Factors associated with large watery stools after out-of-hospital cardiac arrest and their relationship with neurological outcomes: A retrospective observational study

Aim

To evaluate the association between large watery stools—a potential early sign of non-occlusive mesenteric ischaemia—and neurological outcomes.

Methods

We retrospectively analysed data from 495 adult patients with out-of-hospital cardiac arrest admitted to our tertiary centre between April 2015 and March 2024. Individuals who achieved return of spontaneous circulation but remained comatose after resuscitation were included. Large watery stools were defined as ≥300 mL of watery/loose stools occurring at least twice within 24 h after admission. We performed stepwise logistic regression analysis to identify predictors available at admission and assess the relationship between watery stool occurrence and neurological outcomes.

Results

Overall, 161 (32%) patients developed large watery stools within 24 h. Patients with watery stools experienced significantly higher rates of unfavourable neurological outcomes at discharge (favourable in 9% vs. 21% in patients without watery stools). Stepwise logistic regression analysis identified six independent predictors of large watery stools—lactate, low-flow interval, C-reactive protein, activated partial thromboplastin time, noradrenaline use, and creatinine—reflecting disturbances in metabolic status, perfusion, coagulation, and inflammatory responses. The final model demonstrated an ROC_AUC of 0.72 (95% CI [0.68–0.73]). After calibration, the Brier score improved from 0.21 (95% CI [0.20–0.23]) to 0.19 (95% CI [0.17–0.21]).

Conclusion

After out-of-hospital cardiac arrest, early-onset large watery stools is strongly associated with poor neurological outcomes, potentially serving as a clinical indicator of intestinal ischaemia and systemic inflammation. Incorporating coagulation and metabolic markers into predictive models may facilitate early identification of high-risk patients, aiding timely diagnosis and intervention.