Nitrate-induced paradoxical ischemia predicts adverse outcomes in elderly patients with healed myocardial infarcts

Abstract

BACKGROUND:

It is well known that nitrates can induce paradoxical myocardial ischemia.

METHODS AND RESULTS:

Fifty patients (median age 73 years; range 67 to 78 years; 80% male) with healed myocardial infarcts were selected. All patients underwent resting single-photon emission computed tomography (SPECT) and resting baseline gated-SPECT using sestamibi or thallium-201 after the sublingual administration of 5 mg isosorbide dinitrate (ISD). Forty-eight per cent (24 of 50) of the patients demonstrated ISD-induced peri-infarct ischemia as observed by SPECT. Compared with patients without ISD-induced ischemia, patients with ISD-induced ischemia presented larger infarcts as determined by the extent of perfusion defects (mean [± SD] 27±12 pixels versus 11±9 pixels; P<0.0001), lower ejection fractions (39±17% versus 50±15%; P<0.02) and a higher incidence of severe coronary artery disease (P<0.04). At five years, the survival probability on Kaplan-Meier analysis was 42% and 96% for patients with and without ISD-induced ischemia, respectively (HR 5.6 [95% CI 1.6 to 20]; P=0.009).

CONCLUSIONS:

Nitrates may have low efficacy in improving blood flow through the coronary vessels that supply large myocardial infarcts with high-resistance microvascular damage. At the same time, nitrates induce dilation and blood pressure decrease in remotely patent or mildly stenotic vessels. The blood pressure gradient elicited between the high- and low-resistance coronary vessels may provide the force for a blood flow steal from the viable zones of the infarct toward the healthy myocardium. The resultant nitrate-induced paradoxical ischemia could be a silent marker of myocardial instability and adverse outcomes in elderly patients with healed myocardial infarcts.

Evaluating residual myocardial viability (1) is crucial in patients with healed myocardial infarcts. The detection of myocardial viability in such patients is essential for selecting appropriate medical therapy or revascularization, and for prognostic assessment. The use of organic nitrates has been determined to be a safe and valid diagnostic tool for evaluating myocardial vitality when used in combination with single-photon emission computed tomography (SPECT) (2,3). This technique is based on the fact that organic nitrates administered intravenously or sublingually improve perfusion and, thus, improve the incremental isotope uptake of residual viable zones located across the borders of or within the myocardial scar. Nevertheless, several clinical studies report the occurrence of severe paradoxical effects of diverse nitrate compounds in provoking unexpected ischemia (4–11). Therefore, the use of sublingual organic nitrates in patients at rest before cardiac SPECT imaging is a potential alternative for detecting residual myocardial ischemia in asymptomatic patients with healed myocardial infarcts. Fifty such patients were studied via resting cardiac SPECT using technetium-99m sestamibi (MIBI) or thallium-201 after the sublingual administration of 5 mg of isosorbide dinitrate (ISD) and at baseline. The results were evaluated after a long-term clinical follow-up period.

METHODS

A total of 50 asymptomatic patients with a median age of 73 years (range 67 to 78 years; 80% male) and healed myocardial infarcts that occurred one to three years before the study were assessed for myocardial perfusion and left ventricular (LV) evaluation using SPECT imaging after sublingual ISD administration. This technique was performed because these patients were not eligible for exercise testing or commonly used pharmacological stress tests due to an existing high incidence of physical disability, multiple pathologies and advanced age.

Patients were studied using SPECT from 2003 to 2007, and underwent a median of 4.5 years (interquartile range 4.2 to 4.9 years) of clinical follow-up evaluations consisting of periodic medical assessments or inhospital readmissions. A formal ethics review committee was consulted and all patients provided informed consent; the principles outlined in the Declaration of Helsinki were carefully followed.

