Abstract
Background:
The role of myocardial perfusion imaging (MPI) in patients with suspected coronary artery disease (CAD) presenting with syncope is controversial. We aimed to determine diagnostic yield of MPI for evaluation of syncope in patients without known CAD, as a function of pre-test patient risk and test modality (PET vs SPECT).
Methods:
Between 1/2010 and 12/2016, 1324 consecutive patients presenting with syncope without known CAD underwent MPI with PET (n=640) or SPECT (n=684). Rates of abnormal MPI (summed difference score (SDS) >2 or left ventricular ejection fraction (LVEF) reserve ≤0 for PET and SDS > 2 or post-stress LVEF ≤ 45% for SPECT) were determined among patients stratified by pre-test risk. In patients who were referred for coronary angiography, diagnostic yield of obstructive CAD was calculated in the overall cohort as well as in a propensity-matched cohort compared to patients without syncope.
Results:
Abnormal MPI was noted in 36.5% (201/551) of patients who had PET compared with 13.0% (87/671) who had SPECT (p <0.001), which is largely related to higher comorbidity burden and greater pre-test CAD risk in the PET population. Among patients who had an abnormal MPI, 8.5% (47/551) with PET and 0.7% (5/671) with SPECT were found to have obstructive CAD if referred for coronary angiography. Patients at intermediate-high pre-test risk had a higher proportion of abnormal MPIs and obstructive CAD as compared to those at low-risk in both the PET and SPECT cohorts. The rate of abnormal testing and diagnostic yield of PET MPI was similar and proportionate to pre-test likelihood among matched patients with and without syncope.
Conclusions:
Among patients referred for PET MPI with syncope at an intermediate-high pre-test CAD risk, 1 in 3 had an abnormal MPI and 1 in 10 have obstructive CAD. The value of MPI was related to pre-test risk as opposed to the presence of syncope, and MPI testing with PET or SPECT in the low-risk population was low value.
INTRODUCTION
Syncope is a transient loss of consciousness due to global cerebral hypoperfusion. This common condition affects 20-40% of adults in their lifetime, leading to many emergency room visits and hospitalizations with an estimated associated cost of more than $2 billion annually (1-6). A particularly challenging component of syncope management is that up to 40% of patients admitted with syncope do not have an identifiable etiology (1,7). Coronary artery disease (CAD) has been reported as a cause in 4-6% of patients, and in those who have CAD, there is a higher associated mortality (1,7). Cardiac ischemia can cause syncope through depressed cardiac output and hypotension, ventricular arrhythmias, or atrioventricular block (5).
Multi-society appropriate use criteria have supported the use of stress imaging in the diagnostic workup of syncope to assess for obstructive CAD if patients are at intermediate or high risk for CAD (8-10). However, a prior study evaluating this recommendation revealed a low diagnostic yield of myocardial perfusion imaging (MPI), predominantly performed with single photon emission computed tomography (SPECT), in patients with syncope regardless of their pre-test likelihood for CAD (11). Given the lack of evidence regarding the benefit of MPI in evaluation of syncope, most recent guidelines only recommend stress testing, mainly exercise ECG for exercise-induced syncope (12, 13). Positron emission tomography (PET) MPI offers the unique ability to measure perfusion defects, myocardial flow reserve, and left ventricular ejection fraction at peak stress in a single study. PET MPI also has higher spatial resolution and diagnostic accuracy than SPECT MPI. Whether these attributes of PET make it more suitable to assess for ischemia and obstructive CAD as the underlying etiology for syncope is unclear based on the currently available evidence.
Given the limited evidence surrounding the utility of MPI to diagnose CAD as a part of the work-up of patients with syncope, we aimed to determine the yield of MPI in the evaluation of syncope and compared the diagnostic utility of PET with SPECT. We hypothesized that patients who are referred for PET are higher risk, with multiple comorbidities and limited ability to exercise, may have a higher diagnostic yield for CAD, in comparison to what is previously reported in the literature. Our study focused on the group where there is greatest uncertainty, those with syncope and no known CAD.
METHODS:
Study population
We included all consecutive patients who had a stress MPI test performed for the evaluation of syncope at any of the Saint Luke’s Health System nuclear cardiology laboratories between January 2010 and December 2016. Patients with known CAD (defined as a history of myocardial infarction or coronary revascularization), severe valvular disease, or prior heart transplant were excluded. Patients who underwent rest only imaging, viability studies, or who had uninterpretable imaging results were also excluded. The Saint Luke’s Health System includes 4 hospital-based nuclear cardiology laboratories with access to both cardiac PET and SPECT MPI within the Kansas City metropolitan area, and 5 other affiliated, community-based sites with access to SPECT MPI. After the MPI study is performed, all images are electronically transmitted to a central nuclear cardiology laboratory where they are processed by experienced nuclear technologists and interpreted by a core group of 5 physicians.
