risk of side branch occlusion during coronary angioplasty

Risk of Side Branch Occlusion During Coronary Angioplasty

BERNHARD MEIER, MD, ANDREAS R. GRUENTZIG, MD, SPENCER B. KING Ill, MD,

JOHN S. DOUGLAS, Jr., MD, JAY HOLLMAN, MD, THOMAS ISCHINGER, MD,

FRED AUERON, MD, and KATHY GALAN, RN

To assess the risk of side branch occlusion during percutaneous transluminal coronary angioplasty (PTCA), 600 consecutive procedures were analyzed. On the basis of pre-PTCA angiograms of 557 patients in whom the balloon was actually inflated, 365 side branches in 302 patients (54% of patients) were deemed in jeopardy. A total of 122 side branches in 102 patients (18 %) originated from the lesion segment itself, i.e., their take-off was nar- rowed (Group I, 33% of side branches at risk), whereas 243 side branches in 214 patients (38 % ) originated from the immediate vicinity of the stenosis in a way that they were subjected to temporary occlusion during balloon dilatation (Group II, 67% of side branches at risk). Patency of side branches was determined by consensus of 2 observers. Cri- teria for occlusion were disappearance, filling by collaterals, or stagnation of flow. After PTCA, 20 of 365 side branches (5 % ) were occluded and asso-

ciated with chest pain in 5 patients, creatine kinase increase in 6, left anterior hemiblock, septal Q waves and transient atrial fibrillation in 1 and nonsustained ventricular tachycardia in 1 of the 20 patients. Exercise tolerance did not decrease. No local predilection for side branch occlusion was evident. Seventeen of 122 side branches (14 % ) occluded in Group I, compared with 3 of 243 (1% ) in Group II (p <O.OOl).

Thus, more than half of the patients who underwent PTCA had side branches at risk for iatrogenic occlusion. Side branches involved in the narrowing (Group I) must be regarded as high-risk side branches, whereas the more frequent side branches adjacent to the narrowing (Group II) can be considered low-risk side branches. The outcome of side branch occlusions was invariably favorable.

(Am J Cardiol 1984;53:10-14)

Percutaneous transluminal coronary angioplasty (PTCA)l has a number of well-known risks.2 One of them is iatrogenic occlusion of side branches in the proximity of the stenosis attempted. This study was initiated to quantitate the risk of permanently occluding such side branches by PTCA.

Methods

Patients: The films of 557 PTCA procedures were analyzed. They derived from 600 consecutive PTCA procedures performed with the previously described technique’ between October 25, 1982, and May 13, 1983, at Emory University Hospital (519 procedures) and Crawford W. Long Hospital (81 procedures). Sixteen procedures (3%) that resulted in emergency bypass surgery were excluded. Twenty-seven

From the Department of Medicine (Cardiology), Emory University School of Medicine, Atlanta, Georgia. Manuscript received July 5, 1983; revised manuscript received September 14, 1983, accepted September 22, 1983.

Address for reprints: Andreas R. Gruentzig, MD, Interventional Car- diovascular Medicine, Emory University Hospital, 1364 Clifton Road NE, Atlanta, Georgia 30322.

procedures (6%) failed, 3 because of inability to reach the stenosis, 22 because of inability to cross the stenosis and 2 because of failure to find the way through a totally occluded artery. They were also excluded. Ten procedures failed because an improvement in diameter of >20% could not be achieved despite repeated balloon fillings. These patients remained part of the study population because their arteries had been subjected to the expansive forces of the balloon. The total failure rate was 9% (53 of 600 procedures), corresponding to a primary success rate of 91% (547 of 600).

Of the included 557 procedures, 74 were double PTCAs (PTCA of 2 individual stenoses in 1 or 2 vessels) and 1 was a triple PTCA. A total of 633 lesions were assessed. Twenty patients had 2 PTCAs during the study period. They were counted each time as an individual patient for the purpose of this paper. The mean age of the patients was 55 years (range 20 to 81); 78% were men.

Table I shows the distribution of the lesions to the main regions of the coronary vasculature.

In the 547 successful procedures there was a mean change in diameter narrowing assessed as previously described3 from 70 to 25% and a mean change in transstenotic pressure gradient from 50 to 13 mm Hg.

IO

January 1. 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 53 11

TABLE I Location of Lesions

Lesions with Lesions with Coronary Side Branch Occluded

Artery n All Lesions n at Risk n Side Branch

LAD 346 (55%) [ lOO%] 199 (64%) [58%] 11 (55%) R 151 (24%) [iOO%] 76 (24%) [50%] 6 (30%) LC 94 (15%) [ lOO%] 3.5 (11%) [37%] 3 (15%) BPG 42 (6%) [lOO%] 4 (1%) [IO%,1 0 (--) Total 633 (100%) [ lOO%] 314 (100%) [5O%j 20 (100%) [3%]

Figures in parentheses denote column percentages: figures in brackets denote row percentages. BPG = bypass graft; LAD = left anterior descending; LC = left circumflex; R = right.

