LH&fives. This sFudg sought to evaluate the r0k and ittcre- mental value wf atr0pine in a large patient gmup undergOing dobuhmlue press ~h~~~j~~p~y.

I& ‘The use Of atr0pitte to pntentiate dobutamine stress is rtot staudard pructice~ Altltough the utility Of atmpine has been detihed, data On its Iticremental value remain Bimited an exist fOr a r0utine clinical practice setting.

. DOhuhmiue stress e&cardiography was performed in 1,171 patients with use Of a standard prot0&. Atropine (maximal dOse 2.0 mg) was given to 299 pattents (26%) who did not attaiu tatget heart rate. Co:awoaty attgiOgruphy was performed in 10 p&tests (46 received a&opine), 148 ofwlrom were found TV have siguiticrtrtt corutmry artery disease (L7M diameter stetaosis in a maor epicardial vessel, LW% stenosis POr left main c0romuy artery disease). All tests were reviewed iudepettdetttly by experi- eueed observers.

induced Esrhcmia devetopeti~ 70 (DA) required atnqtine beTcue ischemia became evident. Sensitivity Rt detection OT signibnt coronary artery disease was !W% with dolmtamine atoue and 95% after the addition Of atropine. In 66 paFieuts 6th nOrmal nall motion at rest, test sensitivity was 65% before red &% after a&Opine was given. Atropiae use did nut compromise test sue& ficity* New diagnnostie ~~~o~attO& was &t&ted in Ztl (§tt%) of 4O putieuts with angiograpkiic coronary artery disease given atropiue. Fropartioustely more patients with single-vesset disease required a&opine before att ischemic resp~ttse was observed, this appeared related to the higher iscltemic tureshold iu t&se patients.

Resuks. There were ~0 mJOr adverse events. Patients receiving atrpitte Lad a hmer rest heurt rate (65 VS. 74 beats/mitt, p 6 %tWl) and, mOre often received beta+tdwuergir Mockiug agents (4% vs. I&, p < OAltktM). Of 444 patients in when stress-

Concfmim. ~u~e~~tion of heart rate had a modest ingu- en&W On the Overall diaguostic sensitivity of d0butamine stress ~~~a~(io~~p~y in our study cohort. DOwever, it wus particu- larly helpful in putieuts receiving beta-t&t&et% asd thuse with milder ntronaty disease. Despite the use Of 2 t mg Of altlrapine in some patients, this incremental value was not achieved at the expense af :;rCty.

[J .&tt Cdi Cmfid l996;28:5Sl-7)

- -- ----

tiate dabutanrine st-ress is not standard practice (9,X!), possibly because initial validat.imn studies have documcnte.d high diag- nostic sensitivity without the need for a supplementary agent (9,3).

Dobutamiuc strt~s echocardiogranhy has become an cstab- lished noninvasivc test for diagnosing coronary artery disease (l-3). Rcccntly, M&kill et al. (4) proposed atropine USC to augment pharmacologic stress. Because induction of ischemia by r9&tl,ut~tt~litlC IXliCS 1XKlly 011 itdCfjt%tC ChilXlOt~~~9~~ (S-7j1 atropinc should enhance the diagnostic sensitivity of Qbut- amine stress echocardiography. Nowever, data supporting the incremental value of atropine use arc limited, do not exist for a routine clinical practice ond primarily reflect singtc-center experience (4,X). Other investigators remain less convinced of the impottance of heart rate as a determinant of optimal stress levels during dobutamine echocardiography (9). A review of rcccnt publications reveals that the use of atropinr to p&x-

Tia principal purpose of this investigation. which reviews a iarge experience with dobut;unine-atropiill~ stress tcsting~ was to determine the role of adjunctive atropine rise in a cohort represent:?tive of routine stress echocrdrdiugraphic practice, Because benefit should be defined in the context of potential adverse eQcts (12.12), the issue of test safety was also examined.

Mcthards

From the Division of Cwdiwwul;~r Diseases and%ernal Medicine awl Section of Biclstatktics, hkyo Clinic and Mayo Foundatiorr, Rochester, Minne- sota. Dr. Ling w supported by a Fac~ulty Development Program schdmhip from the National University Hospital, Singqnwt! and the MdFo Pwndatitm,

Manuscript rc&ved Scptemher 29,: NY% rzviscd manuscript received April 10, lY!%, accep!ed April 24, lYY6.

&idci~: for correspondence,: Dr, Patti&it A, Pellikkn, Mayo Clinic, 200 %I Street SW, Rochester, Minnesota 55905.