The study was approved by the IRB at Saint Luke’s Hospital of Kansas City. The requirement for informed consent was waived due to the retrospective nature of the study and use of a de-identified dataset for analysis.
Study variables
Trained clinical personnel collected information on patients’ symptoms, demographics, clinical risk factors, and medications through patient interview and chart review, and verified these against the indication for the study at the time of the study. All data were prospectively entered into an electronic database at the time of the test and were verified at the time of study interpretation by the reading nuclear cardiologist. Patients with syncope as the primary symptom were identified by querying this database. Chest pain was categorized into typical, atypical, and non-anginal chest pain using Diamond-Forrester classification (14). For the current study, we examined the yield of stress MPI testing in the overall cohort and in the following two subgroups: 1) Patients presenting with syncope as an isolated symptom, and 2) patients with syncope and other additional symptoms (chest pain or dyspnea). The pre-test probability of obstructive CAD was calculated for all patients using the Clinical CAD Consortium score comprised of baseline patient characteristics that were collected at the time of stress testing (15). Patients’ risks were stratified into low, intermediate, and high pre-test likelihood of CAD defined as <15%, 15-85%, and >85%, respectively (14).
Etiology for syncope was abstracted from the electronic health record using manual chart review. Other cardiac testing done for evaluation of syncope performed within 6 months of the MPI was also abstracted from the medical record.
Stress MPI Protocols
MPI stress testing was performed per standard ASNC guidelines (16, 17). All SPECT studies were performed using either a small or large field of view Anger SPECT (CardioMD, Cardio60™, or CardioEPIC™ all with line-source attenuation correction; Philips Medical Systems, Milpitas, CA; or Symbia series SPECT/CT cameras) system or a D-SPECT Solid State Detector-SPECT camera (Spectrum Dynamics, Sarasota, FL). All PET studies were performed on a dedicated PET (Siemens ECAT ACCEL) or PET/CT camera (Siemens Biograph 16 or 64, Nashville, TN).
For exercise Tc99m SPECT MPI, the Bruce or modified Bruce protocol was used. Patients unable to exercise or those with a suboptimal response (< 8 METS) to exercise received pharmacologic stress using standard doses of IV regadenoson (n=288), IV dipyridamole (n=80), or adenosine (n=56). All patients undergoing Rb82 PET MPI received pharmacologic stress with IV regadenoson (n=405) or dipyridamole (n=235). Patients were monitored for heart rate, blood pressure, rhythm, symptoms, and ECG appearances at baseline, at every stage of the exercise or pharmacologic stress protocol, and for at least 4 minutes following stress or until baseline appearances were present. All studies were ECG-gated in the absence of technical problems. For selected patients undergoing SPECT MPI, a stress only protocol was used (n=354), and patients were only called back for rest images if deemed necessary by the interpreting cardiologist (18). Morbidly obese patients (BMI ≥ 40 kg/m2) and those tested at 2 affiliated community-based hospitals underwent a 2-day rest/stress or stress/rest SPECT gated MPI (n=74). All patients without a prior calcium score who were imaged using a SPECT or PET/CT camera, had a separate CT acquisition post MPI study for coronary calcium scoring (Siemens Syngo Calcium Scoring software, Erlangen, Germany).
The MPI images were quantified using commercial software (Cedars Sinai QPS for SPECT, QPET for PET; Los Angeles, CA) and a 17-segment model and standard 5-point scoring system were used for semi-quantitative interpretation of perfusion images (17). The presence and extent of perfusion defect (summed stress score [SSS], summed difference score [SDS], and summed rest score [SRS]) were quantified. Gated myocardial perfusion images acquired with 8-frame gating were used to calculate rest and stress LVEF. LVEF reserve for PET was estimated as the difference between peak stress LVEF and rest LVEF. Myocardial blood flow (ml/min/g) on PET was calculated both at rest and stress using previously validated software (Imagen Q, Kansas City, MO, USA) (19, 20). MFR was calculated as the ratio of stress to rest myocardial blood flow.
Study Outcomes
To assess the yield of MPI testing for syncope, the proportion of abnormal MPI studies in each group was identified. For the primary analysis, an abnormal PET MPI was defined as any ischemia (SDS >2) or LVEF reserve ≤0, and an abnormal SPECT MPI as any ischemia (SDS > 2) or post-stress LVEF ≤45% (16-17, 21-22). In a sensitivity analysis, to evaluate if the yield would differ based on different definitions for abnormal MPI, we assessed the rate of abnormal MPI using the following alternative definitions: any perfusion defect size >5% (SSS ≥ 3), any ischemia (SDS > 2), significant ischemia (SDS > 7), presence of any fixed perfusion defect (SRS > 0) (11, 16-17). For patients who underwent PET MPI, we incorporated MFR and LVEF reserve information and additionally reported the rates of abnormal MPI using the following cut-offs: MFR ≤ 1.8 and LVEF reserve ≤ 0 (21, 23).
