4. Discussion
This study showed that the COVID-19 protocols of the largest hospitals in Andalusia (Spain) varied widely in several aspects, including antiviral treatment: they used different dosages, durations, and combinations of treatment for the same clinical condition. Corticosteroid and anakinra regimens varied among hospitals, whereas tocilizumab was uniform across all protocols. Although anticoagulant treatment was recommended in all protocols, including prophylactic heparin during hospitalization for all patients and therapeutic heparin for patients with evidence of venous thromboembolism, only a few protocols included higher prophylactic doses for patients at high risk of thrombosis and therapeutic doses for patients in critical condition or who experience a progressive increase in D-dimer levels. On the other hand, all treatment protocols included the same patient groups, defined according to the clinical situation, symptomatic treatment and recommendations for infection prevention and control, supportive care and hospital organization, equivalent empirical antibacterial agents, drug warnings and interactions, and discharge recommendations.
The clinical presentation of the novel COVID-19 may vary from mild cases with fever, fatigue, and cough to moderate–severe cases involving pneumonia and multiorgan failure [10], and several factors may predispose COVID-19 patients to adverse outcomes. The older age, critical disease, and high levels of inflammatory markers have been associated with increased risk of death [11]. Patients with moderate to severe forms of COVID-19 normally require hospitalizations and pharmacological treatments. However, currently, there is no solid evidence that any potential drug improves outcomes in patients with suspected or confirmed COVID-19 [12,13]. Only clinical experience and treatment guidance based on repurposed and experimental treatments are available.
Several antiviral agents have been proposed for use in combatting COVID-19 based on apparent in vitro activity [3]. One of the most widely used agents has been hydroxychloroquine and chloroquine. Although studies from China [14] and France [15] have reported improved radiologic findings, improved viral clearance, and reduced disease progression with the use of chloroquine and hydroxychloroquine compared to standard supportive care, their efficacy remains inconclusive, and further studies are warranted. In addition, it seems that hydroxychloroquine may act synergistically in combination with azithromycin [3,15]. Recently, in a large observational study from Italy, hydroxychloroquine was associated with a 30% lower risk of death in COVID-19 hospitalized patients [16]. Despite these results, these studies had severe methodological limitations, such as lack of randomization, lack of covariate-adjusted analysis, and potential selection bias. RECOVERY trial—the largest randomized controlled study on hydroxychloroquine—has suggested that hydroxychloroquine might not reduce deaths and might increase length of hospital stay [17]. Chloroquine dosage has been 500 mg orally once or twice daily. However, hydroxychloroquine dosage recommendations have varied from a total daily dose of 400 mg (with a loading dose of 400 mg twice daily for 1 day) to 600 mg orally, based on safety and clinical experience for other diseases [3]. With this limited evidence, international guidelines with consensus statements on the treatment of COVID-19 have not included any recommendation about using hydroxychloroquine/chloroquine as a potential treatment [4,18,19,20], but rather have only suggested its use in the context of a clinical trial [21,22]. In our study, monotherapy with 400 mg hydroxychloroquine twice daily the first day followed by 200 mg twice daily from day 2 to day 5, alone and in combination with azithromycin, were the preferred recommended regimens among protocols in Andalusia for treating mild respiratory illness with clinical risk factors. For mild pneumonia, dual therapies consisting of hydroxychloroquine-azithromycin or hydroxychloroquine-lopinavir/ritonavir were the most frequent treatments recommended, with regimens that included different dosages and durations of treatment. For moderate pneumonia, a shorter regimen of hydroxychloroquine in combination with a longer regimen of lopinavir/ritonavir was the most recommended treatment among all protocols. For the most severe form of pneumonia, the triple therapy with a longer regimen of hydroxychloroquine and lopinavir/ritonavir in combination with azithromycin was the most commonly indicated treatment. Thus, more complex and longer antiviral therapies were recommended according to the severity of COVID-19 in our protocols.
Lopinavir/ritonavir has had the same limitations of use as hydroxychloroquine/chloroquine, with no clear benefits beyond standard care [3,23]. Considering the uncertainty and the likely increase in gastrointestinal side effects, principal guidelines around the world have not contemplated its use in patients with COVID-19 [4,18,19,20] or have only considered it in the context of a clinical trial [21,22]. Remdesivir was only considered when patients were enrolled in a clinical trial. Randomized controlled trials have reported that remdesivir reduces the duration of mechanical ventilation and time to symptom resolution, but its effect on mortality and other adverse outcomes remains uncertain [24]. Other antivirals were not routinely included in our protocols. Antiviral agents such as oseltamivir, umifenovir, favipiravir, or ribavirin as well as miscellaneous agents such as interferon-α or interferon-β are being studied for use as possible treatments for COVID-19 [3]. Currently, principal guidelines have either not reported any recommendation for their use [4,18,19,20,21] or have suggested not using them in patients with COVID-19 [22].
