Abnormal cardiac sensitivity in patients with chest pain and normal coronary arteries

JACC Vo1 . 16. No . 6 November 15. 199111159-ra,

1359

Abnormal Cardiac Sensitivity in Patients With Chest Pain and Normal Coronary Arteries RICHARD O . CANNON Iii, MD, FACC . ARSHED A . QUYYUMI . M D . WILLIAM H . SCHENKE, BA, I .AMEH FANANAPAZIR . M D . EBEN E . TUCKER . MD . ALEXANDRA M . GAUGHAN, MA." RICHARD H . GRACELY . PHD .° EDWARD L . CATTAU, JR . . MD, STEPHEN E . EPSFFIN . M D . FACC Belhesda .ILlarvlnoJaed

Wa.ddngla,r . I) .C' .

The causes of chest pain in patients frond to have angin. graphically normal coronary arteries during cardiac catheterization remain controversial . Cardiac sensitivity to catheter manipulation, pacing at various stimulus :nlensities and intracoronary, injection of contrast medium was examined in several groups of patients vcho underwent cardiac catheter nation. Right heart (especially right veatriculor) catheter manipulation and pacing and intrarnronary contrast medium provoked chest pain typical of that previously experienced in 29 (81 %) of 36 patients with chest pain and angiographically normal coronary arteries and 15 (46%) of 33 symptomatic patients with hypertrophic cardiomyopathy . In contrast, only 2 (6%) of 33 symptomatic patients with coronary artery disease experienced their typical chest pain with these sensitivity tests (p < 0 .001). None of 10 patients with valvutar heart disease but without a chest pain syndrome experienced any sensation with these tests .

Chest pain commonly causes patients to seek medical attention because of its association with life-threatening heart disease . However, I0% to 30%r% of patients who undergo coronary angiography for a chest pain syndrome compatible with ischemic heart disease are found to have normal- or nearly normal-appearing coronary arteries I1-4) . Recent studies (5-13) have suggested that many patients with anginalike chest pain but with normal coronary angioFrom the Cardiovascular Diagnosis Section . Cardiology Branch . National Hears, Lung, and Blood Institute . Nallenal Insliwles at Health . Dohesda, Maryland ; the 'Clinical Pain Section . Neombiology and Ancdhcwlology Branch . National Institutes of Dental Research . National Imta utes of Health ; and the *Gaslmeoierology Division . Georgeloun University School or Medicine. Washington. D .C . Manuscript'eveived February 9 . 19'x5.rcatsed manuscript received May 1 . 1990, accepted stay 15 . 1"1. Addressfor reprints: Richard D. Cannon III. M D. Building III. Roam 7015, National Institutes of ttea0h . Bethesda. Maryland 70X9'- . 31990 by the

Amencan College ,rc'wd,ulogy

Cal-on, pain threshold testing demonstrated that patients with chest pain and normal aneeearv arteries had a higher pain threshold to thermal stir .,ulatima compered with patients echo had coronary artery disease or hypertrnphic cardiomyopathv . No relation existed between car . diae sensitivity and cutaneous sensitivity testing . Thus, patients who have chest pain despite angiographically normal coronary arteries may have abnormal cardiac sensitivity to a variety of stimuli . This increased sensitivity may be of causal importance to their chest pain syndrome or may contribute to their perception of ischemia-induced pain . The same phenomenon was also commonly seen in symptomatic patients with hypertraphie cardiomyopathy . Whether this phenomenon represents abnormal activation of pain receptors within the heart or abnormal processing of visceral afferent neural impulses in the peripheral or central nervous system is unknown. (J Am Cull Cardiol :990;16;1359-66)

graphic studies have either myocardial ischemia due to abnormal coronary vasodilator reserve, or esophageal dysmotility, or both . Nonetheless . a considerable number of patients with a chest pain syndrome do not have such abnormalities. and even those with such abnormalities often experience symptoms that appear to be disproportionate to the severity of the documented disease . Several alternative causes, including various types of psychogenic syndromes 114-17) . have therefore been suggested as the basis for the chest pain esperieuced by these patients . Clearly . however, a satisfactory explanation for chest pain in many patients with angiographically normal or nearly normal coronary arteries remains elusive, especially in those patients who have frequent, prolonged and severe pain despite reassurances of a benign prognosis 118.19) . One possible contributing cause of the chest pain experienced by these patients is a decreased cardiac pain threshn