The hospital angiography database and medical chart review were used to identify patients who had a coronary angiography performed within 6 months of an abnormal MPI. However, not all patients with an abnormal MPI underwent coronary angiography and it was at the discretion of the treating physician. Results of angiography were abstracted by manual chart review and validated by 2 authors (M.T and K.K.P). Diagnostic yield was defined as the rate of obstructive disease (left main coronary artery stenosis ≥ 50% or any epicardial coronary vessel stenosis ≥ 70% as per report of the procedural physician) on coronary angiography. All coronary angiograms were performed following stress MPI.
Statistical Analysis
Baseline characteristics were compared between patients presenting with syncope as an isolated symptom and those with syncope and additional symptoms using t-test or Wilcoxon rank sum test for continuous variables and chi-square, or Fisher exact test for categorical variables. Rates of abnormal MPI and obstructive disease on angiography within 6 months of the test were estimated for the overall cohort and both groups separately and stratified by pre-test probability of CAD. These results were compared using chi-square or Fisher exact test.
To assess if symptom of syncope by itself was related to a higher yield compared to patients referred for MPI for other indications, we conducted a matched propensity score analysis to create a cohort of patients with similar characteristics in each imaging modality who did not have syncope. Using logistic regression, a propensity score model was created to calculate the probability of presenting with syncope. The model included pre-test likelihood of CAD, age, sex, diabetes, hyperlipidemia, hypertension, heart failure, peripheral vascular disease, cerebrovascular disease, family history of CAD, smoking status, type of chest pain, and dyspnea. Then, we matched syncope patients with PET testing to patients without syncope with PET testing using a 1:1 ratio by greedy match with a caliper width of no more than 0.2 of the standard deviation. A similar propensity score match was conducted for patients with and without syncope undergoing SPECT testing. Balance of covariates before and after matching was assessed using standardized differences with > 10% being considered clinically relevant. Rates of abnormal MPI and subsequent obstructive coronary artery disease on follow-up coronary angiogram within 6 months of test (if performed) were compared between propensity-matched groups of patients presenting with and without syncope using chi-square or Fisher’s exact test.
All analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC). Two-sided p-values <0.05 were considered to be statistically significant.
Missing Data
For patients who underwent PET MPI, MFR and LVEF reserve data was missing for 86 (13.4%) and 89 (13.9%), respectively. For those who underwent SPECT, post-stress LVEF was missing in 13 (1.9%). These patients were excluded from the analyses in which these factors were considered part of the definition for an abnormal stress test, as they would not have been available for clinical decision making.
RESULTS
Of 2,389 patients who underwent an MPI stress test ordered for syncope, 1,065 were excluded (1,053 with history of known CAD, 3 with known severe valvular disease, 5 non-diagnostic studies, 4 rest only imaging, and 3 viability studies). The remaining 1,324 patients were included in our study cohort (Table 1). The average age of the study cohort was 64.4 ± 14.2 years and 46.8% were male, 940 (71%) presented with syncope and other additional symptoms [712 (53.7%) chest pain and 658 (49.7%) dyspnea]. Overall, a fifth of all patients (n=266, 20.1%) had exercise treadmill SPECT MPI, while the majority (n=1058, 79.9%) were stressed pharmacologically. The most common etiology of syncope was idiopathic, followed by vasovagal/neurocardiogenic syncope (Table 2). A small proportion of patients had exercise-related syncope (2.9%). In addition to MPI, 612 patients (45.8%) underwent echocardiography.