The study cohort included 1,971 consecutive patients who had clinicalfy indicated dobutaminc stress echocardio~r~~hy at the Mayo Clinic from May IY!K! TV, Februaty ‘1994; 572: (74%) received dobutamine on9y (group 19, and 39 (26%) received dobutamine and atropine (group 99). Cardioactive medicatknts wcrc not discontinued before testing. Clinica9ly i~ndicated coronary angiography was performed in 137 group I (96%}~d ,

stress p&&, ‘Fhc implementation of dobutnminr: stress ecllucardiography in our laboratoq has previously been de- scribed (13). Dobutamine was infused incrementall;v from 5 to ~$0 pgJqg body weight per min. If the heart rate was withirt 111 &^;\:5 of’ the tsrget value (85% of predicted maximum for age)t), the iafwh was hcreased to 50 pgkg per min. Othenvise,

atropine sulfate was administered intravenously in 0.25-mg increments at I-min intervals (to a maximal total dose of 1.0 mg during the first 8 months of the study period and 2.0 mg thereaft@ if there were no specific contraindications. Dobut- amine infusion was continiled during atropine administration. End points for lest termination were 1) attajnment cd target ]ycart riitc, 2,) conlpletiorl of slrcss protocol, and 3) obvious uew regional wall motion abnormalities, In the absence of these end points, the test was terminated, at the discretion of the ph$cian, iC there was vcrnricular rachycardia, sustained su- praventricular tachyarrhythmia, a subsrantial increase or de- crease in blood pressure or intolerable symptoms.

Four standard imaging planes were digitized at baseline, low dose infusion (10 &kg per min), prepeak heart rate (typically 10 to 20 beatsimin below maximal hetrt rate) and peak heart rate. These four images and the apical long- and short-axis views were also recorded on 0.75in. videotape. Patients were routinely observed for 30 min after discontiau- atinn of stress testing.

Interpretation. Echocaidingraphic images were reviewed by an experienced stress echocardiographer who was unaware of the clinical data. Wall motion in each of the 16 segments of the left ventricle was scored semiquantitatively as follows: 1 = normal or hyperkinetic; 2 = mild to moderate hypokicesia; 3 = severe hypokinesia or akinesia; 4 = dyskinesia; and 5 = aneurysm (14). Nonvisualized segments were not scored. An ischemic response was diagnosed if stress induced an increase in wall motios score in one or more segments, with the exception of a change from akinesia to ,dyskinesia (15). For correlation of an&graphic and echocardiographic data, either a rest nr a strcss&ducible wall motion abnormality was considered indica- tive of coronary disease. Segments were assigned to a vascular supply on the basis of a previously described model (13) The heart rate at which wall motion abnormalities developed (“isch- cmic threshold”) was recorded, Segmental scores were sutnmed and divided by the number of visualized segments to yield a wall motion score index. Left ventricular ejection fraction was cab- lated (16) or estimated visually ( 17).

Coronary angiogqhy. Coronary angiography was pcr- formed by the Judkins technique. Visual estimation of percent coronary lumen diameter narrowing was coded by consensus of two angiographers in a prospectively maintained data base. Hemodynamically significant coronary disease was defined as ~70% stenosis of a major epicardial artery or primary branch or ~50% narrowing of the left main stem. Multivessel disease was diagnosed if two or more maior epicardial vessels or the left main coronary artery was significantly stenosed.

Statistical analysis. Echocardiographic data were prospec- tively entered into a relational dais base (SAS, Version 6.9). Categoric data were reported as percentages and continuous data as mean value t SD. Categoric variables were com,pared by the chi-square test, and continuous data were analyzed by the two-sample Wilcoxon rank sum test. Likelihood of occur- rence of a variable of interest was calculated as an odds ratio with corresponding 95% confidence inten&. Stepwisc logisric regression analysis was used to identify variables predictive of ventricular arrhythmia occurrence. Variables used in this imnth- ysis are identified in Table 1. The ability of the ischemic threshold to discriminate between single- and multivessel disease was assessed using receiver operating characteristic (ROC) curve analysis. An appropriate cutoff of the ischrmic threshold was chosen at the point where sensitivity was approx- imately equal to specificity. Differences were considered sig- nificant at p < 0.05.

&?SdkS

Baseline characteristics. There were 413 men and 756 women (mean age 6Y 4 12 years, range 26 to 100). Indici~tions for dobutamine stress echocardiography were preoperative assessment in 492 (42%) patients; evaluation of coronary artery disease in 39X (34%); and assessment of chest pain, dyspnea or left ventricular dysfunction in 281 (24%). The clinical profiles of groups I and II were similar (Table 1). There were small differences in age, rest ejection fraction and ‘wall motion score index. The most striking difi’erencc between the groups was more frequent use of beta-adrenergic blocking agents in group II patients (49% vs. 14%* p < 0.0001).

Feasibility of dobbutamine-atropine stress testing. The mean atropine dose was 0.6 mg; 192 patients (64%) received doses ~0.5 mg of a&opine, and 20 (7%) were given >l.O.mg.

Target heart rate was achieved in 6% group I (69’ 3) and 203 group II (68%) patients. The end point of target hcnrt rate:, peak dose infusion or induction of new wall motion ahnorm& ities was rezalized in 93% (807 of 872) and 96% (258 of 299) of patients in groups I and II, respectively.

