Anticoagulation in acute mesenteric ischemia
Survival benefit observed with early anticoagulation in arterial AMI, not in NOMI
Full Title: Anticoagulation management and outcomes in critically ill patients with acute mesenteric ischemia: an international study
Authors: Lakbar, I., Delamarre, L., Tamme, K. et al.
Article Type: Observational Study (combining a retrospective single-center cohort and a prospective multicenter observational cohort)
Clinical Question: Does initiating full-dose anticoagulation within 48 hours improve 30-day survival in critically ill patients with acute mesenteric ischemia (AMI), irrespective of etiology, surgical management, or coagulation status?
Background & Rationale: AMI is a critical condition with low survival rates, and early full-dose anticoagulation therapy is widely recommended. However, there’s very little high-quality data to back that up—particularly in ICU patients, who often complicate things with surgical needs, organ failure, and bleeding risk. Non-occlusive mesenteric ischemia (NOMI) makes things even trickier, with no obvious clot to treat. Most prior work in this space has been small, single-centre, or focused on perioperative aspects, not on what we do with these patients day-to-day in the ICU. These complexities highlight the need for data on the initiation timing of full-dose anticoagulation in this population.
Setting: The study combined data from two cohorts:
MOSAIC: a 15-year single-centre retrospective cohort from Saint Eloi Hospital, Montpellier, France (147 ICU patients)
AMESI: a prospective, 32-site international ICU cohort (223 patients) conducted over a 10-month period (June 2022 to April 2023).
Population: A total of 370 patients with AMI admitted to the ICU were included in the analysis. This comprised 147 patients from the MOSAIC cohort and 223 patients from the AMESI cohort. Patients were grouped by whether they received early full-dose anticoagulation (within 48 hours of diagnosis; n=183) or not (n=187).
Inclusion Criteria: AMI diagnosis either on ICU admission or during stay. ICU admission required for both cohorts.
Exclusion Criteria:
AMI not confirmed on final diagnosis,
Secondary AMI due to strangulated bowel disease,
Missing consent,
Not admitted to ICU,
Missing anticoagulation data in final analysis.
Subgroup Definitions: analysis stratified by AMI subtype (arterial, venous, NOMI, and miscellaneous).
Intervention: Full-dose anticoagulation within 48 hours of AMI diagnosis.
Comparator: No full-dose anticoagulation within that window. This group included everything from prophylactic dosing to delayed or absent treatment. No protocolised approach; practice varied by centre.
Outcomes Measured:
Primary Outcome: 30-day survival.
30-day survival was 53.5% (98/183 patients) in the early full-dose anticoagulation group compared to 41.7% (78/187 patients) in the non-early group (HR 0.69, 95% CI [0.52–0.92], p = 0.01).
In patients with arterial AMI (n=204), early full-dose anticoagulation was associated with improved 30-day survival in multivariate analysis (aHR 0.55, 95% CI [0.35–0.87], p=0.01).
In patients with NOMI (n=83), early full-dose anticoagulation was not associated with 30-day survival in either univariate or multivariate analysis (aHR 0.77, 95% CI [0.36–1.67], p=0.52).
For venous AMI (n=39), subgroup analysis was not possible due to insufficient sample size in the non-early group.
Secondary Outcomes: Duration of mechanical ventilation, ICU length of stay, occurrence of hemorrhagic complications (including digestive hemorrhage, any non-digestive hemorrhagic events, and blood transfusions), and 90-day survival. Notably, data on hemorrhagic complications were only available in the MOSAIC cohort.
The improved survival observed at 30 days in patients receiving early full-dose anticoagulation persisted at 90 days (p = 0.02).
Early full-dose anticoagulation was associated with a longer duration of mechanical ventilation.
The median duration was 8.75 days (± 17.5) in the early full-dose anticoagulation group compared to 5.94 days (± 9.06) in the non-early group (IRR 1.47, 95% CI [1.09–1.99], p = 0.01).
This observation may reflect differences in the clinical trajectory of survivors, or residual confounding. Immortal time bias could also be considered although the absence of a longer ICU stay in this group makes this explanation uncertain.
No difference in ICU LoS.
No statistically significant differences in the occurrence of hemorrhagic complications.
In the MOSAIC cohort, hemorrhage under anticoagulation occurred in 21.2% of patients receiving early full-dose anticoagulation and 17.9% of non-early patients, showing no statistical difference (p = 0.794).
The lack of data on hemorrhagic complications across the full cohort meant this secondary outcome was potentially underpowered.
Authors’ Conclusions: This study suggests a significant survival benefit of early full-dose anticoagulation in ICU patients with acute mesenteric ischemia and no difference in hemorrhagic complications. Early full-dose anticoagulation and revascularization and/or bowel resection were associated with survival.
