Feasibility and Safety of Ad Hoc Percutaneous Coronary Intervention in the Modern Era

ABSTRACT: Background. The frequency of ad hoc percutaneous coronary intervention (PCI) varies among institutions and regions of the country. It is unclear what factors limit use of the ad hoc strategy. Objective. To define factors which limit the use of the ad hoc strategy. Methods. All patients who underwent PCI at our center in 2004 were reviewed. Patients who had emergent PCI for ST-elevation myocardial infarction (n = 188), those who had undergone diagnostic coronary angiography at a referring facility (n = 54), and those who had a repeat PCI after a previous ad hoc PCI (n = 19) were excluded. PCIs performed the same day as diagnostic angiography were considered “ad hoc”; all others were designated “staged”. Demographic and procedural factors through hospital discharge were prospectively recorded. Logistic regression analysis was performed to identify correlates of ad hoc PCI, PCI success, and PCI complications. Results. Of the 580 PCI procedures eligible for analysis, 557 (96%) were ad hoc and 23 (4%) were staged. Patients undergoing staged PCI had more lesions treated, a higher rate of no-reflow and periprocedural myocardial infarction, and higher contrast volumes and fluoroscopic times. Logistic regression analysis revealed that patients with history of heart failure, renal insufficiency and a recent myocardial infarction were more likely to undergo a staged PCI. Patients undergoing a staged PCI and those who had previous bypass surgery were more likely to have an unsuccessful PCI procedure. Conclusion. Most PCI procedures can be performed safely and effectively on the same day as diagnostic coronary angiography. J INVASIVE CARDIOL 2009;21:194–200 The safety and efficacy of percutaneous coronary intervention (PCI) at the same laboratory visit as diagnostic angiography, known as ad hoc PCI, remains unclear. Despite this, ad hoc PCI has come into widespread use.1 The ethical concerns and logistic feasibility involved with performing ad hoc PCI have been discussed.2,3 It has been considered to be safe and cost effective in selected groups of patients, though no randomized studies have been completed.4–16 The Society for Cardiac Angiography and Intervention Statement on Ad Hoc versus the Separate Performance of Diagnostic Cardiac Catheterization and Coronary Intervention states it “is reasonable for many, but not appropriate for all patients, and should not be considered standard therapy.”2,3 The frequency of ad hoc PCI remains quite variable among institutions, types of hospital and regions of the country.1 It is unclear what factors limit the use of the ad hoc strategy for all coronary intervention patients. The Society for Cardiac Angiography and Intervention published recommendations for the performance of ad hoc PCI in 20002 and published further recommendations in 2005.3 Prior to 2004, our medical center established policies for ad hoc PCI nearly identical to the 2005 Society for Cardiac Angiography and Intervention recommendations. We evaluated all PCIs done in 2004 at our institution to evaluate the safety and limitations of ad hoc PCI when performed according to Society for Cardiac Angiography and Intervention recommendations. Methods Our angioplasty center is a 437-bed tertiary care hospital in rural central Pennsylvania. All diagnostic coronary angiograms and PCI procedures at our institution were performed by 4 physicians employed by the hospital. These included 1 invasive, noninterventional physician and 3 interventionists. The 3 interventionists had an average of 15 years of experience post fellowship (range 2–26 years), and an average lifetime experience of over 2,500 interventions (range 600–4,000); all interventionists performed > 200 PCIs per year. All 3 interventionists used same-session ad hoc PCI as their default strategy. Cardiology fellows routinely assisted with diagnostic coronary angiograms and an interventional fellow usually assisted with interventions. All attending physicians and fellows underwent yearly radiation safety training courses. All patients who underwent PCI at our angioplasty center from January 1, 2004 to December 31, 2004 were considered for inclusion in this study. Patients were excluded from this analysis if their PCI was for ST-elevation myocardial infarction (STEMI) (n = 188), if the diagnostic coronary angiogram was performed at an outside institution (n = 54), or if they had a repeat PCI after a previous ad hoc PCI (n = 19) (i.e., the second stage of a two-stage PCI). This retrospective analysis was approved by our center’s institutional review board. Preprocedural sedation routinely included diazepam 5 mg and diphenhydramine 25 mg orally. After