Case Report. A 48-year-old otherwise healthy male underwent coronary angiography for acute coronary syndrome. Treatment prior to angiography included aspirin, clopidogrel, intravenous heparin, nitrates and a bolus followed by infusion of tirofiban. Angiography revealed no significant disease in the right coronary artery and moderate proximal left anterior descending artery (LAD) disease, with the appearance of a ruptured plaque (not shown). The left circumflex artery (LCX) demonstrated critical and complex disease in the mid and distal vessel as well as a critical lesion in the ostium of a large bifurcating obtuse marginal (OM) artery (Figure 1). A decision was made to proceed with PCI of the LAD and LCX. The left main artery was engaged with a 7 Fr XB 3.5 guiding catheter (Cordis Corp., Miami Lakes, Florida). The LAD underwent uncomplicated direct stenting with an everolimus-eluting Xience™ V stent (not shown) (Guidant Corp., Santa Clara, California). Next, the LCX and obtuse marginal (OM) were wired with Balanced Middle Weight™ 0.014 inch x 190 cm wires (Abbott Pharmaceuticals, Abbott Park, Illinois). The initial plan was to perform balloon angioplasty of the mid and distal LCX and implant a stent in the proximal LCX going into the lower branch of the large OM. Dilatation of the distal LCX with a JOCATH Mercury (Abbott Pharmaceuticals) 2.5 x 20 mm balloon at 6 atm resulted in dissection, necessitating stenting of the mid and distal vessel with a Xience V 2.5 x 23 mm stent deployed at 10 atm and postdilated with a noncompliant Powersail® (Guidant) 2.5 x 13 mm balloon at 20 atm (Figure 2). Next, the proximal LCX was stented into the lower branch of the large OM with a Xience V 2.5 x 28 mm stent deployed at 14 atm. A dissection was noted at distal edge of this stent, which was covered with a second Xience V 2.5 x 12 mm stent deployed at 10 atm. These stents were then post-dilated with noncompliant Powersail® 2.5 mm (mid and distal segments) and 3.25 mm (proximal segments) balloons at high pressure taking particular care that the balloons did not extend beyond the distal and proximal edges in multiple cine images. Unfortunately, another dissection became evident at the distal edge of the OM stent (Figure 2). Multiple prolonged low pressure balloon inflations were performed at this edge but the dissection persisted. A third stent, Xience V 2.5 x 15 mm was deployed in distal OM at 8 atm. The stent balloon was then pulled back and dilated at 16 atm to postdilate overlapping zone.
Immediately after the final postdilatation, extravasation of dye was noted from within the third OM stent (Figure 3). The stent balloon was immediately dilated proximal to the perforation to seal it off. The tirofiban infusion was discontinued and intravenous protamine was administered to reverse the effect of heparin, lowering the activated clotting time to 136 seconds. An immediate transthoracic echocardiogram revealed no pericardial effusion and lateral hypokinesis. The patient was hemodynamically stable but experienced chest pain. After 15 minutes, the balloon was let down and angiographic imaging demonstrated significantly reduced (but nevertheless persistent) pulsatile ejection of blood from the distal OM. The balloon was inflated again and over the next 45 minutes, this cycle was repeated multiple times. The degree of extravasation persisted and remained unchanged. Repeated echocardiographic images continued to reveal no pericardial effusion. Attempts at deploying a hand-mounted stent graft were unsuccessful, as the latter would not track beyond the proximal OM and eventually was dislodged, requiring it to be crushed against the wall. The equipment for coil embolization of the coronary artery was not available at that time.
Over the next 2 hours, the patient was observed in the cardiac catheterization laboratory with an occlusive balloon inflated in the OM, deflating it transiently for angiographic imaging. Serial images continued to show the same degree of pulsatile leakage from the vessel, with eventual closure of the upper branch of the OM (Figure 4). Serial echocardiographic images only showed severe lateral hypokinesis. The patient remained stable, complaining of mild chest pain.
After considering all options, a decision was made to proceed with surgical correction of the problem despite the patient’s clinical stability, based on the concern that if the leakage of blood was left uncorrected, he could eventually develop a dissecting intramyocardial hematoma that could lead to catastrophic (even fatal) consequences. The catheters were removed and the patient was transferred to the operating room. He was placed on cardiopulmonary bypass within 45 minutes of leaving the catheterization laboratory. Multiple platelet transfusions were administered prior to starting the operation. Perioperatively, a large tracking intramyocardial hematoma was seen on the lateral wall that was somewhat tense (Figure 5). Given the patient’s stable hemodynamics, this hematoma was not incised or decompressed. The perforated lower branch of the OM was ligated and saphenous vein grafts were applied to the upper branch of the OM, LAD and right posterolateral vessels. Doppler flow in the OM distal to the graft was acceptable. The patient’s subsequent hospital course was uncomplicated. Three months post procedure, the patient was doing well.
