key: cord-0981285-bcfm6d5r authors: Blondon, Marc; Cereghetti, Sara; Pugin, Jérôme; Marti, Christophe; Darbellay Farhoumand, Pauline; Reny, Jean‐Luc; Calmy, Alexandra; Combescure, Christophe; Mazzolai, Lucia; Pantet, Olivier; Ltaief, Zied; Méan, Marie; Manzocchi Besson, Sara; Jeanneret, Séverin; Stricker, Hans; Robert‐Ebadi, Helia; Fontana, Pierre; Righini, Marc; Casini, Alessandro title: Therapeutic anticoagulation to prevent thrombosis, coagulopathy, and mortality in severe COVID‐19: The Swiss COVID‐HEP randomized clinical trial date: 2022-05-18 journal: Res Pract Thromb Haemost DOI: 10.1002/rth2.12712 sha: 4c75a94a5056ae53796a4d0a0d3513642ce1585c doc_id: 981285 cord_uid: bcfm6d5r BACKGROUND: Hospitalized patients with COVID‐19 suffered initially from high rates of venous thromboembolism (VTE), with possible associations between therapeutic anticoagulation and better clinical outcomes in observational studies. OBJECTIVE: To test whether therapeutic anticoagulation improves clinical outcomes in severe COVID‐19. PATIENTS/METHODS: In this multicenter, open‐label, randomized controlled trial, we recruited acutely ill medical COVID‐19 patients with D‐dimer >1000 ng/ml or critically ill COVID‐19 patients in four Swiss hospitals, from April 2020 until June 2021, with a 30‐day follow‐up. Participants were randomized to in‐hospital therapeutic anticoagulation versus low‐dose anticoagulation in acutely ill participants/intermediate‐dose anticoagulation in critically ill participants, with enoxaparin or unfractionated heparins. The primary outcome was a centrally adjudicated composite of 30‐day all‐cause mortality, VTE, arterial thrombosis, and disseminated intravascular coagulopathy (DIC), with screening for proximal deep vein thrombosis. RESULTS: Among 159 participants, 55.3% were critically ill and 94.3% received corticosteroids. Before study inclusion, pulmonary embolism had been excluded in 71.7%. The primary outcome occurred in 4/79 participants randomized to therapeutic anticoagulation and 4/80 to low/intermediate anticoagulation (5.4% vs. 5.0%; risk difference +0.4%; adjusted hazard ratio 0.76, 95% confidence interval 0.18–3.21), including three deaths in each group. All primary outcomes and major bleeding (n = 3) occurred in critically ill participants. There was no asymptomatic proximal deep vein thrombosis and no difference in major bleeding. CONCLUSIONS: Among patients with severe COVID‐19 treated with corticosteroids and with exclusion of pulmonary embolism at hospital admission for most, risks of mortality, thrombotic outcomes, and DIC were low at 30 days. The lack of benefit of therapeutic anticoagulation was too imprecise for definite conclusions. • Vascular thromboses including pulmonary microthromboses are characteristic of severe COVID-19. • This multicenter randomized trial compared different doses of anticoagulation. • Risks of thrombotic complications and death were low in both groups. • The common exclusion of pulmonary embolism before study inclusion may explain these low risks. COVID-19 has affected >300 million humans and caused >5 million deaths. The lack of specific therapies has led to an intense research effort to evaluate drugs of potential benefit, especially for severe COVID-19 patients with an initial mortality of 10% to 30%. 1, 2 Early, an unexpectedly high thrombotic burden was found among COVID-19 patients. Despite standard-dose thromboprophylaxis, venous thromboembolism (VTE), restricted to proximal deep vein thrombosis (DVT), and pulmonary embolism (PE), occurred in about 9% of COVID-19 patients, with greatest risks in the intensive care unit (ICU). 3, 4 A prevalent tendency for systemic intravascular coagulopathy was also noted. 1, 5 In parallel, observational studies reported associations of high-dose heparins with better survival in COVID-19 patients, compared with standard low-dose thromboprophylaxis, 1, 6, 7 leading to empiric decision in some hospitals or guidance documents to administer higher doses of anticoagulants. 8 Heparins may have multiple theoretical benefits in severe COVID-19, with potential reduction of thrombosis, systemic inflammation, and perhaps antiviral effects. 9 However, given the bleeding hazards of therapeutic anticoagulation and the possibility of confounding and bias from observational studies, there is a critical need for high-quality interventional randomized trials evaluating the benefit-risk of different dosages of anticoagulation in patients with COVID-19. To inform the decision of anticoagulation for severe patients, at least 20 randomized trials were started early in the epidemic. 10 Several have been reported already and have brought some answers. In particular, large international multiplatform trials have shown a possibly differential effect of therapeutic anticoagulation over low-or intermediate-dose anticoagulation, with a benefit among noncritically ill but not among critically ill COVID-19 patients, with regard to a combined endpoint of survival and the need for organ support. 