The prevalence of T2DM in our study was 56.5%, with 22.4% of them newly diagnosed with T2DM by using the HbA1c test. Rashid et al. reported the prevalence of newly diagnosed T2DM was 14.7% among 693 acute coronary syndrome (ACS) patients [6]. Besides, a previous study from Pakistan involving 140 ACS reported the prevalence of newly diagnosed diabetes was 7.1% [7]. This discrepancy in prevalence could be attributed to the differing characteristics within the study populations in which our study only involved STEMI patients.
The severity of CAD did not correlate with HbA1c level or fasting insulin. The occurrence of MACE at 1-month follow-up also did not significantly differ between DM and non-DM. Patients with hypertension were significantly associated with a higher prevalence of undiagnosed T2DM among STEMI patients.
The present study showed the glycemic level based on HbA1c did not demonstrate any correlation with the anatomy severity of CAD by using syntax score. This finding contradicts with previous study involving 572 ACS patients showed HbA1c level to be positively associated with SYNTAX score after adjusting confounders (effect size β = 1.09; 95% CI = 0.27–1.91; p = 0.0096) [8]. The difference is probably due to the small sample size and different study populations.
Our study also showed no association between fasting insulin level and anatomical severity of CAD. Contrarily, Yaseen et al reported a significant positive linear correlation between fasting insulin level and Gensini score (ρ = 0.837; p < 0.001), indicating a strong correlation between insulin resistance and severity of CAD in patients with T2DM [9]. The difference is probably because they recruited patients that are for elective coronary angiography which means patients were stable and not in stressful conditions compared to our patients who were acutely admitted and went for angiography in emergency settings. This factor may in turn cause a sudden drop in fasting insulin in response to stress thus rendering our correlation analysis to become falsely insignificant. Other than that, Jia et al demonstrated significant correlations between insulin levels and severity of CAD but the multivariate analysis did not indicate that insulin levels are predictors for the severity of CAD [10]. The inconsistent findings regarding the correlation between insulin levels and CAD severity, as evidenced by previous literature suggest that insulin levels may not be a definitive predictor or reliable marker for assessing severity of CAD [10–12].
The current study revealed a significant association between ESRF and severity of CAD as represented by the SYNTAX score. It showed that ESRF patients are more likely to have a more severe CAD compared to non-ESRF patients. A similar finding was described in a prospective study, which indicated that AMI patients with CKD were more likely to have higher SYNTAX scores (OR = 2.453; 95% CI = 1.080–5.569; p < 0.05) [13].
The difference in the occurrence of MACE at 1-month follow-up between STEMI patients with and without T2DM was not significant, as shown in the correlation analysis.
Our study showed no difference in the MACE between diabetes and non-diabetes patients. This finding is consistent with an Asia Pacific Evaluation of Cardiovascular Therapies (ASPECT) collaboration study encompassing 12144 STEMI patients reported that diabetes was significantly associated with increased 30-day MACE (13.3% vs 8.6%; R = 1.73; 95% CI = 1.44–2.08) compared to non-diabetes, after multivariable adjustment [14].
In this current study, however, it was noted that most of the STEMI patients who experienced MACE were those without T2DM: Two (2.35%) died of cardiovascular events and five (5.88%) had a hospitalization for unstable angina. This is probably due to the small number of patients recruited in the study.
Current results showed that newly diagnosed T2DM was higher in males, ethnicity (Malay) and comorbidities such as dyslipidemia and hypertension. Similar findings were described in several previous studies. Dar et al. reported that 85.1% of male ACS patients had undiagnosed T2DM compared to only 14.9% of female patients [15]. Rashdi et al. with a predominantly male patient (n = 25) over female patients (n = 15) also identified the male gender as a risk factor for undiagnosed T2DM among patients with AMI. The study reported a higher prevalence of 45.0% for undiagnosed T2DM among male patients compared to female patients (15.0%) [16].
STEMI patients belonging to the Malay ethnicity (p < 0.05) were found to be significantly associated with a higher prevalence of undiagnosed T2DM than patients from other ethnicities. Studies that investigated the association between ethnicity and the prevalence of undiagnosed T2DM were scarce. This is probably due to our local population being Malay- dominant with approximately 45–55% of the general population in Kuala Lumpur.
STEMI patients with comorbidities, dyslipidemia and hypertension (p < 0.05) were significantly associated with higher incidence of undiagnosed T2DM. This was consistent with observations described by previous studies, which identified dyslipidemia and hypertension as two of the most common risk factors for undiagnosed T2DM among patients with ACS including STEMI. The Prevalence of undiagnosed T2DM was as high as 85.1% and 73.6% among STEMI patients presented with dyslipidemia and hypertension, respectively [16]. Additionally, over 51.2% of STEMI patients with undiagnosed T2DM had ≥ 3 cardiovascular risk factors, including hypertension and different types of dyslipidemia (elevated low-density lipoprotein, elevated triglyceride and low high-density lipoprotein) [4].
The limitations of this study are the small sample size and limited study duration. The current study only included 85 patients in the study cohort. The small study population may cause the statistical analysis to be underpowered. A relatively short study duration limited to only 6 months prevents long-term observation of the STEMI patients, including changes in the HbA1c and insulin levels as well as the incidence of MACE over a longer period of time. This may lower the reliability of the results, especially concerning the relationship between different variables. Therefore, a larger sample size with a longer study duration should be considered for future research to improve the result reliability.