Nemodynamic respa?ase. Group II patients had a lower rest heart rate than group I patients (65 ? 1 I vs. 74 2 I2 beatsimin, p < O.UOOl). Mean heart rate in group II hefore administration of atropine was lower than the peak heart rate attained with dobutamine in group I (94 2 20 vs. I27 t 18 beats/min, p < OMOI). After a&opine, there were no significant differences in heart rate and rate-pressure product between the two groups. Mean blood pressures in group I and II patients at rest (1Otl vs. 98 mm Hg) and ar peak stress (87 vs. ‘31 mm Idg) were also not significantly diRcrcnt,

Detectinn cd coronary artery disease. Tbc pmportion of patients with angiographic coronary disease was 79% (108 of 137) in group I and 87% (40 of 46) in group II (p = 0.23). and the incidence of multivessel disease in group 1 (79 [73%] of 108) and group II (30 [7S%] c;f 40) patients with angiographic disease was similar. Abnormal rest wall motion was present in IO5 patients (76 in group I, 2Y in group II) with coronary disease. New wall motion abnormalities were observed during dobutamine infusion in 65 of these patients (47 in group I, 18 in group II) and in an additional 10 patients after atropine was giveti. Among patients with normal rest wail motion, new wall motion abnormalities were induced by dobutamine in 2X (26 in group I, 2 in group II) and by atropine in another 8. Table 2 shows the sensitivity and specificity of &butamine-&opine stress echocardiography for the detection of coronary disease in all patients who underwent angiography and in the subset with normal rest wall motion. The value of atropinc was most cvidcnt in the latter group. No false positive test results were attributed to a&opine use.

Detection of isclemia. Stress testing induced ischemia in 329 group I and 115 group 11 patients. In the latter group, 45 patients had new wall motion abnormalities first noted during dobutamine infusion, whereas in 70, ischemia developed only after atropine was given. Therefore, 16% (70 of 444) of all patients who had an ischemic response required atropine before myocardial ischemia became evident. The ischcmic

thresholds in groups 1 and 11 were similar (I 11 2 7,o and I10 2 22 beats/min, respectively).

hong group II patiunts with i~ngir~gmphicall~~ detected disease, ischemic wall motion abnormalities developed after atropine in 5W% (20 of JO). Twelve oF these 20 patients had abnormal rest wall motion; in these patients with presumed previous infarction: new dyssynergic regions bccamc. evident after atropine administration. Of these, seven regions were adja- cent to previously infarcted areas, suggesting peri-infarction isch- emiar in five patients, new wall motion abnormalities occurred remotely, confirming multivessel disease. Of interest, in two of thcsa: five : dtientr;, the new dayssynergic regions aplnnred in a different vascular territory after dobutamine had induced a mild wall motion abnormality. Eight patients had normal wall motion at rest and a negative test response until after atropittc was given. Of tlrcsc eight psticnts, thtee had multivc&sel coronary disease, and five had singlwe& disease (left anterior descending coro- nary artery stentosis in one. circumliex coroniiry artery stenoses in three, right corowdry arteT stenosis in one). The lesion sites were proximal in four of these eight patients and included a 70% ostial left anterior descending artery stenosis in a patient who had previously undergone stberectomy.

Induction of ischemia in single- aad multivessel disease. Ischemic segmental wall motion appeared during dohutamine infusion in 72 of 308 group I patients with angiographically defined coronary artery disease; of these, 16 had single-vessel disease, and 56 had multivessel disease. Ischcmia occurred in 38 of the 40 group II patients with coronary disease. After these 38 patients received &opine, ischemia developed in 8 of IO with single-vessel disease and 10 of 28 with multivessel disease (Fig, 1, top). Therefore, in the entire angiographic subset of patients with detectable limitation of flow reserve, 31% of patients with single-vcsscl and 12% with multivessel disease required the addition of atropine to induce ischemia (Fig. 1, bottom).

The increased likelihood (odds ratio [OR] 7;2, 9% confi- dence interval [CIJ I.3 to 40.7) of ischemia developing o$ly after a&opine in group II patients with single-vessel disease was not explained by beta-blocker therapy-50% (5 of IO) bf patients with single-vessel and 57% (16 of 28) with multiv&$$ disease were receiving this medication (p = &70). The ‘r&k:

,i ..I’: p, .,’

Single vessei CAD Multivessel CAD

1

Single vessel CAD Mukivessel CAD

Pigua 1, Top, Distribution of angiographic coronary artery disease (CAD) in 38 group I1 patients with inducible ischemia during dobutamine-a&opine stress cchocardiography. fschemia developed after atropine administration In a significantly higher proportion of patients with single-vessel coronary artery disease. Bnbtom, Among the llb patients with augiographic coronary artery disease and stress- inducible ischemia, atropine was used relatively more frequently in patients with single-vessel coronary artery disease. Open ban = &hernia before atropine; hatched bars = ischemia after atropine.