11, 12 Intermediate-dose anticoagulation was not beneficial, compared with low-dose anticoagulation, among critically ill patients in Iran in the INSPIRATION trial. 13 In noncritically ill patients, two other trials have suggested a benefit of therapeutic anticoagulation, with a reduction of a combined endpoint of VTE and mortality in the HEP-COVID trial 14 and with a reduction of mortality as a secondary outcome in the RAPID trial. 15 Three other trials of smaller sample sizes have not shown benefits of increased doses of anticoagulation in COVID-19 inpatients. [16] [17] [18] Here, we report the COVID-HEP study, a Swiss multicenter randomized trial comparing different doses of heparins during hospital stay. The COVID-HEP study is a multicenter, superiority, open-label, investigator-initiated, randomized controlled trial of therapeuticdose vs. low-or intermediate-dose anticoagulation for hospitalized patients with microbiologically proven severe COVID-19. The trial was set in two university hospitals (Geneva, Lausanne) and two nonuniversity hospitals (Sion, Locarno) in Switzerland. The local ethics committees of the centers approved the study and informed consent was obtained from all participants or their relatives, if a personal consent was unfeasible. The protocol and statistical analytic plan can be found as Supporting Information. Participants were randomly assigned (1:1) to therapeutic anticoagulation (high-dose) or low-/intermediate-dose anticoagulation, using a central, concealed, web-based randomization system. We used randomly selected block sizes between two and six, stratified by study center and severity (acutely ill medical vs. critically ill). Outcome adjudicators were blinded to the group allocation, but participants and investigators were not. Therapeutic anticoagulation was either enoxaparin 1 mg/kg twice daily, with anti-Xa assay monitoring in case of extreme weight (<50 kg or >100 kg) or renal clearance <50 ml/min, or therapeutic intravenous unfractionated heparin in case of renal clearance <30 ml/ min or upon physician preference, with anti-Xa unfractionated heparin (UFH) monitoring for dose titration. We used different dosages for prophylactic anticoagulation for acutely ill medical participants (hospitalized in medical wards) and for critically ill participants (hospitalized in intermediate care units or ICUs), in agreement with local practices and French recommandations. 19 Acutely ill medical participants received low-dose weight-based once-daily enoxaparin ( Study treatment was stopped at the first occurrence of the fol- The primary outcome was a composite of 30-day VTE, arterial thrombosis, disseminated intravascular coagulation (DIC), and allcause mortality. VTE was defined as an objective segmental or more proximal PE diagnosed on computed tomography pulmonary angiography or lung scintigraphy, and/or objectively diagnosed proximal lower extremity DVT, symptomatic distal lower extremity DVT, or noncatheter-related proximal upper extremity DVT. As such, subsegmental PE and asymptomatic distal DVT were excluded because of their uncertain clinical relevance. Arterial thrombosis was defined as acute myocardial infarction, acute ischemic stroke, or acute limb ischemia using objective imaging and biomarker criteria. We used the International Society on Thrombosis and Haemostasis (ISTH) definition for DIC. 20 The motivation to include DIC in the primary outcome followed early COVID-19 literature suggesting an important role in 10% to 30% of severe cases. 1, 5 Detailed definitions of DIC and the bleeding criteria are found in the Protocol (Supporting Information). Secondary outcomes included the individual components of the primary outcome, asymptomatic proximal DVT diagnosed through screening CUS, duration of hospitalization, of ICU stay, and of mechanical ventilation (MV) support. We also examined clinical deterioration among acutely ill medical, which we a priori defined as an admission to the intermediate/ICU, the use of invasive or noninvasive ventilation, the use of high-flow oxygen or the use of an inspired fraction of oxygen >50%. We had planned as secondary outcomes the occurrence of sepsis-induced coagulopathy, acute respiratory distress syndrome, the sequential organ failure assessment score, and respiratory ratios in the ICU, which were captured partially in the dataset and not analyzed or will be reported in a future manuscript. In a post hoc exploratory analysis, we also considered the risk of invasive MV or death at 30 days, among participants without invasive MV at baseline, as was done in other clinical trials. Safety outcomes were major bleeding and clinically relevant nonmajor bleeding (CRNMB), according to their ISTH definition, 21, 22 and confirmed heparin-induced thrombocytopenia, as defined by the 2020 French recommendations. 