Baseline Clinical Characteristics and Midterm Prognosis of STE‐ACS and NSTE‐ACS Patients with Normal Coronary Arteries

Abstract

Objective: We sought to compare clinical profiles and midterm prognosis of patients with normal coronary arteries presenting with ST‐elevation ACS (STE‐ACS) versus non‐ST‐elevation ACS (nSTE‐ACS).

Background: There are limited data regarding ACS in patients with normal coronary arteries, and especially clinical differences between ST‐ACS and nSTE‐ACS patients have not been evaluated sufficiently.

Methods: The study group comprised 190 patients (mean age: 53.2 years, 63.1% males, 63.6% STE‐ACS) presenting with ACS and normal coronary angiograms. The participants were evaluated in terms of 42 clinical variables. MACE [cardiac death (CD) and hospitalization for angina (HA)] were the study end points.

Results: STE‐ACS in comparison to nSTE‐ACS patients were younger (P < 0.01), were more frequently males (P < 0.01), had more often infection prior to ACS (P < 0.01), higher hsCRP on admission (P < 0.01), and greater infarct size, measured by maximal troponin I (P < 0.01). By multivariate analysis in this subgroup, predictors of outcome were hsCRP (P = 0.03) and raised troponin I (P = 0.02). nSTE‐ACS in comparison to STE‐ACS patients were more obese (BMI, P < 0.01), had higher LDL cholesterol (P < 0.01), fasting glucose (P = 0.03). LDL cholesterol (P = 0.02) and fasting glucose (P = 0.03) emerged as independent predictors of outcome in these patients. Mean follow‐up period was 25.4 months. STE‐ACS patients had twice fewer MACE rate than nSTE‐ACS patients [(1‐CD, 12‐HA; 11%) vs (1‐CD, 16‐HA; 25%), respectively, log rank P < 0.01].

Conclusions: STE‐ACS and nSTE‐ACS patients with normal coronary arteriography have different clinical profiles. In nSTE‐ACS patients more pronounced metabolic abnormalities were identified, while in STE‐ACS patients inflammatory background was more significant.

The most common cause of acute coronary syndrome (ACS) is acute thrombosis at the vulnerable plaque site, ¹ which subsequently impairs the coronary blood flow. However, an overall percentage of patients suffering from ACS with normal coronary arteries is reported to range between 1 and 8.5%. ² , ⁸ , ⁹ , ¹⁰ Patients with ACS and normal coronary arteries have better prognosis and fewer risk factors of atherosclerosis than those with coronary artery stenosis and ACS. ² , ³ Only in one‐third of patients with ACS and normal coronary arteries could potential mechanisms leading to ACS be detected. ⁶ They include vasomotion disorders, coagulation disorders, inflammation factors or emotional distress. ⁷

Previous reports focused on comparisons between patients presenting with ACS with normal versus diseased coronary arteries. However, according to literature, two subpopulations of patients with ACS and normal coronary arteries can be distinguished, namely with ST‐elevation ACS (STE‐ACS) and with non‐ST‐elevation ACS (nSTE‐ACS). ⁸ Other differences between the two groups are unknown. There are limited data regarding etiological factors and long‐term follow‐up in a large series of patients with both STE‐ACS and nSTE‐ACS with normal coronary angiography.

Thus, we compared clinical and mid‐term follow‐up data of patients presenting with both STE‐ACS and nSTE‐ACS, who had normal coronary arteriography. We paid attention particularly to the sets of correlates that would be suggestive of a pathological background of the two syndromes.

MATERIAL AND METHODS

The study group consisted of 190 consecutive, prospective registry patients with normal coronary arteries who were admitted to our institution with symptoms of ACS. The subjects were selected from a total cohort of 4253 patients admitted to our institution with ACS, from June 1, 2001 until December 31, 2004.

The inclusion criteria were as follows.

