Author: Young May Cha, MD
IARS and SOCCA 2026 Annual Meeting coverage
Although patients frequently worry about surgical risk, formal risk-assessment tools are not routinely used by many anesthesiologists and surgeons. Experts at the 2026 Annual Meeting discussed how individualized risk prediction can guide treatment, improve informed consent, support shared decision-making, and potentially improve perioperative outcomes.
Rafael Noriega, MD, MPH, reviewed several commonly available risk-assessment tools. Each provides different information and has important limitations.
The American College of Surgeons National Surgical Quality Improvement Program risk calculator provides individualized, procedure-specific estimates of postoperative complications. Frailty assessments are not procedure-specific but may identify patients at increased risk for poor outcomes that other calculators overlook.
POSSUM and Portsmouth-POSSUM can predict postoperative morbidity and mortality but require intraoperative information, limiting their usefulness during the preoperative period.
The ASA Physical Status classification and Revised Cardiac Risk Index are easier to calculate at the bedside. However, neither is procedure-specific, and the Revised Cardiac Risk Index evaluates only cardiac complications. It may underestimate risk in today’s older and more medically complex surgical population.
Biomarkers that may improve risk assessment include:
- B-type natriuretic peptide
- Troponin
- Cystatin C
- Procalcitonin
Unfortunately, these tests are not consistently available at every institution. No single tool captures every relevant risk, so anesthesiologists may need to combine procedure-specific calculators, frailty assessments, clinical findings, and biomarkers.
Benjamin Houseman, MD, PhD, FASA, discussed risk prediction as part of shared decision-making. Patients generally want to understand their surgical risks, and many would consider prehabilitation or other interventions if those measures could reduce complications.
Meaningful risk communication remains difficult. Patients may struggle to recall information because of anxiety, cognitive overload, limited health literacy, or language barriers. Despite requirements for individualized informed consent, fewer than one in 20 anesthesiologists document specific risks of organ injury or death.
Additional barriers include:
- Limited consultation time
- The anesthesiologist’s late involvement in surgical planning
- A task-oriented perioperative culture
- Complex risk information
- Inconsistent documentation practices
Shared decision-making should be a continuing process that begins before surgery and extends through hospitalization and recovery after discharge.
Emerging technologies may improve patient education. Virtual reality-assisted consent has been well received and may reduce the time specialists need to explain procedures. Artificial intelligence systems, including ChatGPT and Gemini, have also demonstrated an ability to explain general anesthesia complications in understandable language.
Individualized risk calculators could be embedded directly into electronic health records. Documenting and discussing risk could then be tracked as a quality measure, encouraging clinicians to incorporate formal assessment into routine practice.
Kamal Maheshwari, MD, MPH, FASA, emphasized that calculating risk is not enough. The information must lead to changes in treatment that improve outcomes.
More comprehensive predictive models could combine:
- Electronic health record information
- Wearable-device data
- Genetic information
- Laboratory results
- Procedure-specific risks
- Patient preferences and recovery goals
Large language models may eventually synthesize these data into individualized recommendations. These systems could help identify patients who would benefit from prehabilitation, specialized intraoperative management, extended hospitalization, or remote monitoring after discharge.
Risk models are also expanding beyond mortality and cardiac events to predict complications such as acute kidney injury and outcomes associated with specific procedures.
Key Takeaways
Preoperative risk assessment should not be treated as a label placed on the patient. It should guide shared decision-making and lead to meaningful changes in perioperative care.
No single calculator is sufficient for every patient. Anesthesiologists should combine validated tools with clinical judgment, frailty assessment, biomarkers, procedural risk, and the patient’s individual goals.
Risk information must be communicated in a way patients can understand and use when deciding whether to proceed with surgery or pursue risk-reduction strategies.
Embedding individualized calculators into electronic health records may improve documentation and encourage routine use. Artificial intelligence may further personalize prehabilitation, intraoperative care, and postoperative monitoring.
The ultimate purpose of risk prediction is not merely to forecast complications. It is to identify opportunities to prevent them and improve the patient’s quality of recovery.
Thank you to IARS and SOCCA for allowing us to summarize this important coverage from the 2026 Annual Meeting.