Prevention of postcoronary angioplasty restenosis by omega-3 fatty acids: Main results of the Esapent for Prevention of Restenosis ITalian Study (ESPRIT)☆☆☆★

Article Outline

• Abstract
• Methods
  • Laboratory evaluations
  • Evaluation of compliance
  • Statistical analysis
• Results
  • Patient characteristics
  • Patient follow-up
    • Restenosis rate
  • Clinical events
  • Adverse effects
  • Evaluation of compliance
  • Laboratory analyses
• Discussion
• Acknowledgements
• Appendix A
• References
• Copyright

Abstract 

Background and Objective Previous trials of omega-3 fatty acids (ω-3 FA) for restenosis prevention after percutaneous transluminal coronary angioplasty (PTCA) have yielded conflicting results. We tested the hypothesis that long-term administration of ω-3 FA before PTCA may have significant effects on restenosis. Methods We randomized 339 patients in a double-blind, placebo-controlled study of ω-3 FA (as an ethyl ester preparation given as 6 1-g capsules providing 3 g eicosapentaenoic acid and 2.1 g docosahexaenoic acid/d started 1 month before PTCA and given for 1 month thereafter, then continued at half-dose for 6 months) versus an olive oil placebo. Of these, 257 patients (125 on ω-3 FA, 132 on placebo) well matched for risk factors underwent successful balloon-only PTCA (280 total lesions) and were evaluable at 6 months with repeat angiography. Restenosis was defined at quantitative angiography as a recurrence of >50% diameter stenosis in the dilated vessel (Definition I) and as >50% loss of the short-term gain immediately after PTCA (Definition II). Results Restenosis rates per vessel were 29.4% and 31.6% in the ω-3 FA group, and 39.6% and 35.4% in the placebo group according to Definitions I (P = .04) and II (P = not significant), respectively. Restenosis rates per patient were 31.2% and 33.6% in the ω-3 FA group, and 40.9% and 37.1% in the placebo group according to Definitions I (P = .05) and II (P = not significant), respectively. Conclusions With a long treatment before PTCA, ω-3 FA produced a small but significant decrease in the restenosis rate compared with placebo. (Am Heart J 2002;143:e5.)

Epidemiological studies have shown an inverse relationship between dietary intake of omega-3 fatty acids (ω-3 FA) and mortality from cardiovascular disease.¹, ², ³, ⁴ In some experimental models, ω-3 FA supplementation has reduced the incidence of native atherosclerosis as well as balloon-injury-induced atherosclerosis in the cholesterol-fed pig model.⁵, ⁶, ⁷, ⁸ Restenosis after percutaneous transluminal coronary angioplasty (PTCA) has attracted clinical and pathophysiological interest both because of its incidence, which considerably limits the long-term impact of PTCA,⁹, ¹⁰ and because of its possible use as a model of arteriosclerosis. ω-3 FA have been the most studied medical intervention in post-PTCA restenosis.¹¹, ¹², ¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰, ²¹, ²² However, incomplete knowledge of the mode of action of these naturally occurring substances²³, ²⁴ has led to remarkably different designs of clinical studies testing the hypothesis of effects of ω-3 FA on post-PTCA restenosis.¹¹, ¹², ¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰, ²¹, ²² Studies have varied in dosages, mode of administration (triglyceride vs ethyl-ester preparations), and timing of treatment with respect to PTCA.¹¹, ¹², ¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰, ²¹, ²² One meta-analysis combining the results of the first 7 fully published studies²⁵ concluded that the combined data were “compatible with a small-to-moderate effect of fish oil on restenosis” requiring confirmation in larger randomized clinical trials. However, it also noted significant inhomogeneity among the different studies.²⁵ The 2 larger most recent studies of ω-3 FA on restenosis reported negative results.²¹, ²² While those studies were in progress, we hypothesized that a potential reason for the inhomogeneity in the results of previous trials was the timing of onset of ω-3 FA administration in relation to the timing of PTCA. Many of the effects of ω-3 FA require incorporation in membrane phospholipids, a process that depends on the turnover rate of individual tissues.²⁶ Vascular smooth muscle cells injured at the time of PTCA respond with initially rapid proliferation.²⁷ Incorporation of ω-3 FA into membrane phospholipids must occur before this proliferation. This likely requires more than the 2 weeks used in one large study²¹ and certainly more than the 0 to 7 day pretreatment used in most other studies. We therefore set up this trial to evaluate the effect of ω-3 FA supplementation given ≥4 weeks before PTCA on the 6-month occurrence of restenosis, by use of quantitative coronary angiography (QCA) performed at the time of PTCA (before and immediately after) and 6 months later.

Back to Article Outline

Methods 

The Esapent for Prevention of Restenosis ITalian Study (ESPRIT) was a double-blind, randomized study evaluating the efficacy of a concentrated ω-3 FA preparation, given at least one month before PTCA, vs placebo in the prevention of post-PTCA restenosis.

