Background
Normal coronary arteries may coexist with abnormal coronary and systemic endothelial function in patients with chest pain [
1]. The functional status of systemic endothelium somewhat mirrors the coronary endothelial function [
2,
3] and can be evaluated non-invasively with brachial ultrasound [
4]. A novel, provocative paper by the Pisa echo-lab group [
5,
6], suggests that endothelial dysfunction strictly correlates with ST-segment depression (STSD) during pharmacologic (as well as exercise) stress, while epicardial coronary artery stenosis exclusively affects wall motion abnormalities (WMA), although both components may coexist; they demonstrated that abnormal brachial artery flow-mediated dilation (measured with brachial artery ultrasound) was predicted by STSD during SE but not by either presence of WMA during SE or angiographically significant CAD. There was no correlation between brachial artery flow-mediated dilation and extent of angiographical CAD.
Therefore isolated STSD during SE could be used as a surrogate for coronary endothelial dysfunction, in the absence of stress-inducible WMA and in the absence of angiographically-significant CAD.
The majority of cardiovascular benefits of PUFAs are likely to be mediated in the vascular wall and at the vascular endothelium level. The long chain PUFAs have been reported to possess several properties that may positively influence vascular function. These include favorable mediator profiles (nitric oxide, eicosanoids) that influence vascular reactivity, change in vascular tone via actions on selective ion channels [
7], and maintenance of vascular integrity. In addition to direct effects on contractility, PUFAs may affect vascular function by modifying expression of inflammatory cytokines and adhesion molecules. These properties may explain the beneficial cardiovascular protection of this family of fatty acids that have been clearly evident through epidemiological data as well as from few clinical trials [
8,
9].
Although many epidemiological surveys and basic research studies have suggested that "marine fish oil" cardioprotective effect is consistent and at least partly mediated through an action on the endothelium, no large clinical intervention trial has yet addressed the effect of commercially-available PUFAs preparations on a clinically-measurable direct marker of coronary endothelial dysfunction in normocholesterolemic subjects [
10,
11]. A small, randomized study performed in hypercholesterolemic subjects suggests a positive effect of 4-month marine oil supplementation on brachial artery endothelium-mediated vasodilation.[
12]
The present study's hypothesis is that a 4 month-course of oral supplementation with PUFAs can reverse coronary endothelial dysfunction. Eventual "normalization" of endothelial dysfunction was measured, at the end of the supplementation period, by the absence of both significant STSD and WMA during repeat SE.
Methods
Patients were enrolled in the study if they met the following inclusion criteria: (1) chest pain syndrome; (2) reference for SE; (3) presence of STSD (strict ECG criteria were applied: a) STSD ± 1 mm 0.08 s after the J point in the peripheral leads and/or b) STSD ≥ 2 mm 0.08 s after the J point in the precordial leads) during a SE with no changes in wall motion score, (4) coronary angiogram performed within 20 days after SE, (5) no nitrate therapy at the time of SE.
When meeting the aforementioned entry criteria, patients had their total cholesterol measured by means of a standard whole-blood sample.
Subjects were referred for SE because of one of the following: (a) patients with positive exercise-ECG in concomitance with conditions lowering the positive predictive value of the ECG marker of ischaemia (for instance, female gender or arterial hypertension); (b) patients with recurring chest pain in the absence of significant electrocardiographic changes during routine exercise-ECG; (c) patients unable to exercise. All of the patients underwent dipyridamole SE.
The following were exclusion criteria: (1) uninterpretable resting ECG (left bundle branch block, preexcitation, baseline STSD or elevation > 1 mm, or typical digitalis repolarization pattern); (2) technically poor acoustic window; (3) cardiomyopathy or severe valvular disease; (4) previously detected CAD; (5) statin therapy; (6) high total cholesterol levels (defined as total cholesterol ≥ 240 mg/dl as suggested by current NCEP guidelines).
The study was approved by the institutional review board. All patients gave their written informed consent before entering the study.