3c

W".1sasn



CANNON ET AL . ABNORMAL CARDIAC SENSITIVITY AND CHEST PAIN

JACC Vol . 16 . No . 6 November 19, 1990:1359-66

old . This concept is suggested by studies demonstrating a lowered pain threshold in women with "Syndrome X" (20) and an altered right atrial pain perception of patients with chest pain and normal coronary angiogams (21). We also observed that a large number of patients with chest pain who have augiographically normal coronary arteries and a normal left ventricle had an increased sensitivity of the heart to intracardiac stimulation by catheters and intracoronary contrast medium . The present study was therefore conducted to determine whether a diminished cardiac pain threshold might contribute to the chest pain syndrome in patients with several forms of heart disease .

toms and 9 without a history of chest pain underwent cardiac catheterization because of suspected myocardial ischemia by exercise electrocardiography and radionuclide ventriculography (silent ischemia) .

1360

Methods Study groups. Four groups of patients unuciwcirt cardiac catheterization and participated in this investigation and represent a consecutive series of patients studied . The group that prompted the present investigation comprised 36 patients with a chest pain syndrome, all with entirely normal coronary arteries and left ventricle as determined by angiography . The average age was 48 ± 9 years ; 21 (58%) were women . Virtually all had one or more atypical features for angina pectoris associated with coronary artery disease including location (left-sided versus suhsternal), occurrence both at rest and on effort, prolonged duration, severe intensity and variable or transient relief by nitrates . Because of the subjective nature of chest pain and the wide variety of pain descriptions among patients, no attempt was made to further categorize symptoms . In all patients chest pain symptoms persisted despite reassurances of a benign condition. No patient had left ventricular hypertrophy or mitral valve prolapse by M-mode and two-dimensional echocardiography . Four patients (11%) had >1 mm ST segment depression during treadmill exercise testing using the standard Bruce protocol . The second group comprised 44 patients with hypertrophic cardiomyopathy defined as a hypertrophied, nondi-

lated left ventricle in the absence of an identifiable cardiac or systemic stimulus to hypertrophy (22) . The average age was 44 ± 15 years ; 22 (50%) were men . Thirty-three (75%) of these patients reported chest pain symptoms, virtually always atypical for angina pectoris associated with coronary artery disease . All underwent coronary angiography and were found to have normal appearing coronary arteries . Eleven patients gave no history of chest pain and underwent electrophysiologic studies for symptoms of presyncope and syncope . Coronary angiography was perforated in 8 of these I I patients and results were normal in all. The third group comprised 42 patients with acg.ographically significant coronary artery disease defined as >50%

luminal diameter narrowing of at least une major coronary artery . The average age was 60 ± 8 years ; 35 (83%) were men. Thirty-three patients (79%) reported chest pain symp-

Group 4 comprised 10 patients with valvular heart dis-

The average age was 45 ± 21 years; 6 (60%) were men . All 10 had normal coronary arteriograms . Study protocol. Cardiac catheterization was performed at least 2 days or five drug half-lives after the discontinuation of all cardiac medications. All patients were premeditated with diazepam, 10 mg orally . During right heart catheterization a 6F pacing wire was advanced into the right atrium . The patient was asked to report any chest s-nsations and, if chest pain was felt, to decide whether or not it was typical of any pain previously experienced. Although the patient was aware that catheter manipulation within the heart was going to occur, the actual performance and sequence of stimulation were inapparent to the patient because the position of the fluoroscopic tube obscured the patient's view of the monitor and the actions of the catheterizer. The pacing wire was manipulated along the right atrial wall for 5 to 10 s and then positioned in the right atrial appendage. Pacing at 5 beats/min faster than the patient's basal heart rate was then initiated, starting at 1 mA and increasing by I mA increments at 10 s intervals until the patient reported chest pain or until a pacing stimulus of 10 mA was reached. Hand signals were used to initiate pacing, change the pacing stimulus intensity and terminate pacing to prevent the patic .t from knowing when interventions were being started or stopped . After stimulation and pacing from the right atrium, the pacing wire was advanced into the apex of the right ventricle . Pacing was then initiated at 5 beats faster than the patient's basal heart rate and the sequence and performance of pacing were repeated as for the right atrium. Diatrizoate meglumine and diatrizoate sodium (Angiovist 370) were used for cineangiography . During left ventriculography patients were asked whether or not manipulation of the pigtail angiographic catheter or the injection of dye into the left ventricle provoked any chest sensation or pain . During coronary angiography patients were asked to report whether or not chest pain was provoked by the first injection of contrast medium into the left coronary artery, although they were not told when that would occur . Most patients with a chest pain syndrome and angiographically normal coronary arteries then underwent assessment of great cardiac vein flow response to pacing stress before and after ergonovine administration . a s previously described (10) . All patients were participating in protocols approved by the Institute Review Board Subpanel, National Heart, Lung, and Blood Institute, o . were undergoing diagnostic catheterizatisn . Informed consent was obtained from all patients . The testing procedures described added approximately 5 min ease asymptomatic for chest pain .



