Postoperative ICU admission following major elective surgery: A nationwide inpatient database study

Whether the routine use of the ICU after major elective surgery improves postoperative outcomes is not well established.

To describe the association between use of postoperative ICU admission and clinical outcomes for patients undergoing major elective surgery.

Observational study.

Nationwide inpatient database in Japan, July 2010 to March 2018.

Patients undergoing one of 15 major elective orthopaedic, gastrointestinal, neurological, thoracic or cardiovascular surgical procedures.

ICU admission on the day of surgery. ICU was defined as a separate unit providing critical care services with around-the-clock physician staffing and nursing, the equipment necessary for critical care and a nurse-to-patient ratio at least one to two.

In-hospital mortality. Patient-level and hospital-level analyses were performed.

Overall, 2 011 265 patients from 1524 hospitals were assessed. The cohort size ranged from 38 547 patients in 467 hospitals for surgical clipping for cerebral aneurysms to 308 952 patients in 599 hospitals for spinal fixation, laminectomy or laminoplasty. In the patient-level analyses, there were no significant mortality differences among patients undergoing the 12 major noncardiovascular surgical procedures, whereas postoperative ICU admission was associated with trends towards lower in-hospital mortality among patients undergoing coronary artery bypass grafting, risk difference -1.0% (95% CI -1.8 to -0.1) open aortic aneurysm repair, risk difference -0.6% (95% CI -1.3 to 0.1), and heart valve replacement, risk difference -0.7% (95% CI - 1.6 to 0.1). In the hospital-level analyses, similar to the results of the patient-level analyses, a higher proportion of postoperative ICU admission at hospital level was associated with trends toward lower in-hospital mortality for patients undergoing the three cardiovascular surgical procedures.

This nationwide observational study showed that postoperative ICU admission was associated with improved survival outcomes among patients undergoing three types of cardiac surgery but not among patients undergoing low-risk elective surgery.