Diagnosing coronary arterial stent thrombosis and arterial closure.

Coronary artery stents have been a major technical advance in the interventional cardiac catheterization laboratory for “bailout” purposes as well as for primary treatment of de novo and restenotic lesions. The bailout indication for stents is defined as abrupt arterial closure or a suboptimal result after conventional balloon angioplasty in patients at increased risk for abrupt closure, such as a severe dissection. Coronary stents have markedly reduced the rate of abrupt closure after angioplasty, from about 7% to 2%.1 Stent thrombosis is accompanied by a high rate of emergency bypass surgery, myocardial infarction, and death.2 These complications occur most often within the first several days after stent placement. This study reports on the procedural and postprocedural predictors of subacute stent thrombosis and the need for repeat intervention. • • • Coronary artery stents were implanted for elective and bailout indications in 376 patients at Duke University Medical Center between January 1993 and July 1995, 91% for elective indications. Warfarin plus aspirin was given to 200 patients and ticlopidine plus aspirin without warfarin to the remaining 176 patients. Abciximab was not available, but intravascular ultrasound was used to assess stent deployment in 25% of these patients. Of the 376 patients, 40 (10.6%) underwent repeat angiography for postprocedural chest pain and are the subject of this report. The baseline characteristics of these 40 patients are listed in Table I. Palmaz-Schatz stents were placed in 37 patients (93%), and Gianturco-Roubin stents were placed in 3 (7%). Of note, 14 of these patients (35%) had stents placed for bailout indications. Table II contains the procedural and postprocedural predictors that were examined for their ability to predict subacute stent closure. These include anticoagulation regimen, elective versus bailout indication, poststent deployment dilatation balloon pressure, poststent deployment dilatation balloon size, platelet count at the time of restudy, and the presence of “typical” chest pain and persistent chest pain of .30 minutes’ duration. In addition, electrocardiographic changes during the episode of postprocedural chest pain were evaluated in an attempt to predict stent closure. Continuous variables were compared using the Wilcoxon rank sum test. Chi-square testing was used to compare categorical variables. Fisher’s exact testing was used to compare categorical variables if the expected cell frequency was small. A p value of 0.05 (2-tailed) was considered statistically significant. All patients had recurrent chest pain as the primary reason for their repeat cardiac catheterization. The time after stent placement ranged from 1 hour to 10 days (mean 2.5 6 2.1 days). Of the 40 patients undergoing repeat cardiac catheterization for postprocedural chest pain, 10 (25%) had subacute closure, representing an overall subacute closure rate of 2.6% of the entire patient population (10 of 376). When narrowing location was reviewed, 43% (6 of 14) of the right coronary lesions were closed, 28% (4 of 14) of the left anterior descending arteries were closed, and 17% of the circumflex vessels (1 of 6) were closed. Of the 6 vein grafts that underwent relook angiograms, none were closed (0 of 6). The subacute closure rates of the 2 anticoagulation groups were 5% (1 of 19) and 43% (9 of 21) for patients treated with ticlopidine plus aspirin versus warfarin plus aspirin, respectively (p 5 0.003). When expressed for the entire patient population, the subacute closure rates were 0.3% (1 of 376) versus 2.4% (9 of 376), respectively. Subacute closure occurred in 15% of those undergoing elective stent placement (4 of 26) versus 43% in those treated with abrupt closure or threatened closure (6 of 14) (p 5 0.12). The mean balloon pressure for poststent deployment dilatation was higher in the 29 patients whose stents were open at restudy (17.2 6 4.0 atm) than in those in whom they were not (14.6 6 2.1 atm) (p 5 0.01). Poststent deployment maximum balloon size was similar for the 2 groups. The mean measured balloon size in those with open stents was 3.3 6 0.4 mm versus 3.27 6 0.4 mm in the closed group (p 5 0.57). Although the platelet count was higher in the From the Duke University Medical Center, Durham, North Carolina. Dr. Zidar’s address is: Division of Cardiology, Duke University Medical Center, 7405 Hospital North, Box 3290, Durham, North Carolina. Manuscript received October 17, 1997; revised manuscript received April 13, 1998, and accepted April 14, 1998. TABLE I Baseline Characteristics of Patients Undergoing Repeat Angiography