Patients

The presence of established risk factors, therapy type and cardiovascular (CV) events that occurred during the follow-up period were carefully analyzed for all patients. Among all patients, none were smokers. Coronary angiography was performed one to two years before the study. Coronary artery disease (CAD) was classified as severe (=1) when at least two to three blood vessels were ≥70% obstructed, or less severe (=0) in individuals with left anterior descending artery or left main CAD, and in cases of one-vessel disease.

Nuclear techniques

Cardiac SPECT was performed on all patients using a double-headed gamma camera (DST-XL; Sopha Medical Vision International, France).

MIBI

In 52% (26 of 50) of patients, 5 mg of ISD was administered sublingually, followed 3 min to 5 min later by an injection of a mean dose of 370 MBq of MIBI, which is commercially available as Cardiolite (Bristol-Myers Squibb Pharma, Belgium) (12). Cardiac SPECT imaging was performed 30 min later. Three to four hours after cardiac SPECT imaging, a mean dose of 925 MBq of MIBI was injected, and baseline images were acquired 1 h later. The gamma camera was equipped with a low-energy, parallel-hole, high-resolution collimator.

Thallium-201

In 48% (24 of 50) of patients, 5 mg of ISD was administered sublingually, followed within 3 min to 5 min by an injection of 74 MBq to 148 MBq of thallium-201 (Amersham PLC, United Kingdom). The initial SPECT images were recorded 2 min to 3 min after the injection, and the baseline images were obtained 3 h later. A gamma camera was equipped with a high-resolution, parallel-hole collimator (Figure 1). The patients were randomly assigned to sessions with either MIBI or thallium-201. Despite the technical differences between these tracers (13), both perform well and have high sensitivity and specificity for angiographic stenosis with no difference in accuracy (14). For all patients, SPECT images were obtained via a 180° circular orbit from a 45° right anterior oblique to a 45° left posterior oblique, 32-frame step-and-shoot, 60 s/frame, with the patient resting supine. Only the projections that were acquired 3 h to 4 h after ISD administration were electrocardiogram (ECG)-gated for ejection fraction (EF) evaluation, with 16 individual ECG-gated frames per cardiac cycle. The gate tolerance was 100%. All patients in the present study were in sinus rhythm. The isotope dosage was adjusted according to the patient’s body weight. No attenuation or scatter correction was applied. The extent of abnormal perfusion defects (ADS) and the LV EFs were analyzed using Cedars-QGS-CEqual software (Cedars-Sinai Medical Center, USA) (15). Bull’s eye polar maps were obtained using SPECT imaging, and the ADS size was quantified as a percentage of the entire pixels of the LV surface. The reversibility of the defects was calculated as the percentage of the ADS pixels (ischemia). ADS were also evaluated visually (16), and the regional site of ADS was compared with the corresponding regional segmental LV dysfunction detected via echocardiography to avoid misinterpreting attenuation artefacts. All data were evaluated by two independent observers for consensus.

Figure 1).

Thallium single-photon emission computed tomography imaging of a 60-year-old man (patient 1 [Table 2]) reveals the reversibility of septal (white arrows), anterior (yellow arrow), and lateral (green arrow) defects and mild reversibility of the inferior defect, representing the infarct zone (red arrow). The colours representing zones of different isotope uptake suggest that ischemia is mostly peri-infarct: healthy myocardium (red), necrosis (black spot), ischemia (white spot), and damaged peripheral hypoperfused zones (magenta). ISD Isosorbide dinitrate

Statistical analyses were performed using MedCalc version 12.0.3.0 (MedCalc Software, Belgium) for Windows (Microsoft Corporation, USA). Patient age and follow-up durations were expressed as median ± interquartile ranges, all other values were expressed as means ± SDs. The differences between means were evaluated using unpaired or paired Student’s t tests. In cases of unequal variances, the Student’s t test was corrected using the Welch test. The differences between percentages were calculated using Fisher’s exact test. The effect of the risk parameters on all-cause CV events was analyzed using logistic regression. Survival was analyzed using Kaplan-Meier analysis with the log-rank test; P<0.05 was considered to be statistically significant.

