Clinical and angiographic outcome after directional coronary atherectomy

Clinical and Angiographic Outcome After Directional Coronary Atherectomy A QUAlJTATlVE AND QUANTITATIVE ANALYSIS USING CORONARY ARTERIOGRAPHY AND INTRAVASCULAR ULTRASOUND Jeffrey J. Popma, MD,, Gary S. Mintz, MD, Lowell F. Satler, MD, August0 D. Pichard, Kenneth M. Kent, MD, PhD, Ya Chien Chuang, PhD, Fadi Matar, MD, Theresa A. Bucher, RN, Alan J. Merritt, BS, and Martin B. Leon, MD

To assess clinical and angiographic outcome after directional coronary atherectomy, the clinical course of 306 patients undergoing this procedure was reviewed. Directional atherectomy was successful in 290 (94.8%) procedures; compliiions developed in 8 (2.6%) patients. After atherectomy, percent diameter stenosis was reduced from 71 f 14 to 14 -c 14% (p ~0.001) and minimal lumen diameter was increased from 0.87 + 0.42 to 2.55 +: 0.57 mm (p ~0.001). In 128 (42%) patients, adjunct balloon an@oplasty was performed to treat either complications or a residual stenosis MO%. Intravascular ultrasound was also performed in 57 patients after directional atherectomy and demonstrated that a significant amount of residual plaque mass remained in lesions with a calcium arc 190” (17 f 5 mm* vs l2 -c 5 mm2 in lesions without calcium; p = 0.007). Durhrg the 11 + 8 month follow-up period, 69 (28.3%) patients developed recurrent clinical events (death, 5; Q wave myocardial infarction, 8; coronary bypass surgery, 31; coronary angioplasty, 36). Using a proportional haxards model, independent predictors of late clinical events included diabetes mellitus (relative risk [RR] = 1.95; p <0.05), unstable angina (RR = 2.78; p ~0.005) and a prior history of restenosis (RR = 2.21; p ~0.01). We conclude that directional atherectomy is associated with high procedural success rates and infrequent complications in selected lesion subsets, although the From the Department of Internal Medicine (Cardiology Division), Washington Hospital Center, Washington, DC. Supported by a research grant from the Washington Cardiology Center Research Fund. Address for reprints: Jeffrey J. Popma, MD, Angiographic Core Laboratory, Washington Hospital Center, 110 Irving Street, Suite 4B-14, Washington, DC 20010.

MD,

degree of plaque resection may be limited if extensive calcium is present. Late clinical events develop in some (28%) patients after directional atherectomy, related to certain preprocedural clinical risk factors. (Am J Cardiil1993;72:55E-645)

N

ew angioplasty devices have been targeted for lesion subsets demonstrated to have suboptimal results after standard balloon angioplasty. lj2 One of these new devices, directional coronary atherectomy, exerts at least part of its beneficial effect by removing atherosclerotic plaque.3 Because directed plaque excision is possible, this device has been successfully used in eccentric lesions, in those lesions with intimal flaps, thrombus, or ulceration, and in concentric lesions in larger vessels.4,5 In reported nonrandomized series, directional atherectomy results in smoother lumen contours and larger residual dimensions than obtained using standard balloon methods.6,7 Moreover, by removing fibroatheroma, directional coronary atherectomy may alter plaque compliance and, if balloon angioplasty is needed after directional atherectomy, reduce the degree of elastic recoil.8,9 Several clinical and angiographic predictors of procedural success and complications after directional atherectomy have been reported.4,5J0 The identification of these subsets has allowed more appropriate triage of patients for directional coronary atherectomy, particularly when other device options exist. Moreover, despite the occurrence of restenosis following 32-58% of directional atherectomy procedures,3$9J-13recent quantitative data suggest that lower restenosis rates may be possible if the residual diameter stenosis is minimized after A SYMPOSIUM:

DIRECTIONAL

ATHERECTOMY

!5!5E

the procedure. 9~14The comparative advantage of directional atherectomy over balloon angioplasty in less complex lesion subsets awaits the result of a randomized clinical trial.15 Over time, a lesion-specific approach to directional atherectomy has evolved, emphasizing appropriate lesion selection and attainment of the minimal residual stenosis safely achievable using larger devices and adjunct balloon angioplasty, if needed. Notably, the early and late outcome using this strategic approach in centers with access to multiple devices has not been reported. To address this issue, the clinical course of 306 patients treated with directional atherectomy was reviewed. METHODS Patient selection: Over an l&month period, 306 patients underwent directional coronary atherectomy for the treatment of symptomatic or exercise-induced coronary artery disease, representing 17% of the 1,815 patients undergoing new device angioplasty during this period. Patients were selected for directional coronary atherectomy based on vessel size ( 2 2.5 mm reference segment diameter), preprocedural lesion morphology (i.e., lesion eccentricity, ulceration, irregularity, thrombus, or ostial location), and the absence of clinical contraindications (acute myocardial infarction, severe peripheral vascular disease, uncompensated congestive heart failure). Other lesions, less amenable to directional atherectomy, were treated with alternative new devices. Lesions in smaller vessels or those with severe calcification, angulation, or proximal vessel tortuosity were generally referred for rotational atherectomy. Lesions in degenerated saphenous vein grafts were treated with extraction atherectomy. Restenotic lesions and discrete lesions in saphenous vein grafts were generally treated with intracoronary stents. Total occlusions and diffuse ( 2 20 mm) lesions were treated with excimer laser angioplasty. Prior to marketing approval of the directional coronary atherectomy device by the Food and Drug Administration (FDA) in September 1990, all patients gave written informed consent under the guidelines of the Institutional Review Board. Hospital chart review was performed to obtain pertinent patient demographics, such as patient age, gender, and a history of prior coronary angioplasty. Unstable angina was defined as a recent acceleration of angina, including rest pain. A recent myocardial infarction was noted if there was clinical, enzymatic, or electrocardiographic evidence of myocardial necrosis within 6 weeks prior 56E

