Angiographic predictors of neointimal thickening after successful coronary wall healing following percutaneous revascularization

Angiographic predictors of neointimal thickening after successful coronary wall healing following percutaneous revascu larization Jos6 Su~rez de Lezo, MD, Djordje Pavlovic, MD, Alfonso Medina, MD, a Manuel Pan, MD, Jos6 Angel Cabrera, MD, a Mignel Romero, MD, Jos6 Segnra, MD, Enrique Hern~ndez, MD, a Arsenio Gallardo, MD, and Francisco Meli~n, MD a CSrdoba and Las Palmas de Gran Canaria, Spain This study was undertaken to characterize, by intracoronary ultrasound technique, the neointimal thickening at follow-up of treated coronary segments after successful arterial wall repair and to compare the findings with serial angiographic studies. We selected for study 81 patients with single-vessel coronary disease successfully treated by percutaneous revascularization who were angiographically and ultrasonically reevaluated at a mean follow-up time of 22 ± 21 months; 23 had been treated by balloon angioplasty, 27 by directional atherectomy, and 31 by elective Palmaz-Schatz stent implantation. The late maximal neointimal thickness varied between 0.1 and 1.5 mm (mean 0.65 ± 0.31 mm), and the neointimal area ranged between 0.97 and 14.9 mm 2 (mean 5.19 ± 3.14 mm2). The neointimal repair was thinner in patients who obtained a better acute angiographic result immediately after treatment and in stented (3.4 ± 1.8 mm 2) versus dilated (7.8 ± 4.1 mm 2) or resected (5 ± 1.6 mm 2, p < 0.001) segments. On the'contrary, the repaired neointimal layer was thicker in those patients who angiographically exhibited less late luminal loss or even expansion and in those evaluated after a longer time since treatment. The acute gain and the time influence resulted in independent predictors of the degree of neointimal thickness. These findings suggest that two reparative mechanisms of the coronary wall may operate in close relation. (Am Heart J 1997;133:210-20.)

The underlying mechanisms of coronary restenosis after percutaneous treatment are believed to be the combination of both intimal hyperplasia and early or delayed dynamic remodeling of the treated segment. 1 Cell proliferation, as a response to vascular injury, and reactive shrinking of the arterial wall seem to contribute, in a still-discussed proportion, 2-5 From the Hospital Reina Sofia, University of C6rdoba, and aHospital del Ping University of Las Palmas. Received for publication April 22, 1996; accepted Aug. 1, 1996. Supported in part by Fundaci6n Cultural Hospital Reina Sofia-Cajasur, C6rdoba, Spain. Reprint requests: Jos4 Su~rez de Lezo, MD, Servicio de Cardiologia, Hospital Reina Sofia, Avda. Men4ndez Pidal s/n, 14004 C6rdoba, Spain. Copyright © 1997 by Mosby-Year Book, Inc. 0002-8703/97/$5.00 + 0 4/1/78083

210

to the restenosis process. However, little is known about the mechanisms leading to a favorable outcome of the treated coronary segment and, by consequence, to stabilization and long-term luminal patency. Ifrestenosis is considered the exaggeration of a normal healing process, 1, 6 increased knowledge of the normal mechanisms leading to arterial wall repair could help us better understand the abnormal response. From a clinical point of view, a great amount of information has been obtained from serial angiographic studies in treated patients, despite the known limitations of angiographic methods alone. 7 Luminal measurements have been used to evaluate severity and evolution of coronary stenosis, particularly after treatment. Significant reduction of the lumen i to 6 months after treatment is usually associated with myocardial ischemia and clinical recurrence. On the contrary, angiographic luminal enlargement, usually associated with clinical benefit, s has been advocated as the consequence of spontaneous 9 or pharmacologically induced 1°, 11 regression of the plaque, However, serial angiographic interpretation of any coronary arterial process of progression or regression of disease has to be made on the basis of luminal change as the final and clinical consequence of arterial wall repair. With the aid of intracoronary wave ultrasonography we can also evaluate intimal thickening. Thus, by combining luminal and intimal measurements, the arterial process could be better evaluated. The purposes of this study were to compare the serial angiographic luminal changes with the degree of neointimal thickening at long-term follow-up in a group of patients with coronary artery disease being successfully treated percutaneously. In patients having persistent plaque stabilization after treatment, the neointimal layer is the final product of residual plaque and favorable scarring after injury. The late ultrasonic appearance of this inner layer after suc-