• • ACS defined by the presence of at least two of the three following conditions: typical chest pain lasting more than 20 minutes, ECG changes (ST‐segment elevation >1 mm in two or more continuous leads, ST‐segment depression of at least 1 mm or T‐wave inversion in >2 leads), elevated myocardial injury markers (troponin I > 0.1 ng/nl or CK‐MB > 6 U/l).
• • Normal coronary arteries defined by visually smooth contours without any wall irregularities and normal coronary blood flow (TIMI 3).

Exclusion criteria were as follows.

• • Other previous cardiac diseases or events, such as valvular heart disease, previous myocardial infarction, cardiomyopathy, heart failure, previous PTCA, previous CABG, pulmonary embolism, aortic aneurysm, pacemaker or ICD implantation.

In each patient, the following sets of variables were collected prospectively.

• 1Medical history—hypertension, diabetes, smoking (presence of smoking habit and number of cigarettes smoked daily), the presence of infection prior to ACS, and family history of coronary artery disease (CAD). Flu‐like infection was defined as an acute onset of upper respiratory tract infection with temperature above 37.8°C, and two symptoms of the following: cough, sore throat, nasal symptoms, myalgia, headache, or malaise.
• 2Clinical information including: age, sex, body mass index, blood pressure, total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides level, and total number of risk factors. Risk factors were defined as total cholesterol >200 mg/dl, LDL >95 mg/dl, hypertension (SBP>150 mmHg, DBP >90 mmHg), smoking, fasting glucose level >111.6 mg/dl, and family history of CAD.
• 3Electrocardiographic findings such as heart rate, QRS complex duration, the presence of electrocardiographic changes and their localisation, the presence of extension of ECG changes beyond single coronary artery territory and their evolution. The evolution of ECG changes was defined by normalising of ST segment and/or appearance of Q wave assessed at admission.
• 4Baseline laboratory parameters including red blood cell count (RBC), white blood cell count (WBC), hemoglobin level, inflammatory parameters level—erythrocyte sedimentation rate (ESR), and CRP level on admission, highest values of cardiac injury markers—troponin I, CK, CK‐MB, and the number of patients with raised cardiac injury markers.
• 5Echocardiograhic parameters included left ventricular end diastolic diameter (LVEDD), left ventricular end systolic diameter (LVESD), left ventricular ejection fraction (LVEF), and the presence of global and segmental wall motion abnormalities (SMWA).

The patients were followed up for the first occurrence of MACE which included death, stroke, heart failure and hospitalization for angina. ⁶ Follow‐up data were collected by telephone interview according to a prespecified questionnaire.

Hospitalization for angina was defined as at least 24‐hour hospitalization resulting from chest pain.

Statistical Analysis

The results are presented as mean and standard deviation (SD) for normally distributed data and for non‐normally distributed data as median and interquartile range (IR). Continuous clinical characteristics of the patients in two groups were compared by Student's t‐test and categorical variables were compared by chi‐square statistics or Fisher's exact test, when appropriate. Survival curves were assessed by the Kaplan‐Meier method and compared using a log‐rank test. Hazard ratios and 95% confidence intervals were calculated by the Cox proportional hazards model. Multivariate analysis included variables with significance <0.1 in univariate analysis. In measuring the time to an event for cases in which a patient had more than one endpoint, the first event was taken into account. Value of P < 0.05 was considered as statistically significant.

RESULTS

We identified 190 patients (4.4% of a total cohort of patients with ACS) who met the inclusion criteria of our registry. The majority of them had STE‐ACS—121 (63.6%) patients; 69 (36.4%) patients had nSTE‐ACS. The time from the beginning of symptoms of ACS to the admission was 4.78 ± 2.89 hours and the majority of patients (91.1%) were admitted to the hospital before 6 hours from the onset of ACS symptoms.