23 All thrombotic and bleeding events were independently and centrally adjudicated by two expert VTE clinicians, in a blinded fashion. We defined serious adverse events as severe events that were unexpected in severe or critically ill inpatients, excluding the primary and safety outcomes except fatalities. At the time of study design, we hypothesized an incidence of the composite primary outcome of 40% from early data from Wuhan. 24 We estimated a 50% relative reduction with therapeutic anticoagulation, which was associated with a 24% absolute lower mortality among patients with severe COVID-19. 1 The sample size of 200 (100 in each arm) was determined with an alpha of 5%, a power of 80%, and a 10% loss of follow-up. There was no formal interim statistical analysis. We suspended inclusions of ICU patients from December 30, 2020, until February 2, 2021, based on preliminary reports of the multiplatform study suggesting a lack of benefit of therapeutic anticoagulation in ICU patients, 11 and resumed enrollment following the advice of the Data and Safety Monitoring Board. On June 2, 2021, the Steering Committee decided to prematurely stop the trial because of low recruitment, in agreement with the data and safety monitoring board. At that time, 40% of the Swiss population had received at least one dose of mRNA vaccine, and hospitalization rates for COVID-19 had dropped to low levels, affecting the potential for recruitment dramatically. All primary and secondary efficacy outcomes were analyzed according to the intention-to-treat method. Safety outcomes were analyzed in a per-protocol fashion, with only participants who had received at least one appropriate dose of study drug. Outcomes were compared between study groups by using survival data analysis: Kaplan-Meier approach to estimate survival probabilities and Cox regression models to assess hazard ratios (HRs). Participants lost to follow-up or who had withdrawn consent were censored at the day of last news. The 30-day cumulative risks were reported with 95% confidence interval (CI) based on the complementary log-log transformation. 25 For outcomes with no observed event, 95% CIs were obtained by the Clopper-Pearson exact method and ignoring information of censored participants. The 95% CI of 30 days' risk difference between study groups was obtained by a parametric bootstrap approach. The HR for the primary outcome was adjusted on the baseline ward type and on a propensity score to account for potential imbalance in risk factors (age, sex, hypertension, body mass index [BMI], diabetes, active smoking, history of previous VTE, cardiovascular disease, chronic lung disease, chronic renal disease) between study groups. Adjustment was predefined, but simplified during the course of the study, before comparative analyses, in view of the small number of events (see Statistical Analytic Plan as Supporting Information). We also used survival data analysis to compare the times to hospital discharge (among the whole sample), to ICU discharge (among participants included in the ICU) and to MV weaning (among participants with MV at baseline). Based on descriptive analyses showing a low incidence of the primary outcome, we restricted the planned subgroups to acutely ill medical versus critically ill participants at baseline and baseline D-dimer levels with a cutoff of 4 times the norm, to align with other studies (<2000 vs. ≥2000 ng/ml). Missing data were not imputed. All statistical analyses were conducted on R software, version 4.0.2, and all statistical tests were two-sided with a significance level of 5%. Inclusion occurred at medians of 1 day after hospital admission and of 10 days after start of COVID-19 symptoms. Baseline characteristics were well-balanced between groups, with a mean age of 62.5 years and BMI of 28.6 kg/m 2 , and a majority of men (69.8%; Table 1 and Table S1 ). Participants were acutely ill medical (44.7%) and critically ill (55.3%) patients, and 52.2% were treated with high- Overall, 81% of participants had a compliance ≥80%. Mean compliances were 87% and 93% and median durations were 9 days (IQR 5.5-14) and 9 days (IQR 6-14.2), in the therapeutic anticoagulation F I G U R E 1 CONSORT flowchart of the study. Inclusion error is the result of the wrong randomization of a COVID-19 patient who had not been included in the study TA B L E 1 Baseline characteristics of the participants, stratified by randomization groups group and control group, respectively. There were 21/159 participants (13.2%) who did not receive the study drugs at the exact dose according to the protocol for >48 h during the study period, well balanced within both groups (12.5% and 13.9%). Two participants were excluded from per-protocol analyses in the therapeutic anticoagulation group because they