Does percutaneous transluminal coronary angioplasty accelerate atherosclerotic lesions?

Catheterization and Cardiovascular Diagnosis 21:l-6 (1990)

Original Study

Does Percutaneous Transluminal Coronary Angioplasty Accelerate Atherosclerotic Lesions? Khiem P.V. Nguyen, MD, Richard E. Shaw, PhD, Richard K. Myler, MD, John G.Webb, MD, and Simon H. Stertzer, MD Recent reports have suggested that angioplasty may cause or accelerate coronary arterial stenoses secondary to traumatic injury. Ninety-four coronary angiograms performed in a 1 yr period were reviewed in patients who had successful coronary angioplasty 6 to 30 mo (mean 10.7) prior to restudy. Restenosis was found in 43 of 140 dilated lesions (31%) and in 41 of 94 patients (44%). Thirty-three (35%) patients had new or progressive lesions outside the angioplasty site. New or progressive lesions occurred with similar frequency in the arteries that did not have angioplasty (23/155=15%) as in the arteries that did (13/127= 10%; chi-square n.s.). In the arteries which underwent angioplasty, new or progressive lesions occurred as commonly proximal to the PTCA site (7114, 50%) as distal (6/13,46O/0). New or progressive lesions occurred in 29% of patients with concomitant restenosis, and 40% of those without restenosis (chi-square n.s.). No clinical, angiographic, or procedural factors distinguished patients with new and progressive lesions in target vessels from those without these lesions in target vessels. Patients with progressive lesions anywhere in the coronary tree were more likely to have had a shorter duration of anginal symptoms before angioplasty and a family history of coronary disease when compared with patients without progressive atherosclerosis. In conclusion, new and progressive lesions outside the angioplasty site occur after the procedure but appear unrelatedto the restenosis process or traumatic injury by angioplasty instrumentation. Key words: progressive lesions, restenosis, angioplasty

INTRODUCTION

At the present time, the primary success rate of percutaneous transluminal coronary angioplasty (PTCA) is 90% or more, and its complication rate is quite low [I31. Restenosis occurs in approximately 30% of the patients within 6 mo after the procedure, and has been a major limiting problem of this technique [4,5]. Moreover, several reports [6-141 have described the accelerated progression of stenoses in previously dilated coronary arteries. Instrumentation by the guiding catheter, balloon dilatation catheter, or guidewire has been implicated in producing endothelial injury and initiating a fibrocellular response which may lead to subsequent encroachment of the arterial lumen. It has been hypothesized that this injury might produce or accelerate lesions in previously normal or mildly diseased arterial segments. Coronary angiograms performed after angioplasty to evaluate the appearance of new lesions adjacent to and remote from the PTCA sites have yielded conflicting findings [15-171. The purpose of the current 0 1990 Wiley-Liss, Inc.

study was to analyze angiographic follow-up in 94 patients who returned for restudy during a l yr time period to evaluate the extent of new or progressive disease, and to determine if any clinical, morphologic, or procedural variables were predictive of the formation of such lesions after coronary angioplasty .

From the San Francisco Heart Institute, Seton Medical Center, Daly City, California.

Received January 10, 1990; revision accepted April 27, 1990. Address reprint requests to Richard K. Myler, M.D., San Francisco Heart Institute, Seton Medical Center, 1900 Sullivan Avenue, Daly City, CA 94015. Presented in part at the 10th Annual Meeting of the Society for Cardiac Angiography and Interventions, Williamsburg, Virginia. 1989.

2

Nguyen et al.

METHODS Patient Sample

This prospective study encompassed a 12 mo period and consisted of 94 consecutive coronary arteriograms performed at Seton Medical Center in patients following successful angioplasty performed at least 6 mo earlier. This represented 24.7% of the post-angioplasty restudies done during the time period between October 1, 1987, and September 30, 1988. The remaining 75.3% of restudy patients were excluded for one or more of the following reasons: 1) diffuse disease, long lesions (greater than 15 mm), or tandem lesions closer than 2 cm (a balloon length) to each other in the target artery; 2) restudy earlier than 6 mo from the initial angioplasty; 3 ) saphenous vein bypass graft angioplasty; or 4) presentation with acute myocardial infarction. Patients with diffuse coronary disease, long lesions, or close tandem stenoses were excluded because we were interested in the response (evolution) of (relatively) normal arterial segments, proximal and distal to the target stenosis, that were outside of the balloon inflation zone. Patients restudied less than 6 mo from the initial angioplasty were excluded so that we could evaluate possible interrelationships between restenosis and new/progressive lesions. By extending the window of exclusion to 6 mo, we would include (virtually) all patients with restenosis [4,18-221 and allow sufficient time for development of new/progressive lesions. Patients with saphenous vein graft angioplasty were excluded because morphologic changes in vein grafts appear to differ and the window of restenosis (or progression) may be longer than lesions in native coronary arteries [23]. Patients with acute myocardial infarction were excluded since they represent a distinct clinical entity with specific complex morphology [24] and are usually reported separately 1251.

