We here report that, in patients admitted to a coronary care unit because of an ACS, the OGTT including both 1-h and 2-h PG values identifies patients with the most severe in-hospital risk profile, adverse remodeling and longer hospitalization; and that the addition of the 1-h PG-OGTT to the usually assessed 2-h PG-OGTT is of help in such identification.
Despite the previously reported high percentage of undiagnosed dysglycemia among ACS patients  and the relevant prognostic role of 2-h-PG at the OGTT  in the coronary care setting, the use of glucose challenge has been increasingly neglected or actually discouraged [17, 18]. We have here re-evaluated the importance of OGTT in providing prognostic information with respect to the severity of myocardial injury and remodeling, as well as to mid-term outcomes. Our results broadly confirm that IGT at the time of hospital discharge associates with worse “outcomes”, in our case investigated with markers of myocardial injury. In our case, however, it is IGT also including elevated 1-h-PG at the time of hospital discharge to be associated with high levels of NT-proBNP and troponin, as well as a numerically lower LVEF and longer hospitalization.
After a myocardial infarction, the in-hospital diagnosis of IGT through the performance of the OGTT had already demonstrated to provide important long-term prognostic information: IGT is indeed associated with the increased occurrence of MACE in the follow-up [13, 22]. In the acute phase, irrespectively of the presence of diabetes, a hyperglycemic status is associated with a higher risk of in-hospital death, cardiogenic shock and congestive heart failure, and correlates with a more extensive myocardial damage , This association likely reflects a higher degree of inflammation and the concomitant release of counterregulatory hormones [23, 24], resulting in a higher degree of stress hyperglycemia, a condition that has been associated with poorer outcomes in a number of acute conditions, including myocardial infarction [23, 25, 26], stroke , trauma [28, 29] and, more recently, COVID-19 . In this sense, our report confirms previous findings on the prognostic role of the OGTT in patients with an acute myocardial infarction.
We have here also explored, however, to what extent 1-h OGTT PG may also have a prognostic value, both as an isolated defect (i.e., 11-h OGTT PG > 155 mg/dL and 2-h OGTT PG < 140 mg/dL), as well as in combination with IGT (i.e., 1-h OGTT PG > 155 mg/dL and 2-h OGTT PG > 140 mg/dL). Previous work had indeed suggested the 1-h OGTT PG may be a simpler and more effective criterion for identifying people at risk of developing type 2 diabetes, cardiovascular disease and death [6, 8,9,10, 31, 32]. To the best of our knowledge, however, no studies has so far investigated the prognostic potential of 1-h OGTT PG at the time of hospital admission because of an ACS. Our results suggest that, although 1-h OGTT PG may identify people with a more severe cardiac injury and impaired remodeling, it is full IGT that best associates with a worse risk profile. Thus, patients with isolated 1-h IGT had a risk profile that was not different from those with normal glucose tolerance and better than in patients with the combination of the two glucose tolerance defects. This latter finding suggests that a more comprehensive glucose tolerance disturbance, assessed by alterations of both the 1-h and the 2-h PG at the OGTT, best identifies patients with admission and discharge conditions more severe than those in patients with isolated altered 2-h PG. Of note, indeed, patients with full IGT featured higher peak TnT and NT-proBNP values than late IGT patients, i.e., those still conventionally and now broadly classified as with “IGT”. The same full IGT individuals were the ones with a worse left ventricular ejection fraction and longer hospitalization. Because the latter group was small in size, and we could therefore not exclude lack of power, a sensitivity analysis performed by pooling group C and group D confirmed the overall findings.
We recorded new cardiovascular events over 9 months after discharge, but this analysis remains largely exploratory given the relatively small size of each of the glucose tolerance group and the very limited number of events (N = 4). Yet, it is intriguing that 1-h-altered PG patients apparently had no more events compared with those with normal glucose tolerance. Our report, therefore, while broadly confirming older data on the usefulness of the routine safe performance of the OGTT in patients otherwise classified as non-diabetic to refine the prognostic stratification of ACS [15, 20], points to the importance of accruing data on the shape of the OGTT curve.
We acknowledge several limitations in this report, which cannot be considered conclusive. First, the limited sample size: this has to be regarded as the first analysis of this kind, with no a priori effect size postulated and no sample size estimated. In fact, the data here accrued will serve as the basis for a larger, more definitive study, now ongoing. A second limitation is in the non-homogenous distribution of patients among groups might have prevented to document differences among groups beyond those here reported. As shown in Table 1, Group A, B and C patients had absolute lower levels of NT-proBNP, peak and admission hs-TnT values compared with group D patients. Other parameters, however, failed to achieve statistically significant differences at multiple comparisons. In particular we did not observe any statistically significant difference in left ventricular ejection fraction values despite Group D featured the numerically lowest values. For the majority of patients (79.5%), this was the first in-hospital access for an ACS: in these patients, we observed a relevant undertreatment of cardiovascular risk factors: particularly, despite a high prevalence of dyslipidemia, only 15.7% of patients reported daily statin treatment; in particular, in Group D no patient reported any statin treatment at admission. This was likely a chance finding, but can theoretically influence the outcome..Third, we considered laboratory markers, instrumental examinations and hospitalization duration instead of clinical events to evaluate in-hospital outcomes. The choice of surrogate markers of disease severity was dictated by the small sample size and the short duration of the follow-up. Further studies including larger populations will be required to determine to what extent the 1-h OGTT PG may contribute independent of and beyond the prognostic value of 2-h PG. Recruitment of a larger patient cohort with a longer-term follow-up is currently ongoing. At the moment we hypothesize that in-hospital outcomes will be later reflected in adverse outcomes in terms of major adverse cardiovascular events at a longer follow-up, but this will have to be verified. Fifth, we cannot definitely rule-out a carry-over effect of the ACS on the outcomes of the OGTT. We feel, however, that a later evaluation would carry little practical implication for the difficulty of implementation after discharge. Groups were also here too small to detect differences between STEMI and non-ST elevation ACS.
Finally, our findings are at variance from those obtained in a primary prevention setting, where 1-h PG at the OGTT was shown to correlate with the later development of diabetes [4, 5] and adverse cardiovascular outcomes irrespective of the 2-h PG . In our cohort, we failed to demonstrate any difference among Group A and Group B patients, letting us to conclude that, in our setting, the sole 1-h PG with the current cut-off of 155 mg/dL should not be considered as a sufficient substitute for the standard 2-h PG at the OGTT, but rather as potentially yielding complementary information.
In summary: in an ACS setting, the routine use of the OGTT including the 1-h PG evaluation before discharge helps identifying, among newly diagnosed IGT, a subset of patients experiencing adverse in-hospital outcomes. Further research should confirm these data in a larger cohort and assess their translation into harder outcomes.