heart rate (65 i 12 vs. 64 i: 10 beats/min), preatropine heart rate (93 -+ 19 vs. 91 t 19 beatsimin) and peak heart rate (124 2 21 vs. 12.5 i 15 beats!min) were similar in both groups. Patients with single-vessel dLt:asc, howcvcr, tended to have a higher ischemic threshold (I 1~’ fi 24 vs. 95 t- 20 beats,‘min, p = U.05). The area under the RC9C curve was 0.71 (SE 0.09>, indicating a discriminant ability better than chance. Seven (70%) of 10 patients with single-vessel disease manifested ischemia at a heart rate 2110 beatsimin compared with ‘7 (25%) of 28 with muitivessel disease.

Patients receiving beta-binckers. 0f the entitc study co- hort, patients taking beta-blockers were more likely to receive atropine (54% [147 of 2711 vs. 17% [152 of 9001, p < 0.0001). Ischemia developed after the addition of atropine in propor- tionately more group II patients receiving beta-blockers (61% vs. 40%, p = 0.03). There were no significant differences in age, gender, previous myocardial infarction, rest wall motion score index or degree of coronary disease (in 34 patients who had angiography) between the 61 group II patients with and the 54 without beta-blockers, implying that the former were unlikely to have less coronary disease.

Tahie & Incidence of Adveme Effects in 1 .I71 Patients Undergoing Dobutamine Stress Echoeardiogmphy -~ --- -

GAup 1 Gr1Nip II (E = 8721 , (n = EJ9) p Valus

!VLUlSi%d 73 (8) 38(13) 0.03 Hradac~te 66 8 B (9) o/a PYtp~t”ti~JllS S6 (10) 13 (8) 0.27

Chest pain 152 (17) 55 (18) 0.71 @Spi?CX 517 (Yf 14 (I) 0.01 Lightheade,dness 41. is? 38 (6) 0.37 Tremor or shivering St (54) 3’(H) 0.48 Paroxysmal atriai fibrillation $4 (5) 8 (3) 0.09 Supravcntria~lar tachycardio 95 (11) Z(7) 0.06 Complex ventricular ectopic heats 115 (14) 31 (IQ) t.m Ncrsustained ventricular !achycardia 6.5 (7) 7 12) 0.001 -~------- --........-..v-v--w-

Ikua presented arc number (W:) of patients. Group 1 = dohiltamine only; Gmup II = dnhulamine and atropinc.

Adverse effects, With the c,xception of nausea, atropine use was not associated with more adverse effects (Table 3). Ar- rhythmias necessitated test termmatron in 2X group I (2.4%) and 3 group II (1.0%) patients without a defined end point (p = 0.14). Nonsustained ventricular tachycardia developed in more patients in group I (p = 0.001). §ustained ventricular tachycardia did not develop in either group. Acute urinary retention, visual disturbance and neurologic dysfunction were not observed after stress testing. The incidence of adverse effects in patients receiving Cl.0 mg (n = 229) or rl.0 mg of atropine (n = 20) did not differ.

By stepwise selection of covariates, a&opine usage and rest wall motion score index were found to be significant indepcn- dent predictors of ventricular tachycardia, A&opine usage was negatively associated with ventricular tachycardia (QR 0,31, 95% CI 0.14 to 0.69, p = 0.004), and a rest wall motion score index :>I .l was a positive predictor of this arrythmia (OR 2.53, 95% CI 1.52 to 4.14, p = 0.0003). Group II patients who had ischemic wall motion abnormalities before atropine adminis- tration (n = 45) were ii3 more prone to complications than those with ischemia after atropine (n = 72). Myocardial infarction: unstable angina, cardiac arrest or death did not occur in either group.

Discussion The present study dctines the role of atropine supplemen-

tation in a large consccutivc series of patients undergoing routine dobutamine stress e;hocardiography. Atropine was required in one in four patients because of a suboptimal stress response. Its use permitted attainment of a heart rate end point in 69% of patients compared with 52% before a&opine administration. A&opine use increased diagnostic sensitivity for ischemia by 164il and for coronary artery disease by 5%. Chronotropic augmentation was of greatest value in patients taking beta-blockers, those with single-vessel disease and those with normal rest wall motion. In 50% of angiographically examined patients receiving a&opine, new regions of ischemia

