Here, we briefly summarize the common disease courses. Both had a history of complex CHD and severe IAI simultaneously when they tested positive for SARS-CoV-2 with a high viral load. Before the administration of remdesivir, blood pressure and blood oxygen saturation were maintained. Shock occurred within 3 h after remdesivir infusion and manifested as high CVP, indicating cardiogenic shock in addition to sepsis. Poor heart contractility, high lactatemia, and shock did not respond well to the standard supportive care. In addition, after the administration of remdesivir, both patients had ECG abnormalities, that is, QRS widening. The second patient developed a left posterior fascicular block.
Based on the electrophysical changes, clinical course, and temporal correlation between the onset of shock and administration of remdesivir, we hypothesized that remdesivir may have impair cardiac function at high viral loads.
As shown in Fig. 6a, SARS-CoV-2 can inhibit the mitochondrial antiviral signaling pathway in the early stage of infection and prevent host cells from producing type 1 and type 3 interferon, thus delaying the initiation of the antiviral immune response [4, 5]. SARS-CoV-2 then rapidly proliferates. SARS-CoV-2 RNA can interfere and hijack mitochondrial biogenetic machinery after entry into mitochondrial matrix from cytosol [5, 6]. The mitochondrial redox potential then decreases, leading to the assembly of the mitochondrial permeability transition pore (MPTP). Mitochondrial substances are released into the cytoplasm as potential damage-associated molecular patterns (DAMPs). DAMPs trigger and intensify cytokine storms by binding to pattern-recognition receptors (PRRs) [7]. By selectively removing dysfunctional mitochondria, mitophagy is the key to mitochondrial quality control. A study showed that SARS-CoV-2 can prevent the formation of autophagosomes and block mitophagy [8, 9]. As a result, the infected cells are filled with tiny and dysfunctional mitochondria. Under this premise, remdesivir potentiates mitochondrial dysfunction and leads to cytotoxicity. Remdesivir does not, like anti-HIV nucleoside and nucleotide analogs, directly inhibit mitochondrial DNA polymerase. Remdesivir has weak inhibitory activity toward mitochondrial RNA polymerase [10, 11] and therefore does not affect mitochondrial protein homeostasis. Remdesivir, instead, is reported to be a selective, partial agonist for urotensin-II receptor (UTS2R). The half-maximal effective concentration (pEC50) of remdesivir was estimated to be 4.89 ± 0.03 (EC50 = 13 ± 0.9 µM), with the working range of agonistic effects starting at 1 µM based on the response curve [12]. UTS2R, as a GPCR, in turn activates heterotrimeric G proteins, which leads to dissociation of Gα and Gβγ subunit complexes [13]. Possibly through impaired regulation of gene expression or trafficking of ERG potassium channels, field potential, which correlates closely with the QT interval, is prolonged [12]. Remdesivir can also disturb the electrophysiological properties by reducing the spontaneous firing rate, which may lead to disruption in conduction system and development of ventricular premature complex, and the diastolic depolarization rate, which may lead to QRS widening [3]. In addition, through binding to UTS2R, remdesivir can activate AKT/ERK axis [12]. Phosphorylation of ERK can in turn phosphorylate mitochondrial fission dynamic-like protein 1 (Drp1) [14]. Thus, Remdesivir induces mitochondrial fragmentation and dysfunction [3]. Such effect is not tissue-specific [15] and is reversed by Mdivi-1, an inhibitor of Drp1 [3]. Augmented mitochondrial dysfunction can manifest as elevated lactic acid levels, cytokine storm, and poor cardiac contractility, as seen in the two cases above.
What makes the hypothesis more convincing is that the pharmacokinetics of remdesivir are well fitted to the temporal correlation between the onset of shock and the administration of remdesivir. Remdesivir undergoes sequential hydrolysis to the nucleoside metabolite GS-441524, which is transmitted intracellularly and converted into the active metabolite remdesivir triphosphate. Remdesivir has a half-life of one hour. Following a single dose of 200 mg of remdesivir administered, the area under the concentration-time curve from 0 to 24 hours (AUC0 − 24) is 4.8 µM•h for remdesivir and 7.7 µM•h for the nucleoside metabolite [16]. Remdesivir triphosphate has a half-life of up to 43 h [11, 17]. Mitochondrial dynamics occur on a timescale of minutes [18]. The administration of remdesivir can rapidly halt SARS-CoV-2 proliferation and facilitate the resolution of systemic inflammation. Peak concentration of remdesivir is reached within two to three hours after the start of remdesivir infusion. In these two pediatric patients who received usual pediatric dosing of remdesivir, cardiogenic shock occurred two to three hours later, coinciding temporally with the changes in remdesivir concentration. Additionally, based on the AUC0 − 24 and the working range of UTS2R agonistic effects of remdesivir, we can anticipate that the effect of remdesivir inducing myocardial suppression through the UTS2R signaling pathway can last for at least one day or more. Therefore, the suddenness of onset of shock and the protracted nature of progression observed in the two cases above were possibly due to side effects of remdesivir instead of actual clinical course of COVID-19 or concomitant bacterial septicemia.
Moreover, certain CHD can lead to ventricular overload, as shown in Fig. 6b. Under such chronic stress, the high mitochondrial turnover rate ensures that the quality of mitochondria meets the needs of myocardial metabolism [19]. Infection with SARS-CoV-2 disrupts compensatory mechanisms [20]. Thus, cardiomyocytes become increasingly susceptible to remdesivir-induced “metabolic shock.”
Several studies have also shown that SARS-CoV-2 can cause elevated pulmonary vascular resistance (PVR) [21–23]. In patients with right heart failure or single ventricle circulation, elevated PVR causes RV afterload and is detrimental to single-ventricle circulation. It is reasonable to assume that the shock in these two patients was caused by intra-abdominal infection; co-infection with SARS-CoV-2 and the administration of remdesivir caused severe myocardial dysfunction due to mitochondrial shock. In summary, Remdesivir should be used with caution in CHD patients with RV failure and single-ventricle circulation.
We have admitted that no causal relationship can be directly established with the two cases. More evidence from clinical trials and basic research is required to prove our hypothesis.