Protocol Selective coronary arteriography was performed before and immediately following angioplasty. Patients were treated with aspirin, 325 mg daily, and nifedipine, 10 to 40 mg four times daily, or diltiazem, 30 to 60 mg four times daily, for 24 h before and after angioplasty. During the procedure, all patients received 10,000 U heparin intravenously and an extra 3,000 U of heparin for each elapsed hour thereafter. Intravenous, sublingual, or intracoronary nitroglycerin was administered as required for evidence of coronary spasm. Administration of aspirin, dipyridamole. isosorbide dinitrate, and/or a calcium channel blocker was continued for 3-6 mo after angioplasty. Thallium-201 cardiac exercise tests were performed within the first month and again between 3 and 6 mo post angioplasty . Coronary arteriography was obtained for re-

LCX

Fig. 1. Diagram demonstrating PTCA target vs. non-targetvessel, PTCA site, proximal and distal PTCA segments.

current angina and/or abnormal thallium scintigraphy (84%) or as a follow-up procedure in asymptomatic patients as part of subset protocols (16%).

Coronary Angiography and Angioplasty For analysis, the coronary arterial tree was divided into segments (Fig. 1). The “target artery” was defined as the vessel in which angioplasty was performed. The stenosis treated by the dilatation catheter was defined as the “PTCA site.” All dilatation balloons utilized were 2 cm long. A lesion located within 2 cm from the PTCA site, i.e., the segment subjected to balloon dilation, was defined as a “restenosis site.” The proximal and distal segments of the target artery were defined as segments more than 2 cm beyond the PTCA site. Successful angioplasty was defined as improvement of 35% or more in diameter stenosis and a residual narrowing less than 50% [4]. Restenosis was defined as greater than 50% loss of the gain in luminal diameter achieved during the angioplasty [4]. A new lesion was defined as one more than 2 cm from the PTCA site, in which there was a 20% or more stenosis, which developed since the angioplasty in a previously normal segment of a coronary artery. Progression of an existing lesion was defined as a further decrease of 20% or more in luminal diameter in a previously diseased arterial segment not exposed to balloon inflation. Angiographic views were obtained in the same orthogonal projections as in the initial study. and measurements were made in end-diastolic frames. All arteriograms were viewed without knowledge of the site of angio-

Does PTCA Accelerate Atherosclerotic Lesions? TABLE 1. Angiographic Findings at Follow-Up in 94 Patients Restudied During a 1 Yr Period in Whom Angioplasty Had Been Performed More Than 6 Mo Previouslv

TABLE II. Comparison of Time to the occurrence of Proaressive Lesions in Taraet Vs Non-Taraet Arteries' Occurrence of new or Droeressed lesion. No. (70)

Patients Findings

N

%

No restenosis Restenosis New or progressive lesions With restenosis Without restenosis

53 41 33 12 21

56.4 43.6 35.1 29.3* 39.6*

*Chi-square, n.s

3

6 to 12 mo DOSt

New or progressed lesion in PTCA target artery New or progressed lesion in non-target artery

PTCA

Greater than 12 mo uost PTCA

7 (53.8)

6 (46.2)

13 (56.5)

10 (43.5)

*Chi-square, n.s. 6-12 vs. > 12 mos in target and non-target arteries

plasty. Lesions were measured in at least two views independently by two investigators with a precision vernier caliper. The difference in the degree of stenosis was resolved in a conference meeting or by a third independent arbitrator [26].

