Author + information
- Received September 11, 2002
- Accepted October 17, 2002
- Published online February 5, 2003.
- Martial G Bourassa, MD, FACC*,* (, )
- Ady Butnaru, MD*,
- Jacques Lespérance, MD* and
- Jean-Claude Tardif, MD, FACC*
- ↵*Reprint requests and correspondence:
Dr. Martial G. Bourassa, Research Center, Montreal Heart Institute, 5000 Belanger Street East, Montreal, QC, Canada, H1T 1C8.
This review article focuses on the morphological and functional alterations that characterize patients with myocardial bridges (MB) as well as the currently available diagnostic and treatment strategies. Because of incomplete understanding of the pathophysiology of MB, their clinical significance has been the subject of debate for the last quarter century. Investigational tools now available in the cardiac catheterization laboratory have helped clarify why symptoms and signs of ischemia can occur in such patients, especially when the only angiographic finding appears to be systolic compression or milking effect of a coronary vessel. Quantitative coronary angiography and intravascular ultrasound (IVUS) clearly demonstrate that the phasic systolic vessel compression visualized on the angiogram is coupled with a persistent diastolic diameter reduction. Intracoronary Doppler reveals increased flow velocities, retrograde systolic flow, and reduced coronary flow reserve. The clinical diagnosis can be established by significant percent lumen diameter and area narrowing, increased flow velocity, and by characteristic patterns such as the “half moon” phenomenon on IVUS and the early diastolic “finger tip” phenomenon on intracoronary Doppler. Successful medical, interventional, or surgical therapy leads not only to marked improvement or normalization of these alterations but also relief of angina and ischemia.
In 1976, we described pacing-induced angina and myocardial ischemia in a group of patients with isolated myocardial bridging of the left anterior descending coronary artery (LAD) documented at coronary angiography by a milking effect or transient constriction of the artery during systole (1)(Fig. 1). During the next 15 to 20 years, the clinical, hemodynamic, and prognostic significance of this entity has remained controversial, some investigators suggesting that it was a benign condition (2–4), and others reporting a variety of acute coronary syndromes associated with them (5–10). However, important new diagnostic tools have become available in the last decade that have greatly contributed in establishing the clinical relevance of myocardial bridging (11–16). This report reviews the morphological and hemodynamic findings derived from these new diagnostic modalities as well as the functional significance and therapeutic options of this intriguing clinical entity.
Incidence of myocardial bridging
Myocardial bridging occurs when a band of cardiac muscle overlies an intramural segment of a coronary artery, the intramural segment being referred to as a “tunneled” artery. Usually, the coronary vessels course over the epicardial surface of the heart but may dip into the myocardium for varying lengths and then reappear on the heart’s surface. The LAD is by far the vessel most often overbridged; however, diagonal branches are occasionally involved as well as the posterior descending right coronary artery or marginal branches of the circumflex artery.
Although myocardial bridges (MB) were initially recognized by Reyman in 1737 (17)and later by Black in 1805 (18), the first detailed postmortem analysis of a series of patients with this anomaly was reported by Geiringer in 1951 (19). In 1960, Portmann and Iwig (20)published the first radiological description of a transient occlusion in a segment of the LAD during systole. In 1976, we described a milking effect or transient narrowing of the LAD during systole in 27 (0.5%) of 5,250 patients having undergone selective coronary angiography (1). Since then, a wide discrepancy has existed between pathological series, in which the incidence of MB has been reported to vary between 15% and 85% (21–24), and angiographic series in which it ranges from 0.5% to 2.5% (1–4).
This large discordance suggests that only a minority of patients with MB are in fact at increased risk for clinical symptoms and cardiac events. Moreover, among patients with angiographically documented milking effect or systolic narrowing of a coronary artery, a fairly large percentage (roughly half in our series of 27 patients) have concomitant atherosclerotic, muscular, or valvular heart disease, which independently affect clinical outcome as well as treatment strategy (1). Finally, among patients with isolated MB documented at angiography, only about two-third exhibit a >50% narrowing of the vessel during systole (1).
