Detection of individuals at risk of future complication/comorbidity is a fundamental objective of DM treatment and management. As previously discussed, MetSyn represents an important risk stratification strategy that can be used as guide for patient-centered treatment. In this study, the first of its kind in Eritrea, we evaluated the frequency of MetSyn in the largest T2DM follow-up facility in Asmara. In sum, MetSyn was documented in 58.1% of the patients. Available data suggests variable prevalence of MetSyn in different parts of SSA [27]. Hospital-based data applying Adult Treatment Panel III [ATP III]) criteria (NCEP: ATP III) or the closely allied IDF harmonized criteria [13] are available for specific sites in Nigeria 60% [28]; Ethiopia (51.1%) [30]; Ghana (24–78.8%) [31]; Cameroon (60.4%) [32]; Nigeria (62.5%) and [33]. Higher percentages are generally reported if the IDF harmonized definition was applied [31, 32]. The highlighted comparisons should be read with several caveats in mind. Foremost is the fact that the available reports from SSA have mostly provided non-standardised prevalence rates (e.g. disparate definitional criteria). Accuracy of medical examinations, differences in access to clinical care, diagnostic criteria may also undermine direct comparison between studies. Therefore, it is important to seek out the similarities and differences in the study design and patient population in these studies.
Among individual permuted phenotypes, the triplet of hypertriglyceridemia, abnormal HDL-C and HTN were the most frequent. Vice versa, a large number of studies from the region have noted that the most predominant component of MetSyn in patients with T2DM is HTN [30, 31, 33, 34, 35, 36, 37] and central obesity [30, 38, 39]. Regardless, the current data is supported by the fact overt DM, especially T2DM, frequently elevates plasma TG and reduces HDL-C. Indeed, multiple studies, have established that T2DM patients almost invariably manifest serious breakdown in lipid dynamics, reflected by increased flux of non-esterified fatty acids (NEFAs), TG, low-HDL-C and increased amount of apolipoprotein (ApoB) particles with preponderance of smaller, cholesteryl ester depleted LDL-C [40, 41]. Although the pathophysiological mechanisms are complex, and not entirely understood; the preponderance of diabetogenic dyslipidemia components in this setting maybe an indicator of IR and suboptimal control of the disease. Unfortunately, management of diabetic dyslipidemia has been overlooked in most countries in SSA. Beyond that, our data demonstrates that caloric restriction/or dieting and increased physical activity/exercising are under-emphasized in this setting. Unfortunately, and as is often the case in SSA, medications for lipid management are largely inaccessible to most patients due to cost and availability.
Further, analysis of the difference between males and females in multiple profiles appears to suggest that women are disproportionately affected in this setting. Significantly higher mean (SD)/or median (IQR) were observed in women for multiple analytes including HbA1c; LDL-C; TC; Non-HDL-C; BMI, among others. A higher percentage of women were also on insulin injection and had positive CRP result. The age at DM onset was also lower in females. In contrast, men had higher average DBP; TG; alcohol consumption and lower HDL-C, among others. The observed pattern of abnormalities in lipid and anthropometric markers has been documented in multiple investigations [31, 34, 38, 34] and might explain disease risk or differential impact between the sexes. Unlike other studies [37], there was no significant sex difference in the prevalence of HTN. The fact that women are disproportionately impacted (e.g. disease incidence, morbidity and mortality) in the NCD landscape in SSA has been reported in the region [32, 34, 37, 38]. The aggravated impact of MetSyn in women was also evidenced by the fact that a large proportion of women had more than 4 abnormal MetSyn components. Taken together, and cardioprotective effects of estrogens in pre-menopausal women notwithstanding; our data reinforces the well-evidenced assertion that women with DM are at a greater risk of CVD than men with DM.
