Stent thrombosis represents an infrequent but potentially grave complication of percutaneous coronary intervention (PCI). With current antithrombotic regimens, stent thrombosis occurs after fewer than 1% of stent-based interventions,1,2 but may be more frequent in acute coronary syndromes,3,4 multivessel procedures,5 and other high-risk circumstances where rates have approached 3% or more. The clinical manifestations of stent thrombosis are well recognized and include death rates of 20–25% and major myocardial infarction in 60–70%.2,6,7 However, the economic impact of these events is unknown. Therefore, we reviewed a series of cases of subacute thrombosis (SAT) at our institution in order to estimate the magnitude of health care costs generated. Methods Study population. We searched the interventional cardiology database at our institution and identified 32 cases of angiographically documented stent thrombosis out of a total of 3,295 stent procedures, between the years 1998 to 2000. Qualifying events were required to occur after the completion of the initial PCI. For 26 of the 32 cases, complete clinical records and billing data were available; these 26 cases comprise the study population. The study protocol was approved by the Committee on Clinical Investigations of Beth Israel Deaconess Medical Center. Clinical data. For all patients, baseline demographic and health history data were obtained by review of the medical record. Detailed procedural data for both the index revascularization procedures and for subsequent procedures performed to manage SAT were collected from catheterization laboratory reports. Clinical outcomes were determined from chart review and discharge summaries. Determination of health care costs. Direct medical care costs stemming from the occurrence of SAT were determined as follows. For PCI procedures, itemized resource utilization (including the number of balloon catheters, stents, other devices, guiding catheters, guidewires, doses of medications, and contrast volume) were recorded for each procedure, and the cost of each item was estimated on the basis of the mean acquisition cost for the item in 2000. Costs of additional disposable equipment, overhead, and depreciation for the cardiac catheterization laboratory, and non-physician personnel were estimated on the basis of the average cost per procedure at Beth Israel Deaconess Medical Center in 2000 and adjusted for actual procedure duration. Physician fees for PCI procedures were determined using the 2000 Medicare fee schedule for Massachusetts. The accuracy of this cost accounting method for catheterization laboratory procedures has previously been shown.8 All other costs for the initial hospitalizations associated with SAT, including ICU and routine care; pharmacy; emergency department; radiology; blood bank; laboratory; and other ancillary services were obtained from the microcost accounting system (Transition Systems, Inc.) at Beth Israel Deaconess Medical Center, which contained itemized cost and charge data for all patients. Costs incurred prior to the year 2000 were inflated using the medical care component of the Consumer Price Index. Thus, all costs are expressed in year 2000 U.S. dollars. For patients in whom the qualifying SAT event occurred on an outpatient basis, we recorded all costs for the resultant hospitalization. For patients in whom SAT took place during the same hospitalization as their index revascularization procedure, our analysis was restricted to those costs incurred during the additional hospital days required to treat the episode of SAT. Costs related to subsequent hospitalizations to treat delayed complications of SAT (e.g., congestive heart failure, arrhythmias, etc.), outpatient costs, and indirect costs of SAT (e.g., days missed from work), were not included in our analysis. Statistical analysis. Discrete variables are presented as frequencies. Continuous variables are presented as both means ± 1 standard deviation as well as medians, given that most distributions were non-normal. Statistical comparisons were made on continuous variables using t-tests for normally distributed data and Wilcoxon rank-sum tests for non-normal data. For all statistical tests, a 2-tailed p-value of Patient population and index revascularization procedures. Baseline characteristics of the 26 study patients are shown in Table 1. Both women (54%) and diabetics (69%) were over-represented in this cohort relative to their overall proportion among PCI