Long QT Syndrome Leading to Multiple Cardiac Arrests After Posaconazole Administration in an Immune-Compromised Patient with Sepsis: An Unusual Case Report

Background

Posaconazole is the newest orally administered triazole antifungal agent with an extended spectrum of activity [1]. It is indicated for the prophylaxis of invasive Aspergillus and Candida infections in severely immunocompromised patients, for the treatment of oral candidiasis (mainly in HIV+ patients), including cases refractory to itraconazole and fluconazole. It also seems to be a promising antifungal agent for the treatment and prophylaxis of invasive fungal infections (IFIs) [1–3]. Finally, it has favorable clinical outcomes against fungal infections of the central nervous system (CNS) [4]. The drug inhibits lanosterol 14α-demethylase enzyme and ergosterol production, it is well absorbed, and demonstrates dose-proportional pharmacokinetics following oral administration, which are not related to age, sex, or race/ethnicity [5]. Its half-life fluctuates between 20 and 35 h [6,7], and it also has prolonged and extensive peripheral tissue distribution [7]. Posaconazole inhibits the hepatic CYP3A4 enzyme, but compared to other azoles has less interactions with the P450 system, as its only significant interaction is the inhibition of CYP3A4 and it has no effect on the other CYP isoenzymes (1A2, 2C8, 2C9, 2D6, and 2E1), in contrast to voriconazole, which is both a substrate and inhibitor of the CYP3A4, CYP2C9, and CYP2C19 enzymes [1,3,7]. Elimination occurs primarily through fecal excretion (77%) and to a lesser degree through urinary excretion (14%) [3,7]. Studies of long-term posaconazole administration (6–12 months) showed that it is a well-tolerated and safe drug, with minimal adverse effects [6–8. Adverse effects like nausea (8%) and vomiting (6%) are the most frequently reported, long QT syndrome occurred in 1%, and elevated hepatic enzymes in 2% [1].

Torsades de pointes (TdP) is a potentially fatal polymorphic ventricular tachyarrhythmia, related to QT interval prolongation and it may present with palpitations, seizures, syncope, and sudden death, or may be asymptomatic if the duration is relatively short and terminates spontaneously.

Electrocardiographic long QT interval syndrome (normal men QTc ≤0.44 s, women QTc ≤0.45 to 0.46 s) is categorized as either congenital or acquired. Congenital long QT syndrome is defined as occurrence of genetic mutations affecting Na+, K+, and Ca2+ channels. The following related syndromes are described in the literature: Romano-Ward syndrome, Anderson syndrome, Timothy syndrome, and the Jervell and Lange Nielsen syndrome. Acquired long QT syndrome is most commonly drug-related [9] Numerous drugs (antiarrhythmic, antimicrobials, antifungals, antipsychotics, antidepressants, serotonin agonists of triptan class, and antihistamines) have an effect on Ikr, an important K+ channel subtype that participates in ventricular repolarization, causing prolongation of QT interval [9,10] (Table 1).

We report the case of an immunocompromised female patient with sepsis, whose treatment with posaconazole was complicated with severe cardiac events.

Case Report

A 26-year-old woman of Albanian origin was admitted to the Orthopedic Department of our hospital for a scheduled surgical removal of ectopic calcifications in the left hip joint. Her past medical history consisted of a caesarean section at the age of 23 followed by a prolonged hospitalization due to abdominal sepsis and multiple plastic surgeries to reconstruct the abdominal wall. No significant cardiac history was reported, and a recent cardiac echocardiography report, as well as the ECG report on admission, were normal.

She had an uneventful operation; however, on postoperative day 2 she developed sepsis due to surgical wound infection. Intra- and post-operative administration of cefamandole was replaced by meropenem and daptomycin, taking into consideration local susceptibility patterns, but the clinical response was poor.

At that point, the persistent high temperature, the previous medical history of continuous multiple infections, and the inadequate response to antibiotics raised the suspicion of a possible immunodeficiency, which was confirmed by low levels of gamma globulins in electrophoresis and low proportion of B and T lymphocytes in flux cytometry.