did not receive any study drugs. The estimated median times to hospital discharge, ICU discharge, and MV weaning were 11 versus 10 days, 8.5 versus 7 days and 9 versus 9 days, in the therapeutic-dose group versus the control group (survival curves in Figure S1 ). Major bleeding occurred in three participants: 1/77 (1.4%) of the therapeutic-dose group versus 2/80 (2.5%) of the control group (Table 2) . All major bleeding events were nonfatal intracranial bleeding. Five CRNMBs occurred, all in the therapeutic-dose group, without difference in the risk of combined major and CRNMB between groups ( Figure 2 ). These included two gastrointestinal bleeding, two nasal bleeding, and one intramuscular hematoma (Table S2 ). There was no heparin-induced thrombocytopenia. All primary outcome events occurred in participants who were critically ill at baseline. No acutely ill medical inpatient suffered from a thrombotic event or died in either study group. There was no significant difference in the effect of therapeutic anticoagulation between groups of wards at admission and of baseline D-dimer, but with low power because of the small numbers of participants and events in subgroups (Table 4) . For bleeding outcomes, all major bleeding and 3/5 CRNMB occurred in participants who were critically ill at baseline. (Table S3) . None was related to study treatment. In this Swiss multicenter trial comparing different anticoagulation intensities in severe COVID-19, the combined risks of thrombotic outcomes, severe coagulopathy, and all-cause mortality at 30 days and demographic characteristics (greater prevalence of obesity and of Black race). Importantly, our study also differs by a stricter definition for VTE, which excluded subsegmental PE, asymptomatic distal DVT, and catheter-associated DVT and by a central adjudication process. We hypothesize several reasons for the good prognosis in our study. First, almost all participants received corticosteroids, with a greater proportion than the multiplatform studies, 11,12 the HEP-COVID, 14 Third, our trial is the first to report the proportion of exclusion of PE at the time of hospital admission, before study inclusion (72% of participants). This may be a critical explanation of the low VTE risk during the study. Because several reports have suggested that VTE mainly occurs before hospital admission, 28, 29 we hypothesize that some of the VTE found during the follow-up of other studies may already have been present at the time of study inclusion. Further, the open-label design of the COVID-19 anticoagulation trials is prone to diagnostic suspicion bias, with a higher likelihood of suspicion of VTE in the group receiving a lower anticoagulant dose, perhaps exaggerating differences in VTE rates between groups. In other words, our trial mostly tested the effect of anticoagulation In particular, this meta-analysis will increase the precision of the possibly beneficial effect of therapeutic anticoagulation in medical COVID-19 patients and add power to detect effect modification in specific subgroups to better inform the decision of anticoagulation dose in COVID-19 inpatients. The yield of screening for asymptomatic proximal DVT was null in our trial and should not be generalized in clinical practice. This is in agreement with previous prospective studies, the HEP-COVID, and X-COVID-19 trials showing a low prevalence of asymptomatic proximal DVT of 0% to 2.9% among acutely ill medical patients. 14, 17, [30] [31] [32] In the ICU, we did not find the elevated (5%-23%) risks found by previous prospective studies. [33] [34] [35] [36] [37] Our trial has several strengths. We conducted active follow-up after hospital discharge, until 1 month. The use of corticosteroids with a low number of events and an overestimated effect size in reports from the early pandemic phase, the study power to observe a difference was low. Although no signal of a numeric difference between groups was observed, our null results are not precise enough to be conclusive. In conclusion, the occurrence of mortality, severe coagulopathy, and thrombotic outcomes was low in this trial of patients with severe F I G U R E 2 Survival curves (Kaplan-Meier estimates) of (A) the primary efficacy outcome (intention-to-treat analysis) and (B) the safety outcome major or clinically relevant bleeding (per protocol analysis). Censored data are represented by crosses. In the analysis of the safety outcome, follow-up of patients who died was censored at the time of death Marc Blondon drafted the manuscript. All authors revised the manuscript critically for important intellectual content, approved the version to be published, had full access to the data and accepted responsibility to submit for publication. Marc Blondon is the guarantor of the content of the manuscript, including the data and analysis. This work was funded with a grant from the