RESULTS

Overall, ISD-induced peri-infarct ischemia occurring silently and without ST depression was detected in 48% (24 of 50) of patients using SPECT imaging (Figures 1 and 2). In 62% (15 of 24) of these patients, the peri-infarct ischemia was extended to core infarct zones. Only two patients presented with severe transient hypotension and bradycardia within 20 min of ISD administration. The patients with ISD-induced ischemia demonstrated a more severe extent of ADS, lower EF values, more severe CAD and a higher incidence of major adverse CV events and cardiac death compared with patients without ISD-induced ischemia. There was no difference in the number of established risk factors, type of treatment or clinical assessment (Table 1). The clinical details of the patients with and without ISD-induced ischemia are described in Tables 2 and 3, respectively. On logistic regression analysis, ISD-induced ischemia was significantly predictive for cardiac events (OR 9.9 [95% CI 2.4 to 36; P<0.001]) but not the ADS extent of EFs values (categorical values), severity of CAD and established risk factors (all dichotomous values) (P>0.05 for all). Based on a Kaplan-Meier analysis at five years, the survival probability was 42% for patients with ISD-induced ischemia and 96% for those without ISD-induced ischemia (HR with ISD-induced ischemia 5.6 [95% CI 1.6 to 20]; P=0.009) (Figure 3).

Figure 2).

Technetium-99m sestamibi (MIBI) single-photon emission computed tomography imaging of a 70-year-old man (patient 4 [Table 2]) reveals the reversibility of the anterior-apical (yellow arrow), septal (white arrows) and lateral (green arrow) defects and mild reversibility of the inferior defect representing the infarct zone (red arrows). Polar map colours representing zones of different isotope uptake suggest that ischemia is mostly peri-infarct: healthy myocardium (red), ischemia (white spots), damaged peripheral hypoperfused zones (magenta). ISD Isosorbide dinitrate

TABLE 1.

Patient clinical characteristics

Characteristic	ISD-induced ischemia on SPECT		P
	Without (n=26)	With (n=24)
Age, years, median ± SD	73±9	71±7	NS
Diabetes	5/26 (19)	11/24 (45)	NS
Hypertension	24/26 (92)	20/24 (83)	NS
Dyslipidemia	21/26 (81)	20/24 (83)	NS
At least ≥2 risk factors	17/26 (65)	20/24 (83)	NS
CAD severity
Previous PTCA	13/26 (50)	16/24 (67)	NS
Previous coronary bypass surgery	11/26 (42)	7/24 (29)	NS
2 to 3 vessels with ≥70% obstruction	3/26 (11)	9/24 (37)	0.04
Gated-SPECT results
Thallium-201 SPECT	11/26 (42)	13/24 (54)	NS
MIBI SPECT	15/26 (58)	11/24 (46)	NS
Average ADS extent as % of LV total pixels, mean ± SD	11±9	27±12	0.0001
ADS reversibility (ischemia) as % of the ADS pixels, mean ± SD	0.4±1	38±28	0.0001
Ejection fraction, %, mean ± SD	39±17	50±15	0.02
Follow-up
Therapy
Angiotensin-converting enzyme inhibitors	9/26 (35)	14/24 (58)	NS
Angiotensin II receptor antagonists	8/26 (31)	4/24 (17)	NS
Diuretics	8/26 (31)	13/24 (54)	NS
Nitrates	12/26 (46)	13/24 (54)	NS
Nondihydropyridine calcium antagonists	5/26 (19)	8/24 (33)	NS
Statins	15/26 (58)	16/24 (67)	NS
Acetylsalicylic acid	24/26 (92)	21/24 (87)	NS
Oral anticoagulants	2/26 (8)	2/24 (8)	NS
Combined therapy with at least 3 drugs	23/26 (88)	22/24 (92)	NS
Clinical assessment
Outpatients, mean ± SD	10±6	7±6	NS
Inpatients, mean ± SD	1.7±3	2.2±2.5	NS
MACEs	9/26 (34)	20/24 (83)	0.0006
Cardiac death	2/26 (8)	8/24 (33)	0.004

