Increased Thallium Lung Uptake After Exercise in Isolated Left Anterior Descending Coronary Artery Disease

PETER LIU, MD, MARLA KIESS, MD, ROBERT D. OKADA, MD, H. WILLIAM STRAUSS, MD, PETER C. BLOCK, MD, GERALD M. POHOST, MD,

and CHARLES A. BOUCHER, MD

Increased thallium lung uptake during exercise correlates with an exercise-induced elevation In left ventrtcular fllllng pressure. This findlng was analyzed in 48 patients with l-vessel left anterior descending (LAD) coronary artery disease (CAD) before and after percutaneous transluminal coronary angio- plasty. Patients were separated into 2 groups: 13 (27 016 ) patients with increased (group 1) and 35 patients with normal thallium lung uptake (group 2). Compared with group 2, group 1 patients had more severe LAD luminal diameter narrowing (90 f 8% vs 81 f 11% [mean f standard deviation], p <0.003); slower clearance of thallium from the LAD

territory segments (half-life 13.5 f 8.0 vs 8.5 f 5.0 hours, p <0.007); and more abnormal thallium segments per patient by quantitative criteria (p <0.02). After angioplasty, thallium lung uptake became normal in 12 of 13 patients in whom it was previously elevated. Increased thallium lung uptake after exercise occurs in l-vessel LAD disease, and is related to the severity of ischemla by thallium scan and the severity of stenosis by angiography. Therefore, increased thallium lung uptake identifies a subset of patients with l-vessel CAD with a greater amount of myocardium at risk.

(Am J Cardiol 1985;55:1489-1473)

Increased lung uptake after exercise thallium imaging has been correlated with an exercise-induced elevation in left ventricular filling pressure.l12 This phenomenon is associated with multivessel coronary artery disease (CAD), prior myocardial infarction and a worse prog- nosis.a4 Nevertheless, it may also occur in l-vessel CAD,ls which traditionally is considered to have a more favorable prognosis, even though in the subset of pa- tients with left anterior descending artery (LAD) ste- nosis this contention has been challenged.5 Therefore, the significance of this observation in the setting of l- vessel CAD is uncertain. Thus, we studied thallium scans from 48’ patients undergoing percutaneous tran- sluminal coronary angioplasty (PTCA) from l-vessel LAD disease compared patients with increased thallium lung uptake after exercise to those without and exam- ined the change after successful reperfusion with PI’CA.

From the Card&c Unit and Division of Nuclear Medicine, Massachusetts General Hospital, Boston, Massachusetts. This study was supported in part by U.S. Public Health Service Grants HL32953 and HL26215, Natiinal Institutes of Health, Bethesda, Maryland. Dr. Pohost’s current affiliation: Division of Cardiovascular Disease, University of Alabama Medical Center, Biiirgham, Alabama. Manuscript received November 30, 1984; revised manuscript received February 20, 1985, accepted February 22, 1985.

Address for reprints: Charles A. Boucher, MD, Cardiac Unit, Mas- sachusetts General Hospital, Boston, Massachusetts 02114.

The purpose was to determine whether increased thallium lung uptake in l-vessel CAD has similar im- plications as when this finding occurs with multivessel CAD.

Methods Patients: Forty-eight patients with l-vessel LAD disease

who underwent successful PTCA at the Massachusetts Gen- eral Hospital were studied. There were 35 male and 13 female patients, and the average age was 50 f 9 years (mean f stan- dard deviation). Eleven had a previous myocardial infarction in the LAD territory. The patients were receiving a variety of antianginal medications: 24 were receiving /3-blockers, 23 long-acting nitrates and 12 had calcium channel blockers. These medications were not discontinued for the exercise test. Twelve patients had documented systemic hypertension.

Patients were excluded from the study if they had unstable angina, chronic congestive heart failure, recent myocardial infarction, previous cardiac surgery, multivessel CAD, other heart diseases or systemic illness, including obstructive lung disease. All patients were stable enough to undergo exercise testing within 2 weeks before and after PTCA.