procedures, both nationally9 and at our center, which ranges from 25–35%. A minority of patients had undergone PCI or CABG prior to their index procedure. Two patients were identified as having hypercoagulable states during their subsequent hospital stay. Details regarding the initial PCI procedures are outlined in Table 2. Over 90% of the initial procedures were performed for unstable angina or myocardial infarction, and 70% involved the placement of two or more stents. The majority of patients (69%) received intravenous platelet glycoprotein (GP) IIb/IIIa antagonists during the procedures, and all but one of them received it prophylactically. More than 90% of patients received a stent with a diameter of 3.0 mm or less. On average, the total length of stented arterial segment(s) was 37 mm. In nearly half of the cases, an acute procedural complication was noted, although many of these were minor dissections judged to have been successfully treated during the procedure, or jailed sidebranches that were rescued or deemed unimportant. All patients were prescribed a standard regimen of aspirin 325 mg daily and clopidogrel 75 mg daily post-procedure. Presentation and in-hospital outcomes. Tables 3 and 4 review the clinical presentation, management, and outcomes of the SAT events. The median time from index procedure to SAT was 3.5 days (interquartile range of 1–9 days), and 58% of events occurred after hospital discharge. The vast majority of patients presented with chest pain and ST elevations; a minority of them presented with non-diagnostic ECGs and chest pain, congestive heart failure symptoms, or cardiac enzyme elevations. All patients underwent diagnostic angiography, and all but two were treated with repeat PCI. New stents were implanted in almost half of the repeat PCIs (typically to address dissection or residual stenosis at the edges of previously stented segments), and rheolytic thrombectomy with the Angiojet catheter (Possis, Minneapolis, Minnesota) was performed adjunctively in about one-third. Platelet GP IIb/IIIa inhibitors were used in most cases. About three-fourths of the SAT episodes were associated with substantial (> 3x normal) CK-MB elevations, despite prompt revascularization. Three patients (12%) required intra-aortic balloon pump (IABP) support for cardiogenic shock. Two patients went on to coronary artery bypass graft surgery prior to hospital discharge due to persistent ischemia and incomplete revascularization after successful treatment of SAT. There was 1 death in a patient with previous bypass surgery, whose SAT occurred in the hospital one day after initial stenting; this was associated with precipitous hemodynamic collapse. Resource utilization and costs. Hospital-based resource utilization and costs are summarized in Table 5. The mean length of stay after SAT was 4.4 days (median: 3 days), with a mean intensive care unit length of stay of 1.8 days (median: 1 day). The median total cost per patient was $11,134 (interquartile range $7,167–$18,629), with more than half of the costs coming from the catheterization laboratory and pharmacy. Mean total costs ($17,134) were roughly $6,000 higher than the median, reflecting the influence of outliers with long, complicated hospital courses. As one would expect, resource utilization and costs tended to be higher when SAT occurred on an outpatient versus inpatient basis. Compared with inpatients, outpatients tended to have longer median LOS (4 vs. 2 days; p = NS) and higher median total costs ($13,258 vs. $9,968; p = NS), although median costs generated in the catheterization laboratory were similar ($4,907 vs. $5,089; p = NS). Discussion In this contemporary retrospective cohort, we observed that stent thrombosis occurred in ~1% of patients — primarily in high-risk circumstances. Clinically detected subacute stent thrombosis usually resulted in major myocardial infarction requiring repeat revascularization procedures. Median total hospital costs per patient were in excess of $11,000. Risk factors predicting the development of SAT have previously been reported, and include technical factors such as longer stent length, smaller final minimum luminal diameter (MLD), and persistent dissection,2,7,10,11 as well as patient-related factors such as acute coronary syndrome presentation, elevated platelet count, and pre-procedural thrombus.3,4,7 In clinical practice, additional technical factors (e.g., incomplete stent expansion7) and patient factors (hypercoagulable state, non-compliance