All patients were connected to a telemetry monitor for 12 to 24 hours after PTCA and were repeatedly interrogated for chest discomfort by nurses. ECGs were obtained before and shortly after PTCA and again the next day. Blood samples for creatine kinase (CK) determinations were collected immediately after and approximately 8 and 16 hours after PTCA. The CK-MB fraction was determined in all samples. Most patients underwent an exercise test before and 2 days after PTCA (day of discharge).

Side branches: The side branches in jeopardy were determined on the basis of the angiogram done immediately before PTCA. Only side branches with an estimated diameter of 21 mm were considered. When dilating a secondary vessel (e.g., diagonal branch) close to its origin, the distal portion of the main vessel (e.g., left anterior descending coronary artery) could qualify as “side branch” at risk. Likewise, the proximal portion of a bypassed artery was considered a side branch when PTCA of the anastomosis was performed with the balloon encompassing the end of the bypass graft as well as the native vessel distal to it.

The side branches at risk were divided into 2 groups (Fig. 1). Group I included side branches originating from the stenosis itself, which signifies a diseased take-off. Group II included side branches in the immediate vicinity of the stenosis that were certain to be subjected to temporary occlusion while the balloon was filled.

Patency of the side branches after PTCA was assessed by consensus of 2 cardiologists reviewing the tine film. Occlusion of side branch was defined as disappearance, stagnation of flow, and filling by collaterals.

FIGURE 1. The types of side branches investigated. Group I comprises side branches that originate from the very lesion and group II those that originate from the adjacent normal artery but still from the segment encompassed by the balloon.

Statistics: For comparison between the groups, the chi- square test and the Student t test were used.

Results

Table I is a list of the site of all PTCAs, of those with side branches at risk and of those with occluded side branches. A total of 365 side branches in 302 patients (54% of the 557 patients included) were considered at risk. Table II lists the encountered combinations of le-

FIGURE 2. Disappearance (double arrow) of a right ventricular branch (single arrow) due to percutaneous transluminal coronary angioplasty (PTCA) (left anterior oblique projection).

12 SIDE BRANCHES IN CORONARY ANGIOPLASTY

TABLE II Encountered Combinations of Lesion to Be Dilated and Side Branch at Risk

Lesion/St3 n SB Occluded

LAD/DgB LADlSPB RIRVB

115 105

a4 2 (7%j 0 l(ll%)

RILVB 4 0

RIPDA RCG/PDA

Total 365 20 (5%)

DgB = diagonal branch; LAD = left anterior descending coronary artery: LAG = graft to LAD; LC = left circumflex coronary artery; LVB = left ventricular branch; OMB = obtuse marginal branch; PDA = posterior descending coronary artery; R = right coronary artery; RCG = graft to R; RVB = right ventricular branch: SB = side branches; SPB = septal perforator branch.

sion to be dilated and side branch at risk and their re- spective outcome. Twenty of the 365 side branches (5%) were occluded after PTCA, which was successful and otherwise uneventful in all of them. Table III provides details about these 20 patients. Thirteen were male and 7 were female.

Chest pain after angioplasty was recorded in 5 patients (25%) with side branch occlusion. CK elevation was found in 6 (30%). In all 6, CK-MB was abnormal

TABLE Ill Patients with Side Branch Occlusion

(>5% of total CK) but only in 3 of them total CK was elevated (> 170 mu/ml) (Table III, patients 6,7 and 18). The highest value was 244 mu/ml. The only patient with elevated CK-MB and chest pain (Patient 15) had a normal total CK value. In comparison, CK levels were increased in 15% of patients without side branch occlusion, of whom 3% exceeded double the normal value and 3% had elevated CK-MB levels only.

Patient 1 had permanent electrocardiographic changes. He had an occluded septal perforator branch after successful PTCA of a lesion in the left anterior descending coronary artery, with reduction of diameter narrowing from 78 to 13% and reduction of pressure gradient from 60 to 4 mm Hg. Although he lost the R waves in leads Vi to V:3, these changes were probably due to a concomitantly appearing left anterior hemiblock rather than to an anterior scar. The patient did have chest pain after the procedure but no CK elevation, which further supported the assumption that left anterior hemiblock was the main culprit for the electrocardiographic changes.

Ten of the 20 patients with side branch occlusion had a negative stress test after PTCA. In 4 patients it was equivocal. In 2 patients it was positive but had been positive before. Four patients did not have a stress test for nonmedical reasons. Exercise test results both before and after PTCA were available in 11 patients. In 7 they became negative, in 2 they remained negative, and in 2 they remained positive. No deterioration of exercise test results was found.

