profile (total cholesterol, triglyceride, low-density lipoprotein, high-density lipoprotein) were within normal limits. The platelet count was also within normal limits. The prothrombin time, international normalized ratio and partial thromboplastin time were in the desired therapeutic range. Rheumatoid factor, antinuclear antibodies, cryoglobulins, anticardiolipin antibodies, antineutrophil cytoplasmic antibody (cytoplasmic pattern), antineutrophil cytoplasmic antibody (perinuclear pattern), protein S, protein C and antiglomerular basement membrane antibodies showed no significant abnormalities. There was no serological evidence of acute infection by parvovirus B19, hepatitis B or C virus, herpes simplex and cytomegalovirus. The gadolinium-enhanced magnetic resonance angiography of lower extremities did not show any signs of arteriosclerosis and vascular stenosis. An arterial Doppler ultrasound examination of the leg confirmed the permeability of the posterior tibial arterial and its digital pulse. A diagnosis of foot ischaemia and toe skin necrosis was made. The suspected causative agent was IL-2 since this was the only drug that the patient was taking before the symptoms appeared. In view of a possible amputation, the patient was empirically started on aspirin 100 mg/day p.o. and pentoxifylline 400 mg/8 hours p.o. in order to improve local microcirculation. We observed a satisfactory response with a quick resolve of skin lesions. During the next days, the symptoms subsided dramatically. He was discharged days seven after admission. The medication on discharge was tenofovir, lopinavir/ritonavir, lamivudine, pantoprazole 20 mg/day p.o., aspirin 100 mg/day p.o. and pentoxifylline 400 mg/8 hours p.o. At a follow-up visit days after discontinuing the drug, the man reported that his skin symptoms had improved, and reported complete resolution of the ischaemia and toe skin necrosis eight weeks after stopping IL-2.

DISCUSSION

IL-2 therapy increases the CD4 cell count in patients on ART, regardless of the baseline CD4 cell count; it is not associated with virological failure or with antiretroviral resistance. The frequency and severity of adverse events associated with IL-2 administration are dependent upon dose, route and administration schedule. One of the most severe adverse reactions associated with IL-2 is the capillary leak syndrome (CLS). CLS results in increased capillary permeability to protein and fluids and decreased vascular tone. A drop in mean arterial blood pressure occurs within two to 12 hours after beginning the treatment. With continued therapy, CLS can result in clinically significant hypotension, hypoperfusion, oliguria, oedema and adult respiratory distress syndrome. ⁹ CLS can contribute significantly to the development of ischaemia and confusion. ¹⁰ Hypovolemia may require administration of large amounts of intravenous fluids and low doses of dopamine can be used to maintain blood pressure. CLS have been associated with high-dose trials of IL-2. The incidence of CLS in patients receiving lower doses of IL-2 is rare. The regimens used in most current HIV and outpatient oncology IL-2 trials yield cardiovascular symptoms that are milder and less frequent than are those seen with high-dose therapy. Ischaemic reactions due to IL-2 are relatively rare. Skin necrosis associated with IL-2 is very unusual. An extensive review of the literature did not reveal cases of ischaemic reaction and skin necrosis in lower extremities with an induced by IL-2 therapy. Junghans et al. ¹¹ described a case of biventricular cardiac thrombosis during IL-2 infusion in a non-HIV patient. In HIV patients treated with IL-2 as a continuous infusion at a dosage of six to 18 million IU per day for five days every eight weeks during a period of seven to 25 months, any adverse ischaemic reaction was occurred. ¹² Currently, the most common dose of IL-2 in HIV patients is 4.5 mIU subcutaneous twice daily for IL-2 for five days every eight weeks. With this schedule, only 5% of patients developed grade 3 or worse adverse events; any of these were ischaemia adverse reactions. ⁶ Wang et al. reported intestinal infarction due to vascular catastrophe in an HIV-infected patient after the third dose of IL-2. Autopsy revealed severe atherosclerosis in the superior and inferior mesenteric arteries. The stenotic coeliac artery was occluded by a recent thrombus at the aortic ostium. The authors suggested a possible contribution of the IL-2 in the development of the severe thrombosis, although definitive proof is lacking. ¹³ Rarely, some cases of myocardial infarction and stroke related to IL-2 administration were observed. ^14–16

The use of animal models has allowed the elucidation of events leading to damage and/or dysfunction of normal tissues after IL-2 administration. These studies have shown that IL-2 induces platelet and neutrophil–endothelial adherence in the skeletal muscle microcirculation that is associated with the development of macromolecular leakage from the microcirculation. Platelet-derived microvascular thrombosis also contributes to the vascular toxicity of IL-2. ^17,18 Ellison et al. ¹⁹ reported that repeated systemic IL-2 infusion results in altered vasomotor responsiveness within the cerebral microcirculation. IL-2 administration causes pulmonary oedema secondary to an increase in pulmonary capillary pressure. This vasoconstriction is mediated by IL-2 stimulation of thromboxane A2 generation from the lung. ²⁰ Boulanger and Vanhoutte reported that IL-2 induces endothelium-dependent contractions to arachidonic acid in the Wistar–Kyoto aorta that are mediated by an augmented release of a metabolite of cyclo-oxygenase, different from thromboxane A2 but activating thromboxane-endoperoxide receptors. ²¹ Furthermore, treatment with IL-2 produced an increase in coronary perfusion pressure that correlated with the increases in circulating tumour necrosis factor-alpha. ²²

In our patient, the most possible cause of foot ischaemia and toe skin necrosis was considered to be IL-2 because of the temporal relationship between exposure to the drug and the onset of symptoms. IL-2 was the only drug administered before the foot ischaemia and toe skin necrosis developed. Our patient did receive antiretroviral drugs concomitantly, which should be considered an additional contributing factor in masking the ischaemic process; however, the patient was on treatment with these medications for three years before the adverse drug reaction occurred. Furthermore, nowadays the patient continues on lopinavir/ritonavir, tenofovir and lamivudine therapy. Other potential causes of foot ischaemia and toe skin necrosis were ruled out. Nevertheless, there was rapid resolution of the symptoms and signs after IL-2 was discontinued and acetyl salicylic acid and pentoxifylline therapy were started. Based on the Naranjo probability scale, IL-2 could be considered the probable cause of the foot ischaemia and toe skin necrosis. ²³

A possible explanation for this condition is that IL-2 infusion could produce a continuous vasospasm of digital microcirculation together with a microthrombosis that developed ischaemic lesions and skin necrosis. Furthermore, in the angiographic examination atheroscleroses lesions were not observed. Neither an alteration of other vessels nor laboratory abnormalities that suggested an ischaemic vasculitis, embolism or hypercoagulability state were observed. The administration of acetyl salicylic acid and pentoxifylline produced a huge improvement in the patient's clinical situation. Edwards et al. ²⁴ demonstrated that pentoxifylline inhibits alterations in the microvasculature associated with the IL-2 administration. Therefore, pentoxifylline could reduce the IL-2 systemic toxicity. In our case, treatment with pentoxifylline produced a clinical improvement on the patient's ischaemic lesions and he was discharged. Further studies should be conducted in order to confirm the role for pentoxifylline in attenuating the toxic effects of IL-2.

Based on our observations, we conclude that IL-2 was the probable offender in the development of foot ischaemia and toe skin necrosis. If clinical evaluation leads to the suspicion of ischaemic process, therapy with IL-2 should be discontinued immediately.