gaining femoral access, coronary angiography (8 or 9 views) with 4–6 Fr catheters was followed by left ventriculography. Nonionic contrast, typically 130 ml, was used. Patients with a reduced estimated creatinine clearance were routinely treated with n-acetylcysteine 600 mg orally twice daily on the day before and day of angiography, pre- and post-procedure hydration (orally for outpatients or intravenously with normal saline for in-patients), and iodixanol as contrast agent. PCI was typically performed with 6 Fr guide catheters, heparin or bivalirudin anticoagulation, and vascular closure with a suture or intra-arterial collagen plug device. Operators used their discretion regarding pre- and post-dilatation, glycoprotein (GP) IIb/IIIa inhibitor treatment and stent type. Drug-eluting stents (DES) were used in approximately 60% of patients. Clopidogrel was usually first administered as a 600 mg loading dose while the patient was on the catheterization table. After PCI, all patients spent at least 1 night in the hospital. All patients underwent informed consent discussions prior to catheterization.They were informed of the alternatives of medical therapy, PCI and coronary artery bypass graft (CABG) surgery, and the average risks of each strategy. Rarely, patients refused to consider same-session PCI, but we estimate that over 99% of patients gave consent for PCI at the same session, and most expected it. After diagnostic coronary angiography, with the patient on the table, the interventionist immediately reviewed digital images and consulted with the referring cardiologist by telephone, and often in person. Immediate review of diagnostic data was available through an electronic medical record. If the diagnostic coronary angiogram was performed by the invasive noninterventional cardiologist, the interventionist working in an adjacent catheterization room was consulted and the physician, not the patient, changed rooms to perform the PCI when it was indicated. Uncertainty about the physiologic significance of a lesion, if no stress imaging data were available, was typically further evaluated with intracoronary pressure-wire assessment of the lesion. If the referring cardiologist and the interventional cardiologist agreed that PCI was the best strategy, this recommendation was discussed with the patient. The patient was given the choice of proceeding immediately, or leaving the catheterization laboratory to further consider options for treatment or consult with family members. The patient was taken off the table after diagnostic angiography if: 1) PCI was not thought to be the best strategy; 2) there was disagreement or uncertainty about PCI being the best treatment strategy; 3) the patient wished to further consider options or consult with family members; 4) the risk of PCI was estimated to be substantially higher than had been discussed during the informed consent discussion preprocedure; or 5) excessive radiation or contrast was used during the diagnostic procedure or might be required by the anticipated PCI. In cases where off-table discussion was needed to make a decision regarding PCI, sheaths were left in place. If a decision for PCI was reached on the same day, the patient was returned to the catheterization laboratory for PCI. When decisions required more time, PCI was scheduled on a different day. During diagnostic angiography and PCI, real-time monitoring of contrast dose and fluoroscopy times was performed. There were no guidelines in place limiting the dose of contrast or radiation, as this was left to the operator’s discretion. In multivessel disease cases, interventionists routinely attempted to achieve complete revascularization in one session, staging the procedure only if necessary to limit contrast, radiation exposure or patient discomfort. Creatinine kinase and creatinine kinase-MB were checked every 8 hours for 24 hours or until discharge. Patients were discharged on aspirin and clopidogrel. Demographic, procedural, and outcome data through hospital discharge were prospectively entered into the American College of Cardiology’s National Cardiovascular Data Registry (ACC-NCDR). Patients undergoing staged PCI during the study period were identified by review of ACC-NCDR data and digital archives. Definitions for variables in Tables 1–5 are those used by the ACC-NCDR. Patients undergoing staged PCI were retrospectively assessed to identify factors leading to the staging of PCI. After all records were reviewed, categories were constructed to describe reasons for staging. These categories included the following: additional discussion and informed consent concerning diagnostic findings required before PCI; medical therapy attempted, but PCI necessary due to recurrent