Mechanisms of myocardial ischemia in patients with symptomatic MB
Until recently, visual interpretation of coronary angiograms only revealed the milking effect or systolic narrowing induced by significant myocardial bridging of a coronary artery. However, it is well recognized that coronary flow occurs predominantly during the diastolic phase of the cardiac cycle. Therefore, it appeared unlikely that this systolic phenomenon could by itself result in myocardial ischemia. Although a delay in diastolic relaxation of the coronary vessel was suspected in these patients, its presence was hard to demonstrate on the coronary angiogram. In 1981, in a frame-by-frame analysis of cine-angiograms during a complete cardiac cycle, we were able to demonstrate that 17 of 20 patients (85%) with a ≥75% milking effect of the LAD (by visual inspection) had an extension of the obstruction into diastole, which averaged 136 ms or 26% (range 4% to 50%) of diastole (25). In 1986, Navarro-Lopez et al. (26)also showed a time-lag of up to one-third of diastole before flow returned to normal following systolic compression. In an elegant study published in 1983, Rouleau et al. (27)demonstrated that transient systolic compression of the left circumflex coronary artery in dogs resulted in a diastolic time lag of 69 ± 4 ms, which was needed to repressurize the distal coronary bed and which could become the limiting factor for myocardial perfusion during tachycardia and maximal coronary vasodilation.
Finally, Pichard et al. (28)reported a significant transient decrease in great cardiac vein flow during atrial pacing in patients with myocardial bridging of the LAD, and they attributed this flow reduction to a reduced time period of diastole (compared to systole) during tachycardia. However, it was not until the current era of new investigational modalities such as quantitative coronary angiography (QCA), intravascular ultrasound (IVUS), and intracoronary Doppler flow velocity measurements that we were able to better understand the hemodynamic disturbances occurring during the diastolic phase of the cardiac cycle in patients with MB (11–16).
On the basis of these recent studies, two mechanisms are responsible for reduced coronary flow reserve (CFR) in the distal vessel and for clinical symptoms and signs of myocardial ischemia: 1) a phasic systolic vessel compression with persistent mid-to-late diastolic diameter reduction and 2) increased intracoronary Doppler flow velocities with abnormal qualitative flow profiles.
Coronary diameter reduction
Table 1summarizes the QCA findings in four different series of patients with symptomatic MB and >50% diameter narrowing of the LAD by visual inspection. The percent mean lumen diameter (MLD) reduction in the most severe view of the MB during systole ranged from 71% to 83%, and there was a persistent mean diastolic MLD reduction from 34% to 41%. These luminal narrowings were confirmed by cross-sectional area measurements using IVUS (15). The average length of the MB during systole ranged from 23 to 28 mm and was not different from the diastolic values of 22 to 27 mm.
Increased intracoronary Doppler flow velocities
As shown in Table 2, resting average peak-flow velocity (APV) and average diastolic peak flow velocity (ADPV) were higher within the bridged segment than in the corresponding proximal and distal segments. The flow velocity increase within the MB was smallest for the average systolic peak flow velocity (ASPV). Instantaneous maximal peak flow velocity (MPV) was more than twice as high within the bridged segment as compared to the proximal and distal segments. Mean diastolic/systolic flow velocity ratio (DSVR) at rest ranged from 2.4 to 2.9. Thus, intracoronary Doppler flow measurements reveal a significant increase in APV, ADPV, and MPV within the MB at rest, with only minor changes in systolic flow. These flow velocity abnormalities are very consistent across the different studies.
During rapid atrial pacing (Table 3), all flow velocities within the MB were increased. The flow acceleration was largest for MPV and again smallest for ASPV within the bridge segment. There was a significant increment in DSVR within the MB during tachycardia, but no change proximally or distally. Thus, atrial pacing induced a further acceleration in flow velocities within the bridged segment, whereas the flow velocities proximal and distal to the bridge remained unchanged. A significant increase in DSVR appears to be the most prominent flow alteration within the MB during stress. Finally, qualitative analysis of the Doppler flow profile shows a highly characteristic pattern in approximately 90% of the patients with an abrupt early diastolic flow acceleration, which has been termed the “finger tip” phenomenon, a rapid mid-diastolic deceleration and mid-to-late diastolic plateau (Fig. 2). A retrograde flow phenomenon during systole was detected in the majority of patients at the proximal entry site of the bridged segments. Atrial pacing led to further accentuation of the velocity profile, with shortening of the diastolic plateau as well as accentuation of the retrograde flow phenomenon.