stenosis was found in 43 of 140 dilated lesions (31 %) and in 41 of the 94 patients (44%). New or progressive lesions occurred in 33 of the 94 patients (35%). Twelve of 41 patients (29.3%) with restenosis also had new or progressive lesions, whereas 21 of 53 patients (39.6%) withStatistical Analysis out restenosis had new or progressive lesions (chiContinuous variables are expressed as a mean 2 1 square, n.s.). New lesions developed in 15 patients, SD and were compared with use of a nonpaired or paired progression was seen in 15 patients, and three patients Student's t test, as appropriate. Categorical variables had both new and progressive lesions. None of the 11 were compared with the chi-square test with the Yates patients with widely separated tandem lesions developed correction. Analysis of factors related to the time-depen- new or progressive lesions. The average luminal narrowdent occurrence of disease progression was accom- ing of all 36 new and progressive lesions was 66 ? 26%. plished by using the Cox proportional hazards regression The average percent diameter stenosis of the 18 new model. A p value of less than 0.05 was considered sig- lesions was 58 -+ 28%. The 18 progressive lesions innificant. All analyses were done by using SPSSX and creased from 32 & 14% at the time of the initial proceBMDP statistical packages. dure to 75 Z 22% at the followup examination. The incidence of new or progressed lesions in PTCA target vessels was compared to that in non-target vessels. RESULTS New or progressive lesions occurred as commonly in In this study of 94 patients, there were 75 (79.8%) target arteries (13/127 = 10%) as in non-target arteries men and 19 (20.2%) women, with age range from 30 to (23/155 = 15%; chi-square n.s.). Comparing the fre84 yr (mean = 58). The time interval from the original quency of new or progressive lesions occurring during angioplasty to follow-up angiography ranged from 6 to the first 6 to 12 mo after angioplasty with lesions occur30 mo (mean = 10.7). A total of 140 lesions were di- ring beyond 12 mo, there was no difference between lated in 127 arteries: 64 left anterior descending, 25 left patients with progressive lesions in target vs. the noncircumflex and 38 right coronary arteries. Sixty-four pa- target arteries (Table 11). Furthermore, in target arteries, tients had angioplasty in one vessel, 27 patients in two new or progressive lesions occurred as commonly proxvessels, and three patients in three vessels. In nine pa- imal (7/14, 50%) as distal (6/13, 46%) to the PTCA site. tients with one vessel dilated there were two lesions per New or progressive lesions were not detected in left main vessel, and in two patients with angioplasty in two ves- stem coronary artery. Comparisons were made between patients with new or sels there were two lesions dilated in each of the two vessels. The tandem lesions in these vessels were located progressive lesions in PTCA target arteries and patients at the proximal and distal extremes of the target vessel without such lesions. Numerous factors were selected for with more than 2 cm (one balloon length) separating analysis and are listed in Table 111. The two groups of them. The peak balloon inflation pressure averaged 8.7 patients did not differ in any of the clinical, morphoBAR. The mean percent diameter stenosis was 77 before logic, procedural, or pharmacologic factors, save for the and 27 after angioplasty; the mean pressure gradient was finding of more severe PTCA site stenoses in those pa54 and 12 mm Hg before and after angioplasty, respec- tients without new or progressive lesions. In addition, patients with progression of disease anytively. The angiographic findings are shown in Table I. Re- where in the coronary tree were compared to patients

4

Nguyen et al. TABLE IV. Comparison of Patients With or Without Progression of Atherosclerosis Outside Anywhere Outside the PTCA Site

TABLE 111. Comparison of Patients With or Without New Progressed Lesions in PTCA Target Arteries

Variables Age (yr)

Hypertension (%) Diabetes (%) Hypercholesterolemia (%) Family history of coronary artery disease (7%) Prior myocardial infarction ('36) Medications after PTCA (%) Calcium blockers Nitrates Aspirin Dipyridamole Target artery (5%) LAD LCX RCA Diameter stenosis (%) Before PTCA After PTCA Length of lesion (mm) Pressure gradient (nim Hg) Before PTCA After PTCA Peak dilation pressure (BAR) No. of dilations Need to change balloons (%) Need to recross lesion (%)

New or progressed lesions in PTCA vessels (N = 13)

No new or progressed lesions in PTCA vessels (N = 81)

p

5 6 ? LO 30.7 7.1 46.27 76.9

58 t LO 45.0 9.8 40.2 46.3

NS NS NS NS NS

53.9

42.7

NS

76.9 69.2 92.9 84.6

84.1 67.1 75.6 89.0

NS NS NS NS

76.9 15.4 46.2

70. I 25.6 63.4

NS NS NS

14 21 5

78 2 15 27 t 17 9.6 2 8

.05 NS NS

59 2 1s 12 t 6

53 2 15 12 t 10

NS NS

8.5 -C 1.3 4% I 46.1

8.8 2 1.8 5 2 2 40.2

NS NS NS

69 29 9.5

0

2 2

2

2.4

NS

without such progression (Table IV). In univariate tests, duration of anginal symptoms prior to angioplasty and family history of coronary artery disease were statistically significant. In multivariate Cox proportional hazards regression analysis of the time-related occurrence of disease progression anywhere in the coronary system, short duration of anginal symptoms prior to the initial angioplasty (log likelihood -82.3394, -2.041 coefficienthtandard error; p < . O l ) was the only significant predictor. DISCUSSION

Positioning of an angioplasty guiding catheter, advancement of a guidewire, or passage of the dilatation balloon catheter might result in arterial endothelial damage. Intimal splitting and medial dissection combined with stretching of the uninvolved adjacent portion of the artery have been observed after experimental angioplasty in both animal models [27] and cadaver hearts [28,29].