or intractable angina; coronary artery bypass surgery recommended by the cardiologist, but declined by 1 or more cardiovascular surgeons; complications during the diagnostic procedure that would have made PCI unsafe if undertaken on the same day; contrast or radiation exposure that would have been excessive if PCI were performed on the same day; and PCI performed only after additional testing provided a clear indication. PCIs performed during the same day as the diagnostic angiogram were considered “ad hoc.” Patients who underwent coronary angiography and PCI on different days for any reason were considered “staged.” Descriptive statistics including means (standard deviation) and percentages were calculated for the ad hoc PCI and staged PCI groups. Univariate tests, including two-sample t-tests (continuous data), Kruskall-Wallis tests (ordinal data), and chi-square or Fisher’s exact tests (categorical data) were used to compare patient characteristics between groups. Multiple logistic regression was used to identify independent predictors of staged PCI. Items considered for inclusion in the regression model included all demographic variables (age, gender, co-morbidities, medical history), presentation characteristics (chest pain, angina) and angiographic and procedural details (number of lesions, percent stenosis, number of stents, contrast volume and fluoroscopy time). Univariate models were examined for each covariate. Those with p-values Results From January 1, 2004 to December 31, 2004, a total of 841 PCIs were performed at our medical center. PCIs were excluded from this analysis if they were for acute STEMI (n = 188), if the original diagnostic angiography was performed at an outside institution (n = 54), or if the PCI was the second stage of a two-stage PCI (n = 19) (Figure 1). Of the remaining 580 PCIs, 557 (96%) were ad hoc and 23 (4%) were staged. Clinical characteristics of the ad hoc and staged groups are listed in Table 1. Staged PCI patients were older and more frequently had previous MI, prior CABG, history of CHF, history of renal failure and decreased left ventricular systolic function. There was no difference in the presenting syndrome for ad hoc versus staged PCI patients (Table 2). Multivessel PCI was more frequent in the staged group (48%) than in the ad hoc group (20%, p = 0.005) (Table 3). Contrast volumes and fluoroscopy times were similar for ad hoc PCI compared to only the staged PCI. When the diagnostic procedure and staged PCI were combined, however, their contrast volumes and fluoroscopy times were significantly greater than for ad hoc PCI procedures. The incidence of procedural angiographic complications was generally similar for ad hoc and staged PCIs (Table 4). Patients with staged PCI had a higher rate of classic no-reflow (8.7%) compared to the ad hoc patients (1.1%, p = 0.036). The overall incidence of post-PCI complications were similar for both groups (Table 5), however, periprocedural MI occurred more frequently in the staged group (8.7% compared to the ad hoc group (3.8%, p = 0.023). Among patients who underwent staged PCI, the reasons for staging included additional discussion and informed consent concerning diagnostic findings required before PCI (n = 8); CABG recommended after diagnostic angiography, but declined by a cardiac surgeon (n = 4); second stage done due to recurrent or intractable angina (n = 3); complications during the diagnostic procedure that would have made PCI unsafe if undertaken on the same day (n = 3); contrast/radiation that would have been excessive if done on the same day (n = 3); and second stage done only after additional testing provided a clear indication (n = 2). Logistic regression analysis of factors potentially associated with staged PCI identified the following: history of renal failure (odds ratio [OR] = 9.46, 95% confidence interval [CI] = [3.15, 28.44]; p = 0.0001), previous CABG (OR = 3.32, 95% CI = [1.34, 8.17]; p = 0.0093), and multiple lesions treated (OR = 5.41, 95% CI = [2.13, 13.72]; p = 0.0004). Logistic regression analysis of factors potentially associated with any clinical complication included the following: age (for each 10-year increase, OR = 1.38, 95% CI = [1.05, 1.82]; p = 0.022) and history of CHF (for an increase in 1 class, OR = 1.37, 95% CI = [1.02, 1.84]; p = 0.035). Staged PCI was not considered in the model since it was not significant in the univariate analysis (p = 0.13). Logistic regression analysis of factors potentially associated with procedural success included only age (for each 10-year increase, OR = 1.42, 95% CI = [1.08, 1.88]; p = 0.011). Discussion The most important finding of this study is that ad hoc PCI can be performed for the vast majority of patients in a healthcare