The trial was carried out in 17 clinical centers throughout Italy (see Appendix), each recruiting ≥12 patients/center, with the exception of 3, between March 1993 and May 1995. Inclusion criteria were a clinical indication for PTCA (angina or ischemia not adequately responsive to medical treatment or significant silent ischemia after acute myocardial infarction [MI]) in the presence of at least 1 critical coronary artery stenosis (≥70% at visual inspection) amenable to PTCA. Exclusion criteria were age <18 years or >75 years; recent (<15 days) acute MI; the presence of unstable disease not allowing the 4-week pretreatment before PTCA; culprit lesions in the left main coronary artery, in a saphenous vein bypass graft, or in a previously dilated site (restenotic lesions); excessive bleeding risk (recent peptic ulcer or gastrointestinal bleeding, platelet count <100,000/μL, and arterial hypertension not responsive to medical treatment); contraindication to ω-3 FA (such as allergy or hypersensitivity to fish products); requirement for anticoagulant therapy; presence of significant hepatic or renal disease; concomitant disease associated with limited life expectancy; drug or alcohol abuse; or other factors (geographic location or language barriers) possibly precluding follow-up.

After written informed consent, patients were assigned randomly to either ω-3 FA (Esapent capsules of ethyl-ester ω-3 FA concentrates, each containing 85% pure eicosapentaenoic acid [EPA, 50%] and docosahexaenoic acid [DHA, 35%] with 3 mg vitamin E, at a dose of 6 capsules/d; Pharmacia-Upjohn, Milan, Italy) or identically appearing olive oil placebo capsules, which also contained 3 mg vitamin E. Such treatment was continued until 1 month after the qualifying PTCA and subsequently reduced to 3 capsules/d until the end of follow-up (Figure 1).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  ESPRIT timeline.

All patients were treated with either aspirin (100 to 500 mg/d) or indobufen (200 mg Ibustrin tablets twice daily; Pharmacia-Upjohn) beginning at least 48 hours before PTCA. Treatment with the antiplatelet indobufen was allowed instead of aspirin on the basis of a previous study demonstrating similar periprocedural complications and similar absence of effects on post-PTCA restenosis for both drugs.²⁸ All patients received nitrates and anticoagulation with 10,000 IU IV heparin at the start of the PTCA procedure. Heparin administration was subsequently adjusted with 1 or more further boluses or infusion in order to maintain the activated clotting time at >300 seconds and was then tapered off at least 4 hours before sheath removal and up to 24 hours post-PTCA. Patients judged to have had a successful procedure (residual stenosis <50% in at least 1 vessel by visual assessment in the absence of death, periprocedural MI, repeat PTCA, or coronary artery bypass surgery) were re-examined 1 and 6 months later for the assessment of study end points. Coronary angiography was repeated at the 6-month follow-up (or sooner if clinically indicated). If repeated sooner and the previously dilated lesion was found to have <50% diameter reduction by visual assessment, the study drug was continued and every effort was made to obtain another angiogram at the 6-month follow-up. If any study lesion was found to have a ≥50% diameter reduction at angiography before 6 months, the study was considered concluded in that patient.

After PTCA, the use of either aspirin or indobufen was mandatory for 15 days and then left discretionary. Other treatments, such as calcium antagonists, β-blockers, and nitrates were also discretionary.

For QCA, nitroglycerin (200 μg) or isosorbide dinitrate (200-600 μg) were administered into the coronary artery with the largest acceptable catheter in order to perform adequate QCA (size ≥6F) before and after the completion of PTCA and at the 6-month QCA. At least 2 orthogonal views of each lesion to be dilated were filmed and angles and distances for each view were recorded and repeated after PTCA and at the follow-up QCA. Care was taken in centering the contrast dye-filled catheter at the beginning of each injection in order to enable calibration of QCA. Quantitative measurements were made in a single angiographic core-laboratory by a customized Statview Artrek system (Image Comm Systems Inc, Mountain View, Calif), according to a previously validated methodology.²⁹ Catheter dimensions and projections used were recorded by each center in an ad hoc form and sent to the core laboratory together with 35-mm cine films. At the time of QCA measurement, the images were chosen from the projections indicated by the angiographer, digitized, and automatically magnified by the computer. Calibration was carried out by use of the contrast-filled catheter in the same run, in the best-centered position, and the region of interest was then defined. The operator corrected the computer-generated centerline axial to the vessel lumen in the region of interest, allowing edge detection and automated measurement of minimal luminal diameter (MLD) and of a reference vessel diameter. This latter measurement was taken as the mean of several operator-defined “normal” diameters proximal and distal to the lesion. The diameter of each segment was calculated in the view where the lesion was most severe. Film sets (immediately before PTCA, immediately after, and at follow-up QCA [6 months or earlier]) were analyzed at the same time by 3 experienced angiographers blinded to treatment allocation.

The primary study end point was the occurrence of angiographic restenosis at an intention-to-treat analysis. Restenosis was defined according to 2 prespecified criteria: (1) stenosis in the dilated vessel >50% at follow-up QCA (Definition I) and (2) loss of dilated vessel diameter at follow-up QCA >50% compared with the gain at immediate post-PTCA QCA (Definition II). Restenosis was calculated with a patient-based and a lesion-based analysis. In the patient-based analysis, patients were considered to have restenosis if at least one of the dilated vessels met the restenosis criteria. For lesion-based analysis, each successfully dilated lesion was considered a separate observation. Secondary study end points were recurrence of angina, occurrence of acute MI, cardiac death, and myocardial revascularization by either coronary bypass surgery or repeat PTCA during the follow-up period. An analysis of the restenosis rate restricted to patients with objective documentation of compliance to treatment (“on-treatment” analysis), although not intended to be a primary end point of the study, was also performed.