Coronary angioplasty: All 48 patients had significant LAD disease, defined as a 50% or greater diameter luminal narrowing on a coronary arteriogram, as judged visually by 2 independent observers. The narrowing was 50 to 75% in 4, 75 to 90% in 14 and more than 90% in 30. In 43 patients, the lesions were located proximal to the first diagonal branch and

1470 THALLIUM LUNG UPTAKE AFTER EXERCISE

TABLE I Data from 13 Patients with Increased Thallium Lung Activity

Thallium Data

No. of No. of Previous Rest Ex No. of Transient Persistent

Pt MI LVEF LVEF Defects Defects Defects

1 No 0.69 0.72

3 Yes No 0.66 0.66 0.65 0.75

1 0 1

2 i i

x 4 No 0.66 0.57 l 0 5 No No 0.65 0.64 0.62 0.52 : t 1

ii Yes No 0.57 0.60 0.59 0.67 s 32 ii 1 9 No 0.54 0.47 1

:: No No 0.53 0.52 0.57 0.51 f 4 4 f x 12 13

Yes 0.50 0.44 0.47 0.45

5 Yes 4

4 i

1 3

Ex = exercise; LVEF = left ventricular ejection fraction; MI = myocardial infarction.

in 4 they were located distal to the first diagonal branch, 1 patient had both proximal and distal lesions.

The PTCA was perforlped in a conventional manner, with the dilating balloon inflated 3 to 4 times for 30 to 60 seconds each at 5 to 10 atmospheres of pressure. PTCA was considered successful if there was a greater than 20% angiographic lu- minal reduction (previously published criteria from the Na- tional Heart, Lung, and Blood Institu@). The average di- ameter stenosis for patients in this group decreased from 83 f 11% to 29 f 14% (p <O.OOl), and translesional pressure gradient decreased from 52 f 14 to 20 f 7 mm Hg (p <O.OOOl).

Radionuclide image acquisition: Exercise thallium imaging: All patients underwent a symptom-limited treadmill exercise test according to Bruce protocol within 1 week before PTCA. At peak exercise, 2.0 mCi of thallium-201 were injected intravenously, and the patient continued to exercise for one additional minute. Immediately after exercise termination, sequential B-minute images were recorded in the anterior, 45” left anterior oblique and 70” left anterior oblique views as previously reported.7-g A second set of B-minute delayed im- ages was obtained in the same views approximately 3 hours after exercise. Repeat exercise thallium study was performed within 2 weeks after PTCA.

Gated blood pool imaging: After the delayed thallium im- ages were recorded, supine rest and exercise gated blood pool scans were recorded using a previously published protocol, and modified in vivo labeling of the patient’s red blood cells with 20 mCi of technetium-99m pertechnetate was used.‘OJl

Radionuclide image analysis: Thallium scan interpre- tation: The serial thallium images were reviewed by 3 expe- rienced observers unaware of both the clinical data and whether images were acquired before or after PTCA. A pre- viously reported qualitative scoring system was used.8 The left ventricle on each view was divided into 3 segments, and therefore 9 segments were analyzed in each patient. Each segment was graded on a 5-point scale from 0 (no thallium uptake) to 2 (normal thallium uptake), with 0.5-grade incre- ments permitted. The results of the 3 observers were averaged. Persistent defects were defined qualitatively as regions with averaged score of 1.5 or less showing less than a O.bgrade in- crease between the initial and delayed images. Transient de- fects were those that showed a 1.0 or more grade increase on the delayed scans reaching an average grade score of 1.6 or higher.

