Coronary Artery Dissection and Perforation Complicating Percutaneous Coronary Intervention

Jason H. Rogers, MD; John M. Lasala, MD, PhD


J Invasive Cardiol. 2004;16(9) 

In This Article

Coronary Artery Dissection

Percutaneous coronary intervention, which depends upon mechanical dilatation of the artery or ablation of atherosclerotic plaque, is requisitely associated with plaque fracture, intimal splitting and localized medial dissection — these tears may extend into the media for varying distances, and may even extend through the adventitia resulting in frank perforation.

The National Heart, Lung and Blood Institute (NHLBI) classification system for intimal tears, developed by the Coronary Angioplasty Registry, was put forth in the pre-stent era for the classification of dissection types after balloon angioplasty. Dissections in this scheme are graded based upon their angiographic appearances as types A through F7 (Figure 1).

National Heart, Lung and Blood Institute (NHLBI) classification system for coronary artery dissection types. See text for complete description. Types A and B are generally clinically benign, whereas types C through F portend significant morbidity and mortality if untreated.

Type A dissections represent minor radiolucent areas within the coronary lumen during contrast injection with little or no persistence of contrast after the dye has cleared. Type B dissections are parallel tracts or a double lumen separated by a radiolucent area during contrast injection, with minimal or no persistence after dye clearance. Type C dissections appear as contrast outside the coronary lumen ("extraluminal cap") with persistence of contrast after dye has cleared from the lumen. Type D dissections represent spiral ("barber shop pole") luminal filling defects, frequently with excessive contrast staining of the dissected false lumen. Type E dissections appear as new, persistent filling defects within the coronary lumen. Type F dissections represent those that lead to total occlusion of the coronary lumen without distal antegrade flow. In rare cases, a coronary artery dissection my propagate retrograde and involve the ascending aorta.[8]

Numerous studies performed prior to the common use of stents found that, in general, type A and B dissections are clinically benign and do not adversely affect procedural outcome. However, types C through F are considered major dissections and carry a significant increase in morbidity and mortality.[1,9] Acute vessel closure is the most feared complication due to coronary artery dissection, and in the pre-stent era occurred in up to 11% of all elective PTCAs.[10–13] With the advent of coronary stents, the incidence of acute closure in elective PCI is now less than 1%.[14] Ischemic complications in the current era usually occur as manifestations of edge dissections after stenting, which may predispose to stent thrombosis. Given the uncertainty of predicting the behavior of edge dissections after PCI, it is considered prudent to treat these with additional stent deployment, if technically feasible (Figure 2).

Coronary artery edge dissection after stent deployment in an obtuse marginal. Note intimal tear (arrow) distal to stent (dashed line). This dissection was successfully treated with an overlapping stent distally with no angiographically evident residual dissection.

Assuming a guidewire may be passed into or resides concurrently within the true lumen of a dissected coronary artery, dissections in the current era can usually be managed by deployment of stents sufficient to "tack-up" the dissection flap. Given the propensity of dissections to propagate distal with antegrade coronary flow, the operator should seek to contain and cover the distal extent of the dissection as soon as possible with a stent to prevent further extension. If a dissection extends into a terminal branch too small to accommodate a stent, it may be impossible to rescue the entire artery.

Multiple studies in the pre-stent era have identified risk factors for the development of coronary artery dissection. Angiographic predictors include calcified lesions,[15] eccentric lesions,[16] long lesions,[15] complex lesion morphology (ACC/AHA type B or C)[17] and vessel tortuosity.[16] Patients with acute myocardial infarction or unstable angina have evidence of coronary dissection in as many as 28% of cases, which despite prolonged balloon inflations, predisposes to subsequent subacute thrombosis.[18] A balloon to artery ratio > 1.2 also predisposes to dissection.[16] In addition, Amplatz guide catheters are often used for better support during PCI; however, they appear to have a higher rate of coronary dissection, likely due to "deep seating" during engagement of the coronary ostium.[8]

Although rare, it is worth mentioning that coronary artery dissection can occur spontaneously and may involve single or multiple coronary arteries. The incidence of spontaneous coronary dissection occurs at rates of 0.1–0.28% of all angiographic studies.[19,20] Blunt trauma,[21] cocaine use[22] and extension of aortic dissection[23] have also been reported to result in coronary dissection. There are over fifty cases in the literature reporting spontaneous coronary dissection during late pregnancy and the early postpartum period.[24,25] Systemic illnesses may also predispose to coronary dissection such as connective tissue disorders,[26] Marfan's syndrome,[27] Kawasaki's disease[26] and alpha-1-antitrypsin deficiency.[28] In general, spontaneous coronary dissection is a life-threatening condition; however, in those surviving the initial event, the survival rate is reported to be 78%–82%.[29,30]

Specifically, DeMaio et al published a review of 83 cases of spontaneous coronary dissection reported in English language journals. Of these, 62 were diagnosed postmortem, yielding a presumed acute mortality rate of 75%. They subsequently characterized 32 patients with an antemortem diagnosis of spontaneous coronary dissection (21 from the aforementioned published cases and 11 from regional telephone enquiries in the Atlanta, Georgia area). Follow-up was available in 28 patients. At a mean follow up of 38 months (range 1.5–144), 23 patients were alive (82% survival).[29]

More recently, Tsimikas et al reviewed the clinical presentation of 65 cases of spontaneous coronary dissection reported in the literature since 1993 and suggested that outcomes have improved due to earlier recognition of this entity. Of the 65 patients reported,[51] 78% were described as surviving the acute event. Therefore, the inferred acute mortality rate was 22%. This may represent an actual decrease in acute mortality due to earlier detection, or could merely represent recent underreporting of acute mortality cases.[30]

Of note, beta irradiation appears to impair healing of coronary dissections during PCI as determined by intravascular ultrasound follow-up.[31] This is an important implication in our era of in-stent restenosis. Although unproven, it may be prudent to stent any angiographically apparent dissections after brachytherapy, provided an extended course of antiplatelet therapy is administered.[32]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.