Author + information
- Received April 19, 1996
- Revision received October 8, 1996
- Accepted October 25, 1996
- Published online February 1, 1997.
- Mark Reisman, MDA,*,
- Verna Harms, PhDA,
- Patrick Whitlow, MD, FACCB,
- Ted Feldman, MD, FACCC,
- Richard Fortuna, MD, PhDD and
- Maurice Buchbinder, MD, FACCE
- ↵*Dr. Mark Reisman 1221 Madison Ave., Suite 1020, Seattle, Washington 98104.
Objectives. We compared an early registry of rotational atherectomy with a recent registry to examine the evolution of patient profiles, lesion characteristics and procedural outcomes for patients treated with rotational atherectomy.
Background. With increased experience, the selection of patients and lesions treated with a device matures. This study documents the changes in the application of rotational atherectomy.
Methods. The patient characteristics and procedural outcomes from two multicenter patient registries—Registry I: 2,953 procedures, 3,717 lesions from 1988 to 1993; and Registry II: 200 procedures, 268 lesions from 1994—were analyzed and compared.
Results. There was an increase in the average age of the patients (63 vs. 65 years, p < 0.02) and the proportion of patients with unstable angina (42.9% vs. 56.5%, p < 0.01) or previous coronary artery bypass graft surgery (18.8% vs. 24.5%, p < 0.05) in Registry II. Registry II included fewer left anterior descending coronary lesions (46.5% vs. 32.8%, p < 0.01), more type B and C lesions (83.1% vs. 91.8%, p < 0.01), more eccentric lesions (69.0% vs. 79.5%, p < 0.01) and more calcified lesions (50.3% vs. 69.4%, p < 0.01). Complications, including urgent bypass surgery, Q and non–Q wave myocardial infarction, dissection, acute occlusion and perforation, were similar in the two groups. However, mortality increased from 1.0% to 3.0% (p < 0.05) in Registry II.
Conclusions. Comparison of recent and early patients treated with rotational atherectomy revealed an increase in the complexity of patients and lesions. Although the rate of death was increased, the overall rate of major complications was not significantly changed (4.7% vs. 6.0%, p = NS).
(J Am Coll Cardiol 1997;29:353–7)
As a procedure or device progresses from an investigational device to a device routinely used by trained operators, the indications for use, procedure results and incidence of complications may evolve as well. Registries of percutaneous balloon angioplasty over time have revealed significant changes in patient selection, procedural success and complications ([1, 2]). Use in practice will reflect both published information and individual operator experience. Since 1986 several devices for percutaneous revascularization have made the transition from investigational to approved devices.
Rotational atherectomy, first introduced in 1988 and approved for general use in 1993, was evaluated in a multicenter registry at 22 sites whose independent operators collected information on patient selection and procedural success during that period. In 1994, a second registry at eight sites collected similar information. The aim of this study was to assess the differences, if any, in patient profile, clinical success and incidence of complications after treatment with rotational atherectomy in these two groups.
From 1988 to 1993, data from 2,953 procedures (3,717 lesions) treated with rotational atherectomy were collected in a multicenter registry at 22 sites. The original 743 procedures with 874 lesions were analyzed and were the subject of a previous report (). We compared the data from the entire multicenter registry of 2,953 procedures (Registry I) to a new registry collected in 1994 (Registry II), which includes 200 procedures in 268 lesions. Registry II was collected to evaluate minor changes in the device, specifically the removal of the Teflon coating on the distal end of the drive shaft. In this latest series, data were collected from eight sites; seven of the eight sites had participated in the original registry. These seven sites contributed 71% (2,100 of 2,953) of the cases in Registry I and 78% (156 of 200) of the cases in Registry II.
Inclusion and exclusion criteria for the two registries were similar and have been presented previously (). In summary, patients with a coronary lesion >70% in a native coronary artery, lesion length <25 mm and without recent acute myocardial infarction or angiographic evidence of thrombus were eligible. All patients gave written informed consent approved by their institution.