Volume 133, Number2 American Heart Journal

cessful scarring h a s not been d e s c r i b e d in detail. T h u s we tried to ultrasonically characterize this n e o i n t i m a l t h i c k e n i n g a n d to c o m p a r e its degree with serial angiographic l u m i n a l changes. F a c t o r s influencing the degree of thickness could help us b e t t e r u n d e r s t a n d the n o r m a l s c a r r i n g process. METHODS Patients. We retrospectively analyzed the findings obtained in 81 patients with single-vessel disease who had been previously treated by percutaneous revascularization and had angiographic and intracoronary ultrasound studies at follow-up. Selection was made in a prospective manner during a 5-year period. Data were collected from patients being successfully treated and having long-term angiographic reevaluation. Once the absence of restenosis in the previously treated segment was detected by late angiography, the ultrasound study was performed. If the quality of the scanned segment was considered adequate, the patient entered the study. Hence all 81 patients fulfilled the following requirements: (1) single-vessel disease at the time of treatment, (2) adequate serial angiographic study from baseline conditions to follow-up, (3) absence ofrestenosis at the treated segment, and (4) adequate ultrasound study at follow-up. Mean age was 55 ± 9 years. At the time of treatment, 65 patients had an unstable condition (angina at rest, worsening angina, or postmyocardial infarction angina), and 16 had stable angina. In all patients the treatment was initially considered successful; they became asymptomatic and remained free from symptoms for at least 6 months. After this period, all patients were hemodynamically reevaluated because of this study's protocol or because they had late typical or atypical chest pain after a mean follow-up time of 22 _+ 21 months (range 6 months to 7.3 years): At late cardiac catheterization, 21 patients had angiographic documentation of progression of disease at different sites of the coronary tree but showed a nonstenotic arterial segment at the site of previous treatment. The remaining 60 patients also showed late success of the initially treated segment and no significant progression of arteriosclerotic disease in other territories. After completion of the ultrasound study of these healed segments, 17 patients were successfully treated percutaneously on newly developed lesions (7 by percutaneous transluminal coronary angioplasty [PTCA] and 10 by stent implantation), 1 received surgical revascularizati0n , and 3 were managed medically for the progression of the disease in other territories. All patients gave written informed consent for the treatment and for late angiographic and ultrasonic reevaluation. Thus we interrogated by wave ultrasonography 81 previously treated segments that had a satisfactory healing process (residual lesion at follow-up ---40%) after intervention and compared the findings with all serial angiographic information obtained from intervention to follow-up. Patients were divided into three groups according to the type of percutaneous treatment they received. Group 1 was

Sudrez de Lezo et al.

T a b l e !.

Baseline data Group 1 Group2 (PTCA, (stent, n=23) n=31)

Age (yr) Male (%) Risk factors Diabetes Hypertension Current smoker Hyperlipidemia Clinical condition Stable angina Unstable angina Type of lesion Native Restenosis Responsible for previous MI Angiographic data (artery) Left anterior descending Right coronary artery Left circumflex ACC/AHAlesion type A B1 B2 C

211

Group 3 (DCA, n=27)

55 ± 9 83

56 -+ 8 93

54 ± 10 93

2 6 8 7

4 9 14 12

3 12 13 8

5 18

6 25

5 22

21 2 10

28 3 11

18 8 10

12 4 7

12 14 4

24 1 2

7 8

4 8

5 10

6 2

13 6

i0 2

MI, Myocardialinfarction;ACC/AHA. American College of Cardiology/ AmericanHeart Association.

composed of 23 treated by PTCA. Group 2 was composed of 31 treated by elective Palmaz-Schatz stent implantation. Group 3 was composed of 27 treated by directional coronary atherectomy (DCA). Selection of each technique at the time of treatment was made on the basis of different aspects, such as the type and location of the stenosis, vessel diameter, and current availability and experience. Because the mechanisms of each technique are different, we postulated that the degree and type of arterial wall injury caused by a specific revascularization technique could influence, in some way, the subsequent repair process. Thus we compared the neointimal thickening of three groups of patients each treated by a specific technique to analyze the influence of the technique in the vessel response. Table I shows the main baseline data from all three groups of patients. No significant differences were observed in clinical variables, risk factors, type of lesion studied, or severity of disease among groups. However, the diameter of the treated vessel was significantly (p < 0.001) larger in patients treated by stent implantation (3,39 _~ 0.62 mm) or DCA (3.37 _+ 0.44 mm) than in patients receiving PTCA (2.93 ± 0.48 mm). Additionally, the time from treatment to observation ranged markedly from 6 months to 7.3 years (mean 22 ± 2 1 months), being significantly longer (p < 0.001) in patients treated by PTCA (43 -+ 26 months) than in patients treated by stent implantation (11 ± 5 months) or by DCA (15 ± 13 months). Angiographic s t u d i e s . All angiographic procedures were performed by experienced operators under optimally

212

Sudrez de Lezo et al.

February 1997 American Heart Journal

Fig. 1. Example of healed coronary segment late after treatment. A, Neointima and lumen can be measured. B, Lumen is filled by echocontrast (asterisk). L.D., Lumen diameter. standardized conditions. Sublingual nitroglycerin was always administered before angiographic recordings to minimize differences in vascular tone. Angiographic analysis was obtained from every patient before treatment, immediately after treatment, and at follow-up. The projection used, the best one showing the treated segment at baseline, was maintained in the two following observations. Measurements were taken by digital quantitative coronary analysis of end-diastolic images. Proximal reference segment, minimal lumen diameter (MLD), and percent stenosis were measured in each condition. The acute luminal gain (MLD after treatment minus MLD before treatment) and the late luminal loss (MLD after treatment minus MLD at follow,up) were obtained in all patients. The mean reference segment at baseline observation did not significantly differ from that obtained at foUow-up (3.22 -+ 0.53 mm). Aninterobserver variability analysis on angiographic measurements was performed in 10 patients. The results for MLD were observer 2 = 0.048 + 0.96 x Observer 1; r = 0.95, standard error of the estimate (SEE) = 0.13. At follow-up, a segment was considered favorably healed if the segment had a nonsignificant (<40%) residual stenosis at the site previously treated. Smooth intimal borders and absence of intraluminal defects were always noted in all considered segments. Ultrasound studies. All ultrasound studies were per-