Laboratory evaluations 

Red blood cell, white blood cell, and platelet counts; hemoglobin; blood glucose; blood urea nitrogen; uric acid; creatinine; alkaline phosphatase; aspartate aminotransferase; alanine aminotransferase; lactate dehydrogenase; activated partial thromboplastin time; C-reactive protein and lipid profile evaluation including total cholesterol, high-density lipoprotein cholesterol, and triglycerides were monitored at 0-, 1-, and 6-month intervals. All tests were performed by the clinical centers participating in the study.

Evaluation of compliance 

Compliance to treatment was objectively evaluated by assessment of plasma FA at various points in the study in the 189 (74%) patients qualifying for follow-up after PTCA. FA were analyzed by gas chromatography after chloroform-methanol extraction and transmethylation, as described.³⁰ These tests were all performed in one centralized laboratory and served for the “on-treatment” analysis of restenosis rates.

Statistical analysis 

Sample size in this study was calculated hypothesizing a restenosis incidence of 40% in the placebo group and 25% in the active treatment group, a rate of dropouts (after randomization) <30%, α = .05, β = .8, and assuming to dilate at least 1 lesion per patient. With these calculations, the minimum number of patients to randomize was 300 and those available for follow-up QCA evaluation was 210.

Differences between the treatment groups for the primary end points were tested by 1-tailed χ² test. Lesion-based analysis did not include a correction for the possible independence of lesion restenosis rates in the same patient, since the number of patients receiving more than one dilation was negligible. Quantitative data were analyzed by parametric statistics, namely t test and Pearson correlation. Statistical analyses were performed blind of treatments with the SAS statistical package (SAS Institute, Cary, NC).

Back to Article Outline

Results 

Patient characteristics 

A flowchart showing the screening of patients originally randomized to either ω-3 FA or placebo up to the final QCA is shown in Figure 2.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  Status of patients throughout the study. Patients disqualified before PTCA (16) include mostly those whose clinical course demanded early PTCA (without the prolonged pre-PTCA administration required) or alternative treatments (coronary bypass surgery or medical treatment). A few patients discontinued treatment after earlier consent (15). Disqualification at PTCA resulted mostly from stent application at the time of PTCA (17) or unsuccessful PTCA (3). Of the 287 in whom a successful PTCA was performed, 23 were designated “unevaluable dropouts,” mostly because of loss to follow-up (7), refusal of final angiography (7), or adverse events (sudden death, 3). Of the remaining 264 patients, QCA was not possible in 7 because of angiogram quality. Final dropout rate was 10.5% of patients qualified for follow-up after PTCA and 24.2% of patients originally randomized. The 2 treatment groups remained numerically balanced throughout the selection process.

Out of 339 patients originally randomized, 52 were disqualified before or at the time of PTCA. Twenty-three patients were considered unevaluable dropouts, and 7 were eventually excluded from QCA because of qualitatively unsatisfactory angiograms, leaving 257 (89%) patients in the final analysis (Figure 2). Clinical characteristics of these patients are detailed in Table I.

Table I. Clinical characteristics of patients with all 3 QCA determinations available

QCA, Quantitative coronary angiography; FA, fatty acids; TIA, transient ischemic attack.

The 2 groups were successfully randomized for all reported variables, apart from hypertension, which was slightly more prevalent (P <.05) in the ω-3 FA group. There were 280 lesions evaluated by QCA in the final 257 patients. Vessel and lesion characteristics divided according to treatment group are detailed in Table II.

Table II. Vessel and lesion characteristics in patients with all 3 QCA determinations available

QCA, Quantitative coronary angiography; FA, fatty acids; ACC/AHA: American College of Cardiology/American Heart Association classification.

^*Categories with missing values (% calculated over total of available data).

Patient follow-up 

Restenosis rate 

Restenosis rates per vessel were 29.4% and 31.6% in the ω-3 FA group and 39.6% and 35.4% in the placebo group according to Definitions I and II, respectively (Table III). Such rates were significantly different (P = .04) for Definition I but not for Definition II and corresponded to a 26% and an 11% reduction in restenosis rates, respectively. The restenosis rates per patient were 31.2% and 33.6% in the ω-3 FA group and 40.9% and 37.1% in the placebo group according to Definitions I and II, respectively (Table III). Such rates were significantly different (P = .05) for Definition I, insignificant for Definition II, and corresponded to a 24% and a 9% reduction in restenosis rates, respectively.

Table III. Restenosis rate per vessel and per patient with all 3 QCA determinations available

QCA, Quantitative coronary angiography; FA, fatty acids.

The distribution of MLD in the 2 treatment groups at pre-PTCA, early post-PTCA, and 6-month follow-up QCA are shown in Figure 3.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3. 

  Cumulative frequency distribution curves of minimal lumen diameters (MLD) (ie, lumen diameter in mm at the site of maximum stenosis) at the 1-month qualifying angiogram (pre-PTCA, I, curves to the left), the immediate post-PTCA angiogram (II, curves to the right), and the exit follow-up angiogram (III, curves in the middle) in placebo (filled squares) and ω-3 fatty acid (open squares) treatment groups, by lesion. The degrees of stenosis under the 2 treatments were comparable at the pre-PTCA and immediate post-PTCA angiogram but tended to diverge at the exit follow-up angiogram.

