1. IntroductionThe severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that causes the coronavirus disease (COVID-19) has infected more than 610 million people worldwide and is responsible, until today, for more than 6.5 million deaths since its outbreak in China in 2019 [1,2]. Although it is expected that the accumulation of mutations and the currently emerging SARS-CoV-2 variants will likely lead to reduced mortality rates and transform the pandemic into its endemic phase, COVID-19 is still pressing the healthcare systems [3,4,5,6]. The most frequently associated factors for severe illness and hospitalization from COVID-19 are cardiovascular diseases, obesity, dyslipidemias, diabetes, respiratory diseases, and cancer along with aging and lack of vaccination [7,8]. Generally, people with more than one underlying health condition are at increased risk for hospitalization and acute disease [9,10,11]. Hence, COVID-19 is a common underlying condition for so many different clinical cases of patients with comorbidities [12,13]. As medical teams in COVID-19 wards try to follow efficient medication protocols for optimal healthcare provision for COVID-19 hospitalized patients, they should also adjust to each patient’s needs in order to avoid clinically significant drug-drug interactions (DDIs) from co-medications that can cause complications, adverse drug reactions (ADRs), and prolong hospitalization [14].DDIs refer to modulations of the pharmacological profile of a drug from co-administered ones. These alterations can be related to pharmacokinetic (PK) processes of absorption, distribution, metabolism, and elimination (ADME), such as modulation of metabolic enzymes (i.e., cytochrome P450, CYPs) or transporter proteins (P-glycoprotein, P-gp; organic anion transporters, OATP) and plasma proteins (i.e., albumin). DDIs can also be related to pharmacodynamic mechanisms in the site of action and/or in other tissues [15]. Clinically significant DDIs may result in ADRs and side effects that further impair patients’ health, obscuring the treatment outcome and prolonging hospitalization [16,17]. In the case of COVID-19, the scientific community gave an early warning of the risk for DDIs from the several applied protocols [14,18,19,20,21,22,23,24]. Even with the introduction of nirmatrelvir/ritonavir (PaxlovidTM), there were regulatory check lists to assist clinicians in evaluating potential DDIs and other patient factors prior to any administration [25]. In addition, for recently approved monoclonal antibodies such as casirivimab/imdevimab (REGEN-COV) and sotrovimab (Xevudy), no official DDI studies have been performed [26,27]. They are not renally eliminated or metabolized by CYP enzymes; hence, DDIs are unlikely with drugs that are substrates, inducers, or inhibitors of CYP enzymes or excreted through the kidneys. However, caution is advised, and healthcare providers should be aware of this new field of DDIs in COVID-19 if any observation should be reported, (e.g., interactions with COVID-19 vaccinations) in terms of pharmacovigilance. In a recent review, we described the risk for potential DDIs with drugs introduced for COVID-19 for patients with respiratory disorders and presented underlying pharmacological mechanisms, their significance, and possible clinical symptoms that could be recognized by healthcare teams staffing the COVID-19 wards [18].

Advancing our approaches, the aim of this work was to record and analyze the occurrence of DDIs in COVID-19 patients hospitalized over the previous months in the University Hospital of Heraklion in Greece. The study analyzes the prevalence of DDIs among the medications administered to those patients, their clinical significance, and their potential impact on hospitalization.

4. DiscussionHealthcare provision for patients with SARS-CoV-2 infection has proven to be a very complicated issue that will bring transformative changes in how healthcare is provided, especially in intensive care units [33,34]. The pervasiveness of the disease creates numerous and complicated clinical scenarios for COVID-19 patients with chronic diseases and complex therapeutic schemes, which are linked with an increased risk of adverse clinical outcomes [35,36]. One of the main risk factors for ADRs is clinically significant DDIs. This work presented the results of an observational study regarding the prevalence of DDIs among 125 patients hospitalized in the COVID-19 department of the University Hospital of Heraklion in Greece. Comparably to previously published works regarding the clinical characteristics of COVID-19 patients, the patients in this study had an average number of four comorbidities: cardiovascular disease (i.e., coronary artery disease, cardiomyopathies, hypertension, etc.); diabetes (types I and II); obesity; dyslipidemias; respiratory disorders; CNS disorders; cancer (occurred most often) (Figure 1) [12,37,38,39]. Furthermore, an additional factor related to prolonged hospitalization was the increased CRP values (34.4% of the cases). CRP has also been proposed as a prognostic indicator for the assessment of disease severity in COVID-19 [40].Potentially interacting drug pairs were found in 67.2% of patients during admission, 92.8% during hospitalization, and 60% upon discharge. Hospitalization’s high prevalence of interacting drug pairs is mostly due to the co-administration of LMWH with corticosteroids (i.e., dexamethasone), which may increase INR, a potential DDI of moderate significance. Nevertheless, the use of LMHW for anticoagulation and alleviation of inflammation mechanisms, along with dexamethasone’s effect in reducing ARDs risk, represents a prominent treatment for COVID-19 worldwide, endorsed by many medical societies’ guidelines and with a lot of evidence to support it [41,42,43,44]. Thus, careful administration of these medications should be performed while taking into consideration the underlying medical conditions, COVID-19 disease severity, as well as dexamethasone’s side effects of hyperglycemia, hypernatremia, hypertension, and potentiation of anticoagulant effects [41,45].Pharmacological mechanisms of DDIs were related to pharmacokinetic processes in 32% of cases, and the rest (68% of cases) were related to pharmacodynamic pathways. PK-DDIs were mostly related to inhibition of CYP-mediated metabolism from perpetrator drugs, which may alter patients’ systemic drug concentrations, thus modulating its pharmacological action (Table 3 and Table 4). For example, inhibition of clopidogrel’s CYP2C19-mediated metabolism by PPIs (e.g., esomeprazole, omeprazole) may result in reduced concentrations of clopidogrel’s active metabolite and reduced antiplatelet action, whereas the co-administration of escitalopram and PPIs may lead to elevated concentrations and thus enhanced escitalopram’s pharmacological action [46,47]. PD-DDIs were mostly associated with synergistic effects of drugs that may potentiate pharmacological outcomes or increase the risk for side effects (i.e., potentiation of the anticoagulation action of antithrombotic agents or drugs that contribute to QT prolongation) [45,48].Clinically significant DDIs of “Serious-Use alternative” or “Use with caution-Monitor” management were found in 40.3% of cases upon admission, 21% during hospitalization, and 40.7% upon discharge. The increased number of clinically significant DDIs upon admission is in line with previous observations regarding polypharmacy and the occurrence of DDIs among outpatient prescriptions in Greece [49,50]. Clinically significant DDIs at the time of admission were also associated with a longer hospital stay (p48]. On the other hand, a reduction in the clinical significance of DDIs during hospitalization was observed compared to admission. This can be attributed to the fact that specialized and multidisciplinary healthcare teams in medical wards have additional clinical information such as laboratory values and a full medication list, which allows them to be more compliant with evidence-based clinical guidelines and proceed to a better risk/benefit analysis with fewer medication errors [29,51,52,53].The occurrence of DDIs was higher for COVID-19 patients in a polypharmacy state (≥5 drugs). Polypharmacy is a documented risk factor for adverse drug reactions from clinically significant DDIs, especially in clinically ill patients [29,30,54]. Regarding COVID-19, polypharmacy and comorbidities have been described as risk factors for clinically significant DDIs early in the first wave, with the main concern being drug combinations that increase the risk for QT prolongation [14,35,55]. Previous studies have shown an increased prevalence of DDIs in COVD-19 patients undergoing treatment with lopinavir or ritonavir, and recognized risk factors include polypharmacy, age over 65, respiratory or CNS disorders, and dyslipidemias [56]. In our work, lopinavir and ritonavir were not administered according to the medical protocols that were followed. However, risk factors such as polypharmacy and comorbidities were also correlated with the high prevalence of DDIs (Figure 5). In addition, QT prolongation (Table 3) was the second most common potential outcome among the PD-DDIs due to the co-administration of antiarrhythmics (ATC-C01), antibacterials (ATC-J01), and especially azithromycin, CNS medications (ATC-N05, ATC-N06), and drugs for obstructive airway diseases (ATC-R03). In other studies, hypoglycemia and QT prolongation have been reported to be the most common predicted outcomes of DDIs, with risk factors being polypharmacy and comorbidities, whereas a correlation between the average number of drugs and the number of medications was observed [57]. In our study, the risk for hypoglycemia was less frequent (Table 3), but again, we observed an association between the number of drugs and DDIs (Figure 5). Overall, treating physicians in COVID-19 wards should be aware of potential PK-DDIs when antithrombotic agents (ATC-B01), antibiotics (ATC-J01), and antivirals (ATC-J05), as well as PPIs (ATC-A02), are co-administered along with cardiovascular medications (ATC-C). Regarding PD-DDIs, they should consider potential DDIs if arrhythmias occur due to the co-administration of drugs that prolong the QT interval. In addition, if INR values are modulated, it can be related to the co-administration of antithrombotic agents (ATC-B01), antibacterial agents (ATC-J01), corticosteroids (ATC-H02), drugs for obstructive airway diseases (ATC-R03), and psychoanaleptics (ATC-N06). They should also be aware of potential synergistic PD-DDIs that can be associated with the co-administration of CNS drugs (ATC-N) (Table 4 and Figure 4) [14,21,35,55,58,59].Until today, several risk score calculators have been developed to aid clinical decisions for COVID-19 patients [60,61]. Usually, these approaches take into consideration the clinical characteristics and risk factors that, until today, have been recognized for COVID-19. These patients are usually under polypharmacy conditions and thus at risk for ADRs from DDIs. This is a factor that should always be considered and incorporated into risk score calculators. Previous research has demonstrated that COVID-19 patients have a high prevalence of DDIs, especially those on polypharmacy [57,62]. This was evident also in this study, which estimated an exponential correlation between the number of medications and the average number of DDIs. On the subject of the apparent association between clinically significant DDIs and prolonged hospitalization, the current study cannot provide a causality conclusion as to whether DDIs actually prolong hospitalization or occur due to complex treatment regimens in prolonged hospitalized patients. Hence, prospective studies are needed to further clarify the possible causality of the observation for COVID-19 patients [16,17,63]. On the other hand, the reduction in the clinical significance of DDIs reveals that specialized medical teams can reduce the burden of DDIs among hospitalized patients and improve the healthcare provided. Overall, clinical signs, such as excessive or reduced drug action, modulation of INR, QT prolongation, and Torsades de Pointes (TdP), changes in electrolytes (K+, Na+), hyper/hypoglycemia, sedation, respiratory depression, and muscle, kidney, or liver dysfunctions, should be approached from a DDIs point of view for hospitalized COVID-19 patients, similar to those described in other works and in this study.Limitations of the study that can be mentioned are the relatively condensed sample size and that it took place in one hospital. However, the analysis of the data revealed a good correlation between the clinical characteristics of this cohort and the occurrence of DDIs compared with the literature for COVID-19 [64].