The clinical characteristics of the study subgroups are presented in Table 1. In comparison to nSTE‐ACS, STE‐ACS patients were younger [50.3 years (14.3) vs 58.1 years (12.1); P < 0.001], were more frequently males [85 (44.7%) vs 35 (18.4%); P < 0.01)]. They also had higher values of cardiac enzymes [TNT I: 0.45 ng/ml (IR = 0.21–6.0) vs 0.12 (IR = 0.02–0.42); P < 0.01 and CK‐MB level: 7.9 U/l (IR = 3.8–27) vs 2.8 (IR = 2.1–6.0): P < 0.01] and higher levels of hsCRP on admission [0.49 mg/dl (IR = 0.09–3.1) vs 0.25 (IR = 0.02–0.86); P < 0.01]. Flu‐like syndrome before ACS occurred more often in STE‐ACS than nSTE‐ACS patients [44 (23.1%) vs 11 (6.1%); P < 0.01]. SMWA were found in greater prevalence in patients with STE‐ACS (14.8% vs 5.8%; P = 0.03); as was the number of patients with decreased LVEF < 45% (9.1% vs 2.8%, P < 0.01).

Table 1.

Major Clinical Characteristics of Patients in Both Study Groups

Variable	STE‐ACS (n = 121)	NSTE‐ACS (n = 69)	P
Medical history
Family history of CAD (%)	30 (24.7%)	19 (27.5%)	NS
Hypertension (%)	29 (23.9%)	11 (15.9%)	NS
History of diabetes (%)	7 (5.7%)	5 (7.3%)	NS
Flu‐like infection prior to ACS (%)	44 (36.3%)	11 (15.9%)	<0.01
Smoking (%)	85 (70,2%)	53 (76.8%)	NS
Number of cigarettes smoked (packages × years) [SD]	8.5 [2.1]	9.1 [3.2]	NS
Clinical information
Age (years) [SD]	50.3 [14.3]	58.1 [12.1]	<0.01
Patients < 35 years of age (%)	22 (18.1%)	2 (2.9%)	<0.01
Males (%)	85 (70.2%)	35 (50.7%)	<0.01
BMI (kg/m2) [SD]	25.9 [3.7]	30.6 [13.1]	<0.01
Fasting glucose(mg/dl) [SD]	108.0 [50.4]	156.6 [148.1]	0.03
Total cholesterol (mg/dl) [SD]	179.3 [38.7]	189.6 [45.6]	NS
HDL cholesterol (mg/dl) [SD]	57 [26.6]	53.2 [14.8]	NS
LDL cholesterol (mg/dl) [SD]	104.5 [38]	117.8 [45.6]	<0.01
Triglicerides (mg/dl) [SD]	149.5 [149]	173.5 [175.2]	NS
Low HDL (men < 40 mmol/l, women < 50 mmol/l) (%)	26 (21.5%)	14 (20.1%)	NS
Systolic blood pressure (mmHg) [SD]	132.6 [31.6]	137.2 [27.8]	NS
Diastolic blood pressure (mmHg) [SD]	79.8 [13.6]	81.7 [12.8]	NS
Mean blood pressure (mmHg) [SD]	97.4 [18.21]	100.2 [17.04]	NS
Total number of risk factors [SD]	1.5 [0.9]	2.5 [1.1]	<0.01
Electrocardiographic findings
Anterior/anterolateral ACS(%)	62 (51.3%)	38 (55.1%)	NS
Inferior ACS(%)	42 (34.4%)	23 (33.4%)	NS
Lateral ACS(%)	17 (14.3%)	8 (11.6%)	NS
Heart rate (bpm) [SD]	76.96 [18.3]	76.91 [20.09]	NS
QRS (ms) [SD]	98.9 [19.69]	97.4 [17.6]	NS
ECG evolution changes (%)	5 (4.1%)	3 (4.3%)	NS
ECG changes beyond single coronary artery territory (%)	26 (21.5%)	19 (27.5%)	NS
Laboratory parameters
WBC (×103/μl) [SD]	9.7 [3.09]	8.9 [2.8]	NS
RBC (×106/μl) [SD]	4.98 [0.41]	4.76 [0.42]	<0.01
Hemoglobin (g/dl) [SD]	14.07 [1.14]	13.5 [1.35]	<0.01
Number of patients with anemia (%)	16.5	31.8	<0.01
Erythrocyte sedimentation rate (mm/h) [SD]	20.7 [19.2]	16.5 [13.1]	NS
hsCRP (mg/dl) [IR]	0.49 [0.09–3.1]	0.25 [0.02–0.86]	<0.01
Cardiac injury markers
Troponin I (ng/ml) [IR]	0.45 [0.21–6]	0.12 [0.02–0.42]	<0.01
CK (U/l) [IR]	406.12	366.36	NS
	[122.8–1290.8]	[121.2–982.3]
CK‐MB (U/l) [IR]	7.9 [3.8–27]	2.8 [2.1–6]	<0.01
Patients with raised troponin I > 0.1 ng/ml (%)	62 (52.8%)	32 (46.3%)	0.02
Angiographic variable
Presence of slow flow (%)	16 (13.1%)	12 (17.4%)	NS
Echocardiograhic parameters
LVEDD (mm) [SD]	50.2 [0.7]	57.7 [5.3]	NS
LVESD (mm) [SD]	32.5 [0.47]	31.1 [0.39]	NS
LVEF (%) [SD]	60.1 [11.9]	63.2 [9.2]	NS
SWMA (%)	18 (14.8%)	4 (5.8%)	0.03
Patients with LVEF < 45% (%)	11 (9.1%)	2 (2.8%)	<0.01
Patients who were admitted to hospital after	11 (9%)	6 (8.6%)	NS
6 hours from the beginning of ACS symptoms (%)