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72

to atherectomy. The use of adjunct balloon angioplasty, device size, and balloon inflation pressures were recorded from a cardiac catheterization worksheet. Procedural success was defined as <50% final diameter stenosis and the absence of death, Q wave myocardial infarction, repeat coronary angioplasty, or coronary bypass surgery within 14 days of the procedure. The absence of tissue retrieval after directional atherectomy was not a criterion for procedural failure. Minor complications included recurrent ischemia (typical chest pain associated with electrocardiographic changes consistent with transmural ischemia), abrupt vessel closure (Thrombolysis in Myocardial Infarction [TIMI] flow grade 0 or 1)16 and renarrowing (TIM1 flow grade 2 or 3 with a residual stenosis >50%). In addition, elevations of total creatine kinase and its MB fraction were recorded. Directional atherectomy procedure: Unless contraindicated, all patients were treated with aspirin prior to the procedure. After an intravenous bolus of heparin, 15,000 U, directional coronary atherectomy was performed using methods described in detail elsewhere.3-5 Additional heparin boluses were administered during the procedure to maintain the activated clotting time > 300 sec. In the event of a residual stenosis >30%, localized dissection, or intimal flap after 2 1 passes of the atherectomy device, a larger device was used to treat the residual stenosis. If a larger device was not available or could not be used due to oversizing relative to the reference vessel, standard balloon angioplasty was performed. Using this strategic approach, the minimal residual stenosis believed to be safely achievable by the investigator was obtained. Qualitative angiographic analysis Cineangiograms were reviewed at the Angiographic Core Laboratory by an observer blinded to the clinical outcome. In patients undergoing multilesion atherectomy, a single “culprit” lesion was identified for analysis. Standard morphologic criteria were used to assess lesion eccentricity, length (“shoulder to shoulder”), angulation 2 45”, and thrombus.r7 Ostial lesions were those within 3 mm of an epicardial coronary artery. Lesion irregularity included those lesions with ulceration, aneurysmal dilation, or a “sawtooth” pattern. Moderate/severe lesion calcification was defined as mobile luminal opacities demonstrated with cardiac motion. Bifurcation lesions included those lesions with a side branch ~1.0 mm in diameter bracketed by the device during directional atherectomy. Postprocedural

OCTOBER 18, 1993

morphologic analysis included the identification of imaging sheath and measuring the amount of lesion irregularity (“sawtooth” pattern opacifica- plaque circumference (in degrees) occupied by tion of the arterial contour without discrete dissec- each arc of calcium. Notably, lesion calcification tion), intimal flap (a discrete filling defect in may cause shadowing of deeper tissue planes, continuity with the arterial wall), thrombus, and which interferes with measurements of external coronary dissection, with or without contrast stain- elastic membrane area. When this occurred, 2 ing. Side branch loss and coronary perforation types of extrapolation permitted completion of the after atherectomy were also noted. measurements. Provided that the vessels were Quantbtiie angiographic analysis: Selected tubular, circumferential extrapolation was possible end-diastolic cineframes before and after direc- if calcific deposits were small. Otherwise, real-time tional atherectomy and after adjunct balloon angio- longitudinal movement of the transducer just proxiplasty, if needed, were digitized using a tine video mal or distal to the calcification unmasked contiguconverter. Using the guiding catheter as the calibra- ous portions of the artery to permit complete tion standard, normal and minimal lumen diam- measurement of the external elastic membrane eters were determined using an automated, com- area. For purposes of analysis, lesion calcium was puter-assisted edge detection algorithm.18 In classified as none, calcium arc < 90”, and calcium addition, the mean lumen diameter of the largest arc 290”. balloon used for adjunct dilation was recorded. In late clinical outcome: Patients were prospecthese lesions, two indices of plaque compliance tively followed at 3-month intervals after successful were recorded. Lesion stretch (%) was defined as coronary atherectomy. The occurrence of major the difference between the mean diameter of the clinical events (death, Q wave myocardial infarclargest balloon and the preprocedural minimal tion, and lesion-related bypass surgery or repeat lumen diameter, referenced to the normal vessel coronary angioplasty) was recorded. If revasculardiameter.19720Elastic recoil (%) was defined as the ization due to progression of disease or restenosis difference between the mean diameter of the of other lesions not related to the initial procedure largest balloon used and the postprocedural mini- occurred, the event was recorded but not attribmal lumen diameter, referenced to the normal uted to the atherectomy procedure. Event-free vessel diameter.19s20 probability was defined as freedom from death, Q Intravascular ultrasound: Intravascular ultra- wave myocardial infarction, or lesion-related revassound was performed after directional atherec- cularization within 1 year after directional atherectomy in 57 patients using a mechanical intracorotomy. nary imaging system (InterTherapy, Santa Ana, Statiiical analysis: Categorical data are preCA). This system incorporates a 25 MHz trans- sented as prevalence rates and continuous data are ducer-tipped catheter within a 3.9 F short monorail presented as mean + 1 standard deviation. Logisimaging sheath. The transducer is withdrawn at 0.5 tic regression was used to identify the predictors mm/set by a motorized pullback device and is for procedural failure immediately after atherecrotated at 1800 rpm to form planar images in tomy; odd ratios, 95% confidence intervals, and p real-time. Studies are recorded on 0.5-in S-VHS values are provided. An odds ratio = 1 suggests tape for offline analysis. Using computer-assisted that no relation exists between the predictor and planimetry of the 2-dimensional images, the narrow- the occurrence of an atherectomy failure. An odds est cross-sectional area within the length of the ratio > 1 indicates that the presence of the predicatherectomy site was compared to a proximal, tor proportionately increases the risk of procedural angiographically normal reference site.21 Plaque failure, whereas an odds ratio < 1 suggests that the morphology at the atherectomy site was analyzed predictor proportionately decreases the risk of using standardized criteria. Prior studies have dem- procedural failure. onstrated that calcification produces bright echoes The Kaplan-Meier method was used to estimate (brighter than the reference adventitia) with acous- event-free probability. Differences among eventtic shadowing; dense fibrous tissue also produces free survival curves were analyzed using Mantelbright echoes, as bright or brighter than the adven- Cox statistics. The stepwise proportional hazards titia, but without acoustic shadowing.21Soft plaque, model was used to identify those characteristics containing lipid, intimal hyperplasia, loose connec- that were independently associated with event-free tive tissue, or thrombus, is less dense than the survival; relative risk, 95% confidence intervals, reference adventitia. The degree of calcification and p values are provided. A relative risk of 1 was quantified by centering a protractor on the implies that no relation exists between the predicA SYMPOSIUM:

