Factors linked to mortality after coronary artery bypass graft (CABG) include post-operative hypoglycemia and graft source. However, it is unknown if there is an interaction between these two factors, namely whether arterial grafts withstand hypoglycemia better than veins after CABG. In a cohort of 1323 patients undergoing CABG with intensive insulin therapy (IIT) in the ICU, we identified graft sources, ICU glucometrics, diabetes (DM) status and one-year mortality. Glucometrics included mean BG in ICU, and hypoglycemic excursions (Incidence, Severity [nadir <40, 40-69mg/dl], Repetitive Frequency [number of excursions/patient]). The impact of different factors, individually and in combination, was analyzed by Chi-squared Automatic Interaction Detection (CHAID). One year mortality was highest in those with BG<40mg/dl in the ICU (12/26, 46%) vs. those without (67/1283, 5%, p<0.001). No other factor was significant in those with BG<40. Of those without BG<40, repetitive (≥2) excursions with BG<70mg/dL had the highest mortality (17/96, 17%, p<0.001), vs. those with one (13/190, 7%) or no excursions (37/997, 4%). Graft source was dominated by other factors in one year mortality, except for the left internal thoracic artery which was protective only in patients who had no hypoglycemia. Hyperglycemia was a contributor to mortality in patients who had a single hypoglycemic excursion <70 (ICU Mean BG>152mg/dl 4/17, 24% vs. ≤152 9/173, 5%, p=0.039) Preexisting DM was protective in patients with repetitive hypoglycemia (DM 2/43, 5% vs. 15/53 28%, p=0.003). Hypoglycemia in the ICU has a durable impact on one-year mortality post-CABG, superseding graft source, which is only significant without hypoglycemia. Severe hypoglycemia poses the greatest risk, but even mild repetitive BG<70 is a threat to one year survival.

In conclusion, hypoglycemia erases the survival benefit from arterial grafts, suggesting that it may be equally damaging, regardless of graft source, and far more durable than hitherto envisioned.


L. Heller: None. M. Zupa: None. P.L. Perreiah: None. R. Rao: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.