CAD = coronary artery disease; ACS = acute coronary syndrome; HDL = high density lipoprotein cholesterol; BMI = body mass index; HDL = high density cholesterol; LDL = low density lipoprotein cholesterol; WBC = white blood cell count; RBC = red blood cell count; hsCRP = high sensitivity C‐reactive protein; CK = creatinine kinase; CK‐MB = creatinine kinase, cardiac isoenzyme; LVEDD = left ventricle end diastolic diameter; LVESD = left ventricle end systolic diameter; LVEF = left ventricle ejection fraction; SWMA = segmental wall motion abnormities.

Thus, we think that the inflammatory background might play a major role in this younger population or at least in 20% of the STE‐ACS patients, who had ECG changes beyond single coronary artery territory. However, a greater infarct size and ECG changes confined to single coronary artery territory in 80% of the patients may point to other causes of STE‐ACS, first of all to the thrombotic origin with quick dissolution of the thrombus leaving no residual luminal irregularity. Of note, a substantial proportion of STE‐ACS patients (48%) did not release troponin I, they could have shorter episodes of ischemia corresponding to coronary spasm or other unidentified causes. However, in those patients, who released troponin, its concentration did not correlate with inflammatory parameters such as hsCRP (r = 0.12, P = NS), ESR (r = 0.177, P = NS) or leukocyte count (r = 0.196, P = NS).

Yet, nSTE‐ACS patients were more obese [BMI 30.6 kg/m² (13.1) vs 25.9 (3.7); P < 0.01], had higher fasting glucose levels [156.6 mg/dl (148.1) vs 108 (50.4); P = 0.03] and higher LDL cholesterol levels [117.8 mg/dl (45.6) vs 104.5 (38); P < 0.01] than did STE‐ACS patients. The total number of atherosclerosis risk factors was greater in patients presenting with nSTE‐ACS than STE‐ACS patients (2.5 vs 1.5; P < 0.01) (Fig. 1). Thus, in the development of nSTE‐ACS, metabolic factors may be more important.

Figure 1.