Table 1:
Comparison of patient characteristics, by type of MPI test
| PET MPI (N= 640) |
SPECT MPI (N=684) |
P-value | |
|---|---|---|---|
| Age, years | 70.5 ± 13.5 | 60.0 ± 13.5 | < 0.001 |
| Male | 283 (44.2%) | 337 (49.3%) | 0.07 |
| Hypertension | 475 (74.2%) | 436 (63.7%) | < 0.001 |
| Diabetes | 166 (25.9%) | 162 (23.7%) | 0.34 |
| Stroke | 86 (13.4%) | 41 (6.0%) | < 0.001 |
| Cardiomyopathy | 58 (8.9%) | 20 (2.9%) | < 0.001 |
| Smoker | 101 (15.8%) | 140 (20.5%) | 0.03 |
| Hyperlipidemia | 417 (65.2%) | 410 (59.9%) | 0.05 |
| Peripheral vascular disease | 56 (8.8%) | 42 (6.1%) | 0.07 |
| History of VT | 14 (2.2%) | 16 (2.3%) | 0.85 |
| History of atrial arrhythmia | 165 (25.8%) | 83 (12.1%) | < 0.001 |
| ICD device | 13 (2.0%) | 9 (1.3%) | 0.31 |
| Pacemaker | 59 (9.2%) | 29 (4.2%) | < 0.001 |
| Left bundle branch block | 29 (4.5%) | 17 (2.5%) | 0.04 |
| Right bundle branch block | 42 (6.6%) | 39 (5.7%) | 0.51 |
| Interventricular conduction delay | 7 (1.1%) | 16 (2.3%) | 0.08 |
| Baseline ST changes | 50 (7.8%) | 88 (12.9%) | 0.002 |
| LVEF ≤ 40%* | 39 (7.0%) | 14 (4.4%) | 0.24 |
| Family history of CAD | 231 (36.1%) | 323 (47.2%) | < 0.001 |
| Calcium score ≥ 100+ | 217 (49.8%) | 63 (33.7%) | <0.001 |
| Calcium score < 100+ | 219 (50.2%) | 124 (66.3%) | <0.001 |
| Pretest likelihood | < 0.001 | ||
| Low | 89 (13.9%) | 176 (25.7%) | |
| Intermediate | 538 (84.1%) | 506 (74.0%) | |
| High | 13 (2.0%) | 2 (0.3%) | |
| Chest pain | 0.24 | ||
| Typical | 0 (0%) | 3 (0.4%) | |
| Atypical | 334 (52.2%) | 375 (54.8%) | |
| Dyspnea | 288 (45.0%) | 370 (54.1%) | < 0.001 |
Continuous variables presented as mean ± SD and compared using t-test, categorical variables are presented as N (%) and compared using chi square or Fisher Exact test. Other symptoms include chest pain and dyspnea. VT= ventricular tachycardia, SVT= supraventricular tachycardia, PET= Positron Emission Tomography, SPECT= Single Photon Emission Computed Tomography, MPI=myocardial perfusion imaging, ICD=implantable cardioverter-defibrillator.
Missing 11.7% of data in PET and 2.8% of data with SPECT.
Missing 31.9% of data with PET and 72.6% of data with SPECT
Table 2:
Etiology of Syncope
| Etiology | PET (n=640) |
SPECT (n=684) |
Total (n=1324) |
|---|---|---|---|
| Idiopathic | 101 (15.8%) | 293 (42.8%) | 394 (29.8%) |
| Vasovagal/neurocardiogenic | 210 (32.8%) | 171 (25.0%) | 381 (28.8%) |
| SVT | 30 (4.7%) | 52 (7.6%) | 82 (6.2%) |
| Orthostatic | 47 (7.3%) | 27 (3.9%) | 74 (5.6%) |
| Sick sinus syndrome | 49 (7.7%) | 21 (3.1%) | 70 (5.3%) |
| Pulmonary hypertension | 45 (7.0%) | 24 (3.5%) | 69 (5.2%) |
| Bradycardia | 46 (7.2%) | 19 (2.8%) | 65 (4.9%) |
| Ventricular tachycardia | 32 (5.0%) | 24 (3.5%) | 56 (4.2%) |
| Exercise-induced | 11 (1.7%) | 27 (3.9%) | 38 (2.9%) |
| Complete heart block | 25 (3.9%) | 9 (1.3%) | 34 (2.5%) |
| Aortic stenosis | 26 (4.1%) | 8 (1.2%) | 34 (2.5%) |
| Hypertrophic obstructive cardiomyopathy |
7 (1.1%) | 3 (0.4%) | 10 (0.7%) |
| 2nd degree heart block type 2 | 8 (1.3%) | 2 (0.3%) | 10 (0.7%) |
| Wolf Parkinson White | 2 (0.3%) | 4 (0.6%) | 6 (0.5%) |
| Long QT | 1 (0.2%) | 0 (0%) | 1 (0.1%) |
Approximately half the cohort underwent PET MPI (n = 640, 48%) and the other half SPECT (n = 684, 52%). Patients who underwent PET were older and were more likely to have co-morbid hypertension, cerebrovascular disease, cardiomyopathy, and history of SVT, LBBB on baseline EKG, and a history of pacemaker implantation (Table 1). They were also less likely to be smokers, to have ST segment changes on baseline EKG, and to be presenting with dyspnea.
Diagnostic yield of PET MPI
Using an integrated definition of an abnormal PET MPI that incorporated ischemia and LVEF reserve, PET MPI was abnormal in 201 of 551 patients (36.5%): 125 had SDS >2, 118 had LVEF reserve ≤0, and 42 with both. Thirty-four percent of these patients (69/201) underwent an angiogram within 6 months following MPI. Of these, 47 had obstructive CAD, resulting in a diagnostic yield of 8.5% (47/551). Half of these patients had 2- or 3-vessel disease (Table 3). Using SSS ≥ 3 to define an abnormal test as in prior literature, 23.6% (151/640) of PET MPIs were abnormal, and 7.2% (46/640) of patients who were referred to coronary angiography after an abnormal PET had obstructive CAD (11).
