The Comparison of Therapeutic With Prophylactic Anticoagulation on Mortality, Risk of Bleeding, and Thromboembolism in Hospitalized Patients With COVID-19: A Systematic Review

Thromboembolism is one of the most severe manifestations of coronavirus disease 2019 (COVID-19). Thrombotic complications have been reported even with the administration of thromboprophylaxis. This has led many experts to have variable opinions on the most effective prophylactic strategy and to anticipate the discovery of the ideal dosing of anticoagulation to reduce thromboembolic events and related mortality. We performed a systematic review to evaluate whether therapeutic-dose anticoagulation is superior to prophylactic-dose anticoagulation by comparing mortality rates, bleeding risks, and rates of thromboembolism. We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to create our systematic review. Twenty-two records were collected from PubMed, PubMed Central (PMC), and Medical Literature Analysis and Retrieval System Online (MEDLINE), after which they undertook quality appraisals. A total of 124 studies were analyzed in six systematic reviews and meta-analyses, one pooled analysis, two multicenter retrospective cohort studies, one observational study, one retrospective chart review, one evidence-based protocol, and four narrative reviews.


Introduction And Background
The coronavirus disease 2019 (COVID-19) has impacted over 560 million individuals worldwide, resulting in over six million fatalities from its first incidence in December 2019 in China to July 2022 globally [1]. Although severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) predominantly affects the respiratory system, it could also impair other organ systems [2]. The pathophysiology and severity of COVID-19 mainly depend on the degree to which the immune and inflammatory systems are compromised, in addition to direct viral effects. In severe cases, over-accumulating pro-inflammatory cytokines promote hypercoagulation, vascular hyperpermeability, multiorgan failure, and even death. These findings support the increased reports of venous thromboembolism (VTE) and pulmonary embolism (PE) in severe cases of COVID-19, which are considered poor prognostic indicators [2][3][4][5]. Consequently, elevated D-dimer, fibrinogen, and indicators of endothelial dysfunction (von Willebrand factor antigen and soluble thrombomodulin) have been associated with unfavorable prognoses, including death [5,6]. As a result, several clinical trials have emerged and initiated studies designed to improve COVID-19 prognosis with antithrombotic treatment [7]. The incorporation of thromboprophylaxis in hospitalized COVID-19 patients has been found to increase survival [8].
On top of its antithrombotic action, heparin exerts an anti-inflammatory effect, a protective action on endothelial function [9], and a specific antiviral action in the extracellular matrix of tissues [10]. A short retrospective cohort analysis showed that the administration of low-molecular-weight heparin (LMWH) resulted in an increased number of lymphocytes and lower interleukin-6 (IL-6) levels compared to control patients, demonstrating an improvement in coagulation profiles and the restoration of immune functions. In another study, heparin treatment was related to increased oxygenation in 27 individuals with COVID-19 [2]. Despite the administration of thromboprophylaxis, COVID-19 patients are more likely to develop VTE than severely ill COVID-19-negative patients. This occurrence has prompted many experts to propose higher anticoagulant dosages, particularly in patients who are at increased risk of thromboembolism due to significantly elevated D-dimer results and/or other comorbidities [5]. Meanwhile, the most efficient thromboprophylactic strategies for a wide range of COVID-19 disease severity are still undetermined [11]. According to current national and international recommendations, hospitalized patients are indicated for universal pharmacologic thromboprophylaxis with subcutaneous LMWH or unfractionated heparin (UFH) [12]. The National Institutes of Health (NIH) guidelines recommend the use of a therapeutic dose of heparin for nonpregnant patients with D-dimer levels above the upper limit of normal who require conventional oxygen and do not have an increased bleeding risk unless heparin is contraindicated. For other patients, including pregnant females, the panel recommends the prophylactic dose of heparin if there is no contraindication [13].
The lack of consensus about the ideal prophylactic anticoagulant regimen has resulted in diverse expert advice, hospital policies, and clinical judgments regarding various antithrombotic treatment modalities [3]. In continuing randomized controlled trials (RCTs) emphasizing outpatients, hospitalized patients in medical wards, and critically compromised patients with COVID-19, a number of different antithrombotic substances, dosages, and the duration of therapy are being evaluated to establish the optimal thromboprophylactic regimens [11]. We conducted a systematic review to explore whether therapeutic-dose antithrombotic therapy has a beneficial effect over prophylactic-dose antithrombosis in terms of inpatient mortality, risk of bleeding, and venous thromboembolism.