In regard to management of ARDS with evidence of cytokine release syndrome, several adjunctive therapies have been proposed. The rationale for the use of corticosteroids is to reduce host inflammatory responses in the lungs. The potential harm, such as delayed viral clearance, increased risk of secondary infection, and hyperglycemia episodes, have led to caution in their routine use in COVID-19 patients unless the patient had other conditions for which these are indicated [3,4,18,19,20,21]. However, in observational and randomized studies focused on hospitalized patients with COVID-19 pneumonia, the administration of corticosteroids reduced the risk of mortality [25,26,27]. Based on this fact, a recently published guideline suggested using 40 mg methylprednisolone intravenously for 10 days in patients with severe COVID-19 and ARDS [22]. In our study, corticosteroids were included in 87.7% of the protocols. Methylprednisolone was the most frequently included corticosteroid in the protocols (13/15 protocols) with high-dose regimens (≥1 mg/kg/day) and a variable duration of treatment (between 3 and 10 days). High-dose dexamethasone (20 or 40 mg) was included in 53.3% of the protocols and prednisone in just one protocol (40 mg per day for 5 days). Monoclonal antibodies targeting key inflammatory citokines or other aspects of the innate immune response are another potential treatment for COVID-19 [28]. Tocilizumab, an anti-interleukin 6 receptor antibody, has been used in small studies with early reports of success. A dose of 400 mg was associated with clinical improvements and successful discharge, with most patients only receiving a single dose [29,30]. Several randomized clinical trials on tocilizumab, either alone or in combination, in patients with COVID-19 with severe pneumonia are underway, and its use was included in the Chinese national treatment guidelines [30,31]. Likewise, tocilizumab was recommended in all protocols in Andalusian hospitals; the protocols indicate a single dose of 400 mg or 600 mg, according to body weight, and a second dose 12-24 h later, if required. Although tocilizumab is a promising therapy, the data currently available are still too limited to draw any conclusion about its viability. Anakinra, an interleukin 1 receptor antagonist, has also been proposed in order to reduce hyperinflammation and respiratory distress in patients with SARS-CoV-2 infection, but here, too, limited evidence has been published [32]. In this study, less than half of the protocols included anakinra in their recommendations, and six different regimens were described with significant differences in dosage (ranging from 100 to 400 mg per day) and duration of treatment (between 3 and 10 days). Other monoclonal antibodies, immunomodulatory agents, and immunoglobulin therapy are in clinical trials for the treatment of COVID-19-associated cytokine release syndrome [3].
As recent studies have found, severe COVID-19 is commonly complicated with coagulopathy, including disseminated intravascular coagulation and venous thromboembolism [33,34]. For this reason, the administration of heparin has been recommended for COVID-19 patients according to expert consensus [4,31], although its efficacy remains to be validated. In a study of 449 patients with severe COVID-19 and sepsis-induced coagulopathy criteria or with markedly elevated D-dimer levels (higher than sixfold of upper limit of normal), anticoagulant therapy, mainly with low-molecular-weight heparin, was associated with better prognosis [34]. Beyond severe coagulopathy, special attention to venous thromboembolism prophylaxis is necessary in the management of COVID-19 [35]. Low-molecular-weight heparin is preferred over unfractionated heparin in order to reduce patient contact (depending on the patient’s bleeding risk and creatinine clearance) [36]. We found that all protocols analyzed included anticoagulant treatment, with low-molecular-weight heparin as the first option. It was recommended that all hospitalized COVID-19 patients receive prophylactic heparin. This recommendation was extended to after discharge if necessitated by the patient’s clinical condition in 46.6% of protocols. An increase in prophylactic heparin doses was recommended in 53.3% of the protocols for patients at high risk of thrombosis. In five protocols (33.3%), heparin was increased to therapeutic doses if the condition was critical or if there was a progressive increase in D-dimer levels. All protocol included therapeutic heparin if there was evidence of venous thromboembolism.
Our findings show significant differences among local protocols throughout Andalusia (Spain) regarding the management of COVID-19 patients. There are some possible explanations for these disparities, including a lack of strong evidence about medical therapies that have been definitively reported to be effective at this time, the local availability of therapies, and administrative decisions about organization and care planning. Although consensus statements in some international guidelines recommend against routine use of most antiviral agents, the eagerness to help our patients leads healthcare providers to explore different therapeutic strategies, balancing benefits and detriments. The same issue occurs with adjunctive therapies that are only suggested for treating ARDS with evidence of cytokine release syndrome as well as with anticoagulation in patients at high risk of thrombosis. There is greater consensus on the routine use of thromboembolism prophylaxis or anticoagulation for established venous thromboembolism. All protocols analyzed, in accordance with other guidelines, emphasize the prompt implementation of recommended infection prevention and supportive care for complications.
These results are important because they show significant differences among the recommended treatments for COVID-19, according to the treatment protocols of large university hospitals in a region of Spain. They reveal a need for specifically designed randomized clinical trials to determine the most appropriate evidence-based treatment regimen. COVID-19 is having a strong impact in several countries, including Spain, in several aspects, especially health, economic, and social. There is a quickly growing body of evidence on this topic trying to find the best practice for the treatment of symptomatic patients with COVID-19 [37].
This study has several limitations. First, we only analyzed the data provided in the hospitals’ treatment protocols, and no information on treated actually provided to patients was obtained. Furthermore, we did not know the degree of protocol implementation in the hospital or its coverage area. Second, only the protocols of the largest hospitals of Andalusia were analyzed in this study, even though there are many smaller hospitals and healthcare centers. Normally, the treatment recommendations adopted in reference hospitals indicate how their area of influence proceeds, but this cannot be definitively assured. Third, our study focused only on adult patients, and the recommendations were not applicable to pediatric populations. Finally, the recommendations provided in the protocols are based on local expert opinions, clinical experience in managing COVID-19 patients, and current international evidence regarding major proposed treatments, whether repurposed or experimental, for COVID-19. Therefore, no strong therapeutic recommendations could be drawn from the protocols analyzed in this study.