with antithrombotic regimen) may be important in individual cases. Our clinical results are largely consistent with previously published data. As one would expect, based on the known risk factors for SAT, the majority of our patients received multiple stents, had long total stent lengths, small vessel diameters, and high rates of acute presentations and acute procedural complications. We noted a higher proportion of women and diabetics in our cohort than reported in other series.2 Whether this was due to these risk factors, themselves, or to the known propensity of these groups to small vessel diameters12 and longer lesions is unknown. The median time to SAT (3.5 days) in our cohort was longer than that reported in a recent analysis of patients treated in coronary stent trials (1 day)2, and more similar to the early stenting era when more aggressive post-procedure anticoagulation protocols were followed.11,13,14 This finding may be due to the more frequent use of GP IIb/IIIa inhibitors in our series, which may delay rather than prevent SAT in some cases.4,15 The longer observed time to SAT in our series combined with the routinely short post-procedure LOS following PCI resulted in the majority (58%) of our SAT episodes taking place in the outpatient setting, again in contrast to the largest recent series.2 This is a potentially important observation, since outpatients experiencing SAT encounter a longer delay to reperfusion therapy and based on our analysis may generate higher health care costs. Based on our cost data, the overall economic impact of SAT on the U.S. healthcare system can be estimated. Approximately 900,000 percutaneous coronary interventions were performed in the U.S. last year, and data from a large national registry indicate that 73% of all interventions currently employ the use of one or more stents.9 Using a conservative SAT rate of 0.9%2 and our median total cost of $11,100 per episode, direct medical care costs related to treatment of SAT are approximately $65 million per year — an average of approximately $100 per stented patient. The total economic burden of SAT would undoubtedly be higher if outpatient and indirect costs resulting from SAT were added. The role of additional measures beyond optimal stenting techniques and standard post-procedure prophylaxis in preventing SAT is unclear. Platelet GP IIb/IIIa antagonists have been shown to reduce post-procedural MI and urgent target vessel revascularization rates following stent procedures,16,17 but it is unclear how much of this benefit is derived from prevention of SAT. In the recently published Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) trial of PCI for acute MI patients,18 the use of abciximab was associated with a decrease in the rate of SAT from 1.4% to 0.4% but this included PTCA as well as stent patients, and there was no difference in the rate of target vessel occlusion at 7-month angiographic follow-up. Heparin coating has recently been introduced as another strategy for reducing the thrombogenicity of stents, as demonstrated in animal models.19,20 The rates of stent thrombosis for heparin-coated stents have been impressively low in large trials involving both acute-MI (0.9%)22 and stable/unstable angina patients (0.2%);21 however, the superiority of heparin-coated versus conventional stents in preventing SAT has not been demonstrated to date in a human comparative trial. Thus, it is not possible to project the extent to which a heparin-coated stent (or other antithrombotic coating) might reduce the costs associated with SAT. Several important limitations must be acknowledged regarding our study, particularly with respect to the economic analysis. Most importantly, our data were derived retrospectively from a single center. As a tertiary teaching center, our case mix may be more complex than at other centers; thus, our costs may be not be generalizable to all U.S. hospitals. In addition, our sample size was small. This limited our ability to make meaningful statistical comparisons within the group and subjected our cost analysis to potential distortion from outliers. For that reason, median values are reported as the best approximation of “typical” costs. Estimating length of stay and hospital costs for subjects in whom SAT occurred on an in-patient basis involved some subjectivity, and we cannot be certain that all costs captured were directly attributable to the occurrence of SAT. Despite these limitations, our data provide perspective on the current context and consequences of SAT in “real world” clinical practice, and provide a reasonable estimate of the medical care costs associated with these unfortunate events. The clinical and economic price paid when SAT takes place points to a continued need to enhance its prevention, particularly in high-risk clinical scenarios.