DIRECTIONAL

ATHERECTOMY

57E

TABLE I Clinical Predictors of Early and Late Outcome After Directional Coronary Atherectomy Procedural Outcome

Late Clinical Outcome

I Univariate Logistic Regression

No. (%I

Age (yr)t

Total (n = 306)

Failure Success (n = 16) (n = 290)

60a

60 k 10

11

602

11

Odds Ratio

95% Cl

Univariate Cox Regression

No. (%I Total (n = 244)

Event (n = 69)

No Event (n = 175)

Relative Risk

95% Cl

Multivariate Cox Regression* Relative Risk

95% Cl

1.00

0.96-1.05

61 2 11

61 +- 11

60 + 11

1.00

0.98-1.02

-

Female gender?

67 (22)

7 (47)

60 (21)

3.275

1.13-9.41

51 (21)

17 (25)

35 (20)

1.25

0.72-2.17

-

Diabetes mellitust

78 (26)

2 (14)

76 (27)

0.45

0.10-2.09

60 (25)

26 (38)

34 (20)

2.14ll

1.31-3.48

1.958

1.16-3.27

Unstable anginat

199 (68)

12 (86)

187 (68)

158 (68)

56 (85)

102 (62)

3.0011

1.53-5.88

2.78ll

1.41-5.47

2.891: 0.63-13.30

Recent Mlt

57 (19)

1 (7)

56 (20)

0.31

0.04-2.45

52 (22)

18 (27)

34 (20)

1.30

0.76-2.22

History of restenosist

66 (23)

4 (25)

62 (23)

1.14

0.35-3.68

48(21)

19(29)

29 (18)

2.005

1.17-3.42

*Multivariate Cox regression model included 221 patients with complete data. iMean, standard deviation, percentage, odds ratio, relative risk, 95% confidence intervals calculations ip <0.20.

-

2.2111

1.28-3.84

and statistical comparison did not include missing data.

1: 2x;, Cl = confidence intervals; MI = myocardial infarction.

tor and the occurrence of an event within 1 year after atherectomy. A relative risk > 1 indicates that the predictor proportionately increases the risk of a clinical event within 1 year after atherectomy, whereas a relative risk < 1 suggeststhat the predictor proportionately lowers the risk of a clinical event. Variables with a p <0.20 by univariate proportional hazards model were entered into the multivariate model. Backward elimination and maximal likelihood estimation were used for the multivariate model. All statistical analyses were performed using BMDP software (Berkeley, CA). A p < 0.05 was considered statistically significant.

Procedural failure occurred more often in women than in men (10.4% vs 3.4%; p <0.05), although no other clinical variable significantly predicted an early adverse outcome (Table I). Moreover, no preprocedural lesion morphologic factor significantly influenced procedural success (Table II). Angiographic results after directional atherectomy: After atherectomy, percent diameter stenosis was reduced from 71 + 14 to 14 2 14% (p < 0.001); minimal lumen diameter was increased from 0.87 + 0.42 to 2.55 + 0.57 mm (p 30%. In these patients, percent 290 (94.8%) of 306 procedures. Directional ather- diameter stenosis was reduced from 71 ? 14 to ectomy was unsuccessful due to complications in 8 31 -+ 15% after directional atherectomy (p < 0.001) (2.6%) patients (coronary bypass operation, 6; Q and to 14 ? 14% after adjunct balloon angioplasty wave myocardial infarction, 1; death, 2), and due to (p 50% in an addi- creased from 0.89 2 0.43 mm (p 1.5 times normal occurred in Figure 1. 17 (5.6%) patients; in 13 additional patients, the Angiographic complications were infrequent afcreatine kinase MB fraction rose to >5 times the ter directional atherectomy; they included the upper normal limit, although the total creatinine presence of postprocedural lumen irregularities kinase was not elevated. (n = 14), intimal flaps (n = 7), thrombus (n = 3), 58E