Distribution of number of risk factors in patients presenting with STE‐ACS and nSTE‐ACS.

The clinical characteristics of patients with and without flu‐like infection prior to ACS with normal coronary arteries are presented in Table 2.

Table 2.

Major Characteristics of Patients with and without Flu‐like Infection Prior to ACS

Variable	Pts with Flu‐like Infection Prior to ACS (n = 55)	Pts without Flu‐like Infection Prior to ACS (n = 135)	P
ST‐elevation	44 (80%)	11 (8.1%)	0.02
Age (years)	45.2 [16]	53.1 [12.4]	<0.01
Male sex (%)	38 (69.1%)	69 (51.1%)	<0.01
BMI (kg/m2)	25 [3.3]	26.5 [3.88]	0.04
Erythrocyte sedimentation rate (mm/h)	12.1 [2–12]	9.2 [1–10]	<0.01
hsCRP (mg/dl)	0.79 [0.21–2.9]	0.21 [0.01–3.1]	0.02
CK (U/l)	446 [321–763.1]	120 [85–209]	<0.01
CK‐MB (U/l)	23.6 [16.9–54.6]	2.9 [0.2–15]	<0.01
Troponin I (ng/ml)	5.1 [0.2–15.9]	0.23 [0.01–2.1]	<0.01

BMI = body mass index; hsCRP = high sensitivity C‐reactive protein; CK = creatinine kinase; CK‐MB = creatinine kinase, cardiac isoenzyme.

Patients with flu‐like infection prior to ACS were younger (45.2 years vs 53.1; P < 0.01) and more often had STE‐ACS (44 patients vs 11; P = 0.02). They also had a significantly increased level of inflammatory parameters: ESR (12.1 mm/h vs 9.2; P < 0.01] and hsCRP (0.79 mg/dl vs 0.21; P = 0.02). In addition, patients with infection prior to ACS had higher maximal values of cardiac enzymes: CK (446 U/l vs 120; P = 0.001), TNT‐I (5.1 ng/ml vs 0.23; P < 0.001), and CK‐MB (23.6 U/l vs 2.9; P < 0.01]. These data show that patients with flu‐like infection prior to ACS develop a greater infarct size in comparison to patients without flu‐like syndrome prior to ACS.

Follow‐up

Follow‐up data are presented in Table 3. Mean follow‐up time was 25.4 months (range 4–62 months). During the follow‐up, 30 patients (15.7%) experienced MACE. There were two sudden deaths (1‐STE‐ACS, 1‐nSTE‐ACS). Twenty‐eight patients (12 vs 16, respectively, for STE‐ACS vs nSTE‐ACS) were hospitalized for angina. There were neither other events nor cardiac hospitalizations. Both hospitalization for angina (P < 0.01) and overall MACE (P < 0.01) occurred significantly more frequently in patients with nSTE‐ACS in comparison to STE‐ACS.

Table 3.

Comparison of Outcome at Midterm Follow‐up between Patients with STE‐ACS and nSTE‐ACS

Data	STE‐ACS (n = 121)	nSTE‐ACS (n = 69)	P
Death	1 (0.8%)	1 (1.4%)	NS
Hospitalization for angina	12 (9.9%)	16 (23.1%)	<0.01
MACE	13 (10.7%)	17 (24.5%)	<0.01

Kaplan‐Meier event‐free survival analysis revealed a better outcome for STE‐ACS than nSTE‐ACS patients (log‐rank test P < 0.01) (Fig. 2).

Figure 2.

Kaplan‐Meier event‐free survival analysis in both groups of patients.

Demographic, clinical, biochemical, echocardiograhic, electrocardiographic and angiographic variables were entered into a regression analysis model in search for MACE predictors during the follow‐up period in the group of patients either with STE‐ACS or nSTE‐ACS with normal coronary arteries.

