Perioperative determinants of post-induction blood pressure and prophylactic norepinephrine dose estimation in noncardiac surgery: a post hoc analysis of a prospective observational study

Post-induction hypotension is a common complication of general anesthesia and has been associated with adverse postoperative outcomes. We examined factors associated with post-induction minimum mean arterial pressure (MAP) and explored risk-profile–based model-derived norepinephrine infusion rates. This post hoc analysis included 1517 adults from a prospective single-center noncardiac surgery cohort. Post-induction minimum MAP, defined as the lowest MAP recorded between anesthetic induction and surgical incision, was analyzed using multivariable linear regression. To address confounding by indication, two-stage least-squares instrumental-variable (IV) analysis used the attending anesthesiologist’s leave-one-out norepinephrine prescribing preference as the instrument, with standard errors clustered by anesthesiologist. Preoperative calcium channel blocker use, a history of atrial fibrillation, and higher pre-induction MAP were associated with higher post-induction minimum MAP, whereas older age, preoperative angiotensin II receptor blocker or beta-blocker use, and a higher propofol induction dose were associated with lower MAP. In conventional regression, a 0.1 μg/kg/min higher norepinephrine infusion rate was associated with a 9.72-mmHg higher minimum MAP (95% CI, 6.35 to 13.0). Leave-one-out IV analysis included 1482 patients in 62 clusters. The instrument was associated with norepinephrine dose (first-stage Wald F = 355.5). The IV-estimated difference associated with a 0.1 μg/kg/min higher infusion rate was 26.7 mmHg (95% CI, 16.5 to 36.8; P < 0.001). The risk-profile–based model-derived infusion rates corresponding to a 95% model-based probability of MAP ≥ 65 mmHg were 0.087 μg/kg/min (approximate 95% CI, 0.063 to 0.141) for the average-risk profile and 0.106 μg/kg/min (0.077 to 0.171) for the high-risk profile. Patient and anesthetic factors were associated with post-induction minimum MAP. IV analysis further supported an association between higher prophylactic norepinephrine infusion rates and better preservation of postinduction MAP, although interpretation of this finding remains dependent on the validity of the instrumental-variable assumptions. The model-derived infusion rates are exploratory estimates and require prospective external validation before they can inform clinical dosing strategies. UMIN Clinical Trial Registry (UMIN000037147), registered June 24, 2019. ( https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000042261 ).