For Registry I, the technique of rotational atherectomy was described previously (). Premedication included aspirin and a calcium channel blocking agent. Heparin was given to maintain an activated clotting time >300 s, and nitroglycerin was given by topical, intravenous or intracoronary routes at the discretion of the operator. All patients had venous access with a 6 to 8F femoral conduit and arterial access with an 8 to 10F sheath. Commercially available 8 to 10F guide catheters were used. The lesion was crossed with a 0.009-in. (0.023-cm) guide wire (type A or C, Boston Scientific Northwest Technology Center, formerly Heart Technology), and the Rotablator was advanced to the site and activated at 180,000 to 200,000 rpm. After high speed rotational atherectomy with one or more burrs, adjunctive balloon angioplasty was performed at the operator’s discretion.
After the procedure, patients were heparinized for 16 to 24 h and continued to receive aspirin and calcium channel blockers. Patients were discharged at the discretion of the physician, with no specific medical regimen dictated by the protocol.
Procedural success was defined as residual stenosis <50% and the absence of complications, including death, emergency coronary artery bypass graft surgery (CABG) and Q wave myocardial infarction (MI). The incidence of non–Q wave MI was examined but was not included in procedural success. The exact definition of non–Q wave MI evolved with time and was redefined in a stringent manner as an increase in creatine kinase (CK) greater than two times normal with a positive myocardial band. However, in about half of the cases it was not possible to retrospectively analyze for the new non–Q wave definition because of the lack of CK data.
Chi-square analysis using the Yates correction or, in comparisons where fewer than five cases were expected, the Fisher exact method was used to determine significance (p < 0.05) for binary and categoric variables. The continuous variables were compared using an unpaired ttest. The denominators for the demographics and lesion characteristics vary owing to incomplete reporting. These differences were not thought to be systematic. Statistical significance was accepted at p ≤ 0.05.
2.1 Patient characteristics.
Several significant changes occurred in the patient group over time (Table 1). Registry II included older patients (65 vs. 63 years, p < 0.02), a higher proportion of patients with previous CABG (24.5% vs. 18.8%, p < 0.01) and more patients with unstable angina (56.5% vs. 42.9%, p < 0.01).
2.2 Lesion characteristics.
The differences in lesion site treated and lesion characteristics are shown in Table 2. There was a decrease in the number of left anterior descending coronary artery (LAD) lesions treated, from 46.5% in Registry I to 32.8% in Registry II (p < 0.01). There was a concomitant increase in right coronary artery (RCA) and left circumflex coronary artery (LCx) lesions. Treatment of lesions in the left main coronary artery trunk remained unchanged.
Similarly, the number of lesions treated per procedure increased from 1.26 in Registry I to 1.34 in Registry II (p < 0.025). In addition, lesion characteristics associated with rotational atherectomy became more complex; American Heart Association/American College of Cardiology (AHA/ACC) type A lesions accounted for 17% of lesions in Registry I and 8.2% of lesions in Registry II. In Registry II, AHA/ACC type B and C lesions accounted for >90% of the lesions. The proportion of lesions that were eccentric (69% in Registry I vs. 79.5% in Registry II, p < 0.01) or calcified (50.3% in Registry I vs. 69.4% in Registry II, p < 0.01) also increased. The proportion of bifurcational lesions, ostial lesions, lesions >10 mm in length and restenotic lesions remained unchanged.
2.3 Major complications.
The incidence of major complications after rotational atherectomy is shown in Table 3. The incidence of death increased from 1% to 3% (p < 0.025). When comparing the two registries, no other complication occurred with increased frequency. It is important to note that the incidence of non–Q wave MI is not directly comparable because the definition of this complication changed with time and data were incomplete.
Procedural success, defined as <50% residual stenosis in the absence of death, Q wave MI or emergency CABG, was compared for lesion type and lesion location. Despite the increase in lesion complexity, there was no significant change in procedural success in the more recent experience compared with the earlier data, as shown in Table 4. In Registry I, type A and B lesions were treated with greater success than type C lesions. Lesions located in the LCx were treated with greater success than RCA or LAD lesions, but these differences were small and may not be clinically relevant. In Registry II, which had fewer cases, there were no detectable differences in procedural success by lesion location or type.
2.4 Angiographic complications.
The incidences of four angiographic complications are shown in Table 5. None are significantly changed from Registry I to Registry II.
Extrapolating from Registry II, rotational atherectomy is being performed in more complex lesion subsets, with a greater number of lesions being treated per procedure in older patients with more unstable angina and previous CABG. This trend is more apparent when compared with the initial 743 procedures previously published (). This early report included 27% type A lesions (vs. <10%), 1.18 lesions/procedure (vs. 1.34) and 40% calcified lesions (vs. 69%).