Schwarz et al. (13)quantitatively analyzed consecutive stop-frame images from a complete heart cycle in 14 patients and correlated their results with Doppler flow measurements obtained in the same patients at an identical site. There was rapid lumen reduction during systole with the smallest diameter during late systole, the characteristic milking effect. Early diastolic diameter gain was clearly delayed with a persistent mid-diastolic diameter reduction always >30%. The abrupt early diastolic flow acceleration was due to this persistent diastolic diameter reduction. This was followed by rapid diastolic lumen gain, which reduced the flow velocity, and finally stabilization of the lumen diameter leading to a plateau during late diastole. As shown by this study, the speed of early diastolic diameter gain and systolic compression strongly depends on heart rate, contractile state, and peripheral resistance and may vary substantially.
The CFR, defined as the ratio of mean flow velocity achieved at peak hyperemia to mean resting flow velocity, was obtained after intracoronary injection of papaverine. A normal CFR ratio should be above 3.0. Thus, CFR distal to the bridge was abnormally reduced in all these studies, ranging from 2.0 and 2.6 (Table 2). Impaired CFR can be explained, besides systolic vessel compression, by the delayed early and mid-diastolic relaxation with increased diastolic blood flow velocities. The increase in CFR obtained after angioplasty of the bridge segment is described in the following text.
Additional mechanisms of myocardial ischemia
Several additional factors have been shown to influence both the degree of systolic compression and the speed of early diastolic gain and subsequent diastolic relaxation in patients with symptomatic MB (Table 4). They include anatomical variables such as length, thickness, and location of the muscle bridge and presence or absence of left ventricular hypertrophy as well as physiological factors such as increased heart rate, low systolic arterial pressure, coronary vasomotion, and enhanced platelet aggregation.
Both autopsy and angiography reports have shown that the length of MB varies widely (from 4 to 40 mm) (1–4,21–24). As shown experimentally, the length of a coronary arterial narrowing markedly influences coronary hemodynamics (29). This may explain why relatively long bridges are observed in symptomatic patients (Table 1). In autopsy reports, the thickness of the bridge also varies widely (from 1 to 4 mm) (21–24). Intravascular ultrasound studies confirm that the thicker bridges are most frequently observed in symptomatic patients (15). As shown by Ferreira et al. (24), the location and orientation of the muscle bundles over the imbedded coronary artery also affect systolic compression. Many bridges are superficial, crossing the vessel transversely toward the apex of the heart. Others are deeper muscle bridges surrounding the mid or even the proximal LAD. Finally, left ventricular hypertrophy, hypertrophic cardiomyopathy, and aortic stenosis are not infrequently associated with symptomatic MB, and a progressive increase in left ventricular wall tension may explain at least in part why MB present from birth produces symptoms only later in life (1,6,7,30).
Other important factors include the effect of heart rate (1,5,12,13,31)and systemic blood pressure (32,33)on muscular bridges. With slow-resting heart rate, most coronary blood flow occurs in diastole. However, at high heart rates, the diastolic period is significantly shortened. Intracoronary Doppler measurements show that diastolic flow velocity disturbances predominate in patients with symptomatic MB during rapid atrial pacing (Table 3). Changes in systemic arterial pressure and coronary perfusion pressure can also significantly affect the severity of systolic and consequently of early and mid-to-late diastolic compression in patients with MB. For example, sublingual or intracoronary nitroglycerin and sodium nitroprusside increase the coronary narrowing in patients with MB, whereas noradrenaline, phenylephrine, and ergonovine decrease it (4,31,33).
As an alternative mechanism, vasospastic coronary constriction may be associated with MB (9,34–36). The systolic milking of the vessel may produce, especially at high heart rates, endothelial damage, which may stimulate platelet aggregation and coronary vasospasm. Interestingly, however, several investigators have reported an absence of atherosclerotic changes both at the level of intramural coronary arteries and in the distal segment of the bridge at autopsy, as well as at coronary angiography (4,19,23,24), and this has been confirmed by IVUS (11,15). The mechanisms responsible for this protective effect are poorly understood. In contrast, it has been postulated that the systolic compression may increase the trauma sustained by the intima proximal to the bridge (11,21,23). Ge et al. (11), using IVUS, found atherosclerotic plaques in the segment proximal to the bridge in 12 of 14 patients (86%). In a subsequent IVUS study (15), significant stenoses were documented in 12 of 69 patients; these 12 patients subsequently underwent percutaneous coronary intervention (PCI).