The thallium images were also analyzed by computer quantitation.ll An ellipse was initially placed by the operator over 1 of the images, then the serial thallium images were realigned by translational and rotational algorithms. Inter- polative background subtraction was carried out for each pixel, and the valve planes were excluded. The left ventricular myocardium was separated into 5 equal segments and the thallium content within each myocardial segment averaged. Therefore, 15 segments are analyzed. The relative thallium content in each region was calculated after normalizing the counts in the 3 X 3 pixel region with the most counts in the initial images to 100%. Clearance was determined for each myocardial segment by fitting the serial thallium content values to a monoexponential curve, and expressed as a half-life of thallium in the myocardium in hours. For this study, the LAD territory was considered to be segments 2 and 3 (ante- rolateral and apical) on the anterior view, segment 4 (septum) on the 45O left anterior oblique view, and segments 3 and 4 (anterior and apical) on the 70° left anterior oblique view. The lower limit of normal of percent thallium content for the 15 segments in our laboratory is 53,82,79,75 and 44 for segments 1 to 5 in the anterior view, 56,83,77,73 and 46 for segments 1 to 5 in the 50° LAO view and 58,83,77,78 and 54 for seg- ments 1 to 5 in the 70’ LAO view. The clearance is first nor- malized for peak exercise heart rate, and the upper limit of normal is 5 hours.12

Calculation of thallium lung uptake: The thallium lung uptake was calculated from the anterior view of the immediate postexercise images. Using regions of interest manually de- termined by experienced observers, average counts per pixel were obtained from the upper left lung field and the brightest myocardial segment in the anterior view. Lung thallium ac- tivity was taken as the ratio of average lung counts over that of the brightest myocardial segment. The upper limit of nor- mal was defined previously from our laboratory, and varied with the peak exercise heart rate and propranolol use in each individual patient.9 The patients were classified based on these criteria into 2 groups: those with increased thallium lung uptake and those with normal thallium lung uptake for pur- poses of comparison.

Gated blood pool scan analysis: The gated blood pool scans were reviewed by 3 blinded observers and each view was sep- arated into 3 myocardial segments. Wall motion in each seg- ment was rated on a scale from -1 to 3 (-1 = dyskinesia, 0 = akinesia, 1 = severe hypokinesia, 2 = mild hypokinesia and 3 = normal), and the result of the 3 observers were averaged.‘3 The ejection fraction was calculated using a commercially available count-based, semiautomatic program using back- ground subtraction and a second-derivative algorithm for edge detection in the left anterior oblique image.

Statistical analysis: Comparison of the associated factors was done using l-way analysis of variance and the Newman- Keuls multiple-comparison test. Significant factors associated with lung thallium uptake were initially identified using univariate regression analysis, and the independent variables were further ranked by a stepwise logistic regression analysis (BMDP program PLR, BMDP Statistical Software, Inc.). Comparisons of proportions were done using the chi-square or Fisher’s exact test.

Results Increased vs normal thallium lung uptake:

Thirteen of 48 patients (27%) had increased thallium activity. Their individual data are listed in Table I, and they were compared to data from the remaining 35 pa- tients with normal lung activity (Tables II and III). There were no differences between the patients with

June 1,1985 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 55 1471

TABLE II Comparison of Clinical and Anglographlc Data in increased and Normal Thallium Lung Uptake Groups

Increased Uptake (n = 13)

Normal Uptake (n = 35) p Value

Sex Male 11 24 Female

51:9 501: Age (v) 10 History of hypertension 10 History of infarction 1 7 Medications

Beta blockers Calcium blockers

; y;y Qo

1: W&; 0

Coronary angiography Pre-PTCA transluminal gradient 53f 10 52 f 16

(mm 4) Pm-PTCA luminal stenosis (%) 90 f 6 8lfll Location of stenosis

Proximal to first septal 1; w;y 3; W&$) Distal to first septal 0 Both 1 (So/)

Exercise testing Peak heart rate (beats/min) 100 f 13 122f5

Peak systolic pressure (mm Hg) 159 f 22 165 f 25 Exercise duration (min) 7.6 f 2.9 9.4 f 4.1 Maximum ST depression (mm) 1.5 f 2 0.6 f 1.6

NS = not significant; PTCA = percutaneous transluminal coronary angioplasty.