Funding and Conflicts: Backed by the Estonian Research Council. Several authors disclosed presentation fees (Viatris, Medtronic, Dräger, etc.). No commercial ties to specific anticoagulants. Jaber is editor-in-chief of Intensive Care Medicine.
⸻
Claim 1: Early full-dose anticoagulation provides a significant survival benefit in ICU patients with acute mesenteric ischemia (AMI).
The study demonstrates a statistically significant association between early full-dose anticoagulation and improved 30-day survival (NNT=8, p=0.01) in the overall cohort, which persisted at 90 days (p=0.02). However, this association was not uniform across all AMI subtypes: it was observed in arterial AMI but not in NOMI.
Recommendation: Early full-dose anticoagulation for ICU patients with arterial AMI where clinically feasible and bleeding risks are appropriately managed.
Claim 2: Early full-dose anticoagulation is not associated with an increase in hemorrhagic complications.
The study found no statistically significant differences in hemorrhagic complications. However, this finding is significantly limited because data on hemorrhagic complications were only available for the MOSAIC cohort, which represents a minority (147/370 patients) of the total study population. This significantly reduces the power and generalizability of this specific conclusion.
Claim 3: Revascularization and/or bowel resection are associated with survival.
The study identified revascularization and/or bowel resection as an intervention therapy independently associated with 30-day survival in multivariate analysis.
Not thoroughly explored in this paper.
Recommendation: Consider revascularization (surgical or endovascular) and/or bowel resection as a key therapeutic component in AMI management when clinically indicated.
For the joint cohort that included AMESI patients, early full-dose anticoagulation was associated with a survival benefit at both 30 and 90 days, with no observed increase in hemorrhagic complications (though data for this specific outcome was limited to the MOSAIC cohort). A longer duration of mechanical ventilation was noted in the early anticoagulation group. Despite several limitations, the authors highlight the total cohort size of 370 patients as the largest published for ICU patients with AMI, which they argue enhances generalizability.
⸻
Critique
As an observational study, there was no randomization or blinding. This design inherently introduces a high risk of bias, particularly from confounding by indication. The “early” group may simply have been less sick, had fewer coagulopathies, or were selected for anticoagulation because they looked more salvageable.
While the overall cohort size of 370 patients is presented as the largest in this field, enhancing generalizability, it proved insufficient for robust subgroup analyses in certain AMI subtypes (e.g., venous AMI). The limited data on hemorrhagic complications (available only in the MOSAIC cohort) meant this secondary outcome was likely underpowered.
Patient management decisions were made based on local practices and clinical judgment, rather than a standardized protocol. Observed differences between cohorts (MOSAIC having higher Charlson scores and more NOMI patients) highlight potential confounding. Similarly, “full-dose anticoagulation” was defined locally. Without standardized protocols, there’s a lot of wiggle room—and potential misclassification. Differences in local practice, drug choice, dosing, and monitoring dilute consistency.
Nevertheless, the overall finding of a significant association between early anticoagulation and improved 30-day survival (p=0.01) presents a positive signal.
⸻
Barriers to Implementation
Physicians may rightfully hesitant to initiate full-dose anticoagulation, especially in critically ill patients with multi-organ failure, severe coagulation disorders, or those undergoing/recovering from surgery, due to concerns about bleeding risk.
Then there’s the uncertainty of harm: while the study reported no difference in hemorrhagic complications, the data for this outcome were very limited (only from the MOSAIC cohort). This means potential harm is not definitively ruled out, which is a significant concern for clinicians and may limit aggressive implementation.
⸻
Our position
This large international observational study offers useful insights into the management of acute mesenteric ischemia in critically ill patients—but we should interpret its findings with healthy caution.
The headline result—a survival benefit with early full-dose anticoagulation—is compelling, particularly in patients with arterial AMI. That aligns with current practice in many units, where early anticoagulation is already considered standard for thrombotic occlusion. However, the benefit wasn’t seen in NOMI, and the study couldn’t evaluate venous AMI adequately due to small sample size. That’s a key limitation: the overall benefit seems driven mainly by one subgroup, and we shouldn’t generalize it across all AMI types.
The safety signal—no increase in hemorrhagic complications—is reassuring on paper, but we need to keep in mind that this was based only on a minority subset, not the full population. So we can’t be fully confident the risk is negligible, especially in surgical or coagulopathic patients.
Despite those caveats, this is the largest dataset to date on ICU-managed AMI, and it reflects real-world practice across a broad international cohort.
In practice, if you have a patient with arterial AMI, no contraindications, and you’re considering early anticoagulation—this study supports that decision. For venous AMI or NOMI, the data just aren’t strong enough to draw conclusions, therefore anticoagulation remains a case-by-case call. Revascularization or bowel resection remains a cornerstone of care and was also independently associated with survival.