Spontaneous resolution of iatrogenic coronary artery to right ventricular fistula secondary to percutaneous coronary intervention

Iatrogenic coronary artery fistula (ICAF) is a rare complication and could happen during endomyocardial biopsy, open heart surgery or percutaneous coronary intervention. The treatment of ICAF is not well established and is dependent upon the patient's haemodynamic situation. Here we report a case of spontaneous resolution of ICAF during such angioplasty. Following repeated exertion and associated chest pain in May 2011, a man aged 52 years with hypertension was referred for coronary angiography, which revealed a 90% stenosis in the proximal segment of left anterior descending coronary artery and a 75% diffuse, tortuous stenosis in the proximal to middle segment of right coronary artery (Figure 1). The left anterior descending lesion was treated with a 3.5 mm×18 mm sirolimus eluting stent. Because of the tortuous diffused right coronary lesion, two 0.35 mm guide wires were passed across the middle segment of the artery. Unfortunately, the working wire was located in the acute marginal artery and a 2.0 mm × 15.0 mm balloon was inflated by error up to 1215.9 kPa (12 atm) in the LAO view. The angiographic appearances of coronary artery perforation were noted (Figure 2). Prolonged inflation was performed intermittently in the segment with the same balloon at 810.6 kPa (8 atm) for 120 minutes, but the perforation persisted. Yet, the angiogram showed the contrast clearing rapidly with no remaining pericardial staining. The patient remained haemodynamically stable with no clinical or echocardiographic features of pericardial effusion. The diagnosis of an ICAF was made based on leakage into the right ventricle. The right lesion was treated subsequently with three overlapping, sirolimus eluting stents with satisfactory result despite the coronary fistula remaining (Figure 3). The cardiac enzyme was within normal limits and no pericardial effusion or wall motion abnormalities were found by echocardiogram postprocedure. He was discharged three days later without chest pain. Seven months later an angiogram (Figure 4) showed no coronary cameral fistula or instent restenosis of left or right coronary arteries.Figure 1.: Coronary angiogram showing a 75% diffuse tortuous stenosis at the proximal to middle segment of right coronary artery.Figure 2.Coronary artery fistula within the acute marginal (white arrow) showing runoff into a cardiac chamber (black arrow), with wires in this vessel and the right posterior descending artery.Figure 3.Coronary artery fistula within the acute marginal (white arrow) showing runoff into a cardiac chamber (black arrow), after stents implanted at the proximal to middle segment of RCA.Figure 4.Coronary angiogram after 7 months showing no coronary artery fistula or instent restenosis of proximal to mid-segment of the right coronary artery (arrow).Coronary artery perforation is a rare but serious complication of percutaneous coronary intervention that may lead to tamponade, myocardial infarction, emergency surgical intervention or death. The incidence of iatrogenic coronary perforation has been reported as between 0.4% and 0.5%1,2 and slightly higher with the use of atheroablative procedures such as atherectomy, thrombectomy and excimer laser angioplasty. Additional perforation risk factors include hydrophilic guidewire technique, the use of oversized balloons and stents, small vessel size, diffuse vessel disease, chronic total occlusion, old age and female gender. The cardiac tamponade rate is 19%-24% and the death rate is 9%.1,2 In the current case, the cause of coronary perforation was dilating the acute marginal artery by error with the oversized balloon. It implied that several different views are necessary to position accurately the guide wire in the target vessel. Ellis et al2 developed an angiographic classification of three types of perforation. Each successive type is associated with decreasing risks of tamponade, myocardial infarction and death. Type I perforations were associated with high rates of tamponade (63%), urgent bypass surgery (63%) and death (19%). However, type III perforations with cavity spilling (ICAF, which account to 5% of coronary perforations) were associated with far less consequences (no deaths, myocardial infarctions or tamponades). It implied that ICAF might have very different pathophysiological mechanisms, such as shunting of blood from coronary arterial bed. Shunting may lead to cardiac volume load and angina if it is severe enough. The management of ICAF is not well established. The strategy for iatrogenic coronary perforations, depending on the type and severity of perforation, generally involves initial conservative management, with prolonged balloon inflation and consideration of reversing anticoagulation depending on patient's stability. If the perforation is distal, embolization with coils, glue or thrombin should be considered. Proximal lesions are likely managed with a covered stent, with a reported 96.4% successful closure rate,3 but the long term patency of covered stent is not certain. Even so, up to one third of iatrogenic coronary perforations may require surgical intervention. ICAFs drainage into the right ventricle, which occurred in 14 of the 176 heart transplant patients who underwent endomyocardial biopsy,4 favour nonsurgical management because of the lack of untoward events, the generally small size of the fistulas and the occurrence of spontaneous resolution. Iatrogenic left coronary artery fistulas were detected in 9 of the 40 study patients with hypertrophic cardiomyopathy underwent surgical myectomy:5 only one patient's fistula did not heal spontaneously and cause symptoms requiring therapeutic intervention. As to ICAF complicating percutaneous coronary intervention, reported cases show variable courses from stable state to rapid detoriation. Spontaneous resolution of ICAF has been reported but it is rare. The management in this case involved an acute coronary perforation resulting in ICAF. The limited evidence in this area suggests that a small perforation into a cardiac chamber in a haemodynamically stable patient should be managed with observation alone. The angiographic results of seven months follow-up verified the decision. In conclusion, the case presented is instructive for two reasons. First, the diagnosis of ICAF should be considered in patients who remain free of clinical and echocardiographic features of tamponade after percutaneous coronary intervention complicated by coronary perforation. Second, angiographic appearances of these fistulas generally can be initially observed in asymptomatic patients.