Data presented as n/n (%) unless otherwise indicated. ADS Abnormal perfusion defects; CAD Coronary artery disease; HF Heart failure; ISD Isosorbide dinitrate; LV Left ventricle; MACEs Major adverse coronary events; MIBI Technetium-99m sestamibi; NS Not statistically significant; PTCA Percutaneous transluminal coronary angioplasty; SPECT Single-photon emission computed tomography

TABLE 2.

Details of coronary artery disease severity and follow-up outcomes in patients with isosorbide dinitrate-induced ischemia (n=24)

Age, years/sex	Clinical situation before gated SPECT
	MI site	LAD	LCX	RCA	With (1) or without (0) ≥2 vessels with ≥70% obstruction	Follow-up data
						MACEs	Deaths
60/male	Inferior	Occluded	Stent	Occluded	1	Angina recurrence, HF	0
63/male	Inferior	Patent	Patent	Stent	0	Atrial fibrillation, HF	0
75/male	Inferior	PTCA	40% stenosis	Occluded	0	Bypass	0
70/male	Inferior	Stent	Stent	Stent	0	MI	Fatal MI
84/male	Anterior-inferior	Bypass	Bypass	Bypass	0	Angina recurrence, HF	Irreversible HF
85/male	Inferior	PTCA occlusion	≥70% stenosis	≥70% stenosis	1	Angina recurrence, HF	Irreversible HF
65/male	Anterior-inferior	Stent	70% stenosis	Occluded	1	Angina recurrence, HF, LCX stenting	0
77/male	Anterior-inferior	Patent	Patent	50% stenosis	0	HF, pacemaker	0
80/male	Inferior-anterior	Stent	≥70% stenosis	≥70% stenosis	1	LCX stenting	Irreversible HF
75/male	Inferior	≥70% stenosis	≥70% stenosis	Occluded	1	HF	Irreversible HF
86/female	Inferior	≥70% stenosis	patent	Patent	0	LAD stenting	0
64/male	Inferior	Stent	Occluded	40% stenosis	0	ICD	0
66/male	Apical	Bypass	Patent	Patent	0	0	0
69/male	Anterior-inferior	Stent occlusion	Patent	Patent	O	Angina recurrence, HF	Irreversible HF
54/male	Inferior	Stent occlusion	≥70% stenosis	≥70% stenosis	1	MI, bypass	0
81/male	Inferior	Bypass	Bypass	Bypass	0	0	0
87/female	Anterior	Patent	Patent	Patent	0	Angina recurrence	0
81/male	Inferior	patent	≥70% stenosis	≥70% stenosis	1	LCX, RCA stenting	Irreversible HF
68/male	Anterior-lateral	≥70% stenosis	≥70% stenosis	≥70% stenosis	1	Angina recurrence, HF	Irreversible HF
74/female	Inferior	Bypass	Bypass	Bypass	0	0	0
81/male	Inferior	Stent	≥70% stenosis	Occluded	1	0
63/male	Inferior	Patent	30% stenosis	Stent	0	Angina recurrence	0
75/male	Inferior	Patent	Patent	≥70% stenosis	0	RCA stenting	0
68/male	Inferior	Patent	Patent	Minimal obstructions	0	High-grade ventricular arrhythmias	0

HF Heart failure; ICD Implantable cardioverter defribrillator; LAD Left anterior descending artery; LCX Left circumflex artery; MACEs Major adverse coronary events; MI Myocardial infarction; PTCA Percutaneous transluminal coronary angioplasty; RCA Right coronary artery; SPECT Single-photon emission computed tomography

TABLE 3.