formed with intravenous weight-adjusted heparin administration (2 mg/kg) and after a bolus (0.45 mg) of intracoronary nitroglycerin to avoid irritative spasm during inspection. A 64-element, 3.5F coaxial or monorail catheter (EndoSonics) was used in every study. The catheter was guided by angiographic monitoring, and the proximal, distal, and treated segments were scanned at a slow and constant pullback. The position of the ultrasound probe at the sites selected for measurement was recorded by cineangiography to mark measurements at the narrowest part. Ultrasonic gain settings were adjusted for optimal visualization of the arterial wall-lumen interface in normal reference segments. The closest reference segment without atheromatous disease was selected for measurement. In 10 patients showing mild coronary disease in proximal segments, the reference was selected at distal segments with normal intimal echoes. 12 A wide lumen filled by echo contrast was always detected in proximal normal segments and at the treated segment (Fig. 1). Qualitative and quantitative image analysis was performed off line at the treated and reference segments. The selected cross-sectional arterial wall and lumen were analyzed by the Tape Measure V1.0 software (INDEC Systems). End-diastolic static images were selected as representative of the narrowest part of the treated segment, and from the closest nondiseased segment. The results from digitized frames at

Volume133, Number2 AmericanHeartJournal

Sudrez de Lezo et al.

213

Table II. Angiographic results

MLD (ram)~% stenosis

Group 1 (PTCA) Group 2 (stent) Group 3 (DCA)

Vessel diameter (ram)

Basal

After treatment

Follow-up

Acute gain (mm)

Late loss (mm)

2.93 _+0.48 3.39 -+ 0.62 3.37 _+0.44

0.72 -+ 0.46/75 + 15 0.51 -+ 0.43/81 + 21 0.66 -+ 0.32/80 -+ 10

2.38 -+ 0.5/19 -+ 12 3.24 ± 0.6/5 _+ 10 2.94 _+0.47/12 -+ 9

2.58 ± 0.44/13 ± 11 2.83 -+ 0.7/16 ± 12 2.66 +- 0.54/21 -+ 13

1.67_+0.64 2.73 _+0.66 2.29 -+ 0.61

-0.20_+ 0.46 0.41 _+0.36 0.28 _+0.41

Acute gain is MLD after treatment minus MLD before treatment. Late loss is MLD after treatment minus MLD at follow-up.

Table III. Ultrasonic follow-up data

Reevaluation time (too) Reference segment Lumen diameter (mm) Lumen area (ram2) Thickness of intimal leading edge (ram) Treated segment Lumen diameter (ram) Lumen area (ram2) Maximal neointimal thickness (ram) Intimal ratio (intima/MLD) Neointimal area (ram2)

Group 1 (PTCA)

Group 2 (stent)

Group 3 (DCA)

p Value

43 _+26

11 _+5

15 _+ 13

<0.001

3.10 __0.64 8.97 _+3.10 0.27 _+0.15

3.51 _+0.53 10.27 _+3.16 0.20 _+0.13

3.77 _+0.7 11 _+4 0.18 _+0.11

<0.01 NS NS

2.82 _+0.44 7.16 _+2.5 0.84 _+0.24 0.31 _+0.11 7.82 _+4.08

3.09 _+0.49 8 -+ 2.47 0.39 ~ 0.20 0.12 _+0.08 3.39 _+ 1.79

3.13 8.2 0.74 0.25 5.01

NS NS <0.001 <0.001 <0.001

_+0.53 _+3 _+0.3 _+0.11 _+ 1.62

NS, Not significant.

each position were obtained without reference to angiographic m e a s u r e m e n t s . I n all selected patients the i n n e r echogenic layer was clearly delimited, favored by two ultrasonic interfaces. The l u m e n / i n t i m a and i n t e r n a l elastic l a m i n a interfaces were traced a n d planimetered. I n both segments the l u m e n diameter and area, the crosssectional neointimal area, and the maximal thickening of the i n t i m a l i n n e r layer were measured (Fig. 1). I n groups 2 a n d 3 (PTCA and DCA), the well-delimited neointimal area was considered the result of both scarred residual plaque and i n t i m a i proliferation in a n u n k n o w n proportion. On the contrary, the i n n e r layer in group 2 (struts to lumen), could be considered only as neointimal growth. An interobserver variability analysis on the m e a s u r e m e n t of neointimal area was u n d e r t a k e n in 10 patients. Between observers the correlation coefficient was 0.96 and the SEE was 0.4 (observer 2 = -0.36 + 1.15 x observer 1). The neointimal layer at the treated segment was compared with the i n t i m a l echoes recorded at reference segments. For comparative purposes, we considered the cutoffpoint of abnormally increased thickening of the treated segment as m a x i m a l neointimal thickening of>0.5 mm. This value exceeded the upper limit o f i n t i m a l echoes observed at reference segments. Factors associated with a n increased (>0.5 mm) thickening at follow-up were studied. The neointimal ring at the treated segment was considered symmetrically thickened if no differences in its thickness >0.3 m m were observed across the arterial circumference; a n asymmetric neointimal repair was considered when differences in thickness of the ring were >0.3 mm. The brightness of