JACC Vol. 16, Na. 6 N2mher 1 5, 1990 :1759-66

CANNON ET AL . ABNORMAL CARDIAC SENSITIVITY AND CHEST PAIN

to the cardiac catheterization and no complications were encountered . Cutaneous pain sensitivity testing. Such testing was petformed on 16 patients with chest pain and normal coronary arteries (3 men, 13 women ; mean age 47 ± 10 years) : 18 patients with coronary artery disease (16 men, 2 women : mean age 57 ± 18 years ; and 12 patients with hypertrophic cardiomyopathy (4 men, 8 women ; mean age 42 ± 16 years) . A multiple random staircase evaluation of thermal pain sensitivity was used (23). Patients used a 7 point category scale (no sensation ; warm, not painful ; hot, not painful ; mild pain ; moderate pain ; intense pain; eaGCmely intense pain) to rate 72 thermal stimuli of 3 s duration delivered to the volar forearm at 20 s intervals by a I em diameter contact thermode . The computer that delivered the stimuli continuously adjusted the intensity of the stimuli to maintain responses at the same level The program divided the stimuli into three independent staircases corresponding to subjective levels between "hot, not painful" and mild pain" . between "mild pain" and "moderate pain" and between "moderate pain" and "intense pain." On each trail one of these three staircases was randomly chosen by the computer and the stimulus associated with that staircase was delivered, thus reducing the influence of psychophysical bias . These studies were performed and interpreted without knowledge o, the patient's cardiac diagnosis and represent a consecutive series of patients tested . Cardiac sensitivity testing was performed in most of the patients who underwent cutaneous pain sensitivity testing, including I I patients with chest pain and normal coronary arteries, 10 patients with coronary artery disease and 7 patients with hypertrophic cardiomyopathy . Esophageal testing. A subset of 14 patients with chest pain and angiographically normal coronary arteries underwent esophageal study, which was performed and interpreted without knowledge of -he results of any of the other studies described . The equipment, techniques and diagnostic criteria used for the performance of esophageal motility assessment, Bernstein test and balloon distension test for esophageal sensitivity testing have been previously described (12,13,24) . Statistics . Continuous variables are reported as mesa values ± I SD . Multiple comparisons were performed by analysis of variance with Scheffd's or Duncan's method for intergroup comparisons . Discrete variables in prevalence determinations were compared by chi-square analysis or Fisher's exact test where indicated. A p value <0 .05 was considered statistically significant.

Results Chest pain with right heart stimulation . Figure I shows the prevalence of any chest sensation and pain similar to that spontaneously experienced (typical chest pain) felt by pa-

1361

Typical Chest Pain

Figure I . Prevalence of any chest sensation ( top panel) and chest pain typical of symptoms previously experienced (bottom panel) in the right heart stimulation studies by patients with chest pain and angiogvephically normal coronary arteries INCA), hypertrophic cardiomyopathy (HCM), coronary artery disease (CAD) or valvular heart disease (VHD) .

tients in the four study groups doing any of the right heart stimulation tests. No patient reported any chest sensation or pain in the abserce of cardiac stimulation- The highest prevalence of any chest sensation provoked was noted in the 36 patients with chest pain and angiographically normal coronary arteries (86%) ; 29 of the patients (81%) who reported chest pain described this as a painful sensation identical to that spontaneously experienced . Patients with hypertrophic cardiomyopathy, including 15 (46%) of the 33 symptomatic patients, also had a high p,evatence of as, sensation (52%) and typical chest pain (34%) provokes during right heart stimulation . Ten (24%) of the 42 patients with coronary artery disease experienced some chest sensation during right heart stimulation, although this sensation was described as their typical chest pair, by only 2 patients