Clinical management of patients with symptomatic MB
Clinical symptoms and objective signs of myocardial ischemia
Myocardial bridges can be an incidental finding at the time of coronary angiography. Conversly, a wide variety of clinical syndromes including unstable angina, acute myocardial infarction (AMI), life-threatening cardiac arrhythmias, and sudden cardiac death have been associated with MB (5–10). Likewise, following our initial publication in 1976 (1), several series of symptomatic patients have been reported (2–4,11–16). Table 5lists the demographic and clinical characteristics of patients with symptomatic MB in six recent studies (11–16). Typically, these patients are predominantly males, 5 to 10 years younger than patients with symptomatic coronary disease, and they have fairly severe anginal symptoms. Typical angina is present in 55% to 70% of the cases, and atypical angina is often reported in association with rest angina. The mean time interval between onset of symptoms and coronary angiography is >18 months, and there were an average of 2.5 previous hospital admissions for angina or suspected AMI. Some patients had documented prior anterior or septal non–Q-wave AMI. Except for 12 patients (18%) of Ge et al. (15)who had significant stenoses in the proximal segment of the bridged artery, all these selected patients had isolated MB with >50% systolic lumen diameter reduction, without angiographic evidence of significant coronary atherosclerosis or left ventricular hypertrophy.
Symptom-limited exercise electrocardiograms were performed in 73% to 100% of patients. As a rule, patients not undergoing stress testing before cardiac catheterization had unstable symptoms. Significant ischemic ST-segment depression >0.1 mV in the anterior leads was identified in 28% to 67% of patients in whom the test was done. Myocardial scintigraphy was obtained in 38% to 100% of patients. Stress-induced perfusion defects of the anterior wall or septum were documented in 33% to 63% of the patients in whom the myocardial scan was performed.
Therefore, compared to patients with significant single-vessel coronary disease, these selected patients with isolated symptomatic myocardial bridging and >50% systolic lumen diameter narrowing of the LAD probably exhibit a comparable severity of anginal symptoms and of noninvasive signs of myocardial ischemia. This probably also includes a similar incidence of prior or recent unstable angina and AMI. Finally, the long-term outcome of these patients may also be as favorable as that of patients with single-vessel coronary disease or even better, because, in contrast to atherosclerosis, little late progression is to be expected.
Invasive diagnostic methods
Quantitative coronary angiography, usually in the plane showing the most severe narrowing, objectively measures the lumen diameter reduction within the bridge segment as compared to the proximal and distal segments and determines the percent MLD narrowing both during systole and during mid-to-late diastole. As a rule, a significant milking effect shows a ≥70% MLD reduction during systole and a ≥35% MLD reduction during mid-to-late diastole (Table 1). The severity of the systolic and mid-to-late diastolic MLD narrowing can be confirmed by IVUS, which has the advantage of assessing lumen area instead of only diameter in an invariably nonspherical vessel lumen. Ge et al. (15)have described a highly specific “half moon” phenomenon surrounding the MB in all 62 patients in whom the IVUS catheter could be passed through the bridge segment (Fig. 3). This ultrasound phenomenon is only found in the bridge segment and not in the proximal or distal segment or in other coronary arteries. In some cases where the specific “half moon” phenomenon was demonstrated but the typical angiographic milking effect was not obvious presumably because of only slight myocardial bridging, the milking effect could be provoked by intracoronary administration of nitroglycerin.
Intracoronary Doppler flow velocity measurements reveal another highly characteristic diagnostic pattern in patients with MB. In approximately 90% of cases, the flow velocity curve within the bridge segment is characterized by an early diastolic “finger tip” phenomenon, followed by a rapid mid-diastolic deceleration and a mid-to-late diastolic plateau (Fig. 2). The typical “finger tip” phenomenon is produced by an abrupt early diastolic flow acceleration as a result of delayed vessel relaxation within the bridge segment. As a result of the systolic compression, there is also a markedly reduced to absent antegrade systolic flow and quite often a typical local systolic retrograde flow phenomenon directly proximal to the more severe site of the MB. Rapid atrial pacing accentuates these characteristic early diastolic and retrograde systolic flow phenomena while it also shortens the mid-to-late diastolic plateau (Fig. 2). Typically, intracoronary Doppler flow velocities at rest, especially APV, ADPV, and MPV should be higher in the bridge segment than in the proximal and distal segments and they should be accentuated by rapid atrial pacing (Tables 2 and 3). Finally, a <3.0 CFR distal to the MB obtained after intracoronary administration of papaverine reflects a hemodynamically significant mid-to-late diastolic coronary flow reduction as a result of the increased resistance produced by the MB (Table 2).