echoes from the i n n e r echogenic ring was considered as significant (similar to the periadventitial echoes) or mildly intense (less bright t h a n periadventitial echoes) (Fig. 2). Statistics. Data are expressed as m e a n _+ SD. Quantitative data were compared by a Student-Fisher u n p a i r e d t test and qualitative variables with a chi-square or exact Fisher test. Analysis of variance was used to determine whether quantitative ultrasonic follow-up data differed among the three groups of patients. Correlations between quantitative variables were performed by s t a n d a r d lineal regression studies; the linear equation, the correlation coefficient, and its significance were obtained. A multivariate analysis was performed by stepwise logistic regression. Significant i n d e p e n d e n t factors were identified a n d expressed with the p-coefficient and s t a n d a r d error. A p value <0.05 was considered significant. RESULTS A n g i o g r a p h i c observations. T a b l e I I s h o w s t h e a n g i o g r a p h i c r e s u l t s i n all t h r e e g r o u p s a s e v a l u a t e d b y changes observed in MLD and percent stenosis, a c u t e g a i n , a n d l a t e loss i n e a c h g r o u p . T h e i m m e d i a t e r e s u l t w a s s i g n i f i c a n t l y b e t t e r i n s t e n t e d or resected lesions than in those treated by PTCA. The a c u t e g a i n w a s s i g n i f i c a n t l y d i f f e r e n t (p < 0.01) a m o n g groups. A larger l u m i n a l gain was obtained in p a t i e n t s t r e a t e d b y s t e n t c o m p a r e d w i t h t h a t obt a i n e d i n t h e D C A or P T C A g r o u p s . H o w e v e r , t h e p e r c e n t s t e n o s i s a n d M L D a t follow-up w e r e n o t sig-

214 Sudrez de Lezo 8t al.

February 1997 American Heart Journal

Fig. 2. Examples ofneointimal brightness. A, Brightness is similar to that from adventitial echoes. B, Not similar.

Fig. 3. Serial angiographic and late ultrasound images from patient with severe eccentric stenosis in right coronary artery who was treated by PTCA. B, Immediate angiographic result. C, Right coronary artery at late follow-up. At that moment, 1 (reference segment) and 2 (treated segment) show level of ultrasonic interrogation. nificantly different among the three groups. The late lumen loss was significantly higher in patients treated by stent or DCA t h a n in those treated by PTCA (p < 0.01). In fact, a mild luminal expansion at follow-up was noted in 14 (61%) of 23 patients treated by PTCA (-0.46 _ 0.3 mm) and in 5 (19%) of 27 of patients treated by DCA (-0.34 + 0.25 mm). No luminal expansion was observed in patients treated by stenting. Patients who exhibited a late luminal enlargement at follow-up had smaller MLD after treatment (2.5 _+ 0.6 mm) than those who had some degree of late luminal loss (3.1 __ 0.5 m m ; p < 0.001). The mean acute gain of these later expanded lumens was also significantly smaller (1.85 _+ 0.79 mm) t h a n that from patients who showed late luminal loss (2.4 _ 0.68 mm, p < 0.01). Ultrasonic findings. The ultrasound appearance of the arterial wall at the treated segment was charac-

terized by a three-layered pattern. This pattern was similar to t h a t observed in proximal or distal segments. However, the inner echogenic ring was the most apparent transitional change to be observed at the treated segment when the probe was slowly pulled back from normal distal to proximal segm e n t s ) 3 The thickness of intima at the reference segment averaged 0.22 _+ 0.13 mm. Similar values of normal ultrasonic intima have been reported by others.12 However, the maximal neointimal thickness at the treated segment (0.65 _+ 0.31 mm) was significantly (p < 0.001) greater t h a n at the reference segment. Symmetric thickening of the neointimal layer was observed in 55% of treated segments, whereas 45% of them showed an asymmetric inner layer. The brightness of the inner ring varied (Fig. 2). In 43 patients the echogenicity was similar to that from adventitia, and in 38 the neointimal layer showed less

Volume 133, Number 2 American Heart Journal

Sudrez de Lezo et al. 215

Fig. 4. Serial angiograms (before treatment [A], after treatment [B], and follow-up [el) and late ultrasound observations in patient with significant stenosis in proximal left anterior descending coronary artery who was treated by DCA. At follow-up, 1 (reference segment) and 2 (treated segment) show levels of ultrasound analysis. Asymmetric neointimal layer can be observed. density t h a n the external fibrous echoes. The maximal neointima] thickening oscillated between 0.1 and 1.5 mm, and the neointimal area ranged between 0.97 and 14:9 mm 2 (mean 5.19 _ 3.14 mm2). Table III summarizes quantitative ultrasonic measurements obtained at follow-up in all three groups. Patients treated with PTCA showed a frequent (67%) asymmetric neointima (Fig. 3) t h a t was significantly (p < 0.001) thicker t h a n that observed in the stent or DCA groups (Table III). Patients treated by DCA also had a frequent (73%) asymmetric repair, and the intimal appearance and thickness was clearly different

from t h a t observed at the reference segment (Fig. 4). By contrast, patients treated by stenting had a frequent(87%)symmetricandthinner(p < 0.001)neointimal layer uniformly covering the metallic struts (Fig. 5). Study of factors. Factors associated with an increased neointimal thickening at follow-up were studied. Increased thickness was considered when the maximal neointimal layer was >0.5 mm. According to this limit value, 48 patients had an increased thickening (0.88 _+ 0.19 mm), whereas 33 did not (0.35 _ 0.12 mm, p < 0.001). Table IV shows the

216

February 1997 American Heart Journal

Sudrez de Lezo et al.