(which in every case carrespcmded with significant coronary stenosis} were appreciated.

~~~~a~i~~ with p&cm wo& McNelll et al. (4) found that atropinc use i.n 49 of 30 consecutive patients who under- went coronary arteriography increased the overall sensitivity of dobutaminc stress eshQcardiography from 32% ro 7%* After atropine administration, additional diagnostic information was obtained in 20 of 31 patients with coronary disease. The benefit of atropine use in that study was more impressive than in the present study became 80% of the patients were receiving beta-blockers. Despite atropine use, test sensitivity in the study by fv¶cNeill et al. (4) was only 70%~~ probably reflecting the preponderance of patients with single-vessel disease. In con- trast, previous studies using dobutamine as the sole stressor agent reported diagnostic sensitivities of up to %%I (3). The higher test sensitivity in these as well as in the present study may be explained by the greater prevalence of multivessel disease and the inclusion of many patients with rest dyssynergy.

Test specificity was relatively low in our study because patients with “cardiomyopathic” ventricles were not excluded -3r analyzed differently and because of posttest referral bias. Marcovitz and Armstrong (3) reported a similarly low speci- ficity of 66% in their study population, 62% of whom had rest dyssynergy. Other investigators described test specificity only in patients with normal rest wall motion (1) or treated myo- pathic ventricles ditferently (18). With criteria equivalent to those of Sawada et al. (I), sp:cilidty in our study was 87%. Importantly, this specificity was not compromised by atropine use (4).

Detection of single-vessel disease. Thhc known lower sensi- tivity of dobutamine stress testing for detection of milder forms of coronary disease (19) may be enhanced by synergy with an agent such as dipyridamole (20). The present study suggests that atropine usage may likewise improve diagnostic sensitivity. The specific value of a&opine is probably related to the generally higher ischemic threshold in patients with single- vessel disease (21), particularly if this threshold is not attain- able with high doses of dobutamine. Segsr et al. (22) suggested that multivessel disease was likely to be present if new wall motion abnormalities occurred at a heart rate <I25 bcatslmin. In the present study, a heart rate of 110 beats/mm better discriminated single- and multivessel disease. The absence of a @hreshold heart rate that perfectly dichotomizes these groups may reflect the dependence of ischemic threshold on other factors, such as collateral circulation and the limitations of visual estimation in predicting physiologically important coro- nary stenoses (23,24).

Atrogine use im patients receiving beta-blockers. bwada et al. (1) reported that beta-blockade during dobutamine echocardiography did not compromise test sensitivity despite significant limitation of chronotropic response but acknowl- edged the possible confounding influence of more severe coronary disease in Patients receiving beta-b!ockers. In a crossover study by Pioretti et al. (X)> atropine induced new ischemic wall motion abnormalities in I2 of 15 patients during beta-blockade and in only 2 of 14 patients during the drug-free

period. Our observations, arrired at by determining ischemic threshold during pharmacologic stress, provide further support for the e&cac;rr of atropine in patients receiving beta-blocker therapy Although nearly 50% of the patients taking beta- blockers in this study required supplementary attopine for optimization of diagnostic sensitivity, atrdrpine was also needed in 30% of the patients with reduced chronotropic rosponsive- ness to dobutamine (18) who were taking neither beta-blockers nor calcium channet blockers.

Mechanisms dischernia in&don by atropine. The equiv- alent mean ischemic thresholds in groups I and II suggest that it is immaterial whether dobutamine alone or an atropine- dobutamine combination is used, provided that an adequate heart rate respomsc can be achieved, In botfi clinical (21) and experimental (6) studies, tachycardia has heen shown to be an important determinant of inducible ischemia during dobut- amine infusion. By contrast, Santiago et al. (9) found no significant differences in peak heart rate but equivalent test sensitivity in the pa&t groups receiving the drug in a dosage of ~30 @g/kg per min and surmised that attained heart rate was “not a useful end point” for dobutrtmine stress echocardi- ography, Wowever, that study did not involve atropine use and did not account for factors important iii the ischemic response other than the angiographic severity of &ease. Moreover, administration of beta-blockers was routinely discontinued before testing, a practice not followed by most laboratories j1,3,13,25).