Table 3:
Number of Diseased Vessels in Patients with Obstructive CAD on Coronary Angiography
| PET MPI (n=47) |
SPECT MPI (n=5) |
|
|---|---|---|
| 1-vessel | 23 (50.0%) | 0 (0%) |
| 2-vessel | 22 (47.8%) | 2 (33.3%) |
| 3-vessels or left main |
2 (2.2%) | 3 (66.7%) |
Global MFR was abnormal in 40.6% (225/554) of patients who underwent PET, and 60.9% (137/225) of these patients did not have any evidence of abnormal perfusion (SSS <3). Significant reversible perfusion defect (SDS >7) was present in 53 patients (8.3%) and rate of obstructive CAD on follow-up for these patients was 5.3% (34/640). The majority of the patients (10/12, 83.3%) with a smaller perfusion defect (SDS between 3 and 7) who had obstructive CAD on follow-up coronary angiograms, had low MFR (≤1.8) and would have been identified after incorporating MFR along with significant reversible perfusion defect to define abnormality.
Additionally, there was not a statistically significant difference in the diagnostic yield between patients presenting with isolated syncope as compared to those presenting with syncope and additional symptoms suggestive of CAD, regardless of the definition used for an abnormal PET MPI (Supplemental Table 1).
Diagnostic yield of PET MPI stratified by pretest likelihood of CAD
In patients undergoing PET MPI, 15.2% had a low pretest likelihood of CAD (84/551), 82.6% (455/551) had an intermediate pretest likelihood, and 2.2% (12/551) had a high pretest likelihood. Rates of obstructive CAD on subsequent coronary angiograms after an abnormal MPI in these groups were 1.2% (1/84), 9.5% (43/455), and 25.0% (3/12), respectively (Figure 1). In those at intermediate-to-high pretest likelihood, PET MPI identified 1 in 10 patients with obstructive CAD. Diagnostic yield of SPECT did not differ across pre-test likelihood categories (Supplemental Figure 1).
Figure 1: Yield of positron emission tomography myocardial perfusion imaging in patients with syncope, stratified by pre-test risk probability for coronary artery disease.
Diagnostic yield defined as presence of obstructive CAD (left main ≥50% or any epicardial coronary artery ≥70% stenosis) on coronary angiogram within 6 months of the stress MPI test. Pre-test CAD risk calculated using CAD Consortium Risk Calculator. Low risk = 0-14%, Intermediate= 15-85%, High= 86-100%. Cath= coronary angiogram. Abnormal MPI refers to MPI with summed difference score >2 or left ventricular ejection fraction reserve ≤0.
Diagnostic Yield of PET MPI vs SPECT MPI
Overall, there were more abnormal tests among those referred for PET MPI (36.5% [201/551]) as compared to SPECT (13.0% [87/671], p<0.001), and obstructive CAD was more frequently found among patients with an abnormal PET who were referred to coronary angiography (8.5% [47/551]) as compared to those referred after an abnormal SPECT (0.7% [5/671], p<0.001). Rate of obstructive CAD on subsequent coronary angiograms after an abnormal SPECT MPI was similar among low and intermediate-high pre-test risk, 0% (0/175) vs. 1.0% (5/496), p= 0.33 (see Table 5 and Supplemental Figure 1). The higher rate of abnormal MPI and subsequent obstructive CAD by PET vs. SPECT was present regardless of how a positive MPI was defined, in patients with isolated syncope and syncope with additional symptoms, and when stratified by pre-test risk (Table 4 and Supplemental Table 1).
Table 5:
Comparing Diagnostic Yield of stress MPI for Different Pre-test Likelihood of CAD
| PET | SPECT | p-value | |
|---|---|---|---|
| Low pre-test likelihood |
1/84 (1.2%) | 0/175 (0%) | 0.32 |
| Intermediate-to-high pre-test likelihood |
46/467 (9.6%) | 5/496 (1.0%) | <0.001 |
| p-value | 0.01 | 0.33 |
PET = positron emission tomography, SPECT = single photon emission computed tomography
Table 4:
Comparing Diagnostic Yield of stress MPI using Different Definitions for a Positive Test Result
| Abnormal MPI | Obstructive CAD* | |||||
|---|---|---|---|---|---|---|
| PET | SPECT | p-value | PET | SPECT | p-value | |
| Abnormal PET MPI: SDS >2 or LVEF Reserve ≤ 0; Abnormal SPECT MPI: SDS >2 or post- stress LVEF ≤ 45% for SPECT |
201/551 (36.5%) | 87/671 (13.0%) | <0.001 | 47/551 (8.5%) | 5/671 (0.7%) | <0.001 |
| Abnormal perfusion (SSS ≥3) | 151/640 (23.6%) | 98/687 (14.0%) | <0.001 | 46/640 (7.2%) | 6/687 (0.9%) | <0.001 |
| Any ischemia (SDS >2) | 125/640 (19.5%) | 66/687 (9.6%) | <0.001 | 46/640 (7.2%) | 4/687 (0.6%) | <0.001 |
| Significant ischemia (SDS >7) | 53/640 (8.3%) | 18/687 (2.6%) | <0.001 | 34/640 (5.3%) | 2/687 (0.3%) | <0.001 |
| Any fixed defect (SRS >0) | 94/640 (14.7%) | 78/687 (11.4%) | 0.07 | 17/640 (2.7%) | 4/687 (0.6%) | 0.002 |
| Abnormal PET: LVEF Reserve ≤ 0 Abnormal SPECT: post-stress LVEF ≤ 45% |
118/551 (21.4%) | 32/671 (4.8%) | <0.001 | 21/551 (3.8%) | 3/671 (0.4%) | <0.001 |
| Global MFR ≤ 1.8 | 225/554 (40.6%) | --- | --- | 30/554 (5.4%) | --- | --- |
PET = positron emission tomography, SPECT = single photon emission computed tomography, SSS = summed stress score, SDS = summed difference score, SRS = summed rest score, MFR = myocardial flow reserve, LVEF = left ventricular ejection fraction
Numerator is the number of patients with obstructive CAD among those with an abnormal MPI and referred to a coronary angiogram.