By univariate analysis in patients with STE‐ACS, five parameters: male sex (P = 0.02), WBC (P = 0.04), total cholesterol level (P = 0.02), hsCRP (P = 0.03), raised troponin I (P = 0.01) were associated with MACE (Table 4). By multivariate analysis, the outcome predictors were hsCRP (P = 0.03) and raised troponin I > 0.1 ng/ml (P = 0.02) (Table 5).

Table 4.

Prognostic Factors in Univariate Analysis of Patients with ST‐ACS and Normal Coronary Arteries

Variable	Hazard Ratio HR	95% Confidence Intervals	P
Total cholesterol	2.02	1.03–3.92	0.02
HsCRP	1.43	1.01–2.83	0.03
Male sex	0.16	0.02–0.93	0.02
WBC	1.33	1.05–1.69	0.04
Troponin I >0.1 ng/ml	6.83	1.61–28.86	0.01

HR = heart rate; hsCRP = high sensitivity C‐reactive protein; WBC = white blood cell count.

Table 5.

Prognostic Factors in Multivariate Analysis of Patients with STE‐ACS

Variable	Hazard Ratio HR	95% Confidence Intervals	P
hsCRP	2.84	1.01–7.96	0.03
Troponin I >0.1 ng/ml	1.40	0.40–2.00	0.02

hsCRP = high sensitivity C‐reactive protein.

In patients with nSTE‐ACS, variables which predicted the observed outcome in univariate analysis were: total cholesterol (P = 0.03), LDL cholesterol (P = 0.02), diastolic blood pressure (P = 0.04) and troponin I (P = 0.03) (Table 6). Fasting glucose (P = 0.03) and LDL cholesterol level (LDL) (P = 0.02) emerged as independent predictors of outcome in those patients (Table 7).

Table 6.

Prognostic Factors in Univariate Analysis of Patients with nST‐ACS and Normal Coronary Arteries

Variable	Hazard Ratio HR	95% Confidence Intervals	P
Total cholesterol	1.21	1.12–1.81	0.03
LDL cholesterol	1.67	1.32–1.89	0.02
DBP	1.73	0.63–2.54	0.04
Troponin I	1.22	0.78–1.91	0.03

LDL = low density lipoprotein cholesterol; DBP = diastolic blood pressure.

Table 7.

Prognostic Factors in Multivariate Analysis of Patients with nSTE‐ACS

Variable	Hazard Ratio HR	95% Confidence Intervals	P
Fasting glucose	1.20	1.01–1.44	0.03
LDL‐cholesterol	1.52	1.30–1.90	0.02

LDL = cholesterol—low density lipoprotein cholesterol.

To summarize, 10.7% of STE‐ACS patients had MACE during midterm follow‐up indicating that despite normal coronarography, their problem remained unsolved, and they were to be monitored closely. High hsCRP and troponin I were independent poor prognostic markers in this subgroup. With regard to nSTE‐ACS patients, nearly 25% of them had MACE during follow‐up with metabolic factors (fasting glucose and LDL cholesterol) emerging as poor prognostic factors in multivariate analysis.

DISCUSSION

In this study we found major differences between STE‐ACS and nSTE‐ACS patients with normal coronary arteries, both with regard to baseline characteristics, prognostic factors and mid‐term prognosis.

The main results of our study were as follows.

• • STE‐ACS patients, with normal coronary angiograms had more pronounced inflammatory background and a greater infarct size in comparison to nSTE‐ACS patients.
• • nSTE‐ACS patients were older and had more pronounced metabolic background (obesity, higher values of metabolic markers: fasting glucose, LDL cholesterol) in comparison to STE‐ACS patients.
• • STE‐ACS patients versus nSTE‐ACS, had fewer risk factors of atherosclerosis, and a better event‐free survival rate (89.3% vs 75.5%, respectively, P = 0.0079).
• • In multivariate analysis, in patients with STE‐ACS, raised troponin I (P = 0.023) and hsCRP (P = 0.031) were independent prognostic variables, while in nSTE‐ACS patients, the outcome was predicted by fasting glucose (P = 0.037) and LDL cholesterol levels (P = 0.02).