Inclusion Criteria
We identified studies in English, randomized controlled trials, clinical trials, multicenter studies, metaanalyses, systematic reviews, traditional reviews, and evidence-based protocols on patients hospitalized with COVID-19. We incorporated the studies published after November 2019.

Exclusion Criteria
We excluded gray literature, books, documents, case series, case reports, cross-sectional studies, duplicate studies, overlapping studies, and studies before December 2019 and COVID-19 patients not requiring hospitalization.

Results
We identified a total of 4,485 articles after searching the databases (PubMed, PMC, and MEDLINE) by using the keywords and applying inclusion and exclusion criteria. After removing 708 duplicates, we screened the remaining 3,777 records by title and abstract. We assessed the derived 22 articles for quality appraisals and retained 22 reports.
Our review included 124 studies from six randomized controlled trials (RCTs), six systematic reviews and meta-analyses, one pooled analysis, two multicenter retrospective cohort studies, one observational study, one retrospective chart review, one evidence-based protocol, and four narrative reviews with a combination of 57,100 patients. We compared the outcomes of therapeutic antithrombosis with that of regular thromboprophylaxis concerning mortality, risk of major bleeding, and thromboembolism in patients with COVID-19. The sample included in our study was a mixed population of moderate to severely ill hospitalized COVID-19 patients. The term "high-intensity anticoagulation" would be referred to as "high-dose, therapeutic-dose anticoagulation." The term "low-intensity anticoagulation" would be defined as "prophylactic-dose, standard-dose, low-dose, intermediate-dose, usual-care, or routine anticoagulation." A detail of our search strategy can be seen in the PRISMA flow chart (Figure 1) with reasons for excluding the primarily identified records. We compared nine studies (five RCTs and four observational studies), which included a total of 9,517 patients. Of the patients, 5,125 received low-intensity anticoagulation; 4,753 patients received high-intensity anticoagulation; and 361 patients did not receive anticoagulation. We further reviewed 13 articles that additionally covered four RCTs, 85 observational studies, and one casecontrol study. We used revised Cochrane risk-of-bias tool for randomized trials (RoB 2) [15] for quality appraisal of the included RCTs. We pooled the summary of nine studies in Table 2. Author

Discussions
To achieve the most favorable outcomes with pharmacologic thromboprophylaxis, the benefits of intensive thromboprophylaxis must be weighed against the possible adverse outcomes [4].

