You're using an outdated browser. Please upgrade to a modern browser for the best experience.
Please note this is a comparison between Version 1 by Stefanie Reis and Version 2 by Yvaine Wei.
COVID-19 patients are at high thrombotic risk. Moderately affected COVID-19 patients may benefit from therapeutic-dose anticoagulation, but the risk for bleeding is increased.
- bleeding
- death
- COVID-19
- thrombosis
- anticoagulant therapy
- SARS-CoV-2
1. Introduction
In its severe form, COVID-19, the clinical manifestation associated with SARS-CoV-2 infection, is characterized by respiratory failure and high rates of thromboembolic complications [1]. Procoagulant markers, such as elevated D-Dimers are now widely accepted as prognostic factors for severe disease progression [2][3][2,3].
A recent systematic review on viscoelastic techniques, namely thromboelastography and thromboelastometry, showed that severe COVID-19 is further associated with fibrinolysis shutdown and hyperfibrinogenaemia, despite the use of appropriate thromboprophylaxis [4][6]. Given the pro-coagulant status and increased thrombotic risk of COVID-19, the question remains whether intensified prophylactic anticoagulation with intermediate or therapeutic doses can decrease the risk of disease progression, clinical worsening or death without increasing the risk for adverse events, such as major bleedings [5][7].
2. Intermediate-Dose Anticoagulation
INSPIRATION [6][7][21,27] and Perepu 2021 [7][27] were included in the comparison of intermediate-dose anticoagulation versus standard thromboprophylaxis with low-dose anticoagulation (Table 12). Intermediate-dose anticoagulation compared to standard thromboprophylaxis in patients with moderate to severe COVID-19 may have little or no effect on all-cause mortality at 30 days (RR 0.98, 95% CI 0.74–1.32, 763 participants, two studies, low-certainty evidence) and 90 days (RR 1.07, 95% CI 0.89–1.28, 590 participants, one study, low-certainty evidence). Intermediate-dose anticoagulation compared to standard thromboprophylaxis may have little or no effect within 30 days on any thrombotic events or death (RR 1.03, 95% CI 0.86–1.24, 590 participants, one study, low-certainty evidence) and on any thrombotic events (RR 0.99, 95% CI 0.51–1.96, 763 participants, two studies, low-certainty evidence). Intermediate-dose anticoagulation may increase major bleedings compared to standard thromboprophylaxis (RR 1.48, 95% CI 0.53–4.15, 763 participants, 2 studies, low-certainty evidence). Certainty of evidence was downgraded for all outcomes due to serious risk of bias and serious imprecision (Table 12).Table 12. Meta-analyses for intermediate-dose anticoagulation versus standard thromboprophylaxis, including certainty of evidence.
| Outcome | Study Population * | Risk Ratio (M–H, Random, 95% CI) | Risk Ratio (M–H, Fixed, 95% CI) | Heterogeneity | Certainty of Evidence | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All-cause mortality at 30 days | Pooled effect, mixed population (WHO 4–9), 763 participants, 2 studies [8][7][9] | Pooled effect, mixed population (WHO 4–9), 763 participants, 2 studies [22,27,28] | 0.98 (0.74, 1.32) | 1.01 (0.84, 1.21) | Tau | 2 | = 0.02; Chi | 2 | = 1.28, df = 1 (p = 0.26); I | 2 | = 22% | Low-certainty evidence due to serious risk of bias and imprecision |
| All-cause mortality at 90 days | Mixed population (WHO 4–9), 590 participants, 1 study [8][9] | Mixed population (WHO 4–9), 590 participants, 1 study [22,28] | 1.07 (0.89, 1.28) | 1.07 (0.89, 1.28) | NA | Low-certainty evidence due to serious risk of bias and imprecision | ||||||
| Any thrombotic event or death up to 30 days | Mixed population (WHO 4–9), 590 participants, 1 study [8][9] | Mixed population (WHO 4–9), 590 participants, 1 study [22,28] | 1.03 (0.86, 1.24) | 1.03 (0.86, 1.24) | NA | Low-certainty evidence due to serious risk of bias and imprecision | ||||||
| Any venous thrombotic event up to 30 days | Pooled effect, mixed population (WHO 4–9), 763 participants, 2 studies [8][7][9] | Pooled effect, mixed population (WHO 4–9), 763 participants, 2 studies [22,27,28] | 0.99 (0.51, 1.96) | 0.99 (0.50, 1.95) | Chi | 2 | = 0.13, df = 1 (p = 0.72); I | 2 | = 0% | Low-certainty evidence due to serious risk of bias and imprecision | ||
| Major bleeding up to 28 days | Pooled effect, mixed population (WHO 4–9), 763 participants, 2 studies [8][7][9] | Pooled effect, mixed population (WHO 4–9), 763 participants, 2 studies [22,27,28] | 1.48 (0.53, 4.15) | 1.49 (0.53, 4.14) | Tau | 2 | = 0.00; Chi | 2 | = 0.23, df = 1 (p = 0.63); I | 2 | = 0% | Low-certainty evidence due to serious risk of bias and imprecision |
M–H, Mantel–Haenszel; CI, confidence interval. * Patient status according to WHO clinical progression scale.