THE AMERICAN

JOURNAL

OF CARDIOLOGY

VOLUME

72

OCTOBER

18, 1993

TABLE II Preprocedural

Lesion Morphology Procedural Outcome Univariate Logistic Regression

No. (%) Total (n = 306)

Late Clinical Outcome

Failure (n = 16)

Success (n = 290)

49 (16)

3 (19)

46(16)

151 (49)

10 (63)

141 (49)

1.09

61 (20)

2 (13)

59 (20)

0.52

Odds Ratio

Univariate Cox Regression

No. (%) Total (n = 244)

Event (n = 69)

No Event (n = 175)

36 (15)

7 (10)

29 (17)

0.29-4.15

121 (50)

34 (49)

87 (50)

1.57

0.70-3.55

0.08-3.27

49 (20)

10 (15)

39 (22)

1.11

0.42X2.92

95% Cl

Relative Risk

95% Cl

Native artery location Right coronary LAD and left main Left circumflex

-

-

1 (6)

44 (15)

0.37

0.05-2.92

38 (16)

18 (26)

20 (11)

2.19*

1.28-3.76

Eccentricity

235 (77)

13 (81)

222 (77)

1.33

0.37-4.82

192 (79)

53 (77)

139 (79)

0.89

0.51-1.56

45 (15)

SVG

Length (mm)?

7.2 f 3.9

8.0 k 2.9

7.2 f 3.9

1.05

0.93-1.19

7.3 f 4.0

7.3 2 4.6

7.3 2 3.7

1.01

0.95-1.08

Ostial location

81 (27)

2 (13)

79 (27)

0.38

0.08-1.73

64 (26)

22 (32)

42 (24)

1.40$

0.84-2.33

Ulceration

32 (11)

2 (13)

30 (10)

1.24

0.27-5.75

27 (11)

6 (9)

21 (12)

0.76

0.33-1.76

20 (29)

10 (6)

1.24

0.63-2.42

Angulation 145”

39 (13)

4 (25)

35 (12)

2.43$

0.74-7.98

30 (12)

lrregulanty

53 (17)

3 (19)

50 (17)

1.11

0.30-4.05

40 (16)

11 (16)

29 (17)

0.94

0.49-I

Calcium?

60 (20)

4 (25)

56 (19)

1.38

0.43-4.46

49 (20)

10 (15)

39 (22)

0.66

0.34-1.30

.79

Bifurcation

79 (26)

6 (38)

73 (25)

1.78

0.62-5.10

63 (26)

16 (23)

47 (27)

0.93

0.52-1.58

Thrombus

9 (3)

1 (6)

8 (3)

-

8 (3)

2 (3)

6 (3)

*p
TABLE Ill

Procedural

Factors and Postprocedural

Lesion Morphology

Procedural Outcome

Late Clinical Outcome Univariate Logistic Regression

No. (%) Total (n = 306)

Failure (n = 16)

PTCA before DCA*

101 (33)

4 (25)

97 (34)

0.66

PTCA after DCA*

128 (42)

8 (53)

120 (42)

1.59

5F

14 (6)

0 (0)

14 (6)

-

6F

135 (53)

7 (78)

128 (52)

-

108 (53)

26 (46)

7F

107 (42)

2 (22)

105 (43)

-

83 (41)

28 k 9

28 2 8

28 + 9

1.00

0.91-1.10

Irregularity

14 (5)

2 (13)

12 (4)

3.31t

0.67-16.30

lntimal flap

7 (2)

2 (13)

5(2)

-

Dissection

Success (n = 290)

Odds Ratio

Univariate Cox Regression

No. (%) Total fn = 244)

Event (n = 69)

No Event (n = 175)

0.21-2.10

83 (34)

22 (32)

61 (35)

0.79

0.47-1.31

0.56-4.52

101 (42)

25 (37)

76 (43)

0.80

0.49-l

82 (55)

0.52*

0.22-1.27

24 (43)

59 (40)

0.81

0.33-1.98

28 2 9

27 f 9

29 k 8

0.98

0.95-1.01

11 (5)

5 (7)

6 (3)

2.137

0.86-5.31

3 (1)

1 (1)

2 (1)

95% Cl

Relative Risk

95% Cl

Procedural factors

.3 1

Largest device size*

Maximum pressure*

-

14 (7)

6 (11)

8 (5)

Postprocedural morphology

29 (10)

5 (31)

24 (8)

5.02t:

1.60-15.70

19 (8)

4 (6)

15 (9)

Without staining

17 (6)

4 (25)

13 (5)

7.109

2.00-25.20

9 (4)

3 (4)

6 (3)

With starning

10 (3)

4 (25)

6 (2)

5 (2)

0 (0)

5 (3)

i 10 m m length

16 (5)

3 (19)

13 (5)

9 (4)

2 (3)

7 (4)

3 (1)

0 (0)

3(l)

-

3 (1)

l(1)

2 (1)

13 (4)

2 (13)

11 (4)

3.62*

10 (4)

0 (0)

10 (6)

Thrombusg TIMI flow grade <3

4.9211

1.24-19.50 0.73-18.05

0.68

0.25-1.87

-

*Mean, standard deviation, percentage, odds ratlo, relative risk, 95% confidence interval calculations, and stabstical comparison did not include missing data. tp <0.20. tp
PTCA = percutaneous transluminal

coronary angioplasty; TIMI = Thrombolysls in Myocardial Infarctloo.