A separate stepwise multivariate analysis demonstrated that the frequency of MetSyn was associated with age, LDL-C, Non-HDL-C, BMI and IR. The link between age and enhanced MetSyn risk is well documented [19]. Several studies have also demonstrated a clear relationship between age, obesity, T2DM and CVD risk. The observed relationship between LDL-C, Non-HDL-C and MetSyn is equally important. Trial level evidence from a previous study in Mexico indicated that LDL-C levels > 100 mg/dL are observed in 74.8% (95% CI 72.5–76.9%) of previously diagnosed DM patients [42]. In this study, the proportion of patients presenting with elevated LDL-C was 75.4%. Non-HDL-C, a surrogate marker of plasma concentrations of atherogenic Apo B-100 containing lipoprotein particles (VLDL, IDL, LDL, lipoprotein A), was also elevated in most patients (73.3%). Taken as a whole, it’s our position that although elevation in LDL-C, Non-HDL-C, age and insulin injection are not part of the MetSyn diagnostic algorithms, they are important surrogate markers of CVD risk. In this respect, the observed associations may further modify absolute risk in this population.
An interesting peculiarity in the clinical expression of MetSyn in SSA is the disproportional contribution of abdominal obesity (WC > 94/80 cm Male/Females) in MetSyn diagnosis in women from the region [34, 35, 38]. A substantial amount of epidemiological and post hoc analyses of clinical trial data on intra-abdominal fat, including those involving imaging technologies, have demonstrated that WC is a better predictor of T2DM and MetSyn incidence [35]. In particular, Evans and coworkers [43] found that WC, WHtR and a Computer Tomography (CT) - derived measure of visceral adipose tissue (VAT) performed similarly in predicting MetSyn in pre-menopausal women from disparate racial groups in South Africa (SA). Multiple propositions have been advanced to explain the link between visceral adipose tissue and the other components of MetSyn (elevated BP, dysglycemia, inflammation) – portal visceral hypothesis; endocrine paradigm and the ectopic fat storage hypothesis. Briefly, it’s been hypothesized that dysfunctional visceral adipose tissue or/ visceral adipocyte hypertrophy is associated with IR and hypersecretion of bioactive adipokines and proinflammatory cytokines such as tumour necrosis factor-alpha (TNF-α), angiotensin II (AII), interleukin − 6 (IL-6), interleukin-1β (IL-1β), (RBP-4), plasminogen activator inhibitor-1 (PAI-1), heparin binding epidermal growth factor, leptin, resistin and a decrease in adiponectin, to mention a few [44, 45, 46] – a dynamic which has been implicated in MetSyn etiology and pathogenesis [45].
At present, the use of WC measurements, independently or in conjunction with other risk factors, to predict cardiometabolic disease is not well established in SSA [13, 35]. Unfortunately, it has been recognized for many years that the default WC value (94 cm males/80 cm females) used in the region may underestimate the prevalence of MetSyn among men and overestimate its prevalence among women [47]. For this reason, the prominence of WC as a marker of MetSyn, particularly in women in SSA, remains controversial due to the absence of specific cut points for local populations. Therefore, the possibility that this study may have underestimated the prevalence of MetSyn in men is considerable.
The relationship between BMI and MetSyn was also explored in this study. Like WC, BMI is not a refined research or clinical tool particularly for patients presenting with DM. A major limitation of BMI is the inability to discriminate between visceral adipose tissue (VAT) and other forms of adipose tissue distribution [48]. As a matter of fact, the exclusion of BMI in some MetSyn diagnostic specifications is partly based on the fact that visceral distribution of body fat is, as we noted previously; a stronger risk factor for CVD than general obesity [5, 35]. The relationship between BMI and MetSyn, hence CVD risk is demonstrated in Fig. 3. One thing is clear: the greater the weight, the greater the frequency of MetSyn and vice versa. Another important finding was the fact that a large number of patients with MetSyn had low BMI. Interestingly, some authors have argued that MetSyn occurs less frequently in those who are overweight (BMI 25–29.9 kg/m2) and is relatively rare in normal weight individuals [49]. Clearly, the latter assertion is not evident in our study. Concordant with other studies in SSA [50], a sizable proportion of patients with MetSyn had BMI < 25 Kg/m2. One additional observation was the finding that the relationship between BMI and MetSyn was a lot more consistent at BMI > 25 kg/m2. Whilst obesity is evidently important and the lower the BMI, the better; other factors appear to influence susceptibility to MetSyn at lower BMI (< 25 kg/m2). First and foremost, investigators have noted that compared to the Western world, a sizable proportion of T2DM patients in Africa have (< 25 kg/m2) [16, 51].