Fig. 5. Serial right coronary angiograms (before treatment [A], after treatment [B], and follow-up [C]) and late ultrasound images of patient with diffuse proximal disease and critical lesion in middle third who was treated by stent implantation. At follow-up (C), 1 (treated segment) and 2 (reference segment) show level of ultrasound interrogation. Note thin neointima covering metallic struts. univariate analysis, and Table V shows the multivariate analysis. The neointimal repair was thinner in those patients who obtained better short-term angiographic results (higher MLD and luminal gain after treatment) and in stented versus dilated or resected segments. On the contrary, the repaired neointimal layer was thicker in those patients who exhibited less late luminal loss after treatment and in those evaluated longer after treatment. By multivariate analysis (Table V), the short-term gain immediately after treatment and the time elapsed since then resulted as independent predictors of thinner or thicker neointimal repair, respectively. C o r r e l a t i o n s . The neointimal thickness, evaluated by the cross-sectional area of the inner layer, significantly correlated with angiographic parameters reflecting acute luminal result and its later evolution. There was a significant inverse correlation (r = -0.52, p < 0.001) between MLD after treatment and the neointimal area at follow-up; this correlation was even higher (r = -0.78,p < 0.001) in the PTCA group. A similar relation was observed in acute luminal gain at treatment (Fig. 3), suggesting that the greater the luminal gain, the thinner neointimal repair. However, late luminal loss and the degree of neointimal thickening inversely correlated. In fact, the neointimal area was significantly thicker in patients who exhibited angiographic luminal expansion at follow-up (7.5 -+ 3.6 mm 2) than in those who had some degree of lu~ninal narrowing (4.2 _+ 2.4 mm 2, p < 0.001). Finally, the neointimal area also directly correlated with reevaluation time (r = 0.68, p < 0.001), suggesting an influence of the period of time elapsed after treatment on the degree of the neointimal thickness. This relation was also signifi-

cant (r = 0.71, p < 0.001) when patients treated by PTCA were considered separately. DISCUSSION

This study focused on the ultrasonic characterization of neointimal thickening after successful percutaneous revascularization. Although the remodeling and healing process of a treated coronary segment may evolve in a wide range of responses to injury, in most patients favorable scarring leads to stabilization of the arterial wall and persistent luminal patency. The final neointimal features of successfully remodeled coronary segments might give some insights into the process of normal arterial wall repair, which might help us better understand the abnormal response leading to restenosis. Neointimal thickening as a normal response to injury.

In all our patients a three-layer appearance of the treated segment was observed. The neointimal layer covering the inner arterial surface was a symmetric thin ring or leading edge or, alternatively, it exhibited a thicker and asymmetric distribution along the arterial wall circumference. Echogenic appearance of the layer allowed a clear separation from luminal and medial lucencies; however, its brightness also varied among the subjects (Fig. 2). Similar intracoronary ultrasound findings have been observed in heart transplant recipients. 14, 15 Diffuse cardiac allograft vasculopathy is characterized by mild intimal thickening and lack of response to vasoactive drugs. 15 Symmetric or asymmetric mild thickening of the intima has also been ultrasonically observed years after a favorable evolution of the coronary wall in patients with Kawasaki disease. 1~ All these observations could suggest that mild thickening of the

Volume 133, Number 2

217

Sudrez de Lezo et al.

American Heart Journal

Table IV. Factors influencing increased (>0.5 mm) neointimal thickening (univariate analysis) NT > 0.5 m m (n = 48)

Age (yr) 54 -+ 9 U n s t a b l e condition 40 (83%) Type of lesion (de novo) 38 (79%) Symmetric neointima 21 (44%) Bright neointima 26 (54%) Technique PTCA 21/23 DCA 20/27 Stent 7/31 Time from t r e a t m e n t (mo) 28-+ 25 Angiographic d a t a Reference l u m e n diameter 3.16 -+ 0.5 (mm) MLD before t r e a t m e n t (mm) 0.68 -+ 0.4 MLD after t r e a t m e n t (mm) 2.71 -+ 0.6 MLD at follow-up (ram) 2.68 _+ 0.6 Percent stenosis before 77 -+ 15 treatment Percent stenosis after 15 -+ 12 treatment Percent stenosis at follow-up 16 -+ 13 Acute gain (mm) 2.03 + 0.8 Late loss (mm) 0.03 -+ 0.5 Relative gain (gain/vessel 0.63 -+ 0.2 diameter) Relative loss (loss/vessel 0.001 _+ 0.2 diameter)

N T <<0.5 m m (n = 33)

p Value

55 -+ 9 25 (76%) 29 (88%) 24 (75%) 9 (19%)

NS NS NS <0.05 <0.001

2/23 7/27 24/31 12 _ 7 3.39 _+ 0.6 0.52 3.16 2.82 82

<0.001 <0.1

-+ 0.4 <0.1 -+ 0.5 <0.01 + 0.6 NS _+ 18 NS

7 -+ 10 17 2.64 0,35 0.79

<0.001

<0.01

_+ 12 NS _+ 0.6 <0.001 -- 0.3 <0.001 _+ 0.1 <0.001

0.11 _+ 0.1 <0.001

Acute gain is MLD after treatment minus MLD before treatment. Late loss is MLD after treatment minus MLD at follow-up. NT, Neointimal thickening; NS, not significant.