The curves relative to the treatment groups were largely superimposed at the time of the pre-PTCA QCA and immediately after PTCA, but tended to diverge at the 6-month follow-up QCA, suggesting a benefit of ω-3 FA on the MLD. Figure 4 is a plot of the late loss of MLD against the early gain in lumen diameter after PTCA for all lesions in the 2 groups under study.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 4. 

  Plot of relation of “late loss” (restenosis) to “acute gain” resulting from PTCA per lesion. Triangles denote lesions in the ω-3 fatty acid group, and circles denote lesions in the placebo group. The regression line of the scattergram would have a slope of 1 mm if restenosis occurred in all lesions and no slope if no restenosis occurred. The slopes for the 2 treatments are intermediate indicating some occurrence of restenosis. The slope for ω-3 fatty acids (n-3 FA) is smaller (although not significantly) than for placebo, suggesting less restenosis in the ω-3 fatty acid group.

Restenosis in all vessels would have been on the line of identity with a slope of 1 mm, and no restenosis would be on a horizontal line. Although there is great scattering of the values, indicating inhomogeneity in the behavior of individual lesions, there is clearly some restenosis on average in both the ω-3 FA and the placebo group. However, the slope of ω-3 FA is less than for placebo, suggesting some effect of the active treatment in preventing restenosis.

There were minimal differences in lesion diameter and percent stenosis between the treatment groups at the 6-month follow-up QCA. Mean MLD was 0.82 mm in the ω-3 FA group vs 0.80 mm in the placebo group before PTCA and 1.49 mm in the ω-3 FA group vs 1.41 mm in the placebo group at 6 months. When MLD measurements in the follow-up QCA were divided into classes of ≤2 mm and >2 mm, the frequency of MLD in the upper classes tended to be higher in the ω-3 FA group when compared with the placebo group at 6-month follow-up (31/136 in the ω-3 FA group vs 33/144 in the placebo group at baseline and 24/135 in the ω-3 FA group vs 19/143 in the placebo group at 6 months; Figure 3).

Clinical events 

Periprocedural (ie, occurring within 24 hours) complication rates were similar in the ω-3 FA and placebo groups. In the 287 patients qualified for follow-up after PTCA, 9 had periprocedural bleeding. Of these, 6 (2.1%) occurred in the ω-3 FA group and 3 (1.0%) in the placebo group (P = not significant [NS]). None required transfusion. The incidence of acute MI was 0.7% and 0%, early angina recurrence 0% versus 2.8%, thromboembolism 0% versus 0.7% (all differences NS). In total, periprocedural adverse events had a similar frequency (6.9% and 5.6%, respectively). Six-month clinical end points were divided into objective (death, acute MI, coronary artery bypass grafting, and re-PTCA) and subjective (presence of significant angina) categories. Incidence of these end points was not statistically different in ω-3 FA and placebo groups (9.5% vs 8.5% and 9.6% vs 15.1%, respectively).

Adverse effects 

Mild gastrointestinal symptoms (dyspepsia, epigastralgia, or gastric discomfort) were reported in 4/287 patients and were equally distributed between the 2 treatment groups.

Evaluation of compliance 

EPA and DHA levels in total plasma lipids were monitored in 74% of patients (at 11 of the 17 recruiting centers). In the analyzed placebo group receiving olive oil capsules, there was no change in plasma 18:1 n-9 (oleic acid) levels, possibly because of the very small additional intake of this fatty acid (approximately 2 g/d) exceeding that derived from the diet. Plasma concentrations of EPA and DHA were similar at baseline in all participating centers between the ω-3 FA and placebo groups. At the time of PTCA, levels changed substantially and remained significantly different at 6 months in the 2 groups, as shown by the absolute increase in the percentage of EPA and DHA over total FA, the decrease in arachidonic acid (not shown), and the increase in the ratio of ω-3/ω-6 FA (Figure 5).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 5. 

  Ratio of total ω-3 (α-linolenic + eicosapentaenoic + docosahexaenoic acids) to total ω-6 fatty acids (linoleic + arachidonic acids) in the active treatment and placebo groups in the various phases of the study. Asterisks denote significant differences (P <.01) from baseline at repeated analyses of variance in the ω-3 fatty acid group. The points marked by the asterisks also are significantly different (P <.05) in the comparison between the active treatment and placebo groups.

Compliance with treatment, as evaluated individually, was good. Fifteen patients (13.7% of those monitored) in the ω-3 FA group did not adhere satisfactorily to treatment as judged by no increase in their ω-3 FA plasma levels above the mean ± 1.96 baseline distribution. Six of them developed restenosis during the treatment period. Conversely, 3 patients (3.8%) had increased levels of EPA and DHA at baseline despite being in the placebo group, but no restenosis was found. Therefore, analysis of treatment efficacy also was performed “on-treatment,” on the basis of results of plasma FA measurements (Table IV). Restricting the analysis of restenosis only to patients for whom control of incorporation was performed revealed significant effects of ω-3 FA treatment in reducing restenosis rates, both in the analysis per vessel and analysis per patient (Table IV).

Table IV. Restenosis rate per vessels and per patients with all 3 QCA determinations available and with documentation of satisfactory compliance (on-treatment analysis)

QCA, Quantitative coronary angiography; FA, fatty acids.