Details of severity of coronary artery disease and follow-up outcomes in patients without isosorbide dinitrate-induced ischemia (n=26)

Age, years/sex	Clinical situation before gated-SPECT
	MI site	LAD	LCX	RCA	With (1) or without (0) ≥2 vessels with ≥70% obstruction	Follow-up data, n
						MACEs	Deaths
70/male	Inferior	Stent	Bypass	Bypass occlusion	0	0	0
77/male	Anterior	Occluded	Stent	Patent	0	Angina recurrence	0
67/male	Anterior-apical	Occluded	≥70% stenosis	Patent	1	0	Fatal MI
78/male	Inferior	Bypass	Bypass	Bypass	0	0	0
77/male	Anterior	Bypass	60% stenosis	Bypass	0	0	0
68/male	Inferior	Patent	Patent	≥70% stenosis	0	0	0
78/male	Anterior	Bypass	Bypass	Bypass	0	0	Noncardiac death
61/female	Anterior	Bypass	Bypass	Bypass	0	0	0
75/male	Anterior	Bypass	Bypass	Bypass	0	0	0
74/male	Anterior	Occluded	Patent	Patent	0	1	0
68/male	Inferior	Bypass	Bypass	Bypass	0	Carotid stenting	0
72/female	Inferior	60% stenosis	Stent occlusion	40% stenosis	0	LAD PTCA	0
80/female	Inferior	Stent	Patent	Stent	0	LAD stenting	0
70/male	Inferior/lateral/apical	Bypass	Bypass	Patent	0	Pacemaker	0
70/male	Inferior	Patent	Patent	≥70% stenosis	0	RCA stenting	Fatal MI
63/male	Inferior	Bypass	Bypass	Bypass	0	MI	0
62/male	Anterior/inferior	Stent	60% stenosis	Patent	0	0	0
51/male	Lateral	≥70% stenosis	Occluded	Occluded	1	LAD stenting	0
63/female	Anterior/inferior	≥70% stenosis	≥70% stenosis	Patent	1	LAD and RCA stenting	0
76/male	Anterior/lateral	Patent	Patent	Patent	0	0	0
82/male	Inferior	Bypass	Bypass	Bypass	0	0	0
82/male	Inferior	Stent	PTCA	Patent	0	0	0
78/male	Inferior	Patent	PTCA	Occluded	0	0	0
78/male	Inferior	PTCA	Patent	Patent	0	0	0
78/male	Anterior	Bypass	Bypass	Bypass	0	0	0
78/male	Apical	40% stenosis	Patent	Patent	0	0	0

LAD Left anterior descending artery; LCX Left circumflex artery; MACE Major advese cornary event; MI Myocardial infarction; PTCA Percutaneous transluminal coronary angioplasty; RCA Right coronary artery; SPECT Single-photon emission computed tomography

Figure 3).

Kaplan-Meier curve stratification of individuals with cardiac death. At five years, the survival probability was 42% and 96% in patients with and without isosorbide dinitrate (ISD)-induced ischemia, respectively

DISCUSSION

The anti-anginal actions of organic nitrates improve coronary blood flow through vasodilation of capacitance veins and conductive arteries, the reduction in ventricular volumes, preload, afterload and the myocardial blood flow necessary for myocardial oxygen requirements (17). These compounds may elicit supplemental blood flow to the residual viable zones of healed myocardial infarcts via the vasodilation of collateral vessels (18,19). The ISD-induced peri-infarct ischemia observed in patients in the present study, therefore, appears to be paradoxical and an unexpected effect that needs to be explained and analyzed in detail. A severe decrease in arterial blood pressure may occur after sublingual nitrate administration (20,21); the resultant reflex increases in heart rate and cardiac output may extend the region of myocardial ischemia (22). This phenomenon is more likely to be observed in older patients because the nitrate plasma half-life is longer and volume of distribution is larger, and there is greater frequency of nitrate-induced severe hypotension and bradycardia compared with younger patients (23). Hypotension itself, thus, plays a causal role in the diminished coronary perfusion pressure and greater ischemia (24). Following reperfused acute myocardial infarction, there is also a variable degree of microvascular damage, cellular inflammation and alterations in mitochondrial and cytosolic calcium ion homeostasis (25). Such reperfusion-elicited tissue injury may result in hypoxia and microvascular dysfunction with the activation of programmed cell death (26). A cellular mechanism can be postulated as the underlying cause for this phenomenon. The ischemic myocardium is characterized by the presence of an increased level of G protein-coupled receptor kinase-2 receptors (GRK2s) that are deleterious to cardiac myocyte survival and inhibit the activation of nitric oxide production by the G protein-coupled receptors, which protect the myocardium in ischemia/reperfusion injury (27,28). A similar mechanism characterizes the endothelial dysfunction that may cause silent ischemia in diabetic patients (29).