False Positive and False Negatives
For patients who underwent PET MPI, 22 (31.8%) of 69 patients who had a positive test and underwent coronary angiography did not have obstructive CAD (false positive rate). On the other hand, there were 15 (4.3%) of 350 who had a negative test and underwent coronary angiography within 6 months of their test; of these, three had obstructive CAD (false negative rate). False positive rate was higher with SPECT: 16 (76.2%) of 21 patients with an abnormal SPECT and underwent coronary angiography did not have obstructive CAD. One (0.2%) of the 584 patients who underwent SPECT and had a negative test underwent coronary angiography within 6 months and had obstructive disease.
Comparison between patients with and without syncope
All 640 patients who underwent PET MPI for evaluation for syncope were propensity-matched 1:1 to patients without syncope (Supplemental Table 2). After excluding matched pairs with missing PET MPI data, 491 patients with syncope and 491 without were included in this analysis. There was no difference in the proportion of patients with an abnormal PET MPI in patients with or without syncope (169/491 [34.4%] in both, p = 1.00). For those who were referred for coronary angiography within 6 months, there was no difference in the prevalence of obstructive CAD in patients with versus without syncope (41/491 [8.4%] in patients with syncope and 39/491 [7.9%] in patients without syncope, p = 0.82). This finding was similar whether patients had a low or intermediate-to-high pre-test likelihood (Supplemental Table 3).
Similarly, 684 patients who underwent SPECT MPI for evaluation of syncope were matched to patients without syncope (Supplemental Table 4). Of these, 665 matched pairs were included in this analysis after excluding pairs with missing SPECT data. There was no difference in the proportion of patients with an abnormal SPECT MPI in patients with or without syncope (86/665 [12.9%] in patients with syncope and 83/665 [12.5%] in patients without syncope, p = 0.80). Although the prevalence of obstructive CAD was low in both groups among those who were referred to coronary angiography within 6 months, obstructive CAD was more common in patients without syncope compared to those with syncope (14/665 [2.1%] vs 5/665 [0.8%], p = 0.04). This was driven by patients at intermediate-to-high pre-test risk, as no patients with low pre-test risk were found to have obstructive CAD (Supplemental Table 5).
DISCUSSION
In this large single-center study of 1324 consecutive patients with syncope and without known CAD, we found that the rate of abnormal testing and subsequent identification of obstructive CAD on coronary angiography was higher in patients who underwent PET MPI testing as compared to SPECT, regardless of the presence of additional symptoms suggesting ischemia such as chest pain or dyspnea. One in 12 (8.5%) patients who underwent coronary angiography within 6 months of PET MPI were diagnosed with obstructive CAD, while this rate was only 0.7% in patients who underwent SPECT MPI. These findings are likely driven by higher-risk patients undergoing PET as these patients were older and had more comorbidities, often being unable to exercise. However, even among the intermediate-to-high risk patients who underwent SPECT, the rate of obstructive CAD on coronary angiography was low and did not differ significantly from low-risk patients who underwent SPECT. On the other hand, the rate of obstructive CAD in patients who underwent PET MPI was significantly higher in patients at intermediate-to-high pretest likelihood of CAD as compared to low pre-test likelihood. Yield of PET MPI testing was similar and proportionate to pre-test likelihood among matched patients with and without syncope.