delivery system that meets the Society for Cardiac Angiography and Intervention recommendations for ad hoc PCI. While ad hoc PCI is commonly performed in the United States, there are few reports published on this topic. Previous investigators have reported analyses in which ad hoc PCI constituted up to 86% of all PCI; Krone et al noted that in government-run hospitals, 90% of all PCI were ad hoc.1 The 96% incidence of ad hoc PCI is the highest of which we are aware. There are many nonmedical reasons why PCI at other hospitals might be staged. These include logistics, scheduling, inadequate preparation of patients (including medical therapy and failure to obtain informed consent) and unavailability of interventionists. Our institution is the first to report methodical implementation of the policies recommended by the Society for Cardiac Angiography and Intervention for ad hoc PCI, and elimination of all the above barriers to ad hoc PCI. The second important finding of this study involves the causes for staging PCI procedures. Of 23 staged procedures, one-half12 were staged because the operator felt that information obtained during the diagnostic angiography required re-consideration of alternatives to PCI. Of these, 8 underwent further consideration of the risks, which were substantially higher than those anticipated before diagnostic arteriography, prior to undergoing staged PCI. Four others were referred for CABG, but underwent PCI after one or more cardiac surgeons declined to perform bypass surgery because of excessive risk. Of the remaining 11 patients in whom PCI was staged, 4 were not planned, but became necessary due to recurrent or intractable angina despite medical therapy. Three were deferred due to concerns over excessive contrast or radiation exposure. (None of these had subsequent renal failure or radiation injury). Three PCI procedures were deferred due to complications related to the vascular access site that occurred during diagnostic angiography, including femoral artery hematomas and entry of the femoral artery above the inferior epigastric artery. (Both of these were thought to increase the risk of vascular access bleeding unacceptably if anticoagulants were given during the PCI). Finally, in 2 patients, intracoronary assessment of the lesion with a pressure wire was not feasible, and PCI was deferred until further testing could confirm the physiologic significance of the lesion. A strategy of ad hoc (rather than staged) PCI, as employed at our center, does not appear to compromise patient safety. Procedural success occurred with almost equal frequency in ad hoc and staged PCIs, as did intraprocedural complications. There was a trend toward more complications in staged patients, largely due to their two-fold higher incidence of periprocedural MI compared to ad hoc PCI patients. It is unlikely, however, that the staging strategy produced a greater number of complications. Review of the staged PCI patients indicates that interventionists identified a high-risk subgroup that was more likely to have complications due to their underlying disease. While the data are limited by the small number of patients who were in the staged PCI group, our findings confirm those of Krone et al, who reported that compared to staged PCI, ad hoc PCI was not associated with a lower success rate or more frequent complications.1 Other investigators have suggested that ad hoc PCI is associated with increased complications, particularly in high-risk subgroups.5,9,11,13 It is possible that our results, confirming the safety of ad hoc PCI, reflect the benefit of preparation for ad hoc PCI recommended by the Society for Cardiac Angiography and Intervention. Only one other investigator reported an increased frequency of complications associated with staged PCI compared to ad hoc PCI.12 Shubrooks et al reported more complications with staged PCI (1.5%) than ad hoc PCI (0.6%, p = 0.006), attributed primarily to more frequent vascular complications associated with need to reaccess the femoral artery.12 In the current analysis, there was no increase in vascular complications in patients undergoing staged PCI. Recent controversy over the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial has included the question of whether some PCIs are performed unnecessarily.18 Some have suggested that the “oculostenotic” reflex frequently leads to PCI when a lesion is discovered.19 Despite the high incidence of ad hoc PCI in this study, our findings should not be taken as supporting this notion for several reasons. First, it was appropriate preparation and lack of logistic obstacles that led to high rates of ad hoc PCI, not inappropriate utilization of the procedure. Every patient provided thorough