Laboratory analyses 

Red blood cell, white blood cell, and platelet counts; hemoglobin; blood glucose; blood urea nitrogen; uric acid; creatinine; alkaline phosphatase; aspartate aminotransferase; alanine aminotransferase; lactate dehydrogenase; activated partial thromboplastin time; and C-reactive protein were within normal values at baseline, comparable between the 2 treatment groups, and without significant change throughout the study period. There were also no significant changes in lipid parameters. Serum triglycerides were 160 ± 84 mg/dL in the ω-3 FA group and 196 ± 142 mg/dL in the placebo group (P = NS). Serum triglycerides declined insignificantly to 151 ± 72 mg/dL in the ω-3 FA group and 182 ± 114 mg/dL in the placebo group (P = NS).

Back to Article Outline

Discussion 

This was a randomized, placebo-controlled, double-blind study of ω-3 FA in 257 patients undergoing elective PTCA and receiving complete QCA evaluation at 6 months. The sample size for this study was calculated assuming a rate of post-PTCA restenosis at 6 months of 40% and a 15% reduction of such rate in the active treatment group. The angiographic end point of restenosis was evaluated quantitatively by an independent blinded core laboratory, and control of compliance was obtained by the assessment of FA incorporation in total plasma lipids by another independent laboratory. In the overall population studied, the group randomized to receive ω-3 FA showed some reduction of angiographic restenosis at 6 months, which was significant according to one of the 2 predefined criteria in both the per-patient and per-vessel analyses.

In these patients, angiographic occurrence of restenosis in the placebo group was 39.6% and 35.4% according to the 2 definitions used, correlating with the anticipated restenosis rate. Thus, the calculated sample size was certainly adequate for the first definition—for which significance was reached—and possibly slightly undersized for the second definition. These rates of angiographic restenosis compare with the 41.6% rate obtained in the placebo group of the Enoxaparin MaxEPA Prevention of Angioplasty Restenosis (EMPAR) study²² and 46% rate reported in the Fish Oil Restenosis Trial (FORT) study.²¹ In this last study, however, the first criterion for restenosis was less strict (>30% increase in narrowing at the stenosis site) than in our study (≥50%). It seems, therefore, unlikely that differences in the incidence of restenosis in the control population account for different results compared with these 2 studies.

The existence of a significant effect of ω-3 FA on restenosis, as detected by the analysis of 1 of the 2 criteria for restenosis, is further supported by the analysis of the distribution of MLD in the placebo and the ω-3 FA groups at the different time points of angiographic assessment (Figure 3). Distributions of MLD in the dilated vessels were quite similar in the placebo and the ω-3 FA group before and immediately after PTCA, showing the characteristic shift to the right indicative of the increase in lumen diameter as a result of a successful procedure. Both curves shifted back to the left in the 6-month follow-up evaluation as a result of restenosis, but the shift to the left of the ω-3 FA group appears to be less indicative of lesser loss of the short-term gain (Figure 4). This contrasts with similar analyses performed for the EMPAR and FORT studies, in which the MLD distributions at 6 months in the active treatment and placebo groups were virtually indistinguishable, as well as the slopes of short-term gain versus late loss for individual lesions.²¹, ²² Analysis of reference vessel diameter and MLD indicated a marginal benefit of ω-3 FA (1.49 mm in the ω-3 FA group vs 1.41 mm in the placebo group at 6 months), and mostly limited to preservation of a higher number of vessels in the larger-diameter class (>2 mm) in the ω-3 FA group compared with the placebo group. The lower incidence of restenosis in the ω-3 FA group in both the per-patient and the per-vessel analysis, significant according to Definition I and with a consistent trend according to Definition II despite minor differences in MLD, appears therefore mostly accounted for by the dichotomous nature of the restenosis definitions.

There are a number of differences in this study design when compared with previous ones. At least 12 studies of ω-3 FA in the prevention of post-PTCA restenosis have been published to date.¹¹, ¹², ¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰, ²¹, ²² In 8 of these, the end point used was an angiographic assessment of restenosis,¹², ¹³, ¹⁶, ¹⁷, ¹⁹, ²⁰, ²¹, ²² but in only 4 was QCA performed in all patients.¹⁷, ²⁰, ²¹, ²² Restricting discussion to these last 4, doses ranged between 3 g/d²⁰ and 8 g/d,²¹ treatment before PTCA was 1 to 2 days in Bellamy's study,²⁰ 12 to 14 days in FORT,²¹ 21 days in the study reported by Bairati et al¹⁷ and ≥28 days in ours. All studies used triglyceride preparations, apart from FORT²¹ and ours. It seems clear that dose and formulation of ω-3 FA do not explain differences in outcome, although length of pretreatment might. Studies reporting significant improvement of post-PTCA restenosis outcome by QCA have employed a fairly long pretreatment (21 to 28 days [Bairati et al¹⁷ and ours]). FA incorporation was not reached in the FORT study using 2 weeks of pretreatment; only 43% and 25% of steady-state values of EPA and DHA plasma levels were achieved at the time of PTCA.²¹ This contrasts with the levels achieved in plasma lipids at baseline in our study, which were even higher than at 6 months (Figure 4). In addition, in the study by Bairati et al, a relationship was seen between a very low habitual dietary intake of fish and the inhibition of restenosis,¹⁷ reminiscent of the decreased mortality after a first MI found in persons on even a fairly low daily consumption of ω-3 FA (dietary intake of fatty fish at least twice a week) in the Diet And Reinfarction Trial (DART).³¹ Similarly, in our study an on-treatment (post hoc) analysis, including in the ω-3 FA group only those patients in whom a significant increase in ω-3 FA incorporation could be objectively documented, further increased the magnitude of difference in the restenosis rate between treated and placebo groups (Table IV). These findings suggest different sensitivity of target populations (according to the baseline intake of ω-3 FA in the diet and according to compliance to treatment) as possible additional explanations for different outcomes in different studies.