intima may be the result of a favorable neointimal scarring of the coronary wall observed late after different types of injury of the arterial wall. This inner arterial layer may vary in composition, and its frequently dense, echogenic appearance may presumably be caused by dense, fibrous tissue 13, 17 as a result of matrix formation during the healing process. This inner appearance, however, is not exclusive ofintimal repair but can also be observed in patients with mild coronary disease is, 19 or at sites of vasospasm. 12,2° Aging can also influence intimal thickening. 21 Determinants of neointimal thickening. Normally healed segments may show a wide range of thickening. Factors influencing such a wide degree of thickness could help us better understand the scarring process. According to our findings the neointimal thickening may depend on the immediate angiographic result, suggesting that the greater the initial gain in angiographic lumen, the smaller the expected neointimal area at follow-up. This finding is particularly significant in dilated stenoses. Similar find-

Table V. Multivariate analysis

Gain Time from treatment Constant

B-coefficient

Standard error

Improvement chi-square

p Value

-1.083 0.0798

0.46 0.0379

6.58 6.27

<0.010 <0.012

2.084

1.202

--

--

ings were reported in 28 postmortem coronary arteries subjected to angioplasty at an average of 71 weeks before death. 22 However, late changes in angiographic lumen of the treated segment may also be related to tissue growth in an unexpected manner. Paradoxically, the late luminal loss was more marked in patients with a thinner neointimal layer at followup. Patients with expanded lumen at follow-up particularly displayed a thicker neointima] layer than those with some degree of lumen narrowing. These findings suggest that compensatory vessel enlargement or favorable remodeling may be associated with thicker neointimal repair. Similar behavior has been described as an adaptive-reactive process to atherosclerotic plaque growing in native stenosis in which vessel enlargement follows intimal g r o w t h . 23 After treatment, a dynamic process of adaptation of arterial size to the new postintervention demands takes place as a repair mechanism. This adaptation could range from constriction of the artery 3, 24 to effective compensatory enlargement as a response to the resultant neointimal thickness. Our findings suggest that favorable arterial remodeling with vessel enlargement and endoproliferative repair may contribute to the successful arterial wall repair after treatment, but both mechanisms seem to operate in a close relation. An imbalance between both repair mechanisms might lead to restenosis. The time after cardiac transplantation has been identified as an independent predictor of intimal thickening in cardiac allograft vasculopathy. 15 Our findings also suggest that the time after treatment may also influence the degree of thickness of a repaired neointima. Although this observed relation could be due, in part, to the longer observation period in the PTCA group (who exhibited thicker neointima), the correlation was also significant when considering only patients treated by PTCA. This finding could reflect a possible slow growth of the scar in successfully repaired segments, which applies at least to patients undergoing PTCA, in which the longer follow-up period permitted the analysis of time. Although it is well accepted that the healing process does not last >6 months in most patients, later restenosis has also been identified in approxi-

February 1 9 9 7 American Heart Journal

218 Sudrez de Lezo et al.

20

r= -0.47; p
15

y= 9.59-

0

0

A

O4 Elo

0

0

0

0 o

0 00

E

000

0

O1 o

0

iim

(

~

~

2

0

0

°oO? o

1

I¢

0

3

4

A c u t e gain ( m m ) 20|~-

O =m,,

r= -0.42; p
15 ~

o

y= 5.65-

o - -

0

- 0 O0 0

5

I¢

1.93 x

E

o

10 L;~.~

8

°

F

2.88 x

o

o o

o

II==m el

i

em a.,,.

i

i

i

0

•" -

0

i

-1

i

1

Late loss ( m m )

'oF

r= 0.68; p
15

o 0

101::

o

o

8

0

o o

° I

0

t

20

i

I

40

I

I

60

I

I

80

Follow-up time (months) Fig. 6. Scatterplots of neointimal surface at follow-up against angiographic parameters and time since treatment.

mately 3% of patients. 25 Thus if neointimal thickening after treatment may slightly increase as time goes on, longer follow-up studies may be needed to successfully assess healed segments. This idea particularly applies to patients treated by intracoronary stenting who do not have a compensatory vessel segment enlargement to compensate for a possible slow late inner growth. However, this potential minor increase in thickness throughout time does not usually

compromise late luminal patency. Successfully healed arterial segments after PTCA have been observed to have less incidence of newly developed atherosclerotic coronary artery disease than in other coronary territories. 25-27 The residual burden after successful scarring might therefore evolve into a chronic, long-term cycle in which neointimal growth is usually followed by compensatory vessel enlargement. 23 Influence of the technique. S o m e s t u d i e s 2s, 29 h a v e suggested that the MLD immediately after treatment is a major determinant of late success, regardless of the technique used. The degree and type of arterial wall injury caused by a specific revascularization technique may possibly influence the subsequent repair process. However, posttreatment evaluation of the wall injury is difficult. In patients treated with DCA the depth of the cut can be assessed by the identification of media or adventitia layers in the retrieved material. In stented or dilated plaques the depth of injury cannot be estimated in patients, and so the influence of the selected technique in grading the injury remains difficult to eva]uate. Our findings suggest that the type of treatment may also influence the healing process because the late neointimal appearance had some apparent differences among techniques used. Whereas the stented segment normally shows a symmetric, thin ring covering the inner arterial surface at follow-up, dilated or resected segments frequently exhibit an asymmetric and thicker ring across the arterial circumference. Although the neointimal thickness in healed stented segments was found to be thin, this arterial wall may not spontaneously expand or contract. 3° By contrast, segments treated with PTCA or DCA usually have a thick inner ring covering a wall that may expand or contract whether the repair has been favorable (Fig. 6) or unfavorable, 1, 3, 24 respectively. Thus compensatory vessel enlargement in a thick inner scar may be an important factor contributing to the healing process in dilated plaques, whereas in stented segments the response is mainly endoproliferative, covering the struts. 3° In patients treated with DCA, in which part of the plaque is resected and the arterial segment is also dilated, 4, 31, 32 both favorable remodeling and endoproliferative mechanisms seem to operate. Limitations of the study. This is a retrospective study in which we tried to compare the angiographic evolution of patients with late luminal patency with the ultrasound characterization of the neointimal thickening after a favorable healing process. Patients were selected for the study during a 5-year period since the use of intracoronary ultrasound examina-