Daptomycin was replaced by linezolid, and colistin and amikacin were added to the antibiotic regimen. Furthermore, considering the immunocompromised status, liposomal amphotericin B was added to cover possible fungal infection despite negative cultures, resulting in clinical improvement. Nine days later, due to refractory hypokalemia (K+=2.6mEq/L) resistant to external K+ administration, a decision was made to replace liposomal amphotericin B with voriconazole.

After discontinuation of amphotericin B, plasma potassium levels were restored but the patient relapsed due to a new septic episode. Results from blood cultures at that time were positive for multi-resistant Acinetobacter baumannii and modifications were made according to the antibiogram. Tigecycline, doripenem, colistin, and oral posaconazole were administered combined with intravenous immune globulin (IVIG) every 5 days. Moreover, the patient underwent through debridement of the surgical wound daily, and finally the infection begun to recede after 3 days.

Voriconazole was administered for a total of 9 days and was then replaced with oral posaconazole. Thirty-six hours after posaconazole intake, the patient had episodes of bradycardia accompanied by QT prolongation in ECG (Figure 1), while plasma electrolyte levels were normal. Finally, she had an episode of cardiac arrest resolved with cardiopulmonary resuscitation and defibrillation. She was transferred to the cardiology ICU for monitoring and at that time she had continuous incidents of cardiac arrest due to polymorphic ventricular tachycardia (torsade des pointes) (Figure 2) that degenerated to lethal ventricular fibrillation (VF). Posaconazole was immediately halted and a temporary pacemaker was placed due to the continuous VF incidents (Figure 3). The patient was treated in our cardiology ICU for 4 days (Figure 4) and finally was transferred to the orthopedic clinic to resume therapy. She was discharged clinically well and planned for further investigation in the hematology outpatient clinic.

Discussion

Only a few cases of antifungal azole-induced long QT interval syndrome followed by TdP have been reported in the literature, but the responsible drug in all of them was voriconazole.

The first one, published by Alkan et al. in 2004, referred to a 15-year-old girl with acute lymphoblastic leukemia and Fusarium infection, treated with voriconazole. The first episode of QTc interval prolongation followed by TdP was noted 3 weeks after voriconazole initiation. At that time, the patient had a combination of severe risk factors (mild hypokalemia K+: 3.2mEq/lt) and potentially arrhythmogenic co-medication. In that case, QT prolongation was concentration-independent and recurred upon re-challenge with a very small voriconazole dose in the absence of other known pro-arrhythmogenic factors, a fact that strongly suggests a causal relationship between long QT syndrome and voriconazole exposure. Moreover, during the first cardiac event, voriconazole plasma concentrations were in the normal range, and during the re-challenge were very low [11]. The second case was reported by Elbey in 2012 and involved a female patient with a past medical history of nephrotic syndrome and rheumatic mitral valve disease, who was hospitalized and treated for fever. She developed sinus tachycardia and a prolonged QT interval of 580 ms when amphotericin-B was changed to voriconazole [12]. The third case, published in 2007 by Eiden et al., involved a 14-year-old Tahitian girl with acute myeloid leukemia and a suspected mucormycosis, who was treated first with voriconazole and caspofungin and later was changed to posaconazole due to the worsening fungal infection. Immediately after the switch to posaconazole, the patient developed bradycardia and long QT interval. The preceding TdP episode was attributed primarily to voriconazole, not posaconazole, because: 1) voriconazole plasma levels were measured 15 h after the last administration and reached levels of 7 mg/lt, instead of the normal rate of 0.6–1.5 mg/lt; and 2) the patient vomited 30 min after the oral posaconazole administration, thus it was assumed that little of the drug had been absorbed [13].

In our case, the causative factor that probably was responsible for the QT interval prolongation and episodes of TdP seems to be posaconazole, either by itself or combined with voriconazole administration. We tend to believe so, despite the fact that the patient had been treated with voriconazole for 9 days, because bradycardia and cardiac arrest occurred 36 h after the first dose of posaconazole and approximately 48 h after the last voriconazole dose. Given the fact that posaconazole half-life reaches 20–35 h [7], whereas voriconazole has an estimated half-life of 9 h, presumably, after 48 h the voriconazole plasma levels would be minimal or insignificant; therefore, posaconazole seems more likely to be responsible for the QT interval prolongation. The patient also kept having continuous incidents of cardiac arrest due to polymorphic ventricular tachycardia (TdP) that degenerated to ventricular fibrillation even when she entered the Cardiology ICU, and these stopped immediately after posaconazole was removed from the regimen.