1362

JACC W1 . 16, No . 6 Nuvember 15, 1990:1359-66

CANNON ET AL . AENORMAL CARDIAC SENSITIVITY AND CHEST PAIN

RA PROBING

FLA PACING

RV PROBING

RA Pacing

RV Pacing

POLING

NCA Figure 2 . Prevalence of typical chest pain provoked during right atria) (RA) and right ventricular (RV) catheter manipulation and pacing in patients with chest pain and normal coronary arteries (NCA), hypertrophic cardiomyopathy (HCM), coronary artery disease (CAD) or valvular heart disease (VHD). Pacing was performed at 5 beatslmin faster than basal heart rate with a stimulus intensity of I to 10 mA,

(5%). No patient with valvular heart disease experienced any chest sensatic:n during right heart stimulation . The prevalence of typical chest pain provoked during the four components of the right heart stimulation study are shown in Figure 2. During probing of the right atrium and right ventricle . ectopic beats were often provoked. Thus, it was not possible to tell with certainty whether the pain was secondary to mechanical stimulation or to ectopic beats . In all four tests within the right heart chambers, patients with chest pain and angiographically normal coronary arteries had the highest prevalence of typical chest pain provoked among the four study groups . Patients with hypertrophic cardiomyopathy had n higher prevalence of typical chest pain provoked during right ventricular pacing compared with patients with coronary artery or valvular disease . Of the patients who experienced their typical chest pain during the four right heart tests, 62% of patients with chest pain and normal coronary arteries did so during more than one test, as did 47% of patients with hypertrophic cardiomyopathy and the two patients with coronary artery disease . The stimulus intensity at the onset of typical chest pain daring right atrial and right ventricular paring is shown for the three study groups with chest pain responses in Figure 3 . A wide range of stimulus intensities was noted for patients with chest pain and normal coronary arteries and patients with hypertrophic cardiomyopathy . The only two patients with coronary artery disease who experienced their typical chest pain during tine pacing test did so only at the highest pacing stimulus intensity (10 mA) for right atrial pacing and at S and 10 mP. dcri .,- : ,,ht ventricular pacing . Fifteen patient- wrist chest pain and normal coronary

HCM

CAD

NCA

HCM

CAD

Figure 3. Stimulus intensity (in mA) at which patients with chest pain and normal coronary arteries INCA), hypertrophic cardiomyapathy (HCM) or coronary artery disease (CAD) began experiencing their typical chest pain during right atrial IRA) and right ventricular (RV) pacing .

arteries underwent an identical pacing protocol within the coronary sinus. Six (40%) described their typical chest pain during pacing. In all cases the pacing stimulus intensity threshold for provoking chest pain was ?5 mA . Pain provoked during angiography. Placement of the angiographic pigtail catheter into the left ventricle never provoked chest pain in any of the study groups, even when such placement was associated with ectopic beats . Further, the injection of contrast medium during left ventriculography provoked no chest pain . However, injection of contrast medium into the left coronary artery provoked typical chest pain in 20 (56%) of the 36 patients with chest pain and angiographically normal coronary arteries (Fig. 4); all also

Figure 4. Prevalence of typical chest pain provoked during injection of contrast medium into the left coronary artery. Abbreviations as in Figure 1 . Typical Chest Pain



JACC Vnl. IN, Nn. 6 No-is,, 1 5, 1950:175')-As

experienced their typical chest pain during right heart stimulation. Only 12% of patients with hypertrophic cardiomyopathy and 3% of patients with coronary artery disease experienced chest pain with intracoronary contrast medium . No patient with valvular heart disease experienced such chest pain . Cardiac sensitivity testing and microvascular angina . Thirty-two of the 36 patients with chess pain and angiographically normal coronary arteries underwent measurement of great cardiac vein flow during rapid pacing from the coronary sinus (I to 2 mA stimulus Intensity) before and utter treatment with ergonovine, 0 .15 mg intravenously . Nineteen patients fulfilled our criteria for microvascular angina (increase in coronary vascular resistance and myocardial oxygen extraction after ergonovine administration in the absence of epicardial arterial constriction associated with provocation of patient's typical chest pain during rapid atrial pacing) (10). The prevalence of typical chest pain provoked during cardiac sensitivity testing was virtually identical in patients with (15 of 19 . 84%) and patients without (10 of 13, 77%) microvascular angina . Cardiac sensitivity testing and chest pain history . Most, but not all, patients with hypertrophic cardiomyopathy and coronary artery disease gave a history of chest pain . To investigate whether such a history correlated with myocardial sensitivity testing results, the 86 patients with hypertrophic cardiomyopathy or coronary artery disease were analyzed separately according to chest pain history . The 33 patients with hypertrophic cardiomyopathy and a history of chest pain had a higher prevalence of any kind of chest sensation nrovoked during the study (20 patients, 61%), compared with the I I patients without a chest pain history (3 patients, 2'%), although rot of statistical significance (p = 0.117) . Eleven (33%) of 33 patients with coronary artery disease and a history of angina had some type of chest sensation provoked during the testing study . This was a significantly higher prevalence than that observed in the nine patients with coronary artery disease who had noninvasive evidence of ischemia without a history of chest pain . None of these latter patients experienced chest sensation or discomfort with cardiac stimulation (p < 0.05) . Cutaneous pain sensitivity testing . Patients with chest pain and normal coronary arteries required a higher temperature (by 1.2 to 1 .8°C among the three thermal sensitivity staircases) to produce the same level of subjective responses compared with patients with hypertrophic cardiomyopathy or coronary artery disease (Fig. 5) . Separate analyses of variance for each staircase showed significant difference in stimulus temperatures among the three patient groups for the low (F(2,431 = 3 .48, p < 0.05) and middle (F[2,43( = 4 .09W p < 0.05) staircases and approached significance for the high staircase (F12,43j = 2.87, p = 0.067) . This effect was accounted for by higher stimulus temperatures in the chest pain, normal coronary artery group in comparison with the