informed consent before the procedure, and most underwent a second consent process on the table. While this second level of “informed consent” was potentially compromised by mild sedation and the compelling position of being on the catheterization laboratory table, most patients seemed able to understand the issues, ask intelligent questions and make considered decisions. Second, examination of the reasons why procedures were staged suggests that the interventionists were appropriately cautious and were willing to consider alternative strategies instead of proceeding automatically to PCI. The fact that some patients referred to CABG were rejected suggests that the bypass option was not neglected by interventionists. It is our experience that patients strongly prefer ad hoc PCI over staged PCI. During informed consent discussions, our patients frequently request that PCI, if necessary, be performed immediately after the diagnostic angiogram. Some patients decline angiography at nonangioplasty hospitals and travel to our center because of the availability of ad hoc PCI. When a staged PCI is recommended for the reasons discussed above, patients often express frustration and dissatisfaction at not receiving ad hoc PCI. Our experience is that patients are the most vigorous advocates of ad hoc PCI. Comparison of this study to other studies is limited by differences in definitions of “ad hoc” PCI (Table 7). Previous authors have defined “ad hoc” as “at the time of,”4–6 “same-day,”13 “combined,”8,10,14,15 “same-session,”7–11,14,16,17 or “immediately after diagnostic catheterization.”8,12,15,17 Previous authors have defined “staged” PCI as that occurring on a different day from the diagnostic procedure,5,6,9,13 as any PCI not done at the same sitting as the diagnostic procedure,8,10,14,15 or have not offered any definition at all.4,7,11,12,16,17 None of these has explicitly mentioned, or categorized, patients who undergo PCI on the same day, but at a different session than diagnostic catheterization. This study defined “ad hoc” PCIs as those occurring on the same day as the diagnostic catheterization. We propose that future studies categorize patients undergoing diagnostic catheterization and PCI on different days as “staged,” and those occurring on the same day as “ad hoc.” From the standpoint of exposure to radiation, contrast dye and vascular complications, patients with same-session PCI are very similar to those with different-session, same-day PCI. A comparison of our patients’ outcomes to those reported in the NCDR for non-STEMI in 2004 (Table 6) suggests that our strategy of routine ad hoc PCI did not increase adverse outcomes. Most safety endpoints were similar. Our rate of peri-procedural MI, while higher than reported in the NCDR for other hospitals (which are not required to routinely check creatinine kinase-MB enzymes), is similar to that reported in series where creatinine kinase is checked routinely. Study limitations. This is a small, single-center study in a rural tertiary care hospital with a closed catheterization laboratory staff, thus the results may not be generalizable to other regions or types of hospitals. This analysis is based on a retrospective post hoc analysis of data compiled from the ACC-NCDR database and from the review of charts. The data were, however, prospectively collected in the NCDR database prior to the conception of this study, which would reduce any bias regarding ad hoc versus staged PCI patients. Due to the small numbers of patients in this study, we cannot exclude the possibility that the ad hoc PCI strategy is associated with increased risk in some subgroups of patients. Finally, we did not account for patients in whom diagnostic angiography was performed without any subsequent PCI. It is possible that patients who were considered for staged PCI, but were treated medically or with CABG, might have constituted a lower-risk group that, if PCI had been undertaken, would have decreased the overall incidence of adverse outcomes in the staged group. Conclusion In a healthcare system that has implemented the most recent Society for Cardiac Angiography and Intervention recommendations, ad hoc PCI is feasible for most patients requiring PCI and does not increase the risk of PCI. Interventionists are capable of identifying a high-risk subgroup of patients in whom staged PCI is appropriate. It is the responsibility of interventionists to appropriately evaluate and prepare patients before diagnostic angiography, to provide ad hoc PCI where appropriate and recommend staging of PCI where necessary to ensure that patients’ autonomy and safety are preserved. Acknowledgments. The authors thank Pavlo Netrebko, MD, for translation of a Russian language manuscript.