A SYMPOSIUM:

DIRECTIONAL ATHERECTOMY

59E

3.5 2

3.0

8 is 2.5

FIGURE 1. Quantitative Impmwemerits in lumen dimensions in a subset of 128 patients undergoing balloon angloplaety after directional atherectomy to optimize final results. MLD = minimal lumen diameter.

E 2.0 2 c 1.5 B z 1.0 0.5 Reference PreDiameter MLD

Post Atherectomy MLD

Mean’ Balloon Diameter

Final ’ MLD

No Lesion Calcium

FIGURE 2. Intravascular ultrasound findings in 57 patients undergoie lntracoronary ultrasound after directional atherectomy.

Mean Residual Plaque Area

Mean Residual Lumen Area

FIGURE 3. Intravascular ultrasound of a lesion located In the mldportion of the left anterior descending artery. After dlrectional atherectomy, a significant region of calcium remains (arrowhead), atthough the lumen diaqwter wee markedly imxeased.

distal embolization (n = 4), loss of side branch 2 10 m m in length and 10 (3.3%) were associated (n = 2), and p er-fora t ion (n = 1). Coronary dissec- with contrast staining. The presence of a coronary tions were noted in 29 (9.5%) patients after direc- dissection after directional atherectomy was associtional atherectomy; 16 (5.2%) dissections were ated with procedural failure (Table III). 60E

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72

OCTOBER 18, 1993

lntracoronary ultrasound: Intravascular ultrasound was performed in 57 patients after directional atherectomy. No lesion calcium was identified in 19 (33%) lesions, <2 quadrant lesion calcium ( < 90”arc of calcium) was demonstrated in 14 (25%) lesions, and 2 2 quadrant lesion calcium ( 2 90” arc of calcium) was noted in 24 (42%) cases. Many of these patients did not have angiographitally apparent lesion calcification. Mean residual plaque area (in mm2) was significantly higher in lesions with r 2 quadrant lesion calcium (16.9 k 4.8 mm2 vs 11.9 2 4.6 mm2 and 12.6 +- 5.5 mm2 for no and <2 quadrant lesion calcium, respectively; p = 0.007) (F i gure 2). Moreover, residual lumen areas also tended to be smaller in lesions with a calcium arc 290’ (7.5 k 2.5 mm2 vs 10.3 + 4.5 mm2 and 8.5 + 4.7 mm2 in lesions with no and <2 quadrant lesion calcium, respectively; p = 0.08; Figure 3).

60

120

180

240

300

Late clinical outcome: Clinical follow-up was obtained in 244 (84%) of 290 patients with a successful atherectomy procedure. Although patients without clinical follow-up were older than those with follow-up (p = 0.03) no other significant differences were noted in baseline demographics, preprocedural lesion morphology, or procedural outcome in these 2 populations. During the 11 2 6 month follow-up period, 69 (28.3%) patients developed recurrent clinical events (death, 5; Q wave myocardial infarction, 8; coronary bypass surgery, 31; coronary angioplasty, 36). Late clinical events (within 1 year after directional atherectomy) occurred more often in patients with diabetes mellitus (43% vs 24% in nondiabetic patients; Figure 4A), unstable angina (35% vs 13% in patients with stable angina; Figure 4B), a prior history of restenosis (40% vs 26% in de novo lesions; Figure 4C), and saphenous vein graft

360

60

Days After Atherectomy

120

180

300

360

120 180 240 300 Days After Atherectomy

360

Days After

ts

80

2

60

240

Atherectomy

%

i i i i i i -t-t-t--t-i--l--t-t I I I I I I

i

i

I Iw I

5

I

I

E w

60

120

180

240

Days After Atherectomy

300

40

360

20

60

FIGURE 4. Event-lkee probability after directional atherectomy in patiints with diabetes melliius (A), unstable angina (B), prior history of restenosls (C), and wlth a saphenous vein graft (SVG) lesion location (0). A SYMPOSIUM:

DIRECTIONAL

ATHERECTOMY

61E

lesions (47% vs 25% in native vessels; Figure 4D). Importantly, no pre- or postprocedural morphologic features were related to late clinical outcome.