Table IV lists the number of patients with 1, 2 or 3 side branches at risk and shows how side branches of

Proof for ECG Pt L/SB Occlusion CP Changes

Group I 1 LADISPB + +’

3’ RIRVB : -

LADISPB D - -

4 RIRVB -

: RIRVB : + -

RIRVB D - -t

; RIRVB S LAD/DgB D - -

9 LADlDgB S -

:: LADlSPB D - -

LADlDgB - -

:: OMBtOMB :

T

-

~:~~~~B S -

:: LAD/DgB z T

- -

:Y LADISPB -

LADISPB : T -

Peak CK mu/ml

102 140 1:

z: 0 0 2:: 0

779 : 44 0

96 91 : 166 0 75

1:: :

113 7;

:o” 0 0

Group II

RIRVB D - -J 244 11 LAD/DgB

z

- -

1:: 0

20 LClOMB - - 8

l Transient atrial fibrillation, permanent left anterior hemiblock, and precordial R loss. r Ventricular fibrillation during PTCA. t Nonsustained ventricular tachycardia 6 hours after PTCA. Abnormal values are italicized. C = side branch filled by collaterals; CP = chest pain after PTCA; D = disappearance of side branch;

DgB = diagonal branch; LAD = left anterior descending coronary artery; LC = left circumflex coronary artery; L/SB = lesion dilated/side branch occluded; MB = CPK-MB fraction; OMB = obtuse marginal branch; R = right coronary artery; RVB = right ventricular branch; S = stagnation of flow in side branch; SPB = septal perforator branch.

January 1, 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 53 13

Group I and II were represented among these patients.

Group I: A total of 122 side branches (33% of jeopardized side branches) in 102 patients (18% of all patients) were classified into Group I, i.e., they originated from the lesion itself and were stenosed. Occlusion occurred in 17 of these 122 side branches (14%): 3 first and 2 second diagonal branches and 5 proximal septal perforator branches were occluded after PTCA of the left anterior descending coronary artery; 5 right ventricular branches were occluded after PTCA of the right coronary artery; and 2 obtuse marginal branches were occluded after PTCA in the left circumflex system. No patient had more than 1 occlusion. No patient had received PTCA to the same site previously. The mean diameter of the occluded branches was 1.4 mm; the largest was 2 mm. The mean preexisting diameter stenosis of the side branches was 56% (range 30 to 90%). Dissection in the dilat,ed vessel as a likely source of oc-

TABLE IV Patients with Jeopardized Side Branches

Patients (n = 557) %

1 jeopardized SB 242 43 2 jeopardized SB 57 10

3 jeopardized SB 3 No jeopardized SB 255 4:

1 SB in Grouo I 70 13 2 SB in Groub I 3 SB in Group I 1 SB in Group II 2 SB in Group II 3 SB in Group II 1 SB each in Group I and II

16 2 <l:

172 31 27

<: 14 2

SB = side branches.

FIGURE 3. Stagnation of flow (double arrow) in a diagonal branch (single arrow) after percutaneous transluminal coronary angioplasty (PTCA) (lateral and right anterior oblique projection).

FIGURE 4. Filling by collaterals from the right coronary artery branch of the left circumflex system (triple arrow) occluded (single double arrows) during percutaneous transluminal coronary angiopl (PTCA) of an obtuse marginal branch (right anterior oblique prc tion).

of a and

lasty jjec-

14 SIDE BRANCHES IN CORONARY ANGIOPLASTY

elusion was visible in 10 instances. Eight of the side branches had disappeared completely (Fig. 2), 7 were still visible but exhibited stagnation of flow (Fig. 3), and 2 filled through collaterals (Patient 12, obtuse marginal branch filled from right coronary artery [Fig. 41, and patient 17, septal perforator branch filled from right coronary artery) (Table III).

The mean outcome of the dilated vessels was char- acterized by an average reduction of the diameter narrowing from 73 f 12 to 19 f 6% and of the pressure gradient from 50 f 15 to 10 f 6 mm Hg. This is comparable to the general outcome in this study or in a random sample of PTCA patients.*

Group II: A total of 243 side branches (67% of jeopardized side branches) in 214 patients (38% of all patients) were classified into Group II, i.e., they originated from the vicinity of the lesion. Occlusion occurred in only 3 of these 243 side branches (1%): 1 diagonal branch was occluded after PTCA of the left anterior descending coronary artery, 1 right ventricular branch was occluded after PTCA of the right coronary artery, and 1 obtuse marginal branch was occluded after PTCA in the left circumflex system (Table III). None of the 3 patients had received PTCA to the same site previously. All 3 side branches were of small caliber (1 mm). Two had preexisting disease (20% and 60% diameter narrowing) independent from the lesion to be dilated. In all 3, a dissection line accounting for the occlusion could be seen. One side branch had disappeared (Patient 18, no preexisting disease) whereas the 2 with preexisting disease showed stagnation of flow (Table III). The mean outcome of the 3 dilated vessels was again comparable to the general outcome (change of degree of stenosis from 67 f 15 to 25 f 12% and of pressure gradient from 57 f 12 to 15 f 8 mm Hg). None of these patients had chest pain after PTCA, but 2 had elevated CK levels (Table III, Patients 18 and 20).