3.1 Lesion complexity.
The increase in lesion complexity observed in this study is based on operator report rather than on evaluation by a core angiographic laboratory. A large percentage of cases originated in centers participating in both registries, promoting consistency of observer interrogation between the two registries. In addition, an independent core laboratory documented lesion characteristics from a random sample of 500 patients treated between 1988 and April 1992 (), which were consistent with the interpretation of the operators.
In a substudy, when the cases from the seven centers participating in both registries were compared, several observations were apparent. First, these high volume centers had a higher percentage of complex patients in the initial registry (Table 6), and, second, patient complexity at these sites increased further from Registry I to Registry II. An unexpected observation was that the proportion of patients experiencing a CK elevation >300 U/liter was greater in the centers with lower enrollment and fewer complex patients. Whether this difference in CK elevation represents a learning curve or is a result of the lesion, patient and operator characteristics is uncertain.
3.2 Procedural success.
Even though these older patients had more unstable angina, more calcified lesions and fewer AHA/ACC type A lesions, there was no significant change in procedural success. There was, however, an increase in mortality, which remains puzzling and is certainly a stimulus for reflection. Although the cause of this increase is unclear, it may reflect increased lesion and patient complexity. For example, a subset analysis of rotational atherectomy in calcified lesions () demonstrated a slightly higher death rate (1.2% vs. 0.5% in calcified vs. noncalcified lesions, respectively, p < 0.04). In that analysis, calcified lesions included a higher proportion of other angiographic risk factors, with 57% having two or more risk factors, compared with 46.5% for noncalcified lesions (p < 0.001). Although the mortality rate in the current registry is higher than that seen in routine angioplasty data sets ([6, 7]), it is similar to the 2.7% mortality rate reported for “high risk” angioplasty (). Similarly, in a report of angioplasty in patients with a reduced ejection fraction (≤40%), the death rate climbed to 2.6% ().
The description of the patient group currently treated with rotational atherectomy reflects characteristics similar to those observed by Stertzer et al. ()—with older patients, a relatively high incidence of diabetes (31% vs. 15% with percutaneous transluminal coronary angioplasty [PTCA]), previous CABG (31% vs. 19% with PTCA) and restenotic lesions (33% vs. 13% with PTCA). The percentage of calcified (43.1% vs. 69.4%) and eccentric (55.6% vs. 79.5%) lesions appears to be higher for patients in the 1994 registry than that for patients treated by rotational atherectomy reported by Stertzer et al. (), whereas overall lesion complexity appears to be similar with type A lesions, accounting for 10% of the lesions treated. The distribution of lesions within the coronary tree was also similar to that observed by Stertzer et al. ().
3.3 Previous studies of rotational atherectomy.
Several reports of single-center experiences with rotational atherectomy have been published ([11–16]). For the most part, these data were collected before May 1993, when the Rotablator was not approved for commercial distribution in the United States, and thus represent analyses of data collected into the original multicenter registry. Several recent reports include a substantial number of procedures that are not part of the original multicenter registry ([17–19]). In a single-center, randomized trial of rotational atherectomy versus excimer laser versus PTCA in type B2 and C lesions (), procedural success was 93% with rotational atherectomy compared with 83% with PTCA, and major cardiac events, including death, emergency CABG and Q wave MI, were lower with rotational atherectomy (1.5%) than with PTCA (7.0%). Similarly, according to single-center data on 159 patients with type B2 and C lesions, the major cardiac event rate was 2.5%, with another 5.7% suffering from non–Q wave MI (). This study excluded 29.4% of patients treated with rotational atherectomy at this institution who had type A or B1 lesions. Finally, a report on 710 procedures () from late 1990 to 1994 includes 2 years of data after the original multicenter registry. In this report the rate of major cardiac events was 3.7%, with type A and B1 lesions representing 25% and type C lesions 46% of the lesions treated. The rate of major cardiac events in Registry II (6.0%) is higher than that reported in these studies, even though they also included more severe lesions.
3.4 Technique changes.
Registry II incorporated a few of the recent advances in rotational atherectomy technique (), including attempting to monitor and control the deceleration of the atherectomy burr to <5,000 rpm from its platform speed. The role of the guide wire in directing the cutting vector of the burr and the impact of guide wire bias on complications () had not been identified. In addition, although some centers have begun to use a perfusion cocktail to decrease spasm and slow flow ([19, 22]), it was not in routine use for this registry. The importance of heat generation () and platelet activation due to device deceleration () had not been delineated. The role of technique modifications in procedural success and restenosis will await the outcome of the randomized trials in progress.