AHNORl1AI C

VtniAC SENSITIVITY

CANNON Er x' . AND CHEST PAIN

1363

s3 ~ STAIRCASE BETWEEN - MODERATE AND - INTENSE PAIN'

so

Pan,

STAIRCASE BETWEEN -MILD PAINAND -MODERATE PAIN' P a 05
5

0

20

49 ~ STAIRCASE BETWEEN - HOT . NOT PAINraLAND'MILD PAIN-

4B 47

P • 05

4s

43 0

5

10

15

20

25

Minutes Figure

S.

Multiple random staircase evaluation of thermal cutane-

ous pain sensitivity. Threshold and suprathreshold values determined over a

24 min

period are plotted for patients with chest pain

and normal coronary arteries (triangles), hypertrophic cardiomyopathy (open circles) or coronary artery disease (closed circles) . StimaIrs temperatures required to produce three subjective levels of pain sensation are shown in the three panels . These temperature differences indicate decreased cutaneous pain sensitivity in the chest pain . normal coronary artery group, because greater stimulus intensities are required to eveke equivalent pain responses . coronary artery disease group (low, p < 0 .05 ; middle p < 0 .05 ; high. p < 0 .05) and to the hypertrophic cardiomyopathy group (middle . p < 0.05: high, p < 0.05) . In the 11 patients who underwent both cardiac and cutaneous sensitivity testing there was no systematic relation between cardiac pain sensitivity, as measured by pacing current threshold, and cutaneous pain sensitivity determined with thermal stimulation (r = -0 .271 . Esophageal pain sensitivity. Fourteen patients with chest pain and ncrmal coronary arteries underwent esophageal



1364

CANNON ET AL . ABNORMAL CARDIAC SENSITIVITY AND CHEST PAIN

testing. Three patients fulfilled criteria for nutcracker esophagus and two for nonspecific motility disorder (12): only one of these experienced chest pain typical for that patient during this phase of testing. Three patients (including one with nutcracker esophagus associated with chest pain) had their typical chest pain provoked by acid infusion into the distal esophagus (Bernstein testing). Two of these patients and an additional patient without esophageal dysmotility or acid sensitivity had their typical chest pain provoked by balloon distension within the esophagus . Of these four patients whose typical chest pain was provoked during esophageal testing, three also had this pain provoked during right heart stimulation . Conversely, 8 of the 10 patients without chest pain during any phase of esophageal testing had their typical chest pain provoked during right heart stimulation. Discussion Cardiac sensitivity as a cause of chest pain. In a preliminary study Shapiro et al. (21) provoked chest pain by high right atria] catheter manipulation and by bolus injection of saline solution into the right atrium in most of the I I patients with chest pain and angiographically normal coronary arteries they studied. Neither test provoked pain in seven patients with coronary artery disease or nine patients with mitral valve disease . In the present investigation, involving a larger number of patients in similar categories and with the addition of patients with hypertrophic cardiomyopathy, we attempted to elucidate a possible role of decreased cardiac pain threshold as a cause or contributing factor to the chest pain experienced by patients with a persistent chest pain syndrome . Our study demonstrated that a large percentage of patients with chest pain and angiographically normal coronary arteries perceive discomfort or pain during catheter probing or application of electrical currents through a catheter within the right heart chambers and during injection of contrast medium into the left coronary artery . The discomfort was interpreted by most patients as typical of their spontaneously experienced chest pain and was often severe, persirsing for up to several minutes after a particular test and commonly provoked by more than one test . Stimulation of the left heart chambers by a pigtail catheter, even when ectopic beats appeared, never produced chest pain . Moreover, increased intracardiac pain sensitivity was also demonstrable in almost half of the patients with hypertrophic cardior.lyopathy, which is often accompanied by chest pain atypical for classic angina pectoris . However, chest pain was uncommonly provoked in patients with coronary artery disease and was not present in patients with valvular heart disease who had no chest pain history. The nine patients with coronary artery discasc and silent ischemia did not experience chest sensations of any sort during any of the testing procedures employed in this study .