Although induction of ischemia by atropine is primririly related to tachycardia-mediated increase in metabolic demand and myocardial tension (3t$ there may be a role for direct anticholinergic modulation of left ventricular contractility (27). Reecntly, Landzberg et al. (28) showed that intracoronary infusion of atropine potentiated peak rate of increase of left ventricular pressures (-+dPfdt) by 25% during dobutamine infu- sion. In addition, atropine may infuence coronary tone, although this has been demonstrated only in animal studies (29).

Safety issues. The present study afitms the safety of dobutamine-atropine stress testing as previously reported (4,30-32). Despite theoretic concerns (33), stropine use was not as&i&cd with an excess of tachyarrhythmias. Neariy 50% of our patients who had atropine were receiving beta-blockers, which are known to reduce ventricular irritability (34). The t,herapeutic window of atropine may also be wider iu these patients because excessive tachycardia is curtailed (35). Thesti observations ccrrnmend atropine use as a means of circumvent- ing the irnpcmtical and potentially hazardous practice of withdrawing beta-blockers before stress testing (36)

In our study, atrapine was used even in patients with mild wall motion abnormalities if the test was otherwise uncompli- cated. This approach appears to be safe (?.,30), reduces equiv- ocal wall motion assessment and may increase recognition of multives& disease.

Doses of atropinc >I mg were used in 7% of group II patients, Because of differences in body weight (37) and pharmacodynam;c response (38), higher atropine dosage may, be required in some patients, In particular, patients with heart,

556 r.Iw ET AL. WBL7TAMiNE-ATROPINE STRESS TEST

disease ma! have abnormal parasympathetic cardiac control and less c;lrdiaa~crelcratilln with equivalent doses of atropine (39,4(1). C&asiooal USC of higher doses of atropine did not, however, compromise the safety of the test in these patients, Notably, central nervous system toxicity which has been reported with therapeutic doses of a&opine in hypersensitive and older patients after stress e~h~ardio~a~hy (11,41), was not observed in our elderly cohort.

Study limihtiors. There is inherent bias in reporting the value of atropine in patients receiving the drug when it tends to be given if an end point is not reached with dobutamine. When appropriate, we have attempted to avoid this by describing the incremental value of a&opine in the context of ai1 patients receiving dobutamine. The manner in which the test is per- formed and reviewed in our institution (i.e., continuous tape recordings arc routinely reviewed to determine ischemic threshold) should also reduce this potential for bias.

As in other reports, the sensitivity and specificity of stress echocardiography in our study are imfiuenced by referral patterns that select patients with a high posttest prui?ability of coronary artery dbcase ior angiography. However, the purpose of this study was to define the additional value of atropine supplementation, and any posttest referral bias should have no differ&al effect on the results observed with and without atropine. We are also aware of the imperfect correlation between well motion changes and coronary aagiographic ap- pearance (Z,24), Nevertheless, the angiographic interprera- tion in this study is the most widely used worldwide and is based on methods established by large multicenter trials comparing medical and surgical treatment of coronary artery disease <‘2.43), the results of which have largely influenced clinical d&ion making. Intracoronary ultrasound mcasure- merits have also shown that lumen diameter and cross-sectional area (and preSUmblg Row reserve) may be underestimated by quantitati;ic: angiogmphy (44). More recently, Folland et al. (4s) demonstrated that the functional significance of coronary steno- ses in patients with single-vessel disease was better predicted by visual analysis than by quantitative methods.

Because 3 time lag to onset of the central nervous system effects of atropine is recognized (46), some anticholinergic side effects may have been underreported at the time that patients were dismissed from the stress laboratory. All patients, how- ever, were seen by a physician at our institution within 24 to 48 hl and no significant complications were reported within this time frame, Moreover, the dramatic features of central anti- cholincrgic syndrome generally become &anifest early after ‘intoxication (47-49) and are unlikely to be overlooked.