Discussion. Our case highlights the fact that serious complications can occur, even in apparently straightforward PCI cases. Coronary perforation is a rare complication of PCI,1 and the subset of patients who develop a subepicardial hematoma without hemopericardium is even more uncommon. A review of published reports suggests that intramyocardial hematomas can present with devastating complications. Pathophysiologically, an initial bleed may appear “contained”, only to progress in a relentless fashion, dissecting its way through the myocardium and, in the process, avulsing perforating vessels, which in turn bleed further, establishing a self-propagating process.3,4 Furthermore, this propagation may continue despite sealing of the index perforation, as was described in a recent report by Quan et al.4 Such progression may take several hours, therefore, initial clinical stability may convey a false sense of security. Variants of this process have also been described including compression of the right ventricular outflow tract9 and the left atrium5 by the subepicardial hematoma. Table 1 summarizes the predominant body of literature on the subject. While the limited body of literature on the subject reflects the rarity of this condition, it is likely that this entity is underreported, particularly given the lack of obvious “diagnostic” features on echocardiography or angiography. Subepicardial hematoma formation is exceedingly rare in patients without a history of CABG. It is thought that adhesions between the epicardium and pericardium in post-CABG patients prevent blood from accumulating in the pericardial space unlike those who have not been operated on. To date, only 2 cases of subepicardial hematoma have been reported in patients without prior CABG undergoing PCI.7,8 Both of these patients developed cardiogenic shock. Our case represents the third such instance. However, our case is different because there were few manifestations either clinically or on echocardiographic imaging, and the diagnosis was correctly anticipated on pathophysiological grounds. Our thought process was dictated by the continued (albeit reduced) pulsatile ejection of blood from the perforation site, the lack of any pericardial effusion whatsoever (therefore implying that the force of the extravasating blood would be directed into the myocardium) and knowledge of the possibility that dissecting intramyocardial hematomas may self-propagate. The loss of flow in the upper branch of the OM may have been a clue suggesting compression of the vessel in the midst of a hematoma (Figure 4), but a thrombotic occlusion was equally likely. Our suspicions were confirmed perioperatively and, given the size of the hematoma (Figure 5), it is quite possible that late deterioration may have occurred had we elected to observe and treat the patient conservatively.
Several important points must be made regarding the case presented here. First, the reason for the perforation remains unclear. The conventional risk factors for perforation (advanced age, tortuosity, female gender, use of stiff wires and devices)10–12 were absent. High-pressure stent postdilatation was performed multiple times and may have been a contributor, although the final inflation was at a lower pressure (16 atm) than previous dilatations. While difficult to be certain, the use of tirofiban may have increased the propensity of the patient to bleed. However, the use of a glycoprotein IIb/IIIa inhibitor in this setting would be routine and appropriate in most catheterization laboratories. Second, our patient’s lack of hemodynamic instability may well have been due to the strategy of keeping an occlusive balloon inflated in the coronary artery, thereby preventing continued extravasation of blood and propagation of the dissecting hematoma; this was maintained for over 2 hours in the catheterization laboratory, the hope being that the perforation would eventually seal. When the balloon was finally removed, the patient was on cardiopulmonary bypass within 45 minutes. Third, even if the perforation had sealed, judging by the large size of the hematoma seen at surgery, it may have continued to propagate, leading to serious consequences similar to the report by Quan et al.4 Thus, in some ways, the failure to deploy a stent graft (and the unavailability of coil embolization) may have worked in the patient’s favor by compelling surgical exploration.
Conclusion. In summary, we report the third documented case of a dissecting subepicardial hematoma as a result of coronary perforation in a patient with no prior history of CABG. Our case is different from the others because the problem was correctly anticipated in the absence of hemodynamic or echocardiographic manifestations, and surgical correction of the perforation was undertaken. While a single case can never be grounds for recommending a particular strategy, our case emphasizes the need to preemptively think of the entity of intramyocardial hematoma when the setting is right — an obvious perforation without pericardial effusion, particularly (though not necessarily as our case indicates) in patients with a history of prior CABG and especially when there is hemodynamic instability.
Acknowledgements. The authors acknowledge the efforts of the nurses and support staff of the cardiac catheterization laboratory and operating room in managing this challenging case.
- Gunning MG, Williams IL, Jewitt DE, et al. Coronary artery perforation during percutaneous intervention: incidence and outcome. Heart 2002;88:495–498.
- Ellis SG, Ajluni S, Arnold AZ, et al. Increased coronary perforation in the new device era. Incidence, classification, management, and outcome. Circulation 1994;90:2725–2730.
- Shekar PS, Stone JR, Couper GS. Dissecting sub-epicardial hematoma — Challenges to surgical management. Eur J Cardiothorac Surg 2004;26:850–853.
- Quan VH, Stone JR, Couper GS, et al. Coronary artery perforation by cutting balloon resulting in dissecting subepicardial hematoma and avulsion of the vasculature. Catheter Cardiovasc Interv 2005;64:163–168.
- Dardas PS, Tsikaderis DD, Makrigiannakis K, et al. Complete left atrial obliteration due to localized tamponade after coronary artery perforation during PTCA. Cathet Cardiovasc Diagn 1998;45:61–63.
- Furushima H, Matsubara T, Tamura Y, et al. Coronary artery perforation with subepicardial hematoma. Cathet Cardiovasc Diagn 1997;41:59–61.
- Rehders TC, Nienaber CA. [Subepicardial hematoma (haemorrhagia per rhexin) after elective PTCA with consecutive compression of the distal RIVA]. Z Kardiol 1993;82:94–98.
- Misfeld M, Khan SA, Ilsley C, et al. Epicardial haematoma: Rare cause of acute myocardial ischaemia. Eur J Cardiothorac Surg 2002;21:119–120.
- Kawase Y, Hayase M, Ito S, et al. Compression of right ventricular out-flow due to localized hematoma after coronary perforation during PCI. Catheter Cardiovasc Interv 2003:202–206.
- Ajluni SC, Glazier S, Blankenship L, et al. Perforations after percutaneous coronary interventions: Clinical, angiographic, and therapeutic observations. Cathet Cardiovasc Diagn 1994;32:206–212.
- Stankovic G, Orlic D, Corvaja N, et al. Incidence, predictors, in-hospital, and late outcomes of coronary artery perforations. Am J Cardiol 2004;93:213–216.
- Ramana RK, Arab D, Joyal D, et al. Coronary artery perforation during percutaneous coronary intervention: Incidence and outcomes in the new interventional era. J Invasive Cardiol 2005;17:603–605.