Angiographic characteristics of coronary artery disease in patients undergoing diagnostic coronary angiography at a tertiary hospital in Tanzania

The prevalence of obstructive CAD of any degree of coronary artery luminal stenosis was estimated at 24.86% among patients undergoing diagnostic CAG in a tertiary care facility in Tanzania. This is comparable to a similar study conducted in Cameroon, which reported the prevalence of 28.8% among patients who underwent diagnostic CAG [17]. Similarly, our findings are comparable to a study conducted in Iran which revealed the prevalence of 19.1% based on a combination of clinical evaluation and CAG findings [18]. Noteworthy, a similar study by Khanbhai, KS et al. conducted at Jakaya Kikwete Cardiac Institute (JKCI); another centre in a more urbanized population in Tanzania, reported the prevalence of 45.5% [19]. The reasons for the observed discrepancy are not entirely clear, however, two factors could be speculated as the possible explanations. Firstly, it is noteworthy that CAG services are expensive and largely unaffordable to most patients in Tanzania; only affordable to those with high insurance premium coverage. Thus, patients with suspected CAD without high premium insurance coverage are less likely to undergo CAG and subsequent coronary intervention, which implies that the nature of patients who undergo cardiac catheterization are often not the true representative of the population. Moreover, even those patients who are insured, for patients in whom MVD is diagnosed, most of the insurance companies would cover the cost for only one stent at a sitting when percutaneous coronary intervention (PCI) is indicated. Secondly, the two cities – Dar es Salaam and Dodoma – where the two “Cath Lab” centres are located might differ in the level of urbanization and associated risk factors for ASCVD [19, 20]. JKCI, being a national cardiovascular centre, tends to receive referral patients with advanced CAD from across hospitals in the country for diagnostic CAG and coronary intervention. Similarly, the increasing medical tourism from neighbouring countries of patients with established, and possibly with severe CAD could potentially over-represent the prevalence of obstructive CAD at JKCI.

The prevalence of CAD was noted to have male gender preponderance and increasing with age consistent with studies conducted elsewhere, including analysis of global data by Khan MA et al. [5], and in a review by Shehu, MN et al. [11]. Age is a well-established independent risk factor for developing NCDs, CAD being not an exceptional. The notably higher prevalence of CAD among male compared to female is multifaceted. This could partly be attributed to the presence of competing risk factors for CAD which are associated with the male gender, particularly smoking and alcohol consumption which are both known to be higher among men in SSA. For instance, a review by Magitta, NF indicates smoking and tobacco use to be primarily a behavioral risk among men in the majority of countries in SSA; with the prevalence of 17.5% and 2.2% among men and women respectively, a pattern which is similar across countries worldwide [21].

The majority of patients with CAD had hypertension and a few of them had T2D, which are well known independent risk of developing CAD. However, hypertension showed a trend towards CAD risk, though it was not shown to be statistically significant contrary to other previous studies [4, 11]. The reason for this observation remain elusive, however, a small sample size could have an impact on the power of the study. T2D is a well-established risk factor for developing CAD, as affirmed by this study in parallel to other previously reported studies. T2D could operate through multiple mechanisms which include increased oxidative stress, endothelial injury, lipid peroxidation and ultimately atherosclerosis. Moreover, the co-existence of hypertension in the context of metabolic syndrome frequently exert synergistic effects to promote the development of CAD [11, 17].

In this study over three-quarters of patients with angina chest pain were found to have angiographically normal coronary arteries. The majority of these patients could have non-ischemic chest pain arising from other causes. However, a minority of patients with typical ischemic chest pain could potentially have non-obstructive CAD; what is termed as ischemia with non-obstructive coronary artery (INOCA), alias angina with non-obstructive CAD (ANOCA). It is generally well established that up to 70% of patients with CAD may not have coronary artery obstruction, though in patients undergoing diagnostic CAG, the absence of obstruction may either rule out CAD or potentially rule in a possible diagnosis of INOCA [15]. The latter is relatively common among women (up to 50–70%) compared to men (30–50%) [15]. In our study, women contributed about 60% of those without coronary obstruction, indicating a possible diagnosis of INOCA in some of these patients.Thus,  our findings are in parallel to a study conducted in Poland among young patients aged below 40 years with a diagnosis of CAD which revealed angiographically normal coronary arteries in 37.2% of those with CAD and 16.9% of those with ACS [22]. The mechanisms underlying INOCA/ANOCA/MINOCA are diverse but available evidence suggests several endotypes that include coronary microvascular dysfunction, augmented vasoreactivity or spasm, non-obstructive atherosclerosis, pre-heart failure with preserved ejection fraction, and hypercoagulability which are either acting singly or in combination [23, 24]. Moreover, there is a body of evidence on racial variability on the severity of CAD, such that black patients with coronary insufficiency have been shown to be more likely to have ANOCA/INOCA compared to other races [25, 26]. This could therefore partly explain the reason for the observed low prevalence of obstructive CAD in our study compared to the findings in other reports.