NS

NN: NS

R

NS

<0.003

NS NS

<O.Ol

E NS

TABLE iii Comparison of Radionuciide Data Between increased and Normal Thallium Lung Uptake Groups

Increased Normal Uptake Uptake

(n = 13) (n = 35) p Value

Thallium Scan

No. of abnormal TI segments/patient by visual assessment

No. of abnormal TI segments/patient by quantitative TI content % persistent defects

Mean TI washout half-life (hours)

Mean maximal myocardial

3.0 f 1.5 1.6 f 1.6 <0.02

2.5 f 1.1 1.7 f 1.0 <0.02

7% 13.5 f 6.0 6.5yI365.0 Go7

347 f a4 391 f 109 NS counts/pixel

Mean lung region countsloixel

218 f 66 169 f 51 -co.05

Dated Blood Pool Scan

Rest ejection fraction (“/I

Peak exercise ejection fraction (%) % of pts with f0.05 increase in ejection fraction with exercise

0.60 f 0.06 0.62 f 0.09 NS

0.62 f 0.11 0.61 + 0.09 NS

31% 33% NS

NS = not significant; TI = thallium.

high and those with normal thallium lung uptake in age, incidence of previous myocardial infarction, trans- lesional gradient or ejection fraction at rest, and inci- dence of high blood pressure. However, patients with high thallium lung uptake had more severe LAD lumi- nal narrowing, lower peak exercise heart rate, more abnormal LAD thallium segments by visual and quan- titative content criteria, and slower clearance of thal- lium from the LAD myocardial segments. There was no relation between increased thallium lung uptake and the rest or exercise left ventricular ejection fractions.

Stepwise logistic regression analysis: Univariate analysis of a pool of 15 variables identified the following set of variables that correlated with the presence of in- creased lung thallium activity: peak exercise heart rate, LAD luminal stenosis, the number of LAD segments with abnormal thallium content by quantitative criteria, the quantitative thallium clearance rate in the same segments, and the number of abnormal thallium seg- ments by visual criteria. Stepwise logistic regression analysis was then performed based on these variables. All the variables except quantitative thallium clearance

1472 THALLIUM LUNG UPTAKE AFTER EXERCISE

TABLE IV Significant Variables That Correlated with the Presence of Increased Lung Thallium Activity

Univariate Analysis

Variables Peak exercise heart rate

p Value 0.015 _._ .-

TI clearance from LAD segments Pre-PTCA stenosis

0.016 0.017

No. of abnormal LAD segments by quantitative TI content

No. of abnormal LAD segments by visual criteria

0.047

0.05

Stepwise Logistic Regression Analysis

Significant variables found Pm-PTCA stenosis

Final p(x*)

Peak exercise heart rate <0.02 No. of LAD segments with

abnormal TI content by quantitative criteria

LAD = left anterior descending coronary artery; PTCA = percuta- neous transluminal coronary angioplasty; TI = thallium.

and number of abnormal segments by visual criteria were found to be significant and independent predictors of increased lung thallium uptake (Table IV), and were ranked in the following order: peak exercise heart rate achieved, percent LAD stenosis, and the number of abnormal LAD thallium segments by quantitative content criteria, with a goodness-of-fit chi-square probability of <0.02.

Findings after angioplasty: In 12 of the 13 patients who had increased thallium lung uptake before PTCA, including all 9 patients without prior myocardial in- farction, uptake returned to normal after PTCA. One of the 4 patients with prior myocardial infarction uptake remained abnormal (patient 13, Table I). This patient also had persistent extensive qualitative and quanti- tative thallium abnormalities after PTCA, suggesting incomplete revascularization or persistent preexisting mvocardial scar. Figure 1 is an examnle of the changes seen in the thallium lung uptake *before and after PTCA.

