Different COVID-19 treatments’ impact on hospital length of stay

The highly contagious SARS-CoV-2 causes the Coronavirus disease (COVID-19), first discovered in Wuhan, China, in December 2019 [1]. The COVID-19 pandemic has adversely affected world economies, global public health infrastructure, and social behaviors. Despite physical distancing and preventative precautions, such as face masks, tens of thousands of people are still being affected by COVID-19 [2]. While it is a disease that mainly involves the lungs, multiple organs, including neurological symptoms, have been described frequently [3]. Even young adults with seemingly asymptomatic COVID-19 infections can end up suffering a thrombotic stroke that adversely affects their health [4]. Currently, there is no definitive antiviral agent to treat hospitalized patients with severe COVID-19 [5]. However, any treatment that may decrease hospital length of stay will still be beneficial, especially during the height of the pandemic when many hospitals are near or at full capacity [6]. Fortunately, combination therapies with pre-existing agents have shown promising clinical improvement and repurposing FDA-approved drugs may prove beneficial. Current SARS-CoV-2 drug therapies may be divided into those that (1) target the RNA or proteins of the virus and (2) biologics that interact and interfere with host proteins and processes supporting the virus. From March 2020 to October 2021, the recommendation was to use dexamethasone and tocilizumab for hospitalized COVID-19 patients requiring mechanical ventilation and dexamethasone and remdesivir for hospitalized COVID-19 patients not experiencing hypoxia or requiring supplemental oxygen without mechanical ventilation [7, 8].

Dexamethasone is a ubiquitous corticosteroid that is widely available at low cost [9]. It was the first major “breakthrough” therapy in treating COVID-19 and has been shown by the RECOVERY trials to increase the number of ventilator-free days and reduce the risk of death [8, 10]. Remdesivir is an antiviral drug, given emergency use authorization for SARS-CoV-2 in 2020, that was originally used in clinical trials against Ebola [10]. It is currently used to treat hospitalized COVID-19 patients not experiencing hypoxia or those requiring supplemental oxygen without mechanical ventilation. Remdesivir has previously been shown to reduce early stage mortality and the need for high-flow oxygen supplementation and invasive mechanical ventilation amongst hospitalized COVID-19 patients—with an increased clinical recovery rate of 21% on day 7 and 29% on day 14 of treatment compared to the placebo group [11]. Originally approved for rheumatoid arthritis, tocilizumab is also used for COVID-19. Tocilizumab is a recombinant humanized IgG1 monoclonal antibody that reduces the pro-inflammatory effect of IL-6 and has previously been shown to reduce mortality and the need for mechanical ventilation in COVID-19 patients [12, 13]. Prior reports analyzing trials of combination treatments for COVID-19 have had mixed results [14]. Other treatments such as hydroxychloroquine, chloroquine, convalescent plasma transfusions, ivermectin, and combination therapy with protease inhibitors such as Lopinavir + Ritonavir and Darunavir + Cobicistat either had no change or increase in mortality [15, 16].

Different treatments, alone or in combination, not only impacts patient morbidity and outcomes but also hospital LOS [17]. During the peak phase of the pandemic, hospitals were inundated with critical patients requiring inpatient hospital admission resulting in hospital staff, bed, and supply shortages and an increasing number of patients awaiting admission despite hospitals operating at maximum capacity [18]. Therefore, assessing the LOS of admitted COVID-19 patients of varying severity based on different treatment arms would be invaluable in assisting resource management of hospital staff and oxygen supplies [6].

In this retrospective cohort study, we investigated adults (18 +) hospitalized with COVID-19 at a community hospital in northern California from March 2020 to March 2021 and September 2021 to October 2021 to evaluate if the combination treatment of (1) remdesivir + dexamethasone, or (2) remdesivir + dexamethasone + tocilizumab truly decreased hospital LOS for non-hypoxic and hypoxic patients as well as its effect on 30-day mortality. In addition, we include data from treatments such as hydroxychloroquine and convalescent plasma transfusion for the sake of inclusion and academic discussion, and because a significant segment of American society, is still firmly entrenched in various conspiracy theories and medications that have no known benefit to the treatment of COVID-19[19].

1233 COVID-19 patients were hospitalized in Kaiser South Sacramento from March 2020 to March 2021 and September 2021 to October 2021. 53.4% of these patients were male and 46.6% were female but there was no statistically significant difference in the ratio of gender (F:M) in each treatment type (p = 0.080). Similarly, the distribution of ages between different treatment groups was indistinguishable and, therefore, insignificant (p = 0.002).