1. Leon MB, Baim DS, Popma JJ, et al. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. Stent Anticoagulation Restenosis Study Investigators. N Engl J Med 1998;339:1665‚Äì1671. 2. Cutlip DE, Baim DS, Ho KK, et al. Stent thrombosis in the modern era: A pooled analysis of multicenter coronary stent clinical trials. Circulation 2001;103:1967‚Äì1971. 3. Tolleson T, Harrington R, Newby LK, et al. Increased rates of stent thrombosis among patients with acute coronary syndromes: Insights from two large oral IIb/IIIa antagonist trials (Abstr). Eur Heart J 2001;22(Suppl):385. 4. Rinaldi MJ, Piana RN, Caputo RP, et al. Do GP IIb/IIIa inhibitors reduce stent thrombosis? Results of a multicenter study (Abstr). Circulation 2001;104. 5. Serruys PW, Unger F, Sousa JE, et al. Comparison of coronary-artery bypass surgery and stenting for the treatment of multivessel disease. N Engl J Med 2001;344:1117‚Äì1124. 6. Karillon GJ, Morice MC, Benveniste E, et al. Intracoronary stent implantation without ultrasound guidance and with replacement of conventional anticoagulation by antiplatelet therapy: 30-day clinical outcome for the French Multicenter Registry. Circulation 1996;94:1519‚Äì1527. 7. Moussa I, Di Mario C, Reimers B, et al. Subacute stent thrombosis in the era of intravascular ultrasound-guided coronary stenting without anticoagulation: Frequency, predictors and clinical outcome. J Am Coll Cardiol 1997;29:6‚Äì12. 8. Taira D, Seto T, Cohen D. Comparing analytic approaches for the economic evaluation of a clinical trial. The balloon vs. optimal atherectomy trial (Abstr). Med Decis Making 1998;18:486. 9. Anderson HV, Shaw RE, Brindis RG, et al. A contemporary overview of percutaneous coronary interventions. The American College of Cardiology-National Cardiovascular Registry (ACC-NCDR). J Am Coll Cardiol 2002;39:1096‚Äì1103. 10. Nath FC, Muller DW, Ellis SG, et al. Thrombosis of a flexible coil coronary stent: Frequency, predictors and clinical outcome. J Am Coll Cardiol 1993;21:622‚Äì627. 11. Foley JB, Brown RI, Penn IM. Thrombosis and restenosis after stenting in failed angioplasty: Comparison with elective stenting. Am Heart J 1994;128:12‚Äì20. 12. Murphy SA, Chen C, Cannon CP, et al. The impact of gender on angiographic and clinical outcomes after fibrinolytic therapy in acute MI (in press). 13. Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators. N Engl J Med 1994;331:496‚Äì501. 14. George BS, Voorhees WD, Roubin GS, et al. Multicenter investigation of coronary stenting to treat acute or threatened closure after percutaneous transluminal coronary angioplasty: Clinical and angiographic outcomes. J Am Coll Cardiol 1993;22:135‚Äì143. 15. Assali AR, Sdringola S, Ghani M, et al. Timing of coronary stent thrombosis in patients treated with prophylactic tirofiban. J Invas Cardiol 2000;12:460‚Äì463. 16. The EPISTENT Investigators. Randomised placebo-controlled and balloon-angioplasty-controlled trial to assess safety of coronary stenting with use of platelet glycoprotein IIb/IIIa blockade. Lancet 1998;352:8792. 17. The ESPRIT Investigators. Novel dosing regimen of eptifibatide in planned coronary stent implantation (ESPRIT): A randomised, placebo-controlled trial. Lancet 2000; 356:2037‚Äì2044. 18. Stone GW, Grines CL, Cox DA, et al. Comparison of Angioplasty with Stenting, with or without Abciximab, in Acute Myocardial Infarction. N Engl J Med 2002;346:957‚Äì966. 19. Hardhammar PA, van Beusekom HM, Emanuelsson HU, et al. Reduction in thrombotic events with heparin-coated Palmaz-Schatz stents in normal porcine coronary arteries. Circulation 1996;93:423‚Äì430. 20. De Scheerder I, Wang K, Wilczek K, et al. Experimental study of thrombogenicity and foreign body reaction induced by heparin-coated coronary stents. Circulation 1997;95:1549‚Äì1553. 21. Serruys PW, van Hout B, Bonnier H, et al. Randomised comparison of implantation of heparin-coated stents with balloon angioplasty in selected patients with coronary artery disease (Benestent II). Lancet 1998;352:673‚Äì681. 22. Grines CL, Cox DA, Stone GW, et al. Coronary angioplasty with or without stent implantation for acute myocardial infarction. N Engl J Med 1999;341:1949‚Äì1956.