JACC V.I . 1 6. No. 6 Novcinbr, 15, 1990 :1159_66

Patients were asked to decide whether the pain provoked during cardiac sensitivity testing was typical of their chest pain spontaneously experienced before this evaluation . The strikingly high prevalence of positive responses in patients with chest pain and angiographically normal coronary arteries and, to a lesser degree, in patients with hypertrophic cardiomyopathy, suggests that abnormal cardiac sensitivity may either be causally related or contributory to their chest pain syndrome . Cardiac versus cutaneous pain sensitivity. Increased cardiac pain sensitivity was unrelated to cutaneous thermal sensitivity. Indeed, in contrast to Turiel et al . (20), who demonstrated a lowered pain threshold in women with Syndrome X (chest pain, angiographically normal coronary arteries and ischemic appearing ST segment responses to exercise), we found that our patients with chest pain and normal coronary arteries (four of whom had ischemic appearing ST segment responses to exercise stress) actually had a higher cutaneous threshold to thermal pain than did patients with hypertrophic cardiomyopathy or coronary artery disease. Although our patients with coronary artery disease were older and included more met. than the other groups, it is unlikely that age and gender influenced cutaneous pain sensitivity testing . There was a clear difference in cutaneous pain threshold between patients with chest pain and normal coronary arteries and patients with hypertrophic cardiomyopathy despite similar age and gender distribution . Thus, abnormal cardiac sensitivity in patients with chest pain and normal coronary arteries does not appear to be linked with heightened sensitivity to cutaneous stimuli . Generalized abnormality of visceral noclceptlon . Of the 14 patients with chest pain and angiographically normal coronary arteries who underwent esophageal testing, 4 had their typical chest pain provoked by wet swallows, acid infusion or intraesophageal balloon distension, or 3 of these 4 also had their typical chest pain provoked during cardiac sensitivity testing . However, of the 10 patients without typical chest pain provoked during esophageal testing, 8 had their typical pain during cardiac sensitivity testing. Itichter et al. (24) noted increased intraesophageal sensitivity in patients with chest pain and normal or nearly normal coronary arteries regardless of the presence or absence of a motility disorder or abnormal results on Bernstein testing . Our study suggests that a generalized abnormality of visceral nocicep Lion may exist in some patients with chest pain and angiographically normal coronary arteries . Several studies (14-17) have reported a high prevalence of anxiety, depression, neuroticism, somatization, hypochondriasis and panic disorders in patients with chest pain and normal coronary arteries . However, our studies, when taken together, indicate that these patients often have an increased sensitivity to cardiac pain stimuli . The behavioral responses described by others as being of primary etiologic importance may be a seconeary consequence of pain . Alter-