vices, angulated5 and calcified2T5stenoses and those in vessels with proximal vessel tortuosity5 have been associated with an unfavorable outcome after directional atherectomy. Moreover, in the present report, ultrasound studies provide evidence that DISCUSSION the residual plaque volume may be higher and Since gaining FDA marketing approval in Sepresidual stenosis more severe in the presence of tember 1990, the directional coronary atherectomy e.xtensive (2 2 quadrant) calcification, due to the device (AtheroCath; Devices for Vascular Intervenlimited ability of the device to excise fibrocalcific tion, Redwood City, CA) has been used as an plaque. Conversely, as a result of the directed effective alternative method of coronary revascularplaque excision achievable with this device, eccenization. Directional atherectomy may have particular benefit in lesions deemed less suitable for tric lesions, those lesions with irregularity, ulcerballoon angioplasty, e.g., those with eccentricity, ation, or ostial location, and complex, thrombusulceration, or ostial location, recanalized total containing lesions may be particularly suited for occlusions, and those containing thrombus.4,5,22>23directional atherectomy. 4a5These latter lesion subLonger lesions, 4~17 those with marked angulation,5 sets include those specifically used in the present or calcification,4,5J7 and lesions in vessels with series, resulting in a 94.8% procedural successrate. marked proximal tortuosity5 are associated with Notably, female gender was associated with an more frequent complications4j5 and suboptimal adverse procedural outcome in this series, alangiographic results. l7 These are more amenable though no other clinical or angiographic risk facto specifically targeted alternative devices.1,2In the tors were identified. Factors afkcting lumen improvement after present series of 306 patients selected for directional atherectomy based on “ideal” lesion charac- directional atherectomy: Several mechanisms are teristics, a 94.8% procedural success rate was responsible for the improvements in lumen geomobtained. Late clinical events developed in 28.3% etry after directional coronary atherectomy.3J7 Unof patients and were less frequently related to like other currently available technologies, direcpreprocedural lesion morphology than to preproce- tional atherectomy exerts its major effect by excising dural clinical risk factors; these clinical factors are biopsy-sized fragments of the atherosclerotic independent of the specific device used for revascu- plaque, although occasional vessel wall cornponents may also be obtained.3 The unilateral cutting larization. window and contralateral inflation balloon permit Procedural success after directional atherectomy: Depending on the criteria used, prior series directed reorientation of plaque excision, dependhave reported procedural success rates of 88-94% ing on the angiographic localization of the residual after directional atherectomy.3-5J3 These series lesion. Pathologic2’j and intravascular ultrasound have suggested that, as in balloon angioplasty,24>25 studies22have demonstrated that coronary atherecprocedural outcome relates, in part, to the complex- tomy results in discrete defects in the plaque and ity of the lesion, operator experience, and the prior vessel wall. Tissue disruption consistent with balresponse to coronary angioplasty.5 For example, in loon barotrauma is notably absent in these vesa series of 378 consecutive patients treated at 6 sels2$ these factors may account for the smoother centers, procedural outcome was related to lesion angiographic appearance typically noted after direccomplexity,5 as defined by the modified American tional atherectomy. Close examination of the quanCollege of Cardiology/American Heart Associa- titative improvements in angiographic dimensions tion Task Force lesion morphologic criteria.24 Over- after directional atherectomy (averaging 2.3 mm) and the volume of plaque removed (averaging 19 all procedural successwas lower and complication rates were higher with type B2/C lesions (proce- mg) suggests that directional atherectomy may also dural success,75%; complications, 13%) than with mechanically dilate the vessel, due to the relatively less complex type A lesions (procedure success, large size and rigidity of the atherectomy device.? 93%; complications, 3%).5 Notably, the marked Moreover, some degree of vascular barotrauma heterogeneity of the lesion morphology classifica- resulting from balloon inflation, albeit at low 10-30 tion system limits its use in identifying which psi inflation pressures, may also occur. Finally, specific lesion subsets may be suitable, or unsuit- partial removal of the plaque may alter plaque able, for directional atherectomy. compliance and render the segment more responDue to the use of relatively large (5-7 F) and sive to balloon dilation, if performed, reducing the somewhat inflexible prototype atherectomy de- degree of elastic recoil.8 Quantitative angiographic 62E

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72

OCTOBER 18, 1993

data suggest that vascular recoil is lower after directional atherectomy than after standard balloon angioplasty in saphenous vein graft lesions.8 In the present report, elastic recoil was 17 * 14%, comparing quite favorably to the degree of elastic recoil resulting after standard balloon angioplasty (approximately 30%)1g,20and suggesting that plaque excision alters lesion compliance in some patients. Results from the present study also suggest that the magnitude of lumen improvement after directional atherectomy (1.88 mm) is larger than historically noted after balloon angioplasty. Notably, because the angiographic endpoint in this series was attainment of the lowest final residual diameter stenosis (average 14 ? 14%), adjunct balloon angioplasty was used in > 40% of patients, either to treat an angiographic complication or to optimize a residual (> 30%) diameter stenosis if a larger device size was not available or was oversized for the reference vessel diameter. Angiographic complications were infrequent after directional atherectomy, occurring in approximately 10% of patients. Prior nonrandomized studies have suggested that the angiographic lumen is smooth or with fewer flaps and dissections after directional atherectomy than after standard balloon angioplasty. On careful angiographic review of postprocedural angiograms in the present series, dissections > 10 mm in length developed in 4.2% of lesions, thrombus was present in l.O%, loss of side branch developed in 0.7%, and coronary perforation developed in 0.3%. Given the complexity of lesions in the series, these angiographic results seem quite acceptable. Late outcome after directiinal atherectomy: Symptom recurrence develops in 25-33% of patients after directional atherectomy.gJ2J3 Angiographic restenosis ( 2 50% diameter stenosis) has been documented in 32-58% of patients within 6 months after the procedure.gJ2-14 Several clinical and angiographic risk factors for restenosis after directional coronary atherectomy have been identified. Diabetes mellitus, prior restenosis,gJ2J3lesion length 2 10 mm,gJ2 and saphenous vein graft lesion locationgJ2J3 have been associated with increased rates of restenosis. Moreover, smaller ( < 3.0 mm) vessels, noncalcified lesions, and the use of a smaller (6 F) device were associated with restenosis in 1 series.g Based on these results, the strategic approach in this series sought to minimize the final residual diameter stenosis, using larger devices and adjunct balloon angioplasty if larger atherectomy devices were not available or could not be safely used.