Possible explanations of this phenomenon are multiple and debatable. For instance, some imaging studies have shown that important and physiologically significant visceral fat accumulations can be observed in some ethnic/racial groups even in the normal BMI range [52]. Consistent with this, it has been noted that non-obese individuals with MetSyn may have susceptibility factors, particularly ethnic, and genetic/or epigenetic characteristics [53]. Misclassification of patients with latent autoimmune diabetes in adults (LADA) and other atypical DM phenotypes as T2DM is another possible explanation. The plausibility of this proposition is augmented by the fact that measurement of glutamic acid decarboxylase antibodies (GADab) in adults diagnosed with T2DM is not a prerequisite to identify subjects with LADA in Eritrea. In all, detailed studies are needed especially among newly-diagnosed DM patients to fully characterize the disease in the region. Understanding the genetic profile of T2DM in the region is another onerous challenge.
In a separate analysis, we evaluated the relationship between specific non-traditional markers (Non-HDL-C, TG/HDL-C, TC/HDL-C and LDL-C/HDL-C) and MetSyn. Broadly speaking, the relationship between number of MetSyn components and mean values of specific traditional and non-traditional cardiometabolic biomarkers is a rare consideration in publications from the region [19, 29–32]. Regardless, it’s our belief that the use of multiple break-points in risk stratification (e.g. component number) may provide supplementary information which maybe masked by the usual MetSyn/No MetSyn dichotomy.
In this analysis, a positive dose-response gradient between increasing number of MetSyn components and mean values of BMI, TG, TC, WHtR, WHR, WC, HC was established - the higher the component number, the higher the mean values of these risk markers. Reducing eGFR was also associated with increasing component number. A similar parallel increase in Non-HDL-C, TG/HDL-C, TC/HDL-C and LDL-C/HDL-C with increasing number of components was also demonstrated. Altogether, our study demonstrates that increasing number of MetSyn components is associated with higher averages in multiple traditional and non-traditional CVD risk indicators in this population. The latter associations are interesting given the link between these non-traditional markers and CVD risk. For example, compelling evidence indicates that TC/HDL-C is a fairly good index of the relative contribution of atherogenic vs. antiatherogenic lipoproteins to accelerated atherosclerosis [54]. Some investigators consider it to be superior even to the measurements of the ratio of apo B to apo A-I. The utilization of the TG/HDL-C ratio as a marker of IR and a means to estimate the presence of the more atherogenic small dense LDL-C subfractions has also been proposed [55, 56]. More importantly, NCEP ATP III identifies non-HDL-C as a secondary target of therapy in patients with TG > 200 mg/dL. Regardless, no rigorous outcome data are available to place into clinical context the relationship between these non-traditional markers and MetSyn-associated CVD risk in populations from SSA.
Finally, it is worth mentioning that although all components of MetSyn appear to promote CVD, the relationship between component number and CVD risk or outcome are not completely understood. According to a persistent view, quartets and quintets have no higher CVD risk than triplets of MetSyn components [57, 58]. Conversely, there exists overwhelming evidence that disparate MetSyn phenotypes (component mix) bear dissimilar CVD risk burden. Phenotypes containing elevated BP tended to confer higher risk. This is predictable given the fact that epidemiologic, experimental, and randomized controlled clinical study trial-level evidence have consistently indicated that elevated BP is a robust, consistent, independent and etiologpathologically relevant risk marker for CVD and chronic kidney disease (CKD). In particular, a past study utilizing NCEP-ATP III criteria for MetSyn stratification established a connection between the number of MetSyn components and several CVD risk indicators including severity of subclinical atherosclerosis as reflected by thicker carotid plaques, increased common carotid intima-media thickness (IMT) and increased pulse wave velocity [58]. Clearly, more research is required in order to examine, and possibly confirm, phenotype/component mix and adverse event relationships for this setting.