3.5 Study limitations.
There are several limitations to the present study. The data from the more recent registry represent data collected after minor changes in Rotablator design. Although such changes should not affect patient selection, they may impact outcomes. Second, the occurrence of some variables, specifically complications, are infrequent enough that a sample of 200 is not large enough to ensure accurate and reliable sampling. Third, the lesion characteristics are based on operator report rather than on core laboratory analysis, as discussed earlier. Finally, centers participating in these registries are institutions with high-volume operators performing numerous procedures, and thus their outcomes may not reflect a more generalized practice.
Rotational atherectomy is being applied to complex lesions, with a significant increase in calcified and eccentric morphologies. The patients treated with rotational atherectomy represent an older group with a high incidence of diabetes, multivessel disease and unstable angina. The increasing lesion and patient complexity, without a decrease in procedural success, suggests that rotational atherectomy may allow percutaneous treatment of patients who were previously considered poor candidates for intervention. The increased mortality in the recent registry is a concern and may reflect the increased procedural complexity or relatively rare quantum events in a small sample size, especially in light of the low complication rates of other studies with complex lesions ([18, 19]). Ongoing randomized studies should clarify these issues, which include important technique modifications.
The technique of rotational atherectomy has evolved from a procedure performed under strict protocol to a widely used procedure in mainstream percutaneous interventions. Since its introduction in 1988, it has established itself as a technique with high procedural success, despite increasingly complex patient subsets with more difficult anatomy, particularly calcified lesions, for whom it has become the technique of choice.
Investigational Centers and Principal Investigators
Registry I: A complete list of the participants in Registry I appears in reference .
Registry II: Maurcie Buchbinder, MD,1Scripps Memorial Hospital, San Diego, California;Ted Feldman, MD, University of Chicago Hospital, Chicago, Illinois;Martin Leon, MD,1Washington Hospital Center, Washington, D.C.;Robert Kipperman, MD,1Galichia Medical Group, Wichita, Kansas;Simon Stertzer, MD,1Stanford University, Palo Alto, California;Gerald Dorros, MD,1University of Wisconsin, Milwaukee, Wisconsin;Patrick Whitlow, MD,1Cleveland Clinic, Cleveland, Ohio;David Warth, MD,1Providence Medical Center, Seattle, Washington.
☆ Drs. Reisman and Buchbinder have a financial interest in Boston Scientific Northwest Technology Center, formerly Heart Technology, Redmond, Washington, which manufactures the Rotablator.
↵1 Participated in both registeries.
- American Heart Association/American College of Cardiology
- coronary artery bypass graft surgery
- creatine kinase
- left anterior descending coronary artery
- left circumflex coronary artery
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- right coronary artery
- Received April 19, 1996.
- Revision received October 8, 1996.
- Accepted October 25, 1996.
- The American College of Cardiology
- Holmes DR,
- Holubkov R,
- Vliestra RE,
- Kelsey SF,
- Dorros G,
- Cowley MJ
- Myler RK,
- Shaw RE,
- Stertzer SH,
- et al.
- Warth DC,
- Leon MB,
- O’Neill W,
- et al.
- Ellis SG,
- Popma JJ,
- Buchbinder M,
- et al.
- MacIsaac AI,
- Bass TA,
- Buchbinder M,
- et al.
- Ellis SG,
- Roubin GS,
- King SB III.,
- et al.
- Safian RD,
- Niazi KA,
- Strzelecki M,
- et al.
- Stertzer SH,
- Rosenblum J,
- Shaw RE,
- et al.
- Vandormael M,
- Reifart N,
- Preusler W,
- et al.
- O’Murchu B,
- Foreman RD,
- Shaw RE,
- Brown DL,
- Peterson KL,
- Buchbinder M
- Reisman M
- Reisman M,
- Harms V
- Cohen BM,
- Weber VJ,
- Blum RR,
- et al.
- Reisman M,
- DeVore LJ,
- Ferguson M,
- et al.
- Eccleston DS,
- Horrigan MC,
- Cowley MJ,
- et al.