[ACC Vol, 1 6, No. 6 November 15. 1990:1359-66

CANNON ET AL . ABNORMAL CARDIAC SENSITIVITY AND CHEST PAIN

1365

natively, these same disorders might result in heightened awareness and painful interpretation of visceral afferent neural stimuli. Cause of cardiac pain sensitivity . The cause of the heightened cardiac pain sensitivity detected in our study is unknown . Afferent sensory fibers, both myelinated and unmyelinated, follow sympathetic efferent nerves to sympathetic ganglia in the chest and neck, then to the spinal cord with connections to processing centers such as the locus ceruleus, which initiates "fight or flight" responses to the hypothalamus and thalamus, which relay signals to the cortex for cognitive recognition and interpretation and to effector sympathetic nerve fibers, which innervate the same viscera. Afferent sensory nerves also follow the vagus nerve to centers within the hypothalamus and thalamus with similar cortical and autonomic efferent connections (25-27) . Visceral sensitivity that elicits a painful response may he a consequence of abnormal function of any of these relay or processing centers or of their cortical connections. Increased cardiac sensitivity to painful stimuli may explain why the syndrome of coronary microvascular dysfunction, which limits appropriate coronary flow responses to stress but usually appears in association with mild or equivocal ischemia (11), causes chest pain that is often prolonged and severe . Patients with similar abnormalities of cardiac microvascular function unassociated with increased cardiac sensitivity to painful stimuli might go unrecognized . Many patients with chest pain and normal coronary arteries, who had no evidence of cardiac microvascular dysfunction, had heightened cardiac pain perception . Although the mechanisms that initiate chest pain in these patients are unknown, it is possible that their altered pain perception is the primary and sole cause of their chest pain syndrome : . Possible initiating mechanisms might include changes in loading conditions, heart rate, contractility or ectopic beats . Limitations of study . An alternative possibility is that patients with chest pain and normal coronary arteries or hypertrophic cardiomyopathy had normal intracardiac sensitivity to our testing and the remaining patient groups had decreased intracardiac sensitivity . However, the common provocation of pain characteristic for the patient leads us to believe that there is a causal relation of intracardiac pain sensitivity to symptoms in these patients . Because no "normal" patients free of chest pain or heart disease were studied, we do not know "normal" cardiac sensitivity . A further limitation is that not all patients nnderwcnt esophageal or cutaneous pain sensitivity testing because of testing and scheduling conflicts . There was no arbitrary decision to perform or exclude these tests in our patients . Although it is possible that the findings in the subset that did undergo these ancillary tests are not representative of the larger group, this seems unlikely . Conclusions. Abnormal cardiac sensory function is common in patients with chest pain and angiographicelly normal

coronary arteries with or without abnormal microvascular function. Furthermore, this phenomenon is also commonly seen in symptomatic patients with hypertrophic cardiomyopathy. Abnormal visceral sensory function may be causally related to the chest pain experienced by these patients or may contribute to ischemia-induced pain . The abnormal sensory function may also contribute to the atypical features of pain often experienced by these patients . Abnormal psychologic states may be a secondary phenomenon or amplify and exaggerate sensory signals that arrive from viscera into midbrain centers . In addition, activation of neural processing centers may activate the autonomic efferents resulting in somatic symptoms commonly ascribed to anxiety or panic attacks. At the other extreme, diminished or absent cardiac sensory perception may minimize a patient's recognition of myocardial ischemia and contribute to the phenomenon of silent myocardial ischemia . We thank June Moon and Toni Julia far their excellent secrelatidd assistance and gratefully acknowledge the technical assistance and support of Donna Fkagle . Greg Johnson. Kathy Maher. Rica Mincemoyer. Linda Scott. Susan Steete . Tim Stockdalc and lady Wiukler.

References 1.

2.

3.

4.

Precdfit WI .. shiny EK, Sones Flu. Selective eine coranary ancriogocphy: correlation with clinical findings in 1000 paliems. Circuialian 1366; 33 :901-10. CASS Principal lnvestigotars . Coronary Artery Surgery Study (CASSI: a randomized oral of coronary artery bypass surgery . Circulnhon 1983 ;48: 939-50. Mahrer PR, Eshoa N. Outpatient cardiac cathetoaizitioo and coronary angiogmphy . Cathel Cardiovasc Diagn 1981 ;7:355-60. Phibbs

B.

Fleming T . Ewy GA, et at. Frequency of normal coronary

arAriogrems in three academic medical centers and one community hospital . Am 1 Cordial 198$62:472-4 . 5.

Opherk D, Zcbe H, Weihe E, et al . Reduced coronary dilatory capacity

and ultraswaural changes of the myocardium in patients wilt angina 6.

1.

8.

pectoris but normal coronary arteriograms. Circulation 1981 ;63:817-25 . Cannon R0, Watson RM, Riming DR, Epstein SE . Angina caused by reduced vasodilator reserve of the small coronary arteries . I Am Coli Cardiol 1983 ;1 :1359-73 .

VirseslS .Rvidenceafmyomodiatisahemiainpatientswithahcn[pain syndromes and normal coronary angiograen . Acta Med Seand Suppl 1984 ;694:58-68. Gree'Serg MA. tense RM, Neuburger N, Silverman R, Stain JE, Cohen

MV . Impaired coronary vasodilator responsiveness as a cause of lactate production durhig lacing-induced ischemia in patients with antina pee9.

tons and normal coronary arteries . J Am Coll Cardirl 1987;9:743-51 . Legmld V, Hodgsan I Men. Bates ER, et al . Abnormal coronary flow

reserve and abnormal radionuclide exercise test results in patients with normal coronary angiogmms . J Am Colt Cardial 10.