Pre PICA

Post PEA

I N ITIAL DELAYED

Ant

Ant

FIGURE 1. Example of a patient with increased lung thallium uptake immediately after exercise before Percutaneous transluminal coronary angioplasty (PTCA) (top). Uptake returned to normal after PTCA (bot- tom). Ant = anterior.

Discussion Increased lung thallium activity after exercise in

patients with CAD has been attributed to transient exercise-induced left ventricular failure. Either an in- crease in pulmonary transit time or an elevation of left atria1 pressure has been shown experimentally to in- crease thallium extraction by the lung.14 In a wide spectrum of patients with CAD, abnormal increases in thallium lung activity have been correlated with the severity and extent of underlying CAD, and related to increased pulmonary artery wedge pressure during ex- ercise.2 Its presence is also associated with a poorer progr~osis.~~~J~ Although it is known to occur in l-vessel CAD, its prevalence and significance in this setting is not known. In our population of patients with l-vessel LAD disease, increased thallium lung uptake ratio oc- curred in 29% of patients. When present, it also indi- cated more severe CAD, being correlated with greater luminal stenosis by angiography and greater degree and extent of ischemia by thallium imaging. This sug- gests that its significance is consistent with previous observations.

Using stepwise logistic regression analysis, peak heart rate achieved during exercise, luminal stenosis and number of LAD segments with low thallium content were the most significant independent determinants of the probability of increased thallium lung uptake. The correlation with peak heart rate is of interest because of our previous observations that peak heart rate also affected normal thallium lung uptake.g This variable had already been considered in the criteria for deter- mining abnormal lung uptake in these patients. From these data, heart rate therefore appears to have a dual association with thallium lung uptake, suggesting 2 mechanisms of increase in the left ventricular filling pressure. In normal persons thallium uptake is greater at higher heart rates, probably because of shorter pul- monary transit times. In patients with CAD, abnormal increases in thallium lung uptake are more frequent when there is a low achieved heart rate. This is probably the result of an inadequate heart rate response or chronotropic incompetence occurring with a more se- vere ischemic insult.

The increased thallium lung uptake was unrelated to differences in global ejection fraction at rest or during exercise. One explanation for this observation is that thallium lung uptake is more sensitive than the ejection fraction response to ischemia-induced left ventricular dysfunction, as increased lung uptake reflects the combined impact of abnormal systolic and diastolic abnormalities during ischemia, whereas ejection frac- tion reflects global systolic function alone. An alterna- tive explanation could be compensatory hyperkinesia in the non-LAD myocardial segments, masking changes in ejection fraction during exercise in this small, select group of patients with l-vessel CAD.

After PTCA with correction of coronary stenosis, the increased lung activity and, hence, the exercise-induced increase in filling pressure resolved in all but 1 patient. This confirms that increased thallium lung uptake during exercise and the coronary stenosis with associ-

June 1, 1985 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 55 1473

ated &hernia during exercise are probably causally related. It also suggests that exercise-induced elevations in left ventricular filling pressure in our patients mainly results from exercise-induced ischemia with or without scar, but not just from scar alone. However, ischemia may not be essential for increased lung activity in all patients with l-vessel LAD disease. Our sample is bi- ased because only patients with LAD disease referred for PTCA were included. Such patients are likely to have angina. Other patients, such as those with a total LAD occlusion, large anterior wall scar and severe LV dysfunction at rest, could also have increased lung ac- tivity. Such patients would be unlikely to be referred for PTCA and included in our study, but do comprise part of the spectrum of l-vessel LAD disease.

Acknowledgment: We thank Gerry Cotter, Maureen McCarthy and William Shea for their technical assistance.