3. Therapeutic-Dose Anticoagulation
Six studies [10][11][12][13][14][15][23,24,25,26,29,30] were included in the comparison of therapeutic-dose anticoagulation versus standard thromboprophylaxis with low- or intermediate-dose anticoagulation (Table 23). Therapeutic-dose anticoagulation compared to standard thromboprophylaxis may decrease all-cause mortality at 28 days for moderate COVID-19 patients (RR 0.23, 95% CI 0.08–0.67, 465 participants, one study, low-certainty evidence). TWe are uncertain on the effect in severe COVID-19 patients are uncertained (RR 0.33, 95% CI 0.04–2.69, 20 participants, one study, very low-certainty evidence). In studies with mixed COVID-19 population, therapeutic-dose anticoagulation may have little or no effect on all-cause mortality at 28 days (RR 1.07, 95% CI 0.56–2.03, 867 participants, two studies, low-certainty evidence). Contrary to effects for all-cause mortality at 28 days, therapeutic-dose anticoagulation may have little or no effect on in-hospital mortality (RR 0.97, 95% CI 0.79–1.19, 3344 participants, three studies, low-certainty evidence) in all hospitalised COVID-19 patients. Subgroup analysis showed similar results irrespective of disease severity.Table 23. Meta-analyses for therapeutic-dose anticoagulation according to pre-specified subgroups (moderate population and severe population) including certainty of evidence.
| Outcome | Study Population * | Risk Ratio (M–H, Random, 95% CI) | Risk Ratio (M–H, Fixed, 95% CI) | Heterogeneity | Certainty of Evidence | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All-cause mortality (28 days) | Moderately diseased population (WHO 4–5), 465 participants, 1 study [14] | Moderately diseased population (WHO 4–5), 465 participants, 1 study [29] | 0.23 (0.08, 0.67) | 0.23 (0.08, 0.67) | NA | Low-certainty evidence due to very serious imprecision | ||||||
| Severely diseased population (WHO 6–9), 20 participants, 1 study [12] | Severely diseased population (WHO 6–9), 20 participants, 1 study [25] | 0.33 (0.04, 2.69) | 0.33 (0.04, 2.69) | NA | Very low-certainty evidence due to risk of bias and very serious imprecision | |||||||
| Mixed population (WHO 4–9), 867 participants, 2 studies [13][15] | Mixed population (WHO 4–9), 867 participants, 2 studies [26,30] | 1.07 (0.56, 2.03) | 1.08 (0.77, 1.51) | Tau | 2 | = 0.16; Chi | 2 | = 3.54, df = 1 (p = 0.06); I | 2 | = 72% | Low-certainty evidence due to serious heterogeneity and imprecision | |
| Pooled effect, mixed population (WHO 4–9), 1352 participants, 4 studies [12][13][14][15] | Pooled effect, mixed population (WHO 4–9), 1352 participants, 4 studies [25,26,29,30] | 0.68 (0.32, 1.45) | 0.85 (0.62, 1.16) | Tau | 2 | = 0.38; Chi | 2 | = 11.47, df = 3 (p = 0.009); I | 2 | = 74% | Low-certainty evidence due to serious heterogeneity and imprecision | |
| All-cause mortality in hospital | Pooled effect, mixed population (WHO 4–9), 3344 participants, 3 studies [10][11][12] | Pooled effect, mixed population (WHO 4–9), 3344 participants, 3 studies [23,24,25] | 0.97 (0.79, 1.19) | 0.99 (0.86, 1.13) | Tau | 2 | = 0.01; Chi | 2 | = 2.78, df = 2 (p = 0.25); I | 2 | = 28% | Low-certainty evidence due to serious indirectness and risk of bias |
| Worsening of clinical status: Progression to intubation or death (28 days) | Moderately diseased population (WHO 4–5), 2231 participants, 1 study [11] | Moderately diseased population (WHO 4–5), 2231 participants, 1 study [24] | 0.90 (0.72, 1.14) | 0.90 (0.72, 1.14) | NA | Low-certainty evidence due to serious indirectness and risk of bias | ||||||
| Worsening of clinical status: Progression to any mechanical ventilation or death (28 days) | Moderately diseased population (WHO 4–5), 465 participants, 1 study [14] | Moderately diseased population (WHO 4–5), 465 participants, 1 study [29] | 0.63 (0.39, 1.02) | 0.63 (0.39, 1.02) | NA | Low-certainty evidence due to very serious imprecision | ||||||
| Improvement of clinical status: participants discharged alive without clinical deterioration or death at 28 days | Mixed population (WHO 4–9), 614 participants, 1 study [13] | Mixed population (WHO 4–9), 614 participants, 1 study [26] | 0.96 (0.90, 1.02) | 0.96 (0.90, 1.02) | NA | High-certainty evidence | ||||||