Most of these patients, 665 (62.9%), were treated with a combination of dexamethasone and remdesivir. 216 (20.4%) were treated with dexamethasone only. 67 (6.3%) were given remdesivir only. 47 (4.4%) received the combination therapy of dexamethasone, remdesivir, and tocilizumab. 33 patients received remdesivir, dexamethasone, and baricitinib combination therapy (3.1%). A small subset, 30 (2.8%), were treated with “other” methods (see methods section).

In mild COVID-19 patients, the LOS did not show any statistical difference between different treatment arms with a p value of 0.186 (Fig. 1, Table 1). In moderate COVID-19 patients, the combination treatment of dexamethasone plus remdesivir showed a statistically significant reduction in LOS from 6.77 days to 4.98 days with a p value of 0.007 (Fig. 2, Table 1). In severe COVID-19 patients, the combination treatment of remdesivir, dexamethasone, and tocilizumab showed a statistically significant reduction in LOS of up to 8 days (p = 0.0034) when compared to “other,'' nonviable treatments, such as hydroxychloroquine or convalescent plasma transfusion (Table 1). However, upon excluding the “other” category, there was no statistically significant difference between various combination treatments with a p value of 0.116 (Fig. 3). Overall, there was no significant reduction in mortality for severe COVID-19 patients between different combination treatments with a p value of 0.252 (Fig. 4).

Prior studies have indicated that the combination of dexamethasone and remdesivir has modest effects in reducing hospital LOS in moderate-to-severe COVID-19 patients [21]. In our study, only moderate patients (NC only) given dexamethasone and remdesivir were found to have a statistically significant decreased LOS by approximately 1.5 days (Fig. 2).

During this investigation (12/31/2022), there were no conclusive published reports on the efficacy of the three-drug combination therapy of dexamethasone, remdesivir, and tocilizumab on COVID-19 patients [8]. In severe COVID-19 cases, we initially found the three-drug combination treatment (dexamethasone + remdesivir + tocilizumab) to be statistically superior to all other treatments when considering nonviable treatments, such as hydroxychloroquine or convalescent plasma transfusion. (Table 1) This is in accordance with the current literature [22]. However, the statistical power of this finding, a p value of 0.0034, came mostly from the grossly poor LOS associated with the nonviable treatment group. When we removed nonviable treatment data from the analysis, three-drug combination treatments failed to yield a statistically significant decrease in LOS, based on a p value of 0.116 (Fig. 3). We know from other trials that the clinical benefit of dexamethasone plus tocilizumab, while significant is small [8]. It is possible that the three-drug combination (dexamethasone + remdesivir + tocilizumab) offers similar minute benefits that are undetectable by our relatively small sample size. More patient data may yield statistically significant findings.

In addition, there was also no statistically significant difference between the three-drug combination of the tocilizumab group and the baricitinib group (Fig. 3). Perhaps, the high p value is due to the relatively small sample size (32) of the baricitinib group. Additional data samples may improve the p value. Finally, It should be noted that prior literature has suggested that baricitinib may not decrease LOS [23].

Current treatment guidelines, per RECOVERY and other trials, recommend dexamethasone only for hypoxic COVID-19 patients [14, 24, 25]. This is in alignment with our study as dexamethasone—individually or in combination—has no statistically significant impact on hospital LOS for mild patients on RA (Fig. 1). In addition, the remdesivir group’s LOS also shows no statistical reduction in LOS for mild COVID-19 patients (Fig. 1). Perhaps, patients with mild COVID-19 experienced minimal benefit from Remdesivir.

In our study, 47 mild patients were treated with only remdesivir, and 41 mild patients were treated with remdesivir and dexamethasone during their stay. Typically, these patients would receive five doses of remdesivir at $520 per dose, costing the healthcare facility at least $228,800 [26]. This figure does not factor in the logistic cost of administering medications (nursing and pharmacy staff). In addition, patients with rheumatoid arthritis faced drug shortage challenges—including remdesivir and steroids—during the pandemic [27]. In addition to mild patients not benefiting from remdesivir in our study, this financial relevance further suggests against the use of remdesivir on mild COVID-19 patients.

In our study, we found no mortality benefit in any treatment arm in mild, moderate, and severe COVID-19 patients (Fig. 4). These results are inconsistent with prior studies that have suggested some survival benefits for dexamethasone, remdesivir, and tocilizumab [10‐12]. However, these survival benefits, while statistically significant, are often minute. Given our relatively small sample size, it is plausible that our study does not detect such survival benefits. In addition, current treatment guidelines from the ACTT-1 trial show that remdesivir only decreases hospital LOS and not mortality [28].