Private Foundation of the Geneva University Hospitals, which had no role in the design, data collection, analysis, interpretation of the data, or in the writing of the report and the decision to submit the paper for publication. Hospital and Faculty of Medicine, Geneva. We thank Cécile Guillot, Tamara Mann, Pauline Gosselin, Lucie Altheer, Floriane Le Petit, Louise Riberdy, Anne-Claude Hugi We warmly thank also the members of the data and safety monitoring board and of the adjudication committee. Predefined clinical deterioration and its components among acute medical participants at baseline (intention-to-treat analysis) Anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China Proximal deep vein thrombosis and pulmonary embolism in COVID-19 patients: a systematic review and meta-analysis Risk of venous thromboembolism in patients with COVID-19: a systematic review and meta-analysis Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia Anticoagulation, bleeding, mortality, and pathology in hospitalized patients with COVID-19 Association of treatment dose anticoagulation with in-hospital survival among hospitalized patients with COVID-19 Venous thromboembolism in COVID-19: systematic review of reported risks and current guidelines Thromboprophylaxis strategies to improve the prognosis of COVID-19 Anticoagulant interventions in hospitalized patients with COVID-19: A scoping review of randomized controlled trials and call for international collaboration Therapeutic anticoagulation with heparin in critically ill patients with Covid-19 Therapeutic anticoagulation with heparin in noncritically ill patients with Covid-19 Effect of intermediatedose vs standard-dose prophylactic anticoagulation on thrombotic events, extracorporeal membrane oxygenation treatment, or mortality among patients with COVID-19 admitted to the intensive care unit: the INSPIRATION randomized clinical trial Efficacy and safety of therapeutic-dose heparin vs standard prophylactic or intermediatedose heparins for thromboprophylaxis in high-risk hospitalized patients with COVID-19: The HEP-COVID randomized clinical trial Effectiveness of therapeutic heparin versus prophylactic heparin on death, mechanical ventilation, or intensive care unit admission in moderately ill patients with covid-19 admitted to hospital: RAPID randomised clinical trial Standard prophylactic versus intermediate dose enoxaparin in adults with severe COVID-19: a multi-center, open-label, randomized controlled trial Enoxaparin for thromboprophylaxis in hospitalized COVID-19 patients: The X-COVID-19 Randomized Trial Therapeutic versus prophylactic bemiparin in hospitalized patients with nonsevere COVID-19 pneumonia (BEMICOP Study): an open-label multicenter randomized controlled trial Prevention of thrombotic risk in hospitalized patients with COVID-19 and hemostasis monitoring Diagnosis and management of sepsis-induced coagulopathy and disseminated intravascular coagulation Subcommittee on Control of Anticoagulation. Definition of clinically relevant non-major bleeding in studies of anticoagulants in atrial fibrillation and venous thromboembolic disease in non-surgical patients: communication from the SSC of the ISTH Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients Diagnosis and management of heparin-induced thrombocytopenia Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study A note on confidence intervals and bands for the survival function based on transformations Therapeutic versus prophylactic anticoagulation for patients admitted to hospital with COVID-19 and elevated D-dimer concentration (ACTION): an open-label, multicentre, randomised, controlled trial Association between administration of systemic corticosteroids and mortality among critically ill patients with COVID-19: a meta-analysis Prevalence of pulmonary embolism in patients with COVID-19 at the time of hospital admission Venous and arterial thromboembolic complications in COVID-19 patients admitted to an academic hospital in Incidence of asymptomatic deep vein thrombosis in patients with COVID-19 pneumonia and elevated D-dimer levels Systematic screening for a proximal DVT in COVID-19 hospitalized patients: Results of a comparative study Duplex ultrasound screening for deep and superficial vein thrombosis in COVID-19 patients Point of care ultrasound screening for deep vein thrombosis in critically ill COVID-19 patients, an observational study Thromboembolic events and role of point of care ultrasound in hospitalized Covid-19 patients needing intensive care unit admission Venous thromboembolism in critically Ill patients with COVID-19: results of a screening study for deep vein thrombosis Venous thrombosis among critically ill patients with coronavirus disease 2019 (COVID-19) High prevalence of deep vein thrombosis in mechanically ventilated COVID-19 patients All authors declare no competing interests.