One further difference among previous studies was the choice of placebo. This was corn oil in the FORT²¹ and the EMPAR²² studies and olive oil in the study reported by Bairati et al.¹⁷ Arguments for both types of placebos have been raised. The use of corn oil should be preferable if the biological hypothesis being tested is that substitution of ω-6 with ω-3 FA in membrane phospholipids renders the injured vessel wall less prone to develop the proliferative lesion-characterizing restenosis. Also, olive oil per se may not be biologically inert, at least in native arteriosclerosis.³² However, olive oil is clearly the most abundant unsaturated dietary FA in some populations, such as the one studied by us, and is less likely to alter the relative proportions of FA in the patient's habitual diet, as it appeared also from the analysis of plasma FA in the placebo group. Because, if implemented into medical practice, ω-3 FA treatment would have to be added to a normal diet, our comparison is likely to offer the best practical transferability of the scientific information provided by the present results. In addition, the value of our finding on post-PTCA restenosis actually seems reinforced by its occurrence compared with an equivalent supplementation of another potentially active dietary component.

The clinical significance of our findings is conceivably debatable. The percent inhibition of restenosis is far from being an abolition of the phenomenon. In addition, the current practice of PTCA—being performed at, or quickly after, diagnostic angiography—may make adequate pretreatment of a significant number of patients impractical. Also, our study was planned and started before the widespread use of coronary stenting during PTCA. Thus, the applicability of the present results to the general population of patients presently undergoing PTCA may be quite limited. On the other hand, stenting, although preventing detrimental vascular remodeling, clearly has not decreased the intimal smooth muscle cell proliferative component of restenosis. Because smooth muscle cell proliferation appears to be the main target of ω-3 FA action in preventing restenosis,³³, ³⁴, ³⁵, ³⁶ one might speculate possibly even greater efficacy of ω-3 FA in the current PTCA practice. A reassuring confirmation from our study was the lack of any significant side effect of the active treatment, either on bleeding, gastric discomfort or other monitored end points. This is consistent with previous results from large trials of this kind.²¹, ²²

In summary, the present study provides evidence for some reduction of restenosis by pretreatment with ω-3 FA in patients undergoing elective PTCA without stenting. On the basis of our findings, compared with previous ones, we suggest that a long pretreatment (≥4 weeks) is necessary to increase chances of detecting a positive effect of ω-3 FA on preventing restenosis. The clinical applicability of such information remains questionable.

Back to Article Outline

Acknowledgements 

We thank colleagues who performed cardiac catheterizations and PTCAs, to the nurses and technicians in the cardiac catheterization laboratories in participating hospitals, to the technicians and biologists in the core angiographic reading and analytical biochemistry laboratories, and to the monitoring personnel of Pharmacia-Upjohn Italy (formerly Farmitalia-Carlo Erba-Italy), who made this study possible. Special thanks also to Dr. Patrizia Visè, Institute of Pharmacological Sciences, University of Milan, for performing fatty acid analyses on samples from the study.

Back to Article Outline

Appendix A 

Participating recruiting centers and investigators.

Dr Aleardo Maresta, Dr Marco Balduccelli, and Dr Elisabetta Varani, from Divisione di Cardiologia, Ospedale di Faenza; Dr Giovanni Baduini, and Dr Mauro De Benedictis, from Divisione di Cardiologia, Ospedale Mauriziano, Torino; Professor Aldo Bigalli, Dr Enrico Magagnini, Dr Piersilvio Chella, and Dr Andrea Pieroni, from Divisione Medicina Cardiovascolare, Ospedale S. Chiara, Pisa; Professor Giovanni Binaghi, and Dr ssa B. Castiglioni, from Divisione di Cardiologia, Ospedale di Circolo, Varese; Professor Achille Bravi, Dr M. Del Sordo, Dr C. Pierli, and Dr S. Casini, from Divisione di Cardiologia, Policlinico Le Scotte, Siena; Professor Carmelo Cernigliaro, Dr F. Aina, and Dr A. Campi, from Divisione di Cardiologia, Ospedale Maggiore, Novara; Professor Raffaele Chioin, Dr A. Ramondo, Dr R. Razzolini, and Dr G. B. Isabella, from Dipartimento Medicina Clinica, Unità Operativa di Cardiologia Interventistica, Università di Padova; Dr Vittorio Di Luzio, Dr F. De Remigis, and Dr A. DíAroma, from Divisione di Cardiologia II, Ospedale Civile, Teramo; Dr Alberto Dolara, Dr G. Santoro, Dr G. Corti, Dr G. Squillantini, and Dr F. Mazzuoli, from Unità Operativa di Cardiologia, Ospedale S. Luca, Firenze Careggi; Professor Emilio Gatto, and Dr F. Della Rovere, from II Divisione Cardiologia, Ospedale S. Martino, Genova; Dr Luigi Giommi, and Dr G. Risica, from Servizio Emodinamica, Ospedale S.Maria Cà Foncello, Treviso; Professor Mario Marzilli, and Dr. Marco Baratto, from Istituto di Fisiologia Clinica, CNR, Pisa; Professor Attilio Maseri, Dr F. Crea, and Dr A. Buffon, from Istituto di Cardiologia, Università Cattolica S. Cuore, Roma; Dr Giuseppe Richichi, and Dr D. Irini, from Servizio di Emodinamica, Ospedale S. Filippo Neri, Roma; Professor Mario Vincenzi, Dr F. Bedogni, and Dr M. Martini, from Divisione di Cardiologia, Ospedale Civile, Vicenza; Professor Odoardo Visioli, Dr L. Niccoli, and Dr C. Lettieri, from Cattedra di Cardiologia, Ospedali Civili, Brescia; Professor Pietro Zardini, and Dr E. Barbieri, from Cattedra di Cardiologia, Istituti Ospedalieri Borgo Trento, Verona.