Volume 133, Number 2 American Heart Journal

tion in our laboratories is routine. The observed baseline differences among groups were partly due to technical considerations. For instance, the arterial size was smaller in patients undergoing PTCA. However, the long-term follow-up time (in patients treated by PTCA) is only explained by the longer implantation time of the technique. Despite these differences among groups the size of each group permitted a separate analysis to find inner peculiarities of the scarred neointima, depending on the different revascularization mechanisms. However, serial luminal measurements were obtained angiographically, not ultrasonically. The changes in lumen size after treatment were compared with the late neointima] appearance to search angiographic predictors of the degree of inner thickening after successful repair. In addition, ultrasound evaluation of the neoendothelial layer did not allow for a quantitative analysis of echo density. Despite the limitation of a binary qualitative differentiation, we found that thickness was associated with brightness; thus, as suggested by previous publications, i3,17 the increased brightness may reflect accumulation of dense, fibrous tissue in the repaired neointima as a frequent scar appearance. Furthermore, whether the covering neoendothelium of the healed segment is able to react to vasoactive drugs would be important to know. In our study, nitroglycerin administration to avoid irritative spasm during inspection prevented us from obtaining data on vascular reactiveness. Despite these limitations, we think that this study provides additional information on the process of coronary wall repair after successful treatment. The favorable healing process seems to be based on the equilibrium between two reparative mechanisms: endoproliferative repair and dynamic adaptation of the arterial wall to the final inner burden.

Sudrez de Lezo et al.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20. REFERENCES

1. Glagov S. Intimal hypeI~]asia, vascular modeling and the restenosis problem. Circulation 1994;89:2888-91. 2. Kakuta T, Currier JW, Houdenschild CC, Ryan TJ, Faxon DP. Differences in compensatory vessel enlargement, not intimaI formation account for restenosis after angioplasty in the hypercholesterolemic rabbit model. Circulation 1994;89:2809-15. 3. Post MJ, Borst C, Kuntz RE. The relative importance of arterial remodeling compared with intimal hyperplasia in lumen renarrowing after balloon angioplasty. A study in the normal rabbit and the hypercholesterolemic Yucatan micropig. Circulation 1994;89:2816-21. 4. Braden GA, Herrington DM, Dowries TR, Kutcher MA, Little WC. Qualitative and quantitative contrasts in the mechanisms of lumen enlargement by coronary balloon angioplasty and directional coronary atherectomy. J Am Cell Cardiol 1994;23:40-8. 5. Rodriguez A, Santaera O, Larribau M, Sosaliprandi MI, Palacios IF. Early decreases in luminal diameter predict late restenosis after successful coronary angioplasty. Am J Cardiol 1993;71:1391-5. 6. Forrester JS, Fishbein M, Helfant R, Fagin J. A paradigm for resteno-

21.

22.

23.

24.

25.

