This study identified independent prognostic factors that predict elevations in serum admission cTnI, and demonstrated elevated admission serum CK-MB and cTnI as prognostic for short and long term outcomes respectively. As a result, we gained insight into the clinical significance of elevated admission serum cardiac biomarkers in acute decompensated diabetes in the absence of clinically evident ACS. In the future, these data can be used to guide management.
An admission serum pH of less than 7.1 was demonstrated to be an independent prognostic factor for elevations in serum admission cTnI. To our knowledge, this is the first study to demonstrate a relationship between serum pH and elevated serum troponin. Our study supports the concept put forth by Moller et al., who noted that patients with elevated cTnI did not have angiographic evidence of coronary artery disease but had serum admission pH values of less than 6.9 .
What accounts for the correlation between cTnI and serum pH in the context of this study is unclear, and remains to be elucidated. A potential explanation highlights a complex dynamic between pH and intracellular calcium. As a result of severe acidemia, there is an increase in intracellular calcium which activates multiple biochemical pathways including proteolysis and myocardial stunning culminating in increased serum cTnI [23–28].
Approximately 40% of the patients with an elevated serum admission cTnI had a documented prior history of CAD. Regression analysis demonstrated a prior history of CAD as a prognostic factor for elevated cTnI. Acute decompensated diabetes is characterized by increased levels of counter-regulatory hormones which increase myocardial oxygen demand [14–16]. In such a population, the added effect of CAD likely impairs blood flow, exacerbating a supply–demand mismatch, resulting in myonecrosis and elevations in cTnI.
Logistic regression models identified an elevated admission serum cTnI as the only statistically significant variable for the long term composite outcome, MACE. Of interest is the finding that CK-MB was not correlated with the long term MACE outcome. Furthermore, an elevated admission serum cTnI proved to be a better predictor for MACE than a prior history of CAD (p = 0.127). MACE was primarily driven by myocardial infarction, thus it was unexpected that a prior history of CAD was not a statistically significant independent predictor. More likely, MACE incidence represents multiple pathophysiologic mechanisms, independent of a prior history of CAD alone.
One factor, likely playing a major role in this cohort is insulin resistance. Recent literature has recognized more aggressive forms of type 2 diabetes marked by greater insulin resistance [16, 20, 29–34]. When these patients decompensate, they have increased production of ketoacids and more severe acidemia. Over time, insulin resistance leads to increased levels of pro-inflammatory cytokines like CRP and homocysteine that accelerate atherogenesis, plaque rupture, and MACE. This study noted an admission serum pH of less than 7.1 to trend toward statistical significance in an unadjusted logistic regression, and after adjusting for other variables, demonstrated a statistical significance for MACE. Demographically, patients with such aggressive forms of diabetes mellitus tend to be African American, in major urban cities. In comparison, 70% of the patients in our study were African Americans in a major urban city. The data suggest that patients in our study have a more aggressive form of diabetes mellitus that predisposes them to MACE independent of a prior history of CAD, and thus are at increased risk.
Concomitantly, in the context of non-diabetic patients status-post myocardial infarction, Knudsen et al. advance our premise. In an article published in 2010, they present a preponderance of evidence which highlights the connection between pro-inflammatory cytokines and deranged blood glucose control .
Studies by Hernandez et al. noted that 21% of diabetics have silent ischemia, and Zheng et al., suggest patients with diabetes have a chronic level of myocardial injury, and identified a correlation with blood glucose and elevated levels of serum hs-cTnT [36, 37]. Populations such as this cohort, which stand at increased risk for adverse cardiac events underscore the need for future studies that further characterize a complex pathway.
Our findings corroborate those of Al-Mallah et al. who identified cTnI as prognostic for long term cardiovascular events in the context of DKA . This study differed in that we demonstrated a different relationship of MACE to Troponin and to CK-MB. Further, our study identified independent prognostic variables that predict elevations in serum cTnI, lending an objective basis to explain their clinical significance. Further, we assessed short term outcomes, and identified CK-MB as an independent predictor for length of hospital stay. Thus, we define a novel role for CK-MB in the context of acute decompensated diabetes. Regarding patient characteristics, while both studies were relatively small, our study had nearly twice as many participants enrolled (n = 264 v n = 96) and our average length of follow up was nearly twice that of the prior study (40 months v 24 months).
This study was retrospective with a small sample size. In the future there is a need for larger multicenter prospective studies. Additionally, we lacked angiographic evidence at time of admission required to characterize coronary anatomy and thus patients were excluded based upon clinical information alone.