INFORMING INTENSIVE CARE UNIT DIGITAL TWINS: DYNAMIC ASSESSMENT OF CARDIORESPIRATORY FAILURE TRAJECTORIES IN PATIENTS WITH SEPSIS.

Summary

Using unsupervised clustering on 19,177 ICU sepsis patients, the authors identified four robust 14-day cardiorespiratory trajectories—two recovery and two high-mortality decline patterns—separable by comorbidity and severity indices, offering a framework for prognostication and digital twin decision support.

Key Findings

• Four distinct 14-day trajectories: fast recovery (27%, mortality 3.5%), slow recovery (62%, mortality 3.6%), fast decline (4%, mortality 99.7%), delayed decline (7%, mortality 97.9%).
• Trajectories were distinguished by Charlson Comorbidity Index, APACHE III, and day 1/3 SOFA (P<0.001).
• Findings underpin prediction modeling and digital twin decision support tools for sepsis in ICU.

Clinical Implications

Early classification into recovery vs decline trajectories may inform goals-of-care discussions, escalation/de-escalation of organ support, and ICU resource allocation.

Limitations

• Retrospective single health system; generalizability to other systems requires external validation
• Potential residual confounding and unmeasured treatment effects influencing trajectories

Future Directions

Prospective validation with real-time trajectory assignment; integrate biologic markers and treatment policies to enable adaptive digital twin simulations and interventional testing.