Diagnostic yield of MPI for syncope has been evaluated in another large single center cohort previously. In a study by Al Jaroudi, et al., the authors demonstrated a low diagnostic yield of MPI (1.3%) for evaluation of syncope in patients without coronary artery disease across low-, intermediate-, and high-risk patients (11). That study included only patients with syncope as an isolated symptom and less than 5% of patients were evaluated with PET MPI; additionally, half of the patients in the study had a low pre-test likelihood, which is not the patient population identified as appropriate for such testing. Our study expands upon their findings in several ways. First, we evaluated the role of PET MPI in the work up of syncope and compared it with SPECT MPI. To our knowledge, this is the first study to compare PET vs. SPECT MPI for the evaluation of patients with syncope. We found that PET performs better at identifying patients who are at risk for having obstructive CAD on coronary angiography. We also noted that there was a higher yield in patients at intermediate-to-high pre-test likelihood of CAD in those undergoing PET, which was not found in patients evaluated by SPECT, as reported in the prior study (11). Our cohort also included patients who had other symptoms such as chest pain or dyspnea in addition to syncope, which increases physician suspicion for significant CAD as a possible cause of their presentation. However, we found similar results for patients with isolated syncope as the presenting symptom and those with syncope and additional ischemic symptoms.
Another novel insight we gained is that as much as a quarter of patients with syncope undergoing PET had evidence of low myocardial flow reserve without a perfusion defect suggestive of underlying coronary microvascular dysfunction on their tests, regardless of the presence of additional symptoms that would be more typical. This could certainly be related to the higher burden of comorbidities such as hypertension, diabetes, old age, frailty, etc. among these patients. While the relationship between low myocardial flow reserve and syncope is not understood, the high prevalence of coronary microvascular dysfunction among patients presenting with syncope is noteworthy.
Yield of testing was similar with PET for patients with and without syncope, and it was higher for patients without syncope among intermediate-high risk patients referred for SPECT. This suggests that yield of MPI testing is proportionate to patient’s pre-test likelihood of CAD, and the symptom of syncope by itself does not modify the pre-test risk. Conversely, the yield of testing is not low as previously described (11) among patients with intermediate-high pre-test likelihood, even if they have syncope as the only presenting symptom. Therefore, MPI testing, especially with PET, may be reasonable to serve as a gatekeeper for invasive testing to rule out obstructive CAD among patients with syncope at an intermediate to high pre-test risk of CAD. There appears to be little role for MPI testing with any modality in patients at low-risk for CAD presenting with syncope; the yield of testing is low in these patients regardless of the modality used or concomitant symptoms.
Based on the general absence of data, more recent guidelines do not provide recommendations for testing patients with syncope using any stress testing modality, except exercise stress testing in cases of exercise-induced syncope (22, 23). On the other hand, the 2013 Multimodality Appropriate Use Criteria for the Detection and Risk Assessment of Stable Ischemic Heart Disease rates radionuclide imaging as appropriate for evaluation of syncope in patients at intermediate/ high-risk and rarely appropriate in patients at low-risk for CAD (10). Our data support the Appropriate Use Criteria recommendations. In practice, although MPI is generally appropriate for patients with intermediate-to-high pretest likelihood of CAD regardless of presenting symptoms, for patients presenting with syncope, MPI is often performed for low risk patients with syncope to rule out significant CAD in the absence of other causes. The findings from our study help to clarify the appropriate clinical setting in which stress MPI should be considered for the evaluation of syncope. Regardless of whether syncope was the only presenting symptom or other symptoms of suspected ischemia are present in addition to syncope, patients at intermediate- to high-risk for CAD who underwent PET MPI had a higher proportion of abnormal MPI studies and a higher rate of obstructive CAD on follow up angiography, while the patients at a low pre-test likelihood of CAD had a very low yield for both PET and SPECT. This suggests that stress MPI testing for ischemia with any modality should not be a part of routine clinical work-up for patients with syncope without any cardiac risk factors who are otherwise at a low pre-test likelihood of coronary artery disease. However, MPI can be useful in certain settings. PET provides additional information on myocardial blood flow reserve and changes in LVEF peak stress, which increases the diagnostic accuracy for obstructive CAD and also increases physician confidence in a negative test to rule out significant obstructive CAD. In patients with syncope and suspected myocardial ischemia contributing to the presentation, who are unable to exercise and are at an intermediate-high pre-test risk, stress MPI with PET can help identify obstructive CAD, with a yield as high as 9.6%.
While our study provides a comprehensive report on the utility of stress MPI testing for syncope evaluation in a large health system across 9 small and large sites, it has some potential limitations. We were unable to determine whether patients referred for MPI had a single syncopal episode or recurrent syncope. Also, we were unable to determine the medications patients were taking at the time of their syncopal event as our database included only those medications a patient was taking at the time of MPI. Not all patients with an abnormal MPI had a post-test coronary angiogram, which would affect the diagnostic yield. Additionally, patients may have undergone coronary angiography following a positive MPI at a different center leading to the appearance of a lower diagnostic yield; though, this would have been unlikely given the practice pattern within our health system. Some patients with syncope may have been referred for ischemia testing using other modalities such as stress echocardiography, stress magnetic resonance imaging, or direct coronary angiography; these patients could not be identified in this study and are not included. This is especially important for low-risk patients who may be referred for exercise treadmill ECG testing or stress echocardiography only, which could lead to patients with higher pre-test being evaluated by MPI and thus increasing the diagnostic yield of this imaging modality. It is also likely that patients who have other symptoms such as chest pain or dyspnea with syncope are more likely to have other reasons for their syncope, such as pulmonary embolism or dissection, which when detected, would not prompt the physicians to look for ischemia as a possible etiology. Our institutional practice favors stress MPI for testing for CAD and ischemia in patients with syncope; however, different institutional patterns for syncope testing may affect diagnostic yield as well. Larger multicenter and multimodality studies in patients with syncope are needed to improve our understanding on how best to use diagnostic testing for risk-stratification and to optimize resource utilization and outcomes.