On the basis of the previous pathophysiological and clinical observations, three different treatment strategies, medical therapy, PCI, and direct surgical myotomy or coronary bypass grafting, might be beneficial in these patients. Obviously, medical therapy should be the first and principal strategy, and interventions should be limited to patients with refractory angina despite medical therapy. Finally, these treatment options are not strikingly different from those of patients with single-vessel coronary disease, and especially those with single LAD disease.
Symptomatic patients with myocardial bridging must be treated. Medical therapy should include optimal doses of beta-blockers, calcium channel blockers, and antiplatelet agents with the objective of relieving symptoms and signs of myocardial ischemia and/or protecting against the risk of future coronary events. Negative inotropic and chronotropic agents, especially beta-adrenergic receptor blockers, can reduce external vessel compression by lowering systemic and intramural pressures and can improve coronary perfusion by prolonging the period of diastole (37,38). Schwarz et al. (12)demonstrated that the administration of a short-acting beta-blocker during atrial pacing alleviated symptoms and signs of ischemia in patients with symptomatic MB. This effect was mediated by a significant reduction in systolic and diastolic vascular compression as well as a significant reduction in mid-to-late diastolic flow velocities within the bridged segments with a normalization of both DSVR and CFR. Calcium channel blockers have been less often used in these patients (38). They may be particularly useful when there is a contraindication to beta-blocker therapy or as primary therapy when coronary vasospasm is suspected. Nitrates have been used effectively in some patients, possibly because of their capability to reduce preload and to relieve vasospasm (28), although they have been shown to increase the milking effect on angiography mainly as a result of coronary and systemic vasodilation with increased heart rate, cardiac contraction, and reduced coronary pressure distal to the bridged segment (32,33). Finally, in most symptomatic patients, limitation of strenuous physical activity is usually indicated to avoid the deleterious effect of tachycardia.
Stables et al. (39)first demonstrated, in 1995, that intracoronary stent implantation could achieve internal stabilization of the coronary artery lumen against external compression in a patient with a muscular bridge. In 1997, Klues et al. (14)reported hemodynamic, angiographic, and IVUS data immediately and seven weeks after successful coronary stent implantation in three symptomatic patients. Stent placement abolished the phasic lumen compression, the diastolic flow abnormalities, and clinical symptoms. The CFR improved from 2.4 ± 0.5 to 3.8 ± 0.3. Coronary angiography after seven weeks revealed an identical lumen enlargement without any systolic or diastolic diameter reduction and a further increase in CFR. Intravascular ultrasound revealed no signs of neointimal proliferation within or proximal or distal to the stented segments. All patients reported remarkable clinical improvement, with increased physical activity. Favorable short-term clinical and angiographic results after coronary stenting were also reported by others (40,41). Haager et al. (16)recently described successful coronary stenting in 11 patients who subsequently underwent repeat coronary angiography at seven weeks and six months as well as clinical follow-up at two years. Immediately after stent deployment, QCA showed absence of systolic compression, and all flow abnormalities including abrupt early diastolic flow acceleration, rapid mid-diastolic deceleration and mid-to-late plateau, and retrograde flow during systole were normalized. At seven weeks, QCA showed in-stent restenosis in 5 of 11 patients (46%). Four underwent repeat target lesion revascularization; two received coronary angioplasty and two underwent coronary bypass surgery with an internal mammary artery graft to the LAD. At two years, all remained free of angina and cardiac events. As stated by the investigators, the incidence of in-stent restenosis in these patients was not different from that of lesions of 25-mm length and relatively small-vessel calibers in coronary patients (Table 1).
Long-term studies are necessary to evaluate the stability of stent geometry, in-stent restenosis, and clinical outcome in these patients.