Variables Age (yr) Hypertension (70) Diabetes (%) Hypercholesterolemia (%) Duration of angina before FTCA (weeks) Family history of coronary artery disease (96) Prior myocardial infarction (56) Medications after PTCA (%) Calcium blocker Nitrate Aspirin Dipyriddmole Target vessel (%) LAD LCX RCA Diameter stenosis (%) Before PTCA After PTCA Lesion length (mm) Pressure gradient (mm Hg) Before PTCA After PTCA Peak dilation pressure No. of dilations Need to change balloons (5%) Need to recross lesion (%)

Patients with any newiprogressive lesions (N = 32)

Patients without progressive lesions (N = 62)

5 6 2 10 31.3 9.4 28.1 26 2 53

59 -+ 10 48.4 9.7 46.8 79 -C 144

p

NS NS NS NS 01

68.8

40.3

01

43.8

41.9

NS

75.0 65.6 81.3 90.6

87. I 67.7 75.8 87. I

NS NS NS NS

31.3 71.9 56.3

29.0 77.4 62.9

NS NS NS

77 -+ 17 -C 16 9.4 2 8.2

15

56 2 13 15 ? 13 8.9 2 1.5 5?2 34.4 0

77 t 14 27 2 19 9.7 2 7.2

52"

15

NS NS NS

8.7 2 1.9 5'2 43.5

NS NS NS NS NS

3.2

NS

11 2 7

Intentional arterial endothelial damage obtained by using a Fogarty balloon catheter is an accepted method for producing atherosclerotic lesions in the experimental animal [30,31]. Case reports have described new stenotic lesions after successful angioplasty . These new lesions have been described in the left main stem [8-141 and in the segments of the coronary artery proximal [15] and distal [6,7] to the dilated lesion. The occurrence of these new lesions may be coincidental or secondary to the angioplasty procedure. Although new or progressive left main stenoses were not noted in our study, 4 yr ago we had a case of significant late progression of a previously minimal lesion at this site, almost certainly due to deep engagement of a guiding catheter in order to "power" a dilatation catheter across a calcified subtotal left anterior descending stenosis. A few case reports of a similar nature have been published [8-141. Hopefully, with currently available

Does PTCA Accelerate Atherosclerotic Lesions?

soft-tipped guiding catheters and very-low-profile dilatation catheters and balloon-on-a-wire devices, this late complication will occur even more rarely. This study examined a series of patients presenting for angiographic restudy in a 1 yr time period following successful angioplasty . The post-angioplasty interval of at least 6 mo was chosen because this appears to be the window of restenosis [ 18-20]. This interval would allow sufficient time for treated lesions to recur and therefore permit assessment of possible correlations between restenosis and disease progression. In this group of patients with follow-up angiography, restenosis occurred in 3 1% of the dilated lesions and 44% of patients and is similar to that reported elsewhere [4,20-221. New or progressive lesion occurred in 29.3% of our patients with restenosis and 39.6% of those without restenosis, quite similar to another report [17] of 26.8% and 38.3%, respectively. In our study, disease progression outside the PTCA site occurred in 35% of patients at a mean of 10.7 mo after angioplasty. This compares to reports of a 50% incidence of disease progression at a mean of 7 mo after angioplasty [15], a 36% occurrence at a mean of 34 mo [ 161, and a 45% incidence at a mean of 31 mo [17]. To test the hypothesis that technical or procedural factors associated with angioplasty might accelerate atherosclerotic lesions, the incidence of new and progressive lesions in target arteries was compared to the progression in non-target arteries. These lesions occurred as commonly in target as in non-target arteries. In target arteries, new and progressive lesions were distributed as commonly proximally as distally to the PTCA site. Furthermore, one might assume that new or progressive disease related to angjoplasty trauma might emerge sooner rather than later; analyses were performed to assess the time period during which disease progression occurred. These analyses showed that there was no difference in the occurrence of new or progressive lesions in target vs. non-target arteries comparing patients who had these lesions in the first 6 to 12 mo vs. those with progression more than 12 mo after their initial angioplasty. Our findings are in contrast to those of Hwang et al. [ 151, who showed that acceleration of disease tended to occur more commonly in the arteries in which angioplasty was performed (32%) than in the arteries in which it was not performed (12%), and more commonly in the segment proximal (54%) than in the segment distal (31%) to the dilated lesion. Our findings are in agreement with those of Cequier et al. [16], who reported similar rates of disease progression in angioplasty target (17%) vs. non-target (14%) arteries. Bottner et al. [ 171 in a report of recurrent ischemia more than 1 yr after succesful angioplasty, showed that time to ischemia was earlier in patients with restenosis only than in those with

5

disease progression only (20 vs. 38 mo, respectively, p