1. Krone RJ, Shaw RE, Klein LW, et al. American College of Cardiology - National Cardiovascular Data Registry. Ad hoc percutaneous coronary interventions in patients with stable coronary artery disease — A study of prevalence, safety, and variation in use from the American College of Cardiology National Cardiovascular Data Registry (ACC-NCDR). Catheter Cardiovasc Interv 2006;68:696–703.

2. Blankenship JC, Mishkel GJ, Chambers CE, et al. Ad hoc coronary intervention. Catheter Cardiovasc Interv 2000;49:130–134.

3. Blankenship JC, Klein LW, Laskey WK, et al. Society for Cardiac Angiography and Intervention statement on ad hoc versus the separate performance of diagnostic cardiac catheterization and coronary intervention. Catheter Cardiovasc Interv 2004;63:444–451.

4. Feldman RL, MacDonald RG, Hill JA, et al. Coronary angioplasty at the time of initial cardiac catheterization. Cathet Cardiovasc Diagn 1986;12:219–222.

5. Myler RK, Stertzer SH, Clark DA, et al. Coronary angioplasty at the time of initial cardiac catheterization: “Ad hoc” angioplasty possibilities and challenges. Cathet Cardiovasc Diagn 1986;12:213–214.

6. Haraphongse M, Tymchak W, Rossall RE. Coronary angioplasty at the time of initial diagnostic coronary angiography in patients with unstable angina. Cathet Cardiovasc Diagn 1988;14:73–75.

7. Breisblatt WM, Ruffner RJ, Uretsky BF, Reddy PS. Same-day angioplasty and diagnostic catheterization: Safe and effective but riskier in unstable angina. Angiology 1991;42:607–613.

8. O’Keefe JH Jr, Gernon C, McCalllister BD, et al. Safety and cost effectiveness of combined coronary angiography and angioplasty. Am Heart J 1991;122:50–54.

9. Rozenman Y, Gilon D, Zelingher J, et al. One-stage coronary angiography and angioplasty. Am J Cardiol 1995;75:30–33.

10. Kimmel SE, Berlin JA, Hennessy S, et al. Risk of major complications from coronary angioplasty performed immediately after diagnostic coronary angiography: Results from the Registry of the Society for Cardiac Angiography and Interventions. J Am Coll Cardiol 1997;30:193–200.

11. Adele C, Vaitkus PL, Wells SK, Zehnacker JB. Cost advantages of an ad hoc angioplasty strategy. J Am Coll Cardiol 1998; 31:321–325.

12. Shubrooks SJ, Malenka DJ, Piper WD, et al. Safety and efficacy of percutaneous coronary interventions performed immediately after diagnostic catheterization in northern New England and comparison with similar procedures performed later. Am J Cardiol 2000;86:41–45.

13. Varadendra B, Panchamukhi VB, Flaker GS. Should interventional cardiac catheterization procedures take place at the time of diagnostic procedures? Clin Cardiol 2000;23:332–334.

14. Goldstein CL, Racz M, Hannan EL. Impact of cardiac catheterization — Percutaneous coronary intervention timing on in hospital mortality. Am Heart J 2002;144:561–567.

15. Le Feuvre C, Helft G, Beygui F, et al. Safety, efficacy, and cost advantages of combined coronary angiography and angioplasty. J Interven Cardiol 2003;16:195–199.

16. Vural A, Pershukov IV, Batyraliev TA, et al. Efficacy and safety of coronary stenting during diagnostic coronary angiography in patients with stable angina. Vestn Rentgenol Radiol 2005;6:27–31.

17. Feldman DN, Minutello RM, Gaade CL, Wong SC. Outcomes following immediate (ad hoc) versus staged percutaneous coronary interventions (Report from the 2000 to 2001 New York State Angioplasty Registry). Am J Cardiol 2007;99:446–449.

18. Diamond GA, Kaul S. COURAGE under fire. J Am Coll Cardiol 2007;50:1604–1609.

19. Topol EJ. Coronary angioplasty for acute myocardial infarction. Ann Intern Med 1988;109:970–980.