| Improvement of clinical status: survival until hospital discharge without receiving organ support | Moderately diseased population (WHO 4–5), 2219 participants, 1 study [11] | Moderately diseased population (WHO 4–5), 2219 participants, 1 study [24] | 1.05 (1.00, 1.10) | 1.05 (1.00, 1.10) | NA | Low-certainty evidence due to serious indirectness and risk of bias | ||||||
| Any thrombotic event or death | Moderately diseased population (WHO 4–5), 2396 participants, 2 studies [11][15] | Moderately diseased population (WHO 4–5), 2396 participants, 2 studies [24,30] | 0.64 (0.38, 1.07) | 0.72 (0.57, 0.91) | Chi | 2 | = 2.90, df = 1 (p = 0.09); I | 2 | = 66% | Low-certainty evidence due to serious risk of bias and indirectness/heterogeneity | ||
| Severely diseased population (WHO 6–9), 1174 participants, 2 studies [10][15] | Severely diseased population (WHO 6–9), 1174 participants, 2 studies [23,30] | 0.98 (0.86, 1.12) | 0.98 (0.86, 1.12) | Chi | 2 | = 0.09, df = 1 (p = 0.77); I | 2 | = 0% | Low-certainty evidence due to serious risk of bias and indirectness | |||
| Mixed population (WHO 4–9), 614 participants, 1 study [13] | Mixed population (WHO 4–9), 614 participants, 1 study [26] | 1.03 (0.70, 1.50) | 1.03 (0.70, 1.50) | NA | Low-certainty evidence due to serious risk of bias and imprecision | |||||||
| Pooled effect, mixed population (WHO 4–9), 4184 participants, 4 studies [10][11][13][15] | Pooled effect, mixed population (WHO 4–9), 4184 participants, 4 studies [23,24,26,30] | 0.86 (0.71, 1.06) | 0.90 (0.80, 1.01) | Chi | 2 | = 8.61, df = 4 (p = 0.07); I | 2 | = 54% | Low-certainty evidence due to serious risk of bias and indirectness/heterogeneity | |||
| Any thrombotic event | Pooled effect, mixed population (WHO 4–9), 4669 participants, 6 studies [10][11][12][13][14][15] | Pooled effect, mixed population (WHO 4–9), 4669 participants, 6 studies [23,24,25,26,29,30] | 0.58 (0.45, 0.74) | 0.57 (0.45, 0.73) | Tau | 2 | = 0.00; Chi | 2 | = 4.68, df = 5 (p = 0.46); I | 2 | = 0% | Moderate-certainty evidence due to serious risk of bias |
| Major bleeding at 28 days | Pooled effect, mixed population (WHO 4–9), 4650 participants, 5 studies [10][11][13][14][15] | Pooled effect, mixed population (WHO 4–9), 4650 participants, 5 studies [23,24,26,29,30] | 1.78 (1.15, 2.74) | 1.82 (1.19, 2.78) | Tau | 2 | = 0.00; Chi | 2 | = 3.95, df = 5 (p = 0.56); I | 2 | = 0% | Low-certainty evidence due to serious indirectness and risk of bias |
M–H, Mantel–Haenszel; CI, confidence interval. * Patient status according to WHO clinical progression scale.
4. Efficacy and Limitations of Intermediate- and Therapeutic-Dose Anticoagulation
Therapeutic-dose anticoagulation decreased the rate of thrombotic events for all patients independent of disease severity. This outcome was reported in all studies. However, the composite outcome of any thrombotic event or death treating the individual outcomes as competing risks, revealed a reduced risk due to therapeutic anticoagulation only for patients with moderate COVID-19. Additionally, this effect was only statistically significant when using the fixed-effects model. In contrast, for patients with severe COVID-19, there was no effect due to therapeutic anticoagulation shown in the meta-analysis. The effect could be explained by recent findings on COVID-19 associated coagulopathy, suggesting a pro-coagulant state only at the beginning of the infection, which transforms into disseminated intravascular coagulation with increased haemorrhagic risk as the disease progresses [16][36].
Irrespective of disease severity, therapeutic-dose anticoagulation showed a non-significantly higher rate of major bleeding events in the subgroups, and the pooled effect with a sufficient number of events and patients reached statistical significance. Therefore, the risk for bleeding should be taken into account in decision making and anticoagulated COVID-19 patients should be carefully monitored for bleeding events.
Despite this meta-analysis including evidence from 5580 participants, the overall certainty of evidence for intensified thromboprophylaxis in hospitalised patients with COVID-19 remains low. Limitations of the evidence base are the conflicting results and lack of evidence that would warrant high certainty, due to the wide heterogeneity of study settings, populations, and therapeutic approaches.