GoneIusions. The present study defines the role and ben- efit of a&opine use in a large stress echocardiographic practice. Although the incremental diagnostic value of chronotropic augmentation in our study cohort was relatively modest, atro- pine was frequently essential for achievement of an adequate stress level, particularly in patients receiving beta-blockers, and for detection of milder forms of coronary artery disease. This incremental value was paralleled by a remarkable degree of safety.

WI: g~atehdly xknowiedge the advice c?i’ A. lamil Tajjik, ME?, in the preparation of the manuscript.

2. dazeika 1-K, Nadazdin A, Oakley CM. Dohutantine stres cchwardiugraphy for detection and assessment of coronary artery disease. J Am Coil Cardiol 1991?;1Y:1203-Il.

3, Marco&z PA, Armstron@ WF. Accuracv of dobutamine streks echocardiog-

4

5.

6.

7.

8,

9.

10.

11.

12.

13.

14.

1s

raphy in detecting co&&y artery disc&. Am J Cardiol !992;69:12ii9-73. McNeil! Al, Fiorerti PM, &Said SM; Satustri A, Furstm T, Rcelandt JR, Enhanced ssnsitivity for dcleclion of coronary arlev disease by addition of arropine to dobutaminc stress uchucardiogtaphy, Am J Cardiol 1Wz71): 4iMi. Willerson JT, Hutton I, Watson JT, Pldtt MR, Templeton CH. lnliuencc of dobutaminr on regional myocardial Mood Aow and ventricular ncrformnxe during aculc and chronic myocardiat ischrmia in dogs, Circtdation 1976:53 828-33. Kude KB, Iquierdo C; Buja LM, Willcrson JT. Etfccts of inotropic and chronotropic stimuli on acute myocardiai irchcmic injury. 1. Studies with dobutaminc in th:: ancrthetixed dog. Circulation 1982B:l321-8. Vatner SF, Baie H. lmoortance of heart rate in determining the effects of sympatht*nim&c amin& on regional myocardia! i”unction anld blood Row in ronscious dogs with acute mvocardial &hernia. Cixr Res 1979:451793-803. Fioretti PM, Poldermans D, &us&i A, et al. Atropine increases the accuracy of dobutamine stress echocardiography in patients taking beta-blockers. Euc Heart J 1994;15:355-60. Santiago P, Vacck JL: Rosamond TL. Dobutamine sires5 echocardiogrqhy: clinical utility and predictive value at vailrious infusion rates. Am Heart J 19?4;128:804-8. Mairesse FH, Marwick TH: Vanoverschclde JL, et al. How accurate is dobutaminc stress electrocardiography for detection of corona? artcry disease? Comparison with two-dimcxZiooal cchociirdiogiapny and technetium-‘9Ym methoxyi isobutyl isonitrik: (nibi) pcrfusktn scintiqphy, J Am Co11 Cardiol luYJ;24:YX~-7. Picano E, Pingitore A, Conti U, et al. Enxnced sensitivity for detection of coronary artery disease by addition of atropine to dipyridamole echocardi- ography. Eur Heart J 1993;1431216-22. Picano E, Mathias W Jr, Bingitore A, Bigi R, Previtali M. Safety and tolerability of dobutamine-atropinc stress cchocardiography: a prospective, muiticcntre study. Echo Dobutaminr: International Cooperative Study Group. luncet 1994;344:11YO-2. Pellikka PA, Roger VL, Oh JK, Miller F’A, Seward JB, ia~ik AJ. Stress echocardiorraphv. Part II. Dobutamine stress echocardiopranhv: rechniaues,

. _ - 1 -

implement&on, clinical applicaiions, and correlations. Mayo Clin Proc 1995;70:16-27. Schiller NB, Shah PM, CrawFord M, et al. Recommendalions for quantita- tion of the left venrricle bv two-dimensional echocardioeranhv. AmericaB Society of Echocardiograpi?y Committee on Standards, &&c~mmittec on Quantitntinn of Two-dimensional Echocardiograms, J Am Sot Echocardioar 1989;2:358-67, Arncsc M, Worctti PM, Comel JH, Postma-“Fioa J, Rciis AE, Roclandt JR. Akin& becoming dyskincsis during high-d&e dabutamine stress echocnr- diography: a marker of myocardial &hernia or a mechanical phenomenon? Am J Cardiol 1994;73:896-9.

16. Quinoncs MA, Waggoner AD, Reduto LA, et al. A new, simplified and azcuratc method for determining ejection fraction with two.dimcnaional echocardiography. Circulation 1981;63:?44-53.

17. Rich S, Sheikh A, Gallastegui J, Kondos GT, Mason ‘I’, Lam W. I&xl:i- nation of left ventricular ejection fraction by visual estimation during real-time two-dimensional cchocardiography. Am Heart J 198~11)4:603-6.

18. Bach DS, Mullor DW, Gras BJ, Armstrong WF. False positive do&amine stress echocardiograms: characterization nf clinical, echocardiographic and angiographic findings. J Am Co11 Cardiol 1994;24:928-33.

19. lliceto S, Galiuto L, Marange!li V. Rizzon P. Clinical Lse of stress echocar- diography: factors affecting diagnostic accuracy. Eur Heart J 3994;15:672- HO.

JACC Vol. 28, No. 3 Septamhcr 19%X51-7

5.57

20. Ostojic .h4: I’kmno E. B&tin B. et al. L)ip~ridanlok,dobulam3ne echocar- dio~rapbyv: a novel test for the detcctioa of milder forms of eorcmary artery disease. J Am Cot! Card&l 199$23:1115-23.

21. Cohen JL, Greene TO. Gttenweller J, Binenbaum S%, Wilchfort SD, Kim LX. Dubutaminc digital echtxardiography for detect@ coronsty arteF disca%?, Am J Cardirrl 1991;6?:13Ll-8.