Our registry is one of the largest populations of patients with ACS and normal coronary angiograms. In a consecutive series of 4253 patients with ACS–190 (4.4%) of them had normal coronary angiography. In unselected groups of patients with ACS, the prevalence of normal coronary angiograms was reported from 1% to 8.5%. ² , ⁹ , ¹¹

The young age was pointed out in previous studies as a characteristic factor of patients with ACS and normal coronary arteries. ³ , ⁴ , ⁵ The mean age of patients in our study population was 53.2 years. In the articles published previously, the mean age of patients with ACS and normal coronary arteries ranged from 43 to 53 years. ⁵ , ⁶ , ¹⁰ In our study group, the patients with STE‐ACS and normal coronary arteries were significantly younger than those with nSTE‐ACS. Moreover, the percentage of young patients (below 35 years of age) was significantly higher among STE‐ACS patients (18%).

In our study group, ACS with normal coronary arteries was found in 63.1% of males. This is in agreement with previous studies showing the percentage of male sex in the population of patients with ACS and normal coronary arteries to be from 63 to 83%. ⁵ , ⁶ , ⁹ , ¹⁰ , ¹¹ We found that STE‐ACS patients were more often males in comparison to nSTE‐ACS patients. This finding may be age‐related, since nSTE‐ACS patients were significantly older.

STE‐ACS and nSTE‐ACS patients did not differ with regard to family history of CAD, smoking status, hypertension, diabetes and total cholesterol level indicating that both subgroups share some similar features. One‐fifth of our patients had positive family history of coronary artery disease (CAD). De Costa et al. found this rate to be a little lower—15% ⁶ and, together with Ulh and Farrell, ¹⁷ noted that family history of CAD had been observed with greater prevalence in patients with CAD when compared to patients with normal coronary arteries and ACS. There were 64.3% of smokers in the study population. Some investigators, ⁴ , ⁹ , ¹² but not all, ¹³ pointed out low incidence of smokers in the population of patients with normal coronary arteries and ACS. In fact, the percentage of smokers in our study was higher than in it is in all Polish population (44%). ¹⁴

Hypertension, hypercholesterolemia and diabetes were present in 60.5%, 39.4% and 10% of our patients, respectively. In the published literature these co‐morbidities were found in 15.4–60%, 20–67% and 8–11.8% (respectively) of patients with ACS and normal coronary arteries. ¹¹ , ¹² , ¹⁵ , ¹⁶ , ¹⁷ This very wide range of hypertension and hypercholesterolemia rates might be related to mean age, the characteristics of population in the country of origin and smaller study groups.

Of interest, in our study, significantly more risk factors of atherosclerosis were present in patients with nSTE‐ACS in comparison to those with STE‐ACS (P < 0.001); 67.2% of all our study population had more than one risk factor. Sarda et al. found that 70% of their patients, with myocarditis mimicking myocardial infarction, had cardiovascular risk factors. ¹⁸ However, Fournier et al. ⁵ found only 17% patients to have any of these risk factors. These differences may be accounted for different risk factors criteria and the fact that Fournier et al. investigated patients <40 years of age.

At least one‐third of our STE‐ACS patients had flu‐like infection prior to ACS, which could mimick acute myocarditis. MRI was found to be highly useful in distinguishing ACS from acute viral myocarditis. In the study by Codreanu et al. 12 of 27 consecutive patients (44%) presenting with ACS (chest pain, positive troponin‐I) and no coronary stenosis had subepicardial delayed contrast enhancement (DCE), a pattern typical of acute myocarditis. ¹⁹ The same proportion of patients (44%) had an ischemic pattern (transmural or subendocardial focal DCE) in this study. The importance of unrecognized coronary artery disease in patients with dilated cardiomyopathy was highlighted by McCrohon et al., ²⁰ who showed that at least 13% of patients with idiopathic dilated cardiomyopathy and normal coronary angiography had enhanced fibrosis typical of coronary occlusion on gadolinium MRI. Thus, at least a proportion of our patients who had STE‐ACS with ECG changes confined to single coronary artery territory and with significantly elevated markers of myocardial injury might have had transient single coronary artery occlusion that dissolved spontaneously without any luminal irregularity.