Effects of Therapeutic Anticoagulation on Mortality
Many studies revealed that patients with COVID-19 receiving anticoagulation (either prophylactic or low dose; oral, subcutaneous, or intravenous) were associated with a significant reduction in mortality rates [2,22,23]. A multicenter retrospective cohort study reported that unadjusted inpatient mortality rates were comparable between high and low doses of anticoagulant groups. After adjusting for other characteristics associated with mortality, administering a high dose of anticoagulation was independently associated with decreased in-hospital mortality. The association remains consistent in sensitivity analyses, Cox regression analysis (Cox proportional hazards regression analysis), and 21-day mortality [5]. This result was in line with Ionescu et al., who said that anticoagulation therapy was correlated to lower mortality risk in the propensity score-weighted multivariate proportional hazard model. The outcome was dose-related; compared to no thromboprophylaxis, prophylactic dosing was significantly associated with a 65% decline in mortality risk and therapeutic anticoagulation with an 86% reduction [6]. According to a study evaluated by Kamel et al., giving therapeutic dosages of heparin for seven or more days was related to a better prognosis in patients with severe infection who meet sepsis-induced coagulopathy score criteria or with significantly high Ddimer [2].
The three large, open-label, adaptive, multiplatform, randomized clinical trials (REMAP-CAP, ACTIV-4a, and ATTACC) investigated the outcomes of therapeutic anticoagulation in patients hospitalized due to COVID-19. The investigators published a report for 1,098 critically ill patients receiving intensive care unit (ICU)level support [18] and another report for 2,219 patients not initially receiving ICU-level support [19]. The latter study revealed that compared to prophylactic anticoagulation, therapeutic anticoagulation prolonged survival until hospital discharge, free from organ support in noncritically ill patients [5,19,22]. The trial stopped after the superiority of therapeutic anticoagulation was observed [19]. The former study concluded that therapeutic anticoagulation was not associated with increased survival until discharge or reduced duration of organ support compared to standard-dose anticoagulation in critically ill patients. The data safety and monitoring boards halted the trials owing to the futility of the endpoint of the freedom of organ support at 21 days and the future harmful effects of bleeding [11,18]. The inconsistent outcomes between the two studies could be explained by possible advanced pathologic conditions existing in severely ill patients to profit from therapeutic heparin [22]. Al-Banaa et al. elaborated that the difference in treatment response identified between the two studies implies that the time of initiating therapeutic anticoagulation therapy possesses a significant role in determining treatment outcomes [5]. Although the information on the timing of thromboprophylaxis administration regarding the onset of symptoms was lacking in most included studies, overall results favor the early beginning of higher dosing of thromboprophylaxis before the emergence of critical illness in patients with unfavorable prognostic features [5,8].
A meta-analysis of eight RCTs by Kow et al. concluded that there was no statistically significant difference in the death rates between high-intensity anticoagulant and low-dose anticoagulant groups in COVID-19 patients. At the same time, the author mentioned that high anticoagulant doses would be beneficial, especially in severe cases [24]. According to a meta-analysis by Kollias et al., including 17 studies (two randomized and 15 observational studies) focused on therapeutic-dose and prophylactic-dose anticoagulation, the administration of therapeutic dose did not significantly reduce the death rate compared to prophylactic dose [8]. When Kollias et al. performed a meta-analysis of three studies investigating patients solely in the ICU, the pooled adjusted relative risk (RR) for mortality in patients with COVID-19 who received a therapeutic dosage of thromboprophylaxis versus a usual-care dose was 0.58 (95% confidence interval (CI): 0.35-0.94). The author further uncovered the indication bias in the included observational studies since participants with greater risk for severe disease were more likely to receive higher doses [8]. As a trend toward clinical benefits of therapeutic anticoagulation was recognized, higher doses can be selectively recommended on an individualized basis for patients at high or very high thrombotic risk, provided they also have a low risk of bleeding [6,8,24].  [17]. In contrast, Giossi et al. suggested that conventional thromboprophylaxis should be the initial therapy in moderate and severe COVID-19 patients, specifically with a high risk of bleeding [23]. A meta-analysis by Giossi et al. found a comparable reduction in mortality between therapeutic anticoagulation and prophylactic anticoagulation, while the prophylactic-dose cohort showed a significantly decreased risk of overall bleeding [23]. Another retrospective observational study of 311 critically ill COVID-19 patients at Stony Brook University Hospital demonstrated no significant difference in survival curves for prophylactic and therapeutic dosing domains within the first three weeks [16]. In contrast to Al-Banaa et al., this study also reported that more than three weeks of treatment with therapeutic anticoagulation increased the risk of death up to five times compared to routine anticoagulation. Hoogenboom et al. clarified that patients from the therapeutic dosing arm had a higher rate of chronic obstructive pulmonary disease (COPD); high D-dimer, lactate dehydrogenase (LDH), and brain natriuretic peptide (BNP) levels; low lymphocyte count, and partial pressure of oxygen (PaO 2 ). In addition, the characteristics of non-survivors were elderly males with depleted lymphocyte counts and cardiovascular diseases. The author interpreted that variations in sample characteristics and COVID-19 disease severity may account for discrepancies in research results [16].