11 .

1985 ;6:1245-53 .

Carson00 .Epstein5E ."Microvucularanginn"asacause efcheslPain with angiogmphically normal coronary arteries . Am I Cardio] 1988:61 : 1338-43 . unseen AS, Hess OM, Ebedi FR. et al . Abnormal coronary vasomotirm

during exercise in patients with normal coronary arteries and reduced coronary flow reserve . Circulation 1989;79:516-27.



CANNON ET AL . ABNORMAL CARDIAC SENSITIVITY AND CHEAT PAIN

1ACC Vu1. 16 . No. 6 Nuvumbcr I 5 . IM, 1359-6d

. Rich- SE. Eradley LA . Coolell DO, Eaopyapeal chant patio eanen0 enlrnvehiee in pathogen-, dlaonn0Cn and lhcrapy . Ann lnlem Med 1989 .115.66-78 . 13. Cannon R0 . Callao EL . Yakshc P6, Bcniumin 3E . Fps:cio SE . Coronary flaw reeervem esophageal -lily, and chess pain in pa0enin wish angmpraphiswllt normal coronary arteries . Am 1 Med 1990 :88 :2_27-32 . 14. Boss C, Wade C, Hand D, Jackson 6. Patients with angina wish normal

SE . Tnd_i M, Galasai AR, Glazicl2A Kssci JC . Mascu! A, Psalm c'neshold and tale-- la wo,aen with Synd,ome k and women ,nigh noble angina p aiceis. Am 1 Card l01 1987 :617.593-8 .

1366

and n

rumal cn

v arteries: clinical and psychnsorial slate 11

amhsalterangingraphor Br Med J 1983 :287:1505-8 . 15. Wielgo . AT . Plemher RH, MaCants CB, McKinnis RA . Honey TL . Williams PH . Unimproved chest pain in patients with minimal or nn co ronary disease : a behavioral phenomenon . Am Heart J 1984 :!08 :67-72 . IG Beitman BD . h7akerji V . Lamhedi .9W, e1 al . Punic disorder in patients with chant rain and anginpraphically norma! coronary arteries, Am J Candid 19890 :1399-403. 17. Innlinga LL Sprikin RP, klcCroakery JH, Baker MT, Warns, RA . Hill NH. (Joe-yns psychosocial follow-up of patients with chest pain and angiogrphically normal coronary arteries . Am J Cardiol 1989 :52.319-13 . . 18 Ockcrre IS. Shay Mi . Aspen J8 . Weiner BH . Dales JF . Unexplained chest pain in patients with normal coronary anesiegrams . N Engl J Med 1980;303 :1249-52. 19, Inert IN, Salem DN . Bans IS. Levine I!J . Long-term dimcal --of paiiems with normal coronary aneriography : followup study of 121 paum wnh normal or nearly normal coronary arteeiagrnma Am Head J 1991 :102:645-53 .

2t . 3hapka LM, Caike T, Poole-Wilson P1 . 1s altered cardiac --an responsible for chest pain in patients with normal coronary anerlas^. : clinical obsaraanon daring cardiac calheteriaaiierr Er Med 1 19R$295 170-I . 22 . Moron B7 . Epalein SE. Hyp.rtmphio cmrdiomyop,thy : a discussion

of

nnmencialure . Am J Cordial 1979 .43 :1242-4 . 23 . Oracely I1H. Lota L, WaImr DJ . Dabner R. A mahiple random slaieoase

method of psyehophysical pain assessment . Pain 1998;32:55-63. 2d . Richter IE, Bansh CF . Castell DO . Abnormal sensory pereeprion in palrents wilh esophageal chess pain. 6 stroemenalogy i986 :91 :n,4S-57. 25 . btalllan~ A. pagan! M . Lombardi F . Visceral versus io r0i5 mechanisms .

In: Wall PD . Melzack R. ads . Textbook of Pain. London: Churchill Lit Ingsmne. 1989:128-40. 26 . Svenssose TH . Peripheral. autonomic regulation of locus coemleus noradrenergic neurons in brain : palatial implicaloer far psychiatry and psnahophormacolagy. PsychopharmacologylBerlinl 1983 :91:1-? . 27 . Sylvan C . Angina peslnris : clinical charaoaeristiaca and ran ccular necherisms . Pain 1989 :36:145-67.

rmurophysio'ogkal