Stress echo was performed with dipyridamole up to 0.84 mg/kg over 10 min with atropine up to 2 mg when needed; stress echo was repeated using the same protocol after a 4-month course of oral supplementation with commercially available (Seacor®, Esapent®, Eskim®) n-3 PUFA at a dose of 1 g once a day. An imaging system with digital acquisition was used (GE-Vivid 7). All standard echocardiographic views were obtained when possible. The left ventricle was divided into 16 segments, as suggested by the American Society of Echocardiography. Segmental wall motion was graded as follows: normal = 1; hypokinetic = 2; akinetic = 3; and dyskinetic = 4. Inadequately visualized segments were not scored. Stress echo was considered positive when one left ventricular segment was increased by one grade or more at peak stress. Normalization of coronary endothelial dysfunction was defined as isoelectric ST segment 0.08 s after the J point in the peripheral leads and STSD < 0.5 mm 0.08 s after the J point in the precordial leads during repeat SE with no WMA.
Coronary angiography in multiple views was performed according to the standard Judkins or Sones technique.
The percent diameter stenosis was determined by quantitative coronary angiography. A vessel was considered to have significant obstruction if its diameter was narrowed by 50% or more, with respect to the pre-stenotic tract.
Results
Nine consecutive patients were prospectively enrolled according to inclusion/exclusion criteria during the time period between January 2003 and June 2003. Three patients (out of the 12 who fulfilled entry criteria) were excluded because of the presence of at least one exclusion criteria (2 subjects had high total cholesterol levels as per protocol defined and one was already on statin therapy).
Five patients were male (55%); mean age was 60(± 5).
Four patients (44%) had hypertension (treated with ACE-inhibitors, ARBs, diuretics or combinations of them), 3(33%) had diabetes (treated with a combination of metformin and a sulfonylurea), 2(22%) had total cholesterol levels in the range between 200 mg/dl and 240 mg/dl (defined as borderline high by current NCEP guidelines), 3(33%) were active smokers. None was under antiischaemic therapy (beta-blocker and/or non-dihydropiridinic calcium antagonist) at the time of SE testing. All of the patients maintained their usual therapy througout the study time period, with no exceptions.
All but 2 subjects had normalization of STSD during repeat SE performed after the 4 month course of therapy. Of the two subjects still presenting STSD at repeat SE, one was hypertensive and non-diabetic while the other was diabetic but non-hypertensive).
Discussion
We tested the study hypothesis in a small series of consecutive subjects, with the intent to produce a small hypothesis-generating study; testing the hypothesis in a large, randomized, placebo-controlled trial was felt premature, both because we used an innovative, but not definitely validated marker of coronary endothelial dysfunction and because lack of strong clinical data about the effect of PUFAs supplementation on human coronary endothelium.
Our study, as any small "proof of concept" study, has many limitations:
1) It is a prospective, non-randomized, uncontrolled interventional study, 2) it is performed in a very small group of subjects, 3) it is suggestive for a positive-reversal effect of PUFAs on isolated, angiographically-negative STSD during SE, but even if recent data strongly suggest this clinical pattern to be linked with coronary endothelial dysfunction (5,6), this is not a direct measure of endothelial function and, also, no data exist demonstrating that reversal of STSD indicates normalization of coronary endothelial dysfunction.
Of course no subgroup analysis can be performed in such a small group of subjects, even if it is interesting that non-responders (2 out of 9 subjects) consisted of one hypertensive, non-diabetic subject and one diabetic non-hypertensive subject, suggesting that inefficacy of therapy may not be linked to the presence/absence of these clinical conditions.
Even if evidence is strong for a negative prognostic effect of coronary endothelial dysfunction on mortality, atherosclerosis progression and plaque instability [
13], coronary endothelial function is to date difficult to measure on clinical grounds: hence it is obvious cardiologists prefer to dilate what they can clearly see (epicardial stenosis) rather than pharmacologically engage what they can not measure (endothelium).
An easy to obtain, clinically useful marker of coronary endothelial dysfunction such as isolated STSD during SE in the absence of visible CAD would be very welcome in the cardiology arena, while the striking effect exerted by PUFAs in this small study would be even more valuable.
Conclusions
Striking rate of reversion of stress-induced ST-segment depression after oral n-3 PUFAs suggests reversion of coronary endothelial dysfunction; nonetheless these data need to be validated in larger, placebo-controlled studies utilizing direct measurements of coronary endothelial function.
Competing interests
None declared.
Authors' contributions
NG conceived the study and materially performed echocardiographic examinations, VG participated in the coordination of the study and revision of the paper. All authors read and approved the final manuscript.