Despite the fact that hypotension, old age, ischemia/reperfusion phenomenon, cellular mechanisms, endothelial dysfunction and nitrate tolerance (30) are potential undesirable mechanisms that may favour the occurrence of myocardial ischemia or counteract the beneficial actions of nitrates, an alternative explanation for the patients’ ISD-induced ischemia cannot, however, be ruled out. Patients with ISD-induced ischemia, compared with those without, exhibited a more severe degree of CAD and a larger infarct size. The administration of nitrates is unable to elicit an increase in blood flow across occluded or severely stenotic coronary arteries (31,32). Nitrates do not improve blood flow in collateral vessels that supply areas of high-resistance microvascular damage (33–37). but dilate the mildly stenotic or remote patent coronary vessels of the healthy myocardium (38). As a response, the prestenotic infarct-related artery and collaterals generate high blood pressure as a reaction to the high flow resistance, and the healthy myocardium vessels present low blood pressure in relation to the low flow resistance (39). The resultant pressure gradient can, thus, provide the proportional force (40) for accelerating a blood flow steal from the residual viable infarct zones toward the healthy myocardium.

CONCLUSIONS

A healed cardiac infarct is described as a zone of scar tissue without cellular infiltration. Enhanced coronary perfusion may, however, produce myocytes with contraction bands and large quantities of extravasated erythrocytes (41). Previous studies have suggested that peri-infarct border zone heterogeneity represents an important predictive arrhythmogenic substrate for sudden death due to fatal arrhythmias or reinfarction that are not preceded by symptoms and ECG changes (42–44). These hypotheses suggest that ISD-induced peri-infarct ischemia could be an adjunct factor of myocardial instability in the presence of a large-size myocardial infarct, severe CAD and LV dysfunction, which characterized patients in the present study who experienced adverse CV events. The cause of cardiac death in 87% of the cases was irreversible heart failure that was preceded by major adverse CV events with diverse clinical presentations of angina recurrence. Nitrate-induced myocardial ischemia could, therefore, be considered a silent marker for myocardial instability and poor prognosis in patients with apparently stable, healed myocardial infarctions. However, the present study had limitations due to the relatively small patient sample size whose coronary artery angiograms were analyzed retrospectively; therefore, the role of collaterals was not evaluated in detail. In addition, these conclusions were attained by considering the coronary artery system as an ideal and theoretical hydraulic model rather than based on experimental findings. Another limitation of the present study was the use of discrete rather than continuous variables for the evaluation of established risk factors in the survival analyses (45). Blood pressure, plasma glucose and lipid values may exert a dependent effect on CV events (46,47). Despite these limitations, ISD-induced peri-infarct myocardial ischemia, which was clearly documented by SPECT imaging, demonstrated consistent prognostic clinical validity. These results will need to be confirmed in a study with a much larger patient population. Chronic therapy with nitrates has been associated with significantly increased mortality (48–50) or without improved survival (51), toxicity (52) and tolerance (53,54), while the mechanisms of such deleterious effects remain incompletely clarified. The present study suggests that a nitrate-induced coronary blood flow steal could be responsible for eliciting residual peri-infarct ischemia in elderly patients with a large, healed myocardial infarct and severe CAD.