A study previously suggested that men may benefit more than women with 5-day treatment with remdesivir [20]. Another study reported that the male gender is overrepresented in COVID-19 treatment clinical studies [29]. Interestingly, we also have slightly more men in our database than women. Given that there are more women than men in the human population, it is perplexing to see more men hospitalized with COVID-19 than women. Perhaps, there is a behavior or genetic difference that predisposes men to catch COVID-19 more readily than women [30]. Literature on gender analysis with COVID-19 treatment types and hospital LOS is limited and, therefore, should be investigated in future studies.

The treatment of the COVID-19 geriatric population (65 >) is an important subsample to analyze as this population is the greatest to be hospitalized throughout the pandemic, with 189,735 of 418,804 cases nationally [31]. Our study had an average age of 61 in 1233 patients. There is currently not much data on age correlation with COVID-19 treatment efficacy, side effects, or hospital LOS. A recent literature review noted remdesivir and dexamethasone to have inefficiency in geriatric patients, while tocilizumab findings were inconclusive [32]. With enough sampling data, we may be able to further analyze LOS between treatments based on different age groups and the severity of the disease.

Please note, in our hospital experience, patients who did not tolerate non-invasive ventilation (NIV) were intubated. It is a recognized phenomenon that certain populations cannot tolerate NIV [33]. In addition, within the past 2 years, we have seen acute ischemic stroke as a complication of patients with severe SARS-CoV-2 infection due to its precipitation of a hypercoagulable state [34]. There has even been a reported case of stroke in a young healthy woman with an asymptomatic SARS-CoV-2 infection as well [35]. Studies have shown that in-hospital treatment with heparin was associated with lower mortality in severe COVID-19 patients [36]. All COVID-19 patients in this data were given heparin DVT prophylactic doses unless there were contraindications (i.e., bleeding).

Although our sample size and data represent statistical significance, it has limitations. First, the data collected are partial and not continuous; our study is missing patient data from April to July 2021, which includes an additional 1000 data points. We collected and included the patient population from August to September 2021 in our data analysis to observe the preliminary results on the effect of dexamethasone + remdesivir + tocilizumab treatment, as the use of tocilizumab (IL-6 inhibitor) was not granted emergency use authorization (EUA) for COVID-19 treatment in patients under supplemental oxygen or ventilation until June 24, 2021 [37]. In addition, the sample size for this combination treatment is relatively small at 45 patients.

Second, our selection criteria do not exclude or account for specific comorbidities. We found that many patients admitted as “severe” COVID-19 patients to have preexisting comorbidities, such as chronic kidney disease, liver failure, heart disease, etc. There are also many young healthy patients without comorbidities that ended up on the ventilator with severe hypoxic resp failure. Either way, those with severe resp illness received the same medication available during that given period, regardless of comorbidities. Future studies may further delineate treatment outcomes based on comorbidities.

Biomarkers that are associated with predicting clinical progression and outcome in COVID-19, such as D-dimers, CRP, and IL-6, were not included in the study. It is well-known that the level of IL-6 could be used as a predictor of the efficacy of tocilizumab [39]. However, these labs were not ordered consistently on every patient because of the variation in practices of physicians in the hospital and ongoing changes in treatment protocol throughout the pandemic [40].

Drug shortages (i.e., remdesivir) during the pandemic affected treatment options for our patients [41]. This in combination with continuous changes in national treatment protocol over the LOS may also have affected our data.

Finally, the evolution of COVID-19 variants during the pandemic is not taken into account. Viral genome sequencing was not available for every infected patient who was hospitalized, making it difficult to accurately associate the severity of the disease with certain variants. Our data analyzes patients during the three notable COVID-19 variants: alpha, beta, and delta [42]. The variants trended towards an increase in transmissibility and decrease in virulence, indicating that newer SARS-CoV-2 variants may be less virulent [43]. The emergence of vaccines during the pandemic most definitely affected the demographics of hospitalized COVID-19 patients and probably altered the LOS as well [44].

The current COVID-19 regimen of dexamethasone + remdesivir in moderate COVID-19 patients decreases the hospital LOS. Combination therapy of dexamethasone + remdesivir + tocilizumab in severe COVID-19 patients may decrease the length of a patient’s hospital stay. It is unclear if three-drug combination therapies are superior to two-drug ones in severe COVID-19 patients. No treatment modality seems to offer clinical benefit for hospitalized patients with mild COVID-19 in terms of reduction in LOS or mortality. Future studies could investigate patient age, the impact of COVID-19 vaccination status, and any other risk factors.