Outcome

In the published studies with the longest follow‐up period of patients with ACS and normal coronary arteries, Raymond et al. ²¹ and Zimmerman et al. ²² found a respective survival rate of 85 and 91%, during a follow‐up period of 10.5 and 7 years, respectively. There were no significant differences in the survival rate between patients with no atherosclerotic changes and patients who had coronary artery lumen reduction of up to 69%. ¹⁹ In agreement with those studies, we found a high survival rate of 98.9% at 2‐year follow‐up. Although this result seems to be excellent, two sudden deaths did occur. Of concern, recurrences of angina requiring hospitalization were observed in our study group in 14.7%. Other researchers showed that 5 up to 50% of patients had overall complications, in follow‐up periods of 1 and 3 years. ⁵ , ⁶ On the other hand, Maggi et al. showed that the incidence of death and recurrences of angina was 19% in a 12‐year follow‐up. ²³ A five time smaller group of patients and their younger age might be the cause of that discrepancy. The important observation was that in our series STE‐ACS patients had twice fewer complications than nSTE‐ACS ones (9.9% vs 23.1%, P = 0.0079).

In multivariate analysis, in STE‐ACS subgroup of patients the independent predictors of outcome were hsCRP level and raised troponin I level. Of note, according to previous literature an increase in hsCRP levels recorded in patients admitted after 6 hours following AMI symptom onset who also had elevated cardiac injury markers may partially depend on the cardiac injury. ²⁴ , ²⁵ However, in our series more than 90% of the patients were admitted to hospital before 6 hours elapsed from the beginning of ACS, which means that in our study group hsCRP reflects the baseline inflammatory status. Moreover, the concentration of hsCRP and any other inflammatory parameters (leukocyte count, ESR) did not correlate with cardiac injury markers.

On the other hand, fasting glucose level and LDL cholesterol level were identified as independent prognostic factors in patients with nSTE‐ACS. This might suggest that in nSTE‐ACS patients, factors predicting the outcome were also factors responsible for atherosclerosis development, and therefore these metabolic factors should be a target of effective therapy.

LIMITATIONS OF THE STUDY

We did not perform procedures focused on vessel spasm. The limitation is the absence of intravascular ultrasound analysis, due to the fact that coronary angiography is not a reliable method to detect subtle abnormalities in the arterial wall that may not obstruct the artery lumen but may nevertheless provide an important nidus for inflammation and a high‐risk atherosclerotic plaque. Another limitation is that the interview questions concerning baseline characteristics were formulated for routine analysis of ACS patients, and were not specifically directed toward addressing hypotheses suggested by the differences in the two groups of patients in this analysis.

CONCLUSIONS

To sum up, a small proportion of ACS patients (4.4%) had normal coronary angiograms.

STE‐ACS and nSTE‐ACS patients with normal coronary arteriography have different clinical profiles. In nSTE‐ACS patients we found more pronounced metabolic abnormalities, while in STE‐ACS patients the inflammatory background was more significant. Mid‐term prognosis was favourable (98% survival), though two sudden deaths occurred. Of note, the MACE rate in STE‐ACS patients was half as that in nSTE‐ACS ones. Nevertheless, both STE‐ACS and nSTE‐ACS patients who have normal coronary angiograms require careful follow‐up and strict secondary prevention measures, including smoking cessation, and good control of hypertension, dyslipidemia and diabetes.