22. hg:a DS, Bmn SE, S~-:ids SC;, R~II T, ~c~~cnh~~tlm ii. D&utaminc stress rchocardiograpby: correlation with coronztl; lesion severiF as deter- mined by quantitative angiography. J Am Coil Cardiol 1992;19:1197-201.

23. White CW, Wright CB: Doty DB, et al. Does visual interpretation of tbe cOrOnarV arteriogram predict the physiologic imporbmtce of a coronaty stenosis’? N Engl J Med 1934;310:819-24.

24. W&on RF. Marcus Ml, White CW. Prediction uf the nhvsialoeic sin&i- cance of coronary arteriai lesions by quantitative lesion g&tetty yn p&nts with limited ccrr0narv arterv disease. Circulation !987:75:723-32.

25. Marwick T, Willemart B,b.Hondt AM, et al. Selection d the optimal nunexercise stress for the cvaiuathm of ischcmic rcgiormi mpcardial dps- function and malprfusian. Comperisnn nf dobutaminc and adcnosinc usin!: echocardiography and 99mTc-MLWL sin& photnn emission computed tit,, mography. Circulation 1993$7:345-54.

26. Katz AM. Physio1o.m of the Heart. 2nd cd. New York: Kdven Press, 1992319-50.

27. DeCeest H. Levy MN. Zieskc H. Negative inatropic &XI of the vague nerves upon the canine ventricle. Science 1964;1445223-5.

28. Landzberg JS, Parker JD, Gauthier DF, Colucci WS. Etfeets of intracoronary acetylcholine and atrapine on basal and dobutaminc-stimuiatcd left vcntric- ular contractility. Circulation i994:89:164-S.

29. Feigi EO. Parasympathetic nmtrol of coronary blood Sow in dogs. Circ Ucs 1969;25:S@J-19.

30. Poldermans D, Fioretti FM, Boersma E, et al. Safety of dobutamine-atrapine stress echocardiograpby in patients with suspected or proven coronary artery disease. Am J Cardiol 199~73:456-9.

31. Mertes H, Sawada SC, Ryan T, et al. Symptoms, adverse effects. and complications associated with dobutamine stress cchocmdiokraphy. Exp+ri- ence in I1 18 patients. Circuhttion 1993:RLi:lS-9.

12, Mdermans D, Fiuretti PM, ILrnnm E, et al. Dt)hu:amine-:itrtlpine stress cch~~rdio~rapbp in slder!y patients unable to perform an cxcrcisc test. Hcmodynnmic cham&risttcs, safety, and prirp,nostic value. Arch Intern Mcd 199J;154%81-6.

33. Knoebel SB, McHrnry PL, Phillips JF, Widkrusky S. Atropinc-induced cardioacceleration and myocardinl blood tluw in suhjecls with and without cormnny artery disease. Am J Cardiol 1974;33:327-32.

34. Vend&i FJ Jr, Garan H. Ruskin JN. E!ecttupbysi&gic cffccts of beta bbxkers in ventricnlar arrhythmias. Am J Csrdid 198’;6t?:3D-9D.

5. Flesras AP. Rytm TJ. Atropinc%duccd cardieacceleration in patients on chrome propranalol therapy: comparison with the positive ehronotmpic effect <rf isometric exercise. Am Heart J 1983;105:230-4.

36. Rishman WH. Beta-adrene& blocker withdrawal. Atn 5 CarJ? 198’?$: 26F-3ZF.

37. Morton HJV, Thumas ET. Etfect of atronine ox the heart-rate. Lamer 19S812:1313-S.

38. Ncdd JB Jr. A&n XT. Coleman E: Frederickson EL, Goldberg LJ. Card& rate and rh.ythm changes w,ith atropine and metbscopelamine. Clin Phanna- cd Ther 1975;17:2wi-5.

39. Jose AD. Taylor RR. Autonomic blockade br propranniol and attopine IO study intrinsic myocardial functicln in man. J Clin Invest L9bP:IK:2019-~ 31.

40. E&erg DL. Drabins$ M, BraunwaEd E. Iki~*c~k cardiac parasympathetic control in paticnta with hcitrt disease. N Engl J Med 1971;2RS:Ei7?-83.

41. Mylcs I’. Do&amine-atropine stress ccbocardirrgraphy and <cntral anticho- lincrgic syndrome [Letter]. Lancet 1994;344:1636.

42. CASS Principal Investigators and Their .Assuciates. Ccnonary ,Artcq Surgery Study (CASSI: a randomized ttid of coronary artery bypass s~rgcry Smvival data. Circulation 1983;68:939-58.

43. Varnauskas E. Twehe-year fohow-up of WviViL~ in the randomr~cd Euro- pan Coronas Surgery Study, N Engl J Med 1958;3f9:.332-7.

44. Davidson 0, Sheikh KH, Kisslo KB, et al. In!raccxon~ry ultrasotml evaluation of intervcntional technologies. Am J Cardiol 1991:68:1305-9.

45. Foiland ED, Vogel RA. Hartigan P, et al. Relation between corumdry artery stenosis assessed hv visaal, caliper. and computer methods and exercise capacity in patients&tb single-v&eel coronary -artery disease. The Vetcrans Atfairs ACME Investigators. Circulatian 1994:S9:2(n5~-14.

46. Ellinwood EH Jr, Nikaido AM, Gupta SK, Heath&y DG. Nishita JK, Comparison of central ncrvons system and pcriphcral phanacodynamics IO &opine pbnimacokineti~~. J Phirrmdcol Esp Tbcr Iw(J:2%:1133~-9,

47. Rodger WG. Casr of acute poisoning after instillation of a small dose of atropine into the eye. Glasgow Med J 19tD$Xk10?-S.

43. Shader RI. Grecnblott 135. Dses and tolticitv of brll:~douua alk&iJs and synthetic alrtidldinerllics. Scmin 1’.;8chir&v ~971;.1:4#)-~76.

49. &wtnms P. Lamhr&ht ii, Sci&ms 1;. Vanweldzn 1% Verbacgcn H. Anticholinergie intt~~ici~tion with enmmcrciuiiy avai!able thorn apple tea. J To&l Clin TrnU 1994;32:58!1-92.