Back to Article Outline

References 

1. Bang HO, Dyerberg J, Hjoorne N. The composition of food consumed by Greenland Eskimos. Acta Med Scand. 1976;200:69–73
   • View In Article
   • MEDLINE
   • CrossRef
2. Hirai A, Terano R, Saito H. Eicosapentaenoic acid and platelet function in Japanese. In:  Lovenburg W,  Yamori Y editor. Nutritional Prevention of Cardiovascular Disease. New York: : Academic Press; 1984;p. 231–239
   • View In Article
3. Kromhout D, Bosschieter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N Engl J Med. 1985;312:1205–1209
   • View In Article
   • MEDLINE
   • CrossRef
4. Dolecek TA. Epidemiological evidence of relationships between dietary polyunsaturated fatty acids and mortality in the Multiple Risk Intervention Trial. Proc Soc Exp Biol Med. 1992;200:177–182
   • View In Article
   • MEDLINE
5. Davis HR, Bridenstine RT, Vesselinovitch D, et al.  Fish oil inhibits development of atherosclerosis in rhesus monkeys. Arteriosclerosis. 1987;7:441–449
   • View In Article
   • MEDLINE
6. Parks JS, Kaduck-Sawyer J, Bullock BC, et al.  Effect of dietary fish oil on coronary artery and aortic atherosclerosis in African green monkeys. Arteriosclerosis. 1990;10:1102–1112
   • View In Article
   • MEDLINE
7. Weiner BH, Ockene IS, Levine PH, et al.  Inhibition of atherosclerosis by cod-liver oil in a hyperlipidemic swine model. N Engl J Med. 1986;315:841–846
   • View In Article
   • MEDLINE
   • CrossRef
8. Kim DN, Ho HT, Lawrence DA, et al.  Modification of lipoprotein patterns and retardation of atherogenesis by a fish oil supplement to a hyperlipidemic diet for swine. Atherosclerosis. 1989;76:35–54
   • View In Article
   • Abstract
   • Full-Text PDF (2107 KB)
   • CrossRef
9. Committee on Percutaneous Transluminal Coronary Angioplasty. Guidelines for percutaneous transluminal coronary angioplasty . A report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures. J Am Coll Cardiol. 1993;22:2033–2054
   • View In Article
   • Full-Text PDF (4246 KB)
   • CrossRef
10. Holmes DR, Schwartz RS, Webster MW. Coronary restenosis: what have we learned from angiography?. J Am Coll Cardiol. 1991;17:14B–22B
    • View In Article
    • MEDLINE
11. Slack JD, Pinkerton CA, VanTassel J, et al.  Can oral fish oil supplement minimize re-stenosis after percutaneous transluminal angioplasty [abstract]. J Am Coll Cardiol. 1987;9:64A
    • View In Article
12. Dehmer GJ, Popma JJ, van den Berg EK, et al.  Reduction in the rate of early restenosis after coronary angioplasty by a diet supplemented with ω-3 fatty acids. N Engl J Med. 1988;319:733–740
    • View In Article
    • MEDLINE
    • CrossRef
13. Grigg LE, Kay TW, Valentine PA, et al.  Determinants of restenosis and lack of effect of dietary supplementation with eicosapentaenoic acid on the incidence of coronary artery restenosis after angioplasty. J Am Coll Cardiol. 1989;13:665–672
    • View In Article
    • Abstract
    • Full-Text PDF (932 KB)
    • CrossRef
14. Milner MR, Gallino RA, Leffingwell A, et al.  Usefulness of fish oil supplements in preventing clinical evidence of restenosis after percutaneous transluminal coronary angioplasty. Am J Cardiol. 1989;64:294–299
    • View In Article
    • MEDLINE
    • CrossRef
15. Reis GJ, Boucher TM, Sipperly ME, et al.  Randomised trial of fish oil for prevention of restenosis after coronary angioplasty. Lancet. 1989;2:177–181
    • View In Article
    • MEDLINE
16. Nye ER, Ilsley CDJ, Ablett MB, et al.  Effect of eicosapentaenoic acid on restenosis rate, clinical course and blood lipids in patients after percutaneous transluminal coronary angioplasty. Aust NZ J Med. 1990;20:549–552
    • View In Article
17. Bairati I, Roy L, Meyer F. Double-blind, randomized, controlled trial of fish oil supplements in prevention of recurrence of stenosis after coronary angioplasty. Circulation. 1992;85:950–956
    • View In Article
    • MEDLINE
18. Kaul U, Sanghvi S, Bahl VK, et al.  Fish oil supplements for prevention of restenosis after coronary angioplasty. Int J Cardiol. 1992;35:87–93