Noteworthy, about 5.36% of patients were found to have coronary vessel abnormalities which include ecstatic, tortious and sluggishly flowing vessels, in agreement with previous reports [27]. These aberrant vessels could be associated with inadequate blood supply to meet the myocardial demand resulting in ischemic pain. The mechanisms for the vascular aberrations are not well understood. However, developmental anomalies during embryological morphogenesis due to hitherto unknown causes could be the underlying insults [27]. Recently, overexpression of vascular endothelial growth factor – beta (VEGF-β) in mice has been shown to promote the proliferation of endothelial cells and coronary vessels in the embryological heart development [28]. This signaling pathway is currently being explored as a potential therapeutic target for reperfusion therapy. We further postulate that in SSA, where there is widespread, repeated malaria infection particularly during childhood, the associated oxidative stress and pro-inflammatory cytokines could trigger subtle changes through activation of nuclear factor kappa – beta (NF-κβ) pathway in the vascular system including coronary vessels that could lead to their developmental aberration [29].

Over a half (53.59%) of patients had single vessel disease, slightly lower compared to 61.9% reported in a study conducted in Poland [22]. A substantial proportion of patients (46.41%) displayed MVD, of which 3.87% were found to have triple vessel disease. Likewise, this population of patients further demonstrated a severe degree of vessel stenosis which revealed about 76.88% of patients to have coronary artery luminal stenosis of at least 50% while those who had over 70% degree of stenosis alone accounted for 56.65%. The reasons for this observation are not readily discernable. However, we speculate that patients in this study might have had symptoms for longer time without seeking medical attention or inadvertently could have sustained unrecognized or missed myocardial infarction without being evaluated for CAD. Similarly, this observation could further be attributable to late diagnosis partly due to lack of access to CAG services in Tanzania. It should be recalled that CAG services are relatively new in the majority of countries in SSA, with only two centres in Tanzania – with our centre being operational for less than three years [16]. It is well settled that the presence of multiple risk factors in an individual patient contributes substantially to the vessel pathology which orchestrate the development of ASCVD.

The most common affected vessel in this study was an RCA – which was involved in  36.84% and at 55.67% when single vessel disease was considered. This finding is in agreement with other studies conducted elsewhere. For instance, a study conducted in Egypt among patients with ST segment elevation myocardial infarction (STEMI) indicated RCA to be most commonly affected vessel in 34.6% of patients [30]. However, this finding is contrary to other studies which revealed LAD as the predominant affected vessel. For instance, studies conducted in Cameroon, China and Poland reported LAD artery as the most commonly affected vessel, accounting for 42.28%, 45.5% and 61.6% respectively [17, 22, 31]. The type of the affected vessel could have a clinical significance as the occlusion of RCA could potentially affect the blood supply to the sinoatrial node and the conducting system of the heart resulting in rhythm disturbances and conduction defects. The reason for the observed differences in the predilection of CAD affected vessel remains unknown. However, the existing body of evidence indicates subtle anatomical variation in the coronary circulation among human races [32]. Though, in his seminal work, R. Singer in 1959 did not find any significant racial differences in the major coronary branches [33], skeptics exist on the human variability in the coronary artery luminal diameters. However, it is not known whether any potential anatomical variation could entirely explain the observed inter-racial coronary atherosclerotic predilection and disease severity.

There are two major limitations for this study. Firstly, the study being retrospective it was susceptible to incompleteness or loss of data. This happened because specific variables were not initially considered important and were therefore, not collected or due to limited hospital archiving capacity, some variables were not available particularly when hardcopies were used for initial data collection, though the hospital largely uses electronic data capture and storage platforms. Secondly, CAG services are expensive and not readily accessible or affordable to the majority of patients without adequate insurance premiums. Thus, the data largely consist of patients who are either ensured or those from high socio-economic status, hence without adequate representativeness of the population and with limited generalizability to the wider population.