Before the current era of coronary stenting, surgical myotomy was regarded as perhaps the treatment of choice for patients with persistent symptoms despite intensive medical therapy. Indeed, cleavage of the overlying muscle fibers eliminates the phasic compression of the coronary vessel. Thus, supra-arterial myotomy for myocardial bridging and milking effect of the LAD was reported by several surgical centers (42–45). However, this approach is more invasive than that of a catheter-based intervention and probably also carries a higher risk of postprocedural complications. In addition, the unpredictable intramural course of the coronary artery may require deep incision of the ventricular wall, potentially leading to subsequent ventricular wall aneurysm. Alternatively, internal mammary artery anastomosis to the LAD may be the treatment of choice in patients with unsuccessful coronary stenting or in-stent restenosis (16). Likewise, patients with significant coronary disease in other vessels who require coronary bypass surgery might be suitable candidates for internal mammary artery grafting.
Recent techniques available in the cardiac catheterization laboratory, such as QCA, IVUS, and intracoronary Doppler, have shown that significant myocardial bridging of a coronary artery, especially the LAD, is characterized by the following morphological and hemodynamic alterations: 1) phasic systolic vessel compression; 2) persistent diastolic lumen diameter reduction; 3) increased blood flow velocities; 4) retrograde systolic flow; and 5) reduced CFR. These alterations, particularly in association with aggravating factors such as increased heart rate, decreased systolic blood pressure, and coronary vasospasm, may explain the occurrence of symptoms and myocardial ischemia in patients with symptomatic MB. As a rule, these patients present with symptoms and signs of ischemia, as well as a prior history of unstable angina and non–Q-wave AMI in some patients. In this respect, they are not strikingly different from patients with significant single-vessel coronary disease. The severity of coronary vessel narrowing during both systole and diastole can be established by QCA and IVUS and the degree of flow velocity acceleration by intracoronary Doppler. In addition, highly specific morphological and flow velocity patterns include the “half moon” phenomenon on IVUS and the “finger tip” phenomenon on intracoronary Doppler. Finally, the treatment options include medical therapy in the majority of patients and stent implantation, surgical myotomy, or internal mammary artery bypass grafting in patients with persistent symptoms despite medical therapy. The long-term outcome of such interventions, however, is not well known and requires further study.
- acute myocardial infarction
- average diastolic peak flow velocity
- average peak flow velocity
- average systolic peak flow velocity
- coronary flow reserve
- diastolic/systolic flow velocity ratio
- intravascular ultrasound
- left anterior descending coronary artery
- myocardial bridge(s)
- mean lumen diameter
- maximal peak flow velocity
- percutaneous coronary intervention
- quantitative coronary angiography
- Received September 11, 2002.
- Accepted October 17, 2002.
- American College of Cardiology Foundation
- Morales A.R.,
- Romanelli R.,
- Boucek R.J.
- Bestetti R.B.,
- Costa R.S.,
- Zucolotto S.,
- Oliveira J.S.
- Bauters C.,
- Chmait A.,
- Tricot O.,
- Lamblin N.,
- Van Belle E.,
- Lablanche J.M.
- Ge J.,
- Erbel R.,
- Rupprecht H.J.,
- et al.
- Schwarz E.R.,
- Klues H.G.,
- vom Dahl J.,
- Klein I.,
- Krebs W.,
- Hanrath P.
- Schwarz E.R.,
- Klues H.G.,
- vom Dahl J.,
- Klein I.,
- Krebs W.,
- Hanrath P.
- Klues H.G.,
- Schwarz E.R.,
- vom Dahl J.,
- et al.
- Ge J.,
- Jeremias A.,
- Rupp A.,
- et al.
- Haager P.K.,
- Schwarz E.R.,
- vom Dahl J.,
- Klues H.G.,
- Reffelmann T.,
- Hanrath P.
- Reyman H.C.
- Black S.
- Ferreira A.G.,
- Trotter S.E.,
- Konig B.,
- Décourt L.V.,
- Fox K.,
- Olsen E.G.J.
- Bourassa M.G.,
- Bernard P.,
- Brevers G.,
- Petitclerc R.,
- Dyrda I.
- Rouleau J.R.,
- Roy L.,
- Dumesnil J.G.,
- Dagenais G.R.
- Stables R.H.,
- Knight C.J.,
- McNeill J.G.,
- Sigwart U.
- Bayes A.,
- Marti V.,
- Augé J.M.
- Prendergast B.D.,
- Kerr F.,
- Starkey I.R.
- Binet J.P.,
- Guiraudon G.,
- Langlois J.,
- Piot C.,
- Vachon J.,
- Grosgogeat Y.