Recurrent clinical events have been reported in 24-30% of patents within 6 months of the procedure; coronary bypass surgery has been used to treat restenosis in 14-19% of patients and repeat coronary angioplasty has been used in lo-17% of patients.gJ3 In our series, recurrent clinical events developed in 28% of patients within 1 year after atherectomy; coronary bypass surgery was performed in 13% and coronary angioplasty was performed in 15%. Four factors were associated with recurrent clinical events after directional atherectomy in this series. Patients with diabetes mellitus had a 2-fold higher risk of late clinical events than nondiabetic patients (Figure 4A); patients with unstable angina had a 3-fold higher chance of developing late clinical events than patients with stable angina (Figure 4B); patients with restenotic lesions had a 2-fold higher incidence of late clinical events compared with patients with de novo lesions (Figure 4C). Finally, lesions located within saphenous vein grafts also were associated with a 2-fold higher incidence of late clinical events compared with lesions in native vessels (Figure 4D). Notably, preprocedural lesion morphology did not appear to affect late clinical outcome significantly. Limitations: This study has several important limitations. First, lesions were selected for specific device use based on the assessment of the preprocedural lesion morphology by a senior investigator. Directional atherectomy was chosen because the investigator thought that this device would yield the best immediate angiographic result compared with both balloon and other new device techniques. The selection bias introduced into this series by choosing “ideal” lesions for directional atherectomy may limit its applicability to a broader population of patients in centers where multiple devices are not currently available and directional atherectomy is performed because it is preferable to the other available modalities, i.e., the balloon alone. Second, in the present series, follow-up angiography was available in only 50% of patients after directional atherectomy, generally performed for recurrence of symptoms or exercise-induced ischemia. Given the 20% incidence of asymptomatic restenosis after directional atherectomy,9J3 estimates of late outcome using clinical event-free probability cannot be used to extrapolate the true incidence of angiographic restenosis. Finally, during the study period, a marked evolution of directional atherectomy catheter designs and profiles occurred. Guiding catheters have been designed with softer tips and more coaxial support; atherectomy catheters have become lower in profile with A SYMPOSIUM:

DIRECTIONAL

ATHERECTOMY

63E

comparisonof the immediatesuccessof coronaryangioplasty,coronaryatherectomy and endohuninal stenting.Am J Cardiol1990;6&93N42. 8. Hong MK, Popma JJ, Leon MB, Pichard AD, Satler LF, Cospito PD, Keller MB, Kent KM. Vascular recoil in saphenousvein graft stenosesafter investigational angioplasty.(Abstr.) JAm Coil Cardiol 1992;19:263A. 9. Hmohara T, Robertson GC, Selmon MR, Vetter JW, Rowe MI& Braden LJ, McAuley BJ, Sheehan DJ, SimpsonJB. Restenosisafter directional corenaly atherectomy.JAm Co11Cardiol1992;20:62%32. 10. Popma JJ, Topol EJ, Ellis SG, Holmes DR, Pinkerton CA, Whitlow P, King SB, GhazzalZMB, Top01 ET, Garratt KN, KereiakesDJ. Abrupt closure after directional coronary atherectomyJAm Coil Cardiol1991;18:1183-1189. 1F Garratt KN, Holmes DR, Jr., Bell MR, Bresnahan JF, Kaufmann UP, Vlietstra RE, Edwards WD. Restenosisafter directional coronary atherectomy: Differences between primary atheromatous and restenosislesions and i&uence of subintimaltissueresection.JAm Coil Cardiol1990;16:1665-1671. 12. Popma JJ, DeCeswe NB, Pinkerton CA, KereiakesDJ, Whitlow P, King SB, Top01 EJ, Holmes DR, Leon MB, Ellis SG. Quantitative analysisof factors influencing late lumen loss and restenosisafter directional coronary atherectomy.Am J Cardioll993; in press. l3. Garratt KN, Holmes, DR Jr., Bell MR, Berger PB, Kauiinann UP, Bresnaban JF, Vlietstra RE. Resultsof directional atherectomyof primary atheromatous and restenoticlesionsin coronary artery and saphenousvein grafts.Am .I Cardioi 1992,70:449-454. 14. Kuntz RE, Safian RD, Levine MJ, Reis GJ, Diver DJ, Baim DS. Novel approach to the analysisof restenosisafter the use of three new coronav devices.JAm CON Cardiol1992;19:1493-1499. l5.Hohnes DR Jr, Garrrett KN, Top01 EJ. Coronary angioplasty versus excisionalatherectomy:CAVEAT. (Editorial.) Int J Cardiol1992;35:143-146. 16. The TIMI Study Group. The thrombolysisin myocardialinfarction (TIMI) trial. N Engl JMed 1985;312:932-936. 17. Popma JJ, De CesareNB, Ellis SG, Holmes DR, Pinkerton CA, Whitlow P, King SB, Ghazzal ZMB, Top01 EJ, Garratt KN, Kereiakes DJ. C&nical, angiographicand procedural correlatesof quantitative dimensionsafter directional Coronaryatherectomy.JAm CON Cardiol1991;18:118?-1189. I.S.Mancini GBJ, Simon SB, McGillem MJ, LeFree MT, Friedman HZ, Vogel RA. Automated quantitative coronary arteriography: morphologic and physiologicvalidation in viva of a rapid digital angiographicmethod. Circulation 1987;75:452-460. 19. Hermans WRM, Rensing BJ, StraussBH, Serruys PW. Methodological problems related to the quantitative assessmentof stretch, elastic recoil, and balloon-arteryratio. Catheter Cardiovasc Diag 1992;25:174-185. 20.Rensing BJ, Hermans WRM, Vos J, Beatt KJ, Bossuyt P, Rutsch W, Sermys PW. Angiographic risk factors of luminal narrowing after coronary balloon angioplastyusing balkxn measurementsto reflect stretch and elastic recoil at the dilatation site.Am J Cardiol1992,69:584-591. 21 Potkin BN, Btiorelli AL, Gessert JM, Neville RF, Almagor Y, Roberts WC, Leon MB. Coronary artery imaging with intravascular high-frequency REFERENCES l. Top01 EJ. Promisesand pitfalls of new devicesfor coronary artery disease. ultrasound.Ciulation 1990;81:1575-1.585. Circulntion 1991;83:68%94. 22. Popma JJ, Dick RJL, HaudenschildCC, Top01 EJ, Ellis SG. Atherectomy 3. King SB. Role of new technologyin balloon angioplasty.Circulafion1991&k of right coronaryostial stenoses:initial and long-term results,technicalfeatures 25w2579. and histologicfindings.Am J Car&l 1991;67:431-133. 3.Saiian RD, Gelbtlsh JS, Emy RE, Schnitt SJ, Schmidt DA, Bairn DS. 23. Dick RJL, Haudenschild CC, Popma JJ, Ellis SG, Muller DW, Top01 EJ, Coronary atherectomy.Clinical, angiographic,and histologicfindings and obser- Directional atherectomyfor total coronary occlusions.Comwy A&v Disease vations regarding potential mechanisms.Circulation1990,82:6%79. 1991;2:1sw99. 4. Hinohara T, Rowe MH, Robertson GC, Seknon MR, Braden L, Leggett 24. Ellis SG, Vandormael MG, CowleyMJ, Deligonul U, Top01EJ, Bulle TM. JH, Vetter JW, Simpson JB. Effect of lesion characteristicson outcome of Coronary morphologic and clinical determinants of procedural outcome with directional coronary atherectomy.JAm CoUCardial 1991;17:1112-1120. angioplastyfor multivesselcoronary disease:implicationsfor patient selection. 5. Ellis SG, De Cesare NB, Pinkerton CA, Whitlow P, King SB, Ghazzal Circulation 1990;82:1139-1202. ZMB, KereiakesDJ, Popma JJ, Menke KK, Top01 EJ, Holmes DR. Relation 25. Myler RK, Shaw RE, Steer Sq Hecht HS, Ryan C, Rosenbhun J, of stenosismorphology and clinical presentation to the procedural results of Cumberland DC, Murphy MC, Hansell HN, Hidalgo B. Lesion morphology directional coronary atherectomy.Circulatin 1991;84:644&3. and coronary angioplasty:current experienceand analysis.J Am CoU Cardiol 6. Rowe MH, Hmohara T, White NW, Robertson GC, Sehnon MR, Simpson 1992;19:1641-1652. JB. Comparison of dissectionrates and angiographic results following direc- 20.Garratt KN, Edwards WD, Vlietstra RE, Kaufman UP, Holmes DR. tional coronaryatherectomyand coronaty angioplasty.Am J Card!& 1990,66:49- Coronary morphology after percutaneousdirectional coronary atherectomyin 53. humans: autopsy analysisof three patients. J Am Coil Cardiol 1990,16:14327. Muller DWM, Ellis SG, Debowey DL, Top01 ET. Quantitative angjographic 1435.

more flexible nose cones; new wider angle and shorter cutting window catheters have been tried to improve the degree of catheter flexibility and enhance plaque excision. These changes in catheter designs may expand the lesion profiles that can be effectively treated with directional atherectomy. Clinical implications: As new alternative devices for coronary angioplasty become available, appropriate device selection for specific lesion subsets becomes an increasingly significant and complex issue. In this series, the procedural results after directional atherectomy are reported for lesions deemed “ideal” for this intervention. These lesion subsets included eccentric lesions, lesions with ulceration, intimal flaps, irregularity, ostial location, or involving a side branch, and concentric lesions in larger ( > 3 mm) vessels. Diffise lesions and those lesions with marked calcification, angulation, inaccessibility, degenerated vein graft location, or with recurrent restenoses were treated with other new angioplasty devices. Using these selection criteria, procedural success rates were high (94.8%) and complications were infrequent (2.6%) after directional atherectomy. A strategic approach to minimize the final residual diameter stenosis after the procedure was pursued using a combination of coronary atherectomy and balloon angioplasti, when needed. The attainment of a small final residual stenosis (14%) may have contributed to the relatively low late clinical event rate (28%) noted in these patients.

64E

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72

OCTOBER 18, 1993