219

sis based on cell biology: clues for the development of new preventive therapies. J Am Cell Cardiol 1991;17:758-69. Selzer RH. Precision and reproducibility of quantitative coronary angiography with application to controlled clinical trials: a sampling study. J Clin Invest 1989;83:520-6. Weintraub WS, Ghazzal ZMB, Cohen CL, Douglas JS, Liberman H, Morris DC, et al. Clinical implications of late proven patency after successful coronary angioplasty. Circulation 1991;84:572-82. Mehta VY, Jorgensen MB, Raizner AE, Wolde-Tsadik G, Mahrer PR, Mansukhano P. Spontaneous regression ofrestenosis. A~ angiographic study. J Am Cell Cardiol 1995;26:696-702. Brown G, Albers JJ, Fisher LD, Schaefer SM, Lin JT, Kaplan C, et el. Regression of coronary artery disease as a result of intensive lipid-lowering therapy in men with high levels of apolipoprotein B. N Engl J Med 1990;323:1289-98. Waters D, Higginnson L, Gladstone P, Kimball B, Le May M, Boccuzzi SJ, et al. Effects on monotherapy with an HMG-CoA reductase inhibitor on the progression of coronary atherosclerosis as assessed by serial quantitative arteriography: the Canadian Coronary Atherosclerosis Intervention Trial. Circulation 1994;89:959-68. Yamagishi M, Miyatake K, Tamal J, Nakatani S, Koyama J, Nissen SE. Intravascular ultrasound detection of atherosclerosis at the site of the focal vasospasm in angiographically normal or minimally narrowed coronary segments. J Am Cell Cardiol 1994;23:352-7. Su~rez de Lezo J, Romero M, Medina A, Pan M, Pavlovic DJ, Vaamonde R, et al. Intracoronary ultrasound assessment of directional coronary atherectomy: immediate and follow-up fmdings. J Am Cell Cardiol 1993;21:298-307. A~derson TJ, Meredith IT, Uehata A, Mudge Gift, Salwyn AP, Ganz P, et el. Functional significance of intimal thickening as detected by intravascular ultrasound early and late after cardiac transplantation. Circulation 1993;88:1093-100. Escobar A, Ventura HO, St~pelton DD, Mehra MR, Ramee SR, Collins TJ, et al. Cardiac allograft vasculopathy assessed by intravascular ultrasonography and nonimmuno]ogic risk factors. Am J Cardiol 1994;74:1042-6. Sugimura T, Kate H, Inoue O, Fukuda T, Sate N, Ishii M, et el. Intravascular ultrasound of coronary arteries in children. Assessment of the wall morphology and the lumen after Kawasaki disease. Circulation 1994;89:258-65. Rasheed Q' Dhawale P' Anders°n J' Hodgs°n JM' Intrac°r°nary ultrasound-defined plaque composition: computer-aided plaque characterization and correlation with histologic samples obtained during directional coronary atherectomy. Am Heart J 1995;129:631-7. Porter TR, Sears T, Xie F, Michels A, Mata J, Welsh D, et al. Intravascular ultrasound study of angiographically mildly diseased coronary arteries. J Am Cell Cardiol 1993;22:1858-65. Alfonso F, Macaya C, Goicolea J, Ifiiguez A, Hernandez R, Zamorano J, et al. Intravascular ultrasound imaging of angiegraphically normal coronary segments in patients with coronary artery disease. Am Heart J 1994;127:536-44. Zeiher AM, Schachinger V, Hohnloser SH, Saurbier B, Just H. Coronary atherosclerotic wall thickening and vascular reactivity in humans. Elevated high-density lipoprotein levels ameliorate abnormal vasoconstriction in early atherosclerosis. Circulation 1994;89:2525-32. Fitzgerald PJ, St. Gear FG, Connolly AJ, Pinto FJ, Billingham ME, Popp RL, et al. Intravascular ultrasound imaging of coronary arteries: Is three layers the norm? Circulation 1992;86:154-8. Farb A, Virmani R, Atkinson JB, Anderson PG. Long-term histologic patency after percutaneous transluminal coronary angioplasty is predicted by the creation of a greater lumen area. J Am ColI Cardiol 1994;24:1229-35. Glagov S, Weisenberg E, Zarins CK, Stankunavicurs R, Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 1987;316:1371-5. Mintz GS, Kovach JA, Pichard AD, Kent K, Satler LF, Popma JJ, et al. Geometric remodeling is the predominant mechanism of late lumen loss after coronary angioplasty [abstract]. J Am Cell Cardio] 1994;23: 138A. Ghazzal ZMB, King SB, Douglas JS, Weintraub WS. Late angiographic status of coronary angioplasty site which was <50% narrowed 4 to 12 months after successful angioplasty. Am J Cardiol 1994;73:892-4.

220

Sudrez de Lezo et al.

26. Kitazume H, Kubo I, Iwama T, A~eishi Y. Long-term angiographic follow-up of lesions patent 6 months after percutaneous coronary angioplasty. Am Heart J 1995;129:441-4. 27. Medina A, Meli~n F, Su~rez de Lezo J, Pan M, Romero M, Hern~ndez E, et al. Protective effect of successful wall healing after balloon angioplasty on progression of coronary atherosclerosis [Abstract]. Eur Heart J 1995;16:180. 28. Kuntz RE, Safian RD, Carrozza JP, Fishman RF, Mansour M, Balm DS. The importance of acute luminal diameter in determining restenosis after coronary atherectomy or stenting. Circulation 1992;86:182735. 29. Kuntz RE, Gibson M, Nobuyoshi M, Balm DS. Generalized model of

February 1997 American Heart Journal

restenosis after conventional balloon angioplasty, stenting and directional atherectemy. J Am Coll Cardiol 1993;21:15-25. 30. Painter JA, Mintz GS, Wong SC, Popma JJ, Pichard AD, Kent K_M,et al. Serial intravascular ultrasound studies fail to show evidence of chronic Palmaz-Schatz stent recoil. Am J Cardiol 1995;75:398-400. 31. Matar FA, Mintz GS, Farb A, Douek P, Pichard AD, Kent KM, et al. The contribution of tissue removal to lumen improvement after directional coronary atherectemy. Am J Cardiol 1994;74:647-50. 32. Tenaglia AN, Buller CE, Kisslo KB, Stack RS, Davidson CJ. Mechanisms of balloon angioplasty and directional coronary atherectomy as assessed by intracoronary ultrasound. J Am Col] Cardiol 1992;20:68591.

AVAILABILITY OF JOURNAL BACK ISSUES As a service to our subscribers, copies of back issues of the AMERICAN HEART JOURNAL for the preceding 5 years are maintained and are available for purchase from Mosby until inventory is depleted at a cost of $12.00 per issue. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one-third off on quantities of 24 or more. Please write to Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, or call (800)453-4351 or (314)453-4351 for information on availability of particular issues. If unavailable from the publisher, photocopies of complete issues may be purchased from UMI, 300 N. Zeeb Rd., Ann Arbor, MI 48106, (313)761-4700.