Association of fast-track cardiac management and low-dose to moderate-dose glucocorticoid administration with perioperative hyperglycemia

Objective: To delineate associations between preoperative risk factors and clinical processes of care and perioperalive glucose tolerance in patients managed on a fast-track cardiac surgery clinical pathway with prebypass methylprednisolone administration. Design: Retrospective sequential cohort study. Setting: University-affiliated Department of Veterans Affairs medical center. Participants: Fast-track patients In = 293; n = 72 low-dose methylprednisolone [100-125 mg]; n = 221 moderate-dose methytprednisolone [500 mg]) plus pre-fast-track patients (n = 258; no methylprednisolone) undergoing cardiac surgery with cardiopulmonary bypass. Interventions: None. Measurements and Main Results: Multivariate linear regression was used to model the association of 17 preoperative risk and intraoperative process-of-care variables with serum glucose concentration on arrival in the intensive care unit. Preoperative serum glucose concentrations were not significantly different among the pre-fast-track, fast-track with low-dose methylprednisolone, and fast-track with moderate-dose methylprednisolone cohorts (129 ± 54, 137 ± 55, 127 ± 4.6 mg/dL [mean ± SD]). Postoperative serum glucose concentrations were significantly different (171 ± 58, 223 ± 56, 250 ± 75 mg/dL; p < 0.03, for all pairwise comparisons). Using backward elimination from the full 17-variable multivariate model (R-square = 0.63), 4 variables remained significant (all p < 0.0001; R-square = 0.60): (1) Preoperative diabetes status (adjusted mean postoperative glucose level, mg/dL; [95% confidence interval (Cl)]): no treatment, 193 (188-199); oral agent, 276 (262-291); insulin requiring, 301 (283-320); (2) steroid group: pre-fasttrack, 201 (195-209), fast-track with low-dose methylprednisolone, 271 (256-287); fast-track with moderate-dose methylprednisolone, 295 (284-306); (3) volume of glucosecontaining cardioplegia (β coefficient, 95% Cl): 2.22% (1.37-3.10) increase per 100 mL; and (4) intraoperative epinephrine infusion: none, 231 (224-239); yes, 276 (264-288). No significant interactions were identified. No significant effect of opioid dose was observed. Conclusion: At this institution, implementation of the fast-track pathway was associated with a deterioration of glucose tolerance. Preoperetive diabetes, pre-cardiopulmonary bypass administration of steroids, volume of glucosecontaining cardioplegia solution administered, and use of epinephrine infusions were significantly associated multivariate factors. Copyright (C) 2000 by W.B. Saunders Company.