Discussion

Side branches originating from a stenosed segment of a coronary artery are indeed in some jeopardy during PTCA of that segment. About 14% of them became occluded in the described patients. Thus, they may be called high-risk side branches, in contrast to low-risk side branches i.e., side branches not directly involved in the lesion but within the segment encompassed by the dilating balloon. Only 1% of low-risk side branches became occluded secondary to PTCA (p <O.OOl).

Dissection of the dilated artery involving the take-off of the branch seemed to be the prime pathogenetic mechanism. Dissection is more likely to occur in a diseased than in a normal arterial wall. This explains the low incidence of occlusion in side branches originating from a normal segment. In all 3 occlusions of these side branches a dissection line extending from the lesion itself to the take-off of the side branch was visible. Thus, the occlusion was not due to the impact of the balloon at the take-off of the side branch itself. The presence of disease in 2 of these side branches was probably unre- lated to the occlusion, since it did not involve the immediate origin.

Other predilections for side branch occlusion: The dilatations with side branches at risk and those

with occluded side branches were not different in terms of anatomic location (Table I); neither was there a statistically significant difference in sex distribution between patients with side branch occlusion and the entire study population (65% and 78% males, respectively).

The average outcome of the dilated artery itself in terms of stenosis improvement and pressure gradient reduction was not different between the groups.

Sequelae of side branch occlusion: There was an apparent paucity of sequelae in the 20 patients with side branch occlusion. Chest pain was reported in 5 (25%) and elevated CK levels were recorded in 6 (30%). None of the values exceeded normal values by >50%. In only 3 patients were both CK and CK-MB levels elevated (Table III). The incidence of CK elevation was higher in patients with than in those without side branch occlusion, but the difference was not statistically significant.

Chest pain occurred with all 3 most frequently occluded side branches (diagonal, septal perforator and right ventricular branch). Four of 6 patients with elevated CK levels had an occluded right ventricular branch and a good result of the dilated right coronary artery, emphasizing the relative importance of these branches.

Only 1 patient had electrocardiographic changes compatible with myocardial infarction. The fact that these changes may have been attributable to a new left anterior hemiblock and the absence of CK elevation speak against a significant amount of myocardial ne- crosis.

Extension of the hospital stay because of side branch occlusion was exceptional and never exceeded 1 addi- tional day.

Implications: Of course, the described cohort of patients was selected regarding suitability for PTCA. Anticipation of problems due to side branch occlusion was a selection criteria. Therefore, patients with stenoses in or around bifurcations of large vessels were disqualified if complete revascularization of both vessels by PTCA was unlikely. In such cases, bypass grafts to both vessels were recommended.

With this in mind, some conclusions can still be drawn from the described results. Only side branches intimately involved in the lesion to be dilated (high-risk side branches) are at a relevant risk for iatrogenic occlusion. Occlusion of side branches not involved in the stenosis (low-risk side branches) just by contact with the dilating balloon is rare. Significant consequences of side branch occlusions are unlikely if selection of patients is guided by the relevance of the side branches at risk.

References

Gruentzig AR, Senning A, Siegenthaler WE. Nonoperative dilatation of coronary-artery stenosis. N Engl J Med 1979;301:61-68. Dorros G, Cowley MJ, Simpson J, Bentlvoglio LG, Block PC, Bourassa M, Detre K, Gosselln AJ, Gruentzig AR, Kelsey SF, Kent KM, Mock MB, Mullin SM, Myler RK, Passamanl ER, Stertzer SH, Willlams DO. Percutaneous transluminal coronary angioplasty: report of ,compli+ions from the National t+l,e;r, Lung, and Blood lnstltute PTCA Registry. Clrculatlon 1983;67:723-

Meier B, Gruentzig AR, Goebel N, Pyle R, von Gosslar W, Schlumpl M. Assessment of stenoses in coronary angioplasty. Inter- and intraobserver variability. Int J Cardiol 1983;3:159-169. lschinger T, Gruentzlg AR, Hollman J, King S Ill, Douglas J, Meler B, Bradford J,Tankersley R. Should coronary arteries with less than 60% diameter stenosis be treated by angioplasty? Circulation 1983;68:148-154.

risk of side branch occlusion during coronary angioplasty

Documents