Intermediate vs. Standard-Dose Prophylactic Anticoagulation in Critically Ill Patients With COVID-19: An
Open-Label Randomized Controlled (INSPIRATION) trial (N=562 ICU patients) demonstrated no significant differences in the primary outcome (a composite of venous or arterial thrombosis, treatment with extracorporeal membrane oxygenation, or mortality within 30 days) between both arms [3,22]. Kollias et al. mentioned that microvascular thrombi might already exist in critically ill patients, leaving increased dosing of anticoagulation ineffective [8]. The Therapeutic Anticoagulation versus Standard Care as a Rapid Response to the COVID-19 Pandemic (RAPID) trial, including 465 randomized moderately ill COVID-19 patients, did not detect a significant lowering in the 28-day composite of death, invasive mechanical ventilation, noninvasive mechanical ventilation, or ICU admission in the therapeutically anticoagulated patients; however, the occurrence of all-cause mortality was lowered by 78% in the therapeutic counterpart [17,21,22]. One study reviewed by Reis et al. concluded that therapeutic-dose anticoagulation might have little or no impact on deteriorating clinical outcomes within 28 days, as measured by advancing to intubation or death compared to conventional thromboprophylaxis. Yet, another study reviewed by the same author showed a reduction in progression to mechanical ventilation or death with therapeutic anticoagulation [7]. The Therapeutic versus prophylactic anticoagulation for severe COVID-19: A randomized phase II clinical trial (HESACOVID), a single-center study of 20 patients, published that therapeutic-dose anticoagulation significantly raised partial pressure of oxygen/fraction of inspired oxygen (PaO 2 /FiO 2 ) and long ventilator-free days compared to standard-dose anticoagulation [11,25]. The association between anticoagulant dosage and mortality remains ambiguous, given the unadjusted estimates generated by varying outcomes, and should be evaluated with caution [2].
A large, prospective, multicenter, open-label, randomized controlled comparative safety and efficacy research is funded by the Icahn School of Medicine at Mount Sinai (the FREEDOM COVID-19 Anticoagulation trial) [22]. The trial is expected to enroll up to 3,600 patients, randomly assigned to one of three anticoagulation therapies (prophylactic enoxaparin, therapeutic-dose enoxaparin, or therapeutic-dose apixaban) in a 1:1:1 fashion. It is anticipated that the FREEDOM COVID-19 Anticoagulation trial is powerful enough to detect significant differences in outcomes [22]. The results from the FREEDOM COVID-19 Anticoagulation trial and other ongoing studies are crucial in inferring whether therapeutic-dose anticoagulation is superior in minimizing thrombotic events, preventing intubation, or prolonging survival compared to prophylactic-dose anticoagulation in hospitalized patients with COVID-19. Table 3 summarizes the included studies on the mortality rates of patients receiving different doses of anticoagulation.

Hiking Up of Bleeding Events in Full-Dose Anticoagulation
Although the general bleeding rate was more notable in patients receiving higher anticoagulation doses, the risk of severe bleeding events was similar in both groups [5]. The meta-analysis of Reis et al. showed that therapeutic-dose anticoagulation might promote serious hemorrhage within 30 days relative to routine thromboprophylaxis, regardless of disease severity [7]. These findings coincided with analyses of several other studies that highlighted that higher-dose thromboprophylaxis was associated with a higher occurrence of major bleeding complications than routine thromboprophylaxis [3,6,[17][18][19][20][21][22][23][24][25]. Kollias [12]. Sholzberg et al. also reported a low risk of major bleeding in patients allocated to therapeutic-dose anticoagulation [17]. With an assumed control risk of 0.014 (the frequency of significant bleeding in the control group was 14 per 1,000), approximately 90 patients would require higher-intensity anticoagulation dosing to experience one major bleeding event [20]. Table 4 recapitulates the studies discussing the bleeding events associated with prophylactic-and therapeutic-dose anticoagulation.