References Kushner FG, Okada RD, Klrshenbaum HD, Boucher CA, Strauss HW, Pohost GM. Luna thallium-201 Wake after stress testino in oatients with coronary artery disease. Circulation 1981;63:341-347. - ’ Boucher CA, Zir LM, Beller GA, Okada RD. McKuskk KA, Strauss HW, Pohost GM. Increased luna uotake of thallium-201 durlna exercise mvo- cardiai imaging: clfnlcal, h&&odynamic and angiographi; implications in patients with coronary artery disease. Am J Cardiol 1980;46:189-198. Wllson RA, Okada RD, Boucher CA, Strauss HW, Pohost QM. Radionu- clk&+termined changes in pulmonary blood volume and thallium lung ;&ak;4: patients with coronary artery drsease. Am J Cardiol 1983;51:

Gibson RS, Watson DD, Carabello BA, Holt ND, Belier GA. Clinical impli- cations of increased lung uptake of thallium201 during exercise scintigaphy

5.

6.

7.

0.

9.

10.

11.

12.

13.

14.

15.

2 weeks after myocardial infarction. Am J Cardiol 1982;49:1586-1593. Rahlmtoola SH. Left main equivalence is still an unproved hypothesis but proximal left anterior descending coronary artery disease is a “high-risk” lesion (Editorial). Am J Cardiol 1964;53:1719-1721. Kent KM, Bentlvoglio LO, Block PC, Cowley MJ, Dorros G, Gosselln AJ, Gruntzlg A, Myler RK, Simpson J, Stertzer SH, Wllllams DO, Fisher L, Glllesple MJ, Detre K, Kelsey S, Yullin SM, Mock MB. Percutaneous transluminal coronary angioplasty: report from the registry of the National Heart, Lung, and Blood Institute. Am J Cardiol 1982;49:2011-2020. Brown KA, Boucher CA, Okada RD, Gulney TE, Newell JB, Strauss HW, Pohost GM. Prognostic value of exercise thallium-201 imaging in patients presenting for evaluation of chest pain. JACC:1983;1:994-1001. Okada RD, Boucher CA, Kkshenbaum HK, Kushner FG, Strauss HW, Bbck PC, McKuslck KA, Pohost GM. Improved diagnostic accuracy of thallium- 201 stress test using multiple observers and criteria derived from interob- server analysis of variance. Am J Cardiol 1980;46:619-824. Brcwn KA, Bcucher CA, Gkada RD, Strauss HW, Pohost GM. Guantificaticn of lung thallium-201 activity following upright exercise in normal patients: the importance of peak heart rate and propranoloi usage in defining normal values. Am J Cardiol 1984,53:1678-1682. Callahan RJ, Froalkh JW, McKuslck KA, Leppo J, Strauss HW. A modified method for the in-vivo labeling of red blood cells with Tc-99m: Concise communication. J Nucl Med 1982;23:315-316. Llm YL, Okada RD, Chesler DA, Block PC, Boucher CA, Pohost GM. A new approach to quantitation of exercise thallium201 scintigraphy before and after an intervention: application to define the impact of coronary an- $oolasty on regronal myocardral perfusron. Am Heart J 1984;108:917-

Kaul B, Boucher CA, Newell JB, Chesler DA, Strauss HW, Pohost GM. Computer analysis of thallium images: correlates of coronary artery disease (abstr). Circulation 1964;7O:suppl il:ll-286. Okada RD, Klrshenbaum HD, Kushner FG, Strauss HW, Dlnsmore RE, Newell JB, Boucher CA, Block PC, Pohost GM. Observer variance in the qualitative evaluation of left ventricular regional wall motion and the quantitation of left ventricular ejection fraction using rest and exercise multigated blood pool imaging. Circulation 1980;61:128-136. Blngham JB, McKuslck KA, Strauss HW, Boucher CA, Pohort GM. Influ- ence of coronary artery disease on pulmonary uptake of thallium201. Am J Cardiol 1980;46:821-826. Glbson RS, Watson DD, Craddock GB, Crampton RS, KaNer DL, Denny MJ, Seller GA. Prediction of cardiac events after uncomplicated myocerdtal infarction: a prospective study comparing predischarge exercise thal- lium-201 scintigraphy and coronary angiography. Circulation 1983;68: 321-336.