Risk factors for long QT interval syndrome are well described [10,14–16]. Predisposing risk factors identified in our patient were: female sex, antifungal azoles (voriconazole and posaconazole), drug interactions, and arrhythmogenic drugs (metoclopramide and antifungals). Female sex has a predominance in the development of QT interval prolongation and associated arrhythmias, which is attributed to differences in specific cardiac ion densities [12]. Studies show that posaconazole is a well-tolerated drug [6,8,14], and even after a 6-month to 1-year administration, long QT syndrome prevalence is 1% [8]. In cases of congenital long QT syndrome, posaconazole acts synergistically, leading to a higher incidence of long QT interval and TdP.

Furthermore, all known azoles (fluconazole, ketoconazole, itraconazole, voriconazole, and posaconazole) may prolong QT interval directly by blocking Ikr channels, as well as by delaying CYP450-dependent metabolism of other drugs that prolong QT interval [17]. The possible interaction between voriconazole and posaconazole must also be take into account, as both the Tahitian girl with leukemia [13] and our patient had episodes of bradycardia and finally cardiac arrest after shifting from voriconazole to posaconazole. This suggests an unknown interaction between these 2 antifungal drugs that enhances the incidence of long QT prolongation and TdP. A drug-drug interaction cannot be excluded according to the mechanisms of the hepatic metabolism of both drugs. Serious interactions may occur in cases when drugs like posaconazole and voriconazole, whose metabolism is related to the cytochrome P450 enzymic system, are concomitantly administered. The enzyme CYP3A4 is polymorphically expressed, and its allelic variants have less functional phenotypic implications compared to the alleles of the CYP2C9 and CYP2C19 enzymes, indicating a clinical polymorphism [9]. There is diminished clearance, increased concentration, and prolonged exposure of the poorly metabolized medications via the CYP2C9 and CYP2C19 enzymes. Therefore, patients who express the phenotype of poor metabolism of drugs via the CYP pathway are more likely to develop higher incidence of TdP cardiac arrhythmia. Based on these aforementioned mechanisms, since posaconazole is metabolized via CYP3A4 and voriconazole via CYP3A4, CYP2C9, and CYP2C19, a possible drug interaction could affect the clearance and, finally, plasma concentrations of both drugs [1,18].

In our patient, the co-administered drugs were tigecycline, doripenem, colistin, intravenous immune globulin, lansoprazole, and metoclopramide. Among these drugs, only metoclopramide has been related to cardiac adverse effects such as supraventricular tachycardia, atrial-ventricular block, and QT prolongation, but it was not administered simultaneously with posaconazole. Moreover, our patient was treated with lansoprazole during the cardiac incident. Lansoprazole may augment plasma concentrations of drugs metabolized by CYP3A4 (such as posaconazole and voriconazole).

We believe that our case report provides a meaningful addition to the literature on use of antifungal agents and severe cardiac events, like prolonged QT syndrome and development of lethal cardiac arrhythmias. Although our institution lacked financial resources to measure antifungal agent plasma levels, we think that the cardiac events in our case may have been due to posaconazole administration or to a posaconazole–voriconazole interaction, as no other related drug or clinical conditions related to these incidents existed.

We believe this is the first published report identifying posaconazole as being solely responsible for a patient’s cardiac events. Regarding the use of posaconazole, an analysis from 428 patients with refractory invasive fungal infection or febrile neutropenia who received long-term posaconazole treatment showed a generally favorable safety profile of the medication. The results from this phase II/III open-label clinical trial indicated that long-term therapy with posaconazole did not increase the risk of any individual adverse event, and no unique adverse event was observed with longer exposure to posaconazole [8].

Conclusions

Although posaconazole is considered a safe and effective antifungal medication, its administration may be complicated by occurrence of long QT syndrome and associated lethal arrhythmias. Close cardiac monitoring and modification of the coexisting risk factors predisposing to cardiac incidents are indicated when posaconazole is administered. Drug interactions should also be taken into account when therapeutic decisions are made, because concomitant administration of medications that have metabolism related to cytochrome P450 may contribute to these cardiac incidents.