Study Author Year Type of study Patients
Purpose of the

Slimming of Thromboembolic Events With Therapeutic-Dose Anticoagulation
A systematic review and meta-analysis highlighted that the total incidence of venous thromboembolism (VTE) among COVID-19 inpatients was reported to be 17% (95% CI: 13.4-20.9), with variations depending on research design and method ascertainment; ICU patients had a fourfold greater incidence rate than those in non-ICU settings (28% versus 7%) [11]. Patell et al. supplemented these findings by stating that severely ill patients have an elevated risk of thrombosis, partly due to a combination of prothrombotic risk factors such as chronic immobility and hyperinflammatory conditions [12]. Hasan et al. stated that prophylactic LMWH was associated with subtherapeutic anti-factor Xa levels in critically ill COVID-19 patients [26]. The author also mentioned that monitoring anti-factor Xa levels in patients on UFH was associated with a better achievement of therapeutic anticoagulation than monitoring activated partial thromboplastin time (APTT).
In patients receiving high doses of UFH to reach the goal APTT, the risks of life-threatening hemorrhagic events might be higher without monitoring antithrombotic activity via an anti-factor Xa assay. His metaanalysis reported a lower prevalence of VTE in patients allocated to the mixed anticoagulation approach (prophylactic and therapeutic) compared to patients allocated to prophylactic anticoagulation only. Hasan et al. hypothesized that a lower occurrence of VTE in the former arm might be due to a higher rate of the achievement of desired anti-factor Xa level from the administration of therapeutic-dose anticoagulation [26]. Similar to Farkouh et al. [22], Al-Banaa et al. [5] agreed that full-dose anticoagulation was correlated with a decrease in thromboembolism. Reis et al. analyzed that the risk of any thrombotic events may be lower in therapeutically anticoagulated participants in contrast to prophylactically anticoagulated counterparts independent of disease severity [7]. In addition, the meta-analysis of Kow et al. reported that the reduction in the risk of thrombotic complications with intermediate or therapeutic anticoagulation was statistically significant compared to standard thromboprophylaxis [24]. Tacquard et al. disclosed that cumulative exposure to higher-dose antithrombotic therapy was significantly associated with a decreased risk of thrombotic events, highlighting the potential benefits of a higher-dose anticoagulant regimen in severe COVID-19 patients [25].
Nevertheless, the ACTION and INSPIRATION trials did not detect a significant difference in the risk of VTE between intermediate-dose and standard-dose strategies [3,20]. Similarly, the REMAP-CAP, ACTIV-4a, and ATTACC Investigators and the RAPID trial reported that the difference in the incidence of thromboembolic events in therapeutic and prophylactic groups was not statistically significant [16][17][18]

Limitations
Our systematic review has a number of limitations. We obtained our data from only three databases (PubMed, PMC, and MEDLINE) and retrieved the records providing free access to full texts published in English. There is a chance that unincluded research may provide different results. We did not have a statistician to analyze the included data; therefore, we had to rely on the presented results.

Conclusions
The majority of studies included in our systematic review indicate that therapeutic-dose anticoagulation was associated with a lower rate of thromboembolism and an increased risk of bleeding. However, whether a high-dose strategy can reduce mortality remains inconclusive. Many publications have different interpretations of the superiority of therapeutic anticoagulation, while the certainty of evidence from available data is still low. Globalized, large-scale randomized controlled trials powerful enough to detect statistically and clinically significant differences between high and low doses are required to clarify the ideal dosage and timing of anticoagulation.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.