    • View In Article
    • MEDLINE
    • CrossRef
19. Franzen D, Schannwell M, Oette K, et al.  A prospective, randomized, and double-blind trial on the effect of fish oil on the incidence of restenosis following PTCA. Cathet Cardiovasc Diagn. 1993;28:301–310
    • View In Article
    • MEDLINE
    • CrossRef
20. Bellamy CM, Schofield PM, Faragher EB, et al.  Can supplementation with omega-3 polyunsaturated fatty acids reduce coronary angioplasty restenosis rate?. Eur Heart J. 1992;13:1626–1631
    • View In Article
    • MEDLINE
21. Leaf A, Jorgensen MB, Jacobs AK, et al.  Do fish oils prevent restenosis after coronary angioplasty?. Circulation. 1994;90:2248–2257
    • View In Article
    • MEDLINE
22. Cairns JA, Gill J, Morton B, et al.  Fish oils and low-molecular-weight heparin for the reduction of restenosis after percutaneous transluminal coronary angioplasty. The EMPAR Study. Circulation. 1996;94:1553–1560
    • View In Article
    • MEDLINE
23. Leaf A, Weber PC. Cardiovascular effects of ω-3 fatty acids. N Engl J Med. 1988;318:549–557
    • View In Article
    • MEDLINE
    • CrossRef
24. Simopoulos A. Omega-3 fatty acids in health and disease and in growth and development. Am J Clin Nutr. 1991;54:438–463
    • View In Article
    • MEDLINE
25. O'Connor GT, Malenka DJ, Olmstead EM, et al.  A meta-analysis of randomized trials of fish oil in prevention of restenosis following coronary angioplasty. Am J Prev Med. 1992;8:186–192
    • View In Article
    • MEDLINE
26. Galli C, Tremoli E, Sirtori C. ω-3 Fatty acids: incorporation into tissue lipids and interactions with dietary components. In:  De Caterina R,  Kristensen SD,  Schmidt EB editor. Fish oil and vascular disease. London: : Springer Verlag; 1992;p. 35–41
    • View In Article
27. Schwartz SM. Smooth muscle migration in atherosclerosis and restenosis. J Clin Invest. 1997;100(11 Suppl):S87–S89
    • View In Article
    • MEDLINE
28. Dalla Volta S. on behalf of SIAC. Antiplatelet treatment in the prevention of restenosis after PTCA: a multicenter study [abstract]. Blood. 1989;74:176a
    • View In Article
29. Mancini GB, Simon SB, McGillem MJ, et al.  Automated quantitative coronary arteriography: morphologic and physiologic validation in vivo of a rapid digital angiographic method. Circulation. 1987;75:452–460
    • View In Article
    • MEDLINE
    • CrossRef
30. Marangoni F, Angeli MT, Colli S, et al.  Changes of ω-3 and n-6 fatty acids in plasma and circulating cells of normal subjects, after prolonged administration of 20:5 (EPA) and 22:6 (DHA) ethyl esters and prolonged washout. Biochim Biophys Acta. 1993;1210:55–62
    • View In Article
    • MEDLINE
31. Burr ML, Fehily AM, Gilbert JF, et al.  Effects of changes in fat, fish, and fibre intakes on death and myocardial reinfarction: Diet And Reinfarction Trial (DART). Lancet. 1989;2:757–761
    • View In Article
    • MEDLINE
32. Carluccio MA, Massaro M, Bonfrate C, et al.  Oleic acid inhibits endothelial activation: A direct vascular antiatherogenic mechanism of a nutritional component in the Mediterranean diet. Arterioscl Thromb Vasc Biol. 1999;19:220–228
    • View In Article
33. Fox PL, DiCorleto PE. Fish oil inhibits endothelial cell production of platelet-derived growth factor-like protein. Science. 1988;241:453–456
    • View In Article
    • MEDLINE
34. Kaminski WE, Jendraschak E, Kiefl R, et al.  Dietary omega-3 fatty acids lower levels of platelet-derived growth factor mRNA in human mononuclear cells. Blood. 1993;81:1871–1879
    • View In Article
    • MEDLINE
35. Sassen LM, Lamers JM, Verdouw PD. Fish oil and the prevention and regression of atherosclerosis. Cardiovasc Drugs Ther. 1994;8:179–191
    • View In Article
    • MEDLINE
    • CrossRef
36. Soei LK, Lamers JMJ, Sassen LMA, et al.  Fish oil: a modulator of experimental atherosclerosis in animals. In:  Kristensen SD,  Schmidt EB,  De Caterina R, et al. editor. ω-3 Fatty Acids: Prevention and Treatment in Vascular Disease. London: : Springer Verlag; 1995;p. 55–76
    • View In Article