CONCLUSION
Among patients presenting with syncope who underwent MPI testing to rule out significant CAD across 9 sites in a large health system, 1 in 12 patients undergoing PET had evidence of obstructive CAD on follow-up. Diagnostic rate of obstructive CAD was higher in patients undergoing PET compared to SPECT, likely due to higher risk patients undergoing PET. The rate of abnormal testing and diagnostic yield of PET MPI was related to pre-test likelihood of CAD as opposed to the presence of syncope as a symptom. Our study suggests that use of MPI, especially PET, as a gatekeeper for invasive testing might be reasonable among patients with syncope at an intermediate-high pre-test risk. Conversely, MPI with either modality, PET or SPECT, is a low value test for patients with syncope and a low pre-test likelihood.
New Knowledge Gained:
One in 3 patients with syncope without known CAD referred for PET had an abnormal MPI finding. Among these patients with abnormal PET who underwent subsequent coronary angiograms, 1 in 12 patients had obstructive CAD, with higher yield in patients at intermediate-to-high pre-test likelihood. This diagnostic rate is higher than what was previously reported and higher compared to patients referred to SPECT MPI within our system. The value of MPI was related to pre-test likelihood of CAD as opposed to the presence of syncope, and MPI testing with PET or SPECT in the low-risk population was low value.
Supplementary Material
Supplemental Figure 1: Yield of single photon emission computed tomography myocardial perfusion imaging in patients with syncope, stratified by pre-test risk probability for coronary artery disease
Table 6:
Comparison of Propensity Matched Patients with and without Syncope
| Syncope | No Syncope | p-value | |
|---|---|---|---|
| PET MPI | |||
| Abnormal MPI * | 169/491 (34.4%) | 169/491 (34.4%) | 1.00 |
| Obstructive CAD | 41/491 (8.4%) | 39/491 (7.9%) | 0.82 |
| SPECT MPI | |||
| Abnormal MPI † | 86/665 (12.9%) | 83/665 (12.5%) | 0.80 |
| Obstructive CAD | 5/665 (0.8%) | 14/665 (2.1%) | 0.04 |
SDS >2 or LVEF Reserve ≤ 0.
SDS >2 or post-stress LVEF ≤ 45%
Acknowledgments
The authors have no acknowledgements or sources of funding to disclose for this manuscript.
Disclosures:
Merrill Thomas, Krishna K. Patel, Poghni Peri-Okonny, Firas Al Badarin are supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number T32HL110837 (Thomas, Patel, Peri-Okonny, Al Badarin) and 1 R01HL123980 (Chan). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
John A. Spertus receives research grant support from Abbott Vascular, Novartis and is the PI of an analytic centre for the American College of Cardiology. He serves as a consultant to United Healthcare, Bayer, Janssen, AstraZeneca, and Novartis. He has an equity interest in Health Outcomes Sciences.
Timothy M. Bateman receives research grant support from Astellas and GE Healthcare. He serves as a consultant to Curium and GE Healthcare. He has an equity interest in Cardiovascular Imaging Technologies. He has intellectual property rights for Imagen PET and SPECT software.
Brett W. Sperry, A. Iain McGhie, Ibrahim M. Saeed, Kevin F. Kennedy, Paul Chan, Randall C. Thompson have no disclosures in regards to this project.
Abbreviations:
- CAD
coronary artery disease
- MPI
myocardial perfusion imaging
- PET
positron emission tomography
- SPECT
single photon emission tomography
- SSS
summed stress score
- SDS
summed difference score
- SRS
summed rest score
- MFR
myocardial flow reserve
- LVEF
left ventricular ejection fraction
References:
- 1).Moya A, Sutton R, Ammirati F, Blanc JJ, Brignole M, Dahm JB, et al. Task Force for the Diagnosis and Management of Syncope; EuropeanSociety of Cardiology (ESC); European Heart Rhythm Association (EHRA); Heart Failure Association (HFA); Heart Rhythm Society (HRS), Guidelines for the diagnosis and management of syncope (version 2009): The Task for the Diagnosis and Management of Syncope of the European Society of Cardiology. Eur Heart J. 2009;30(21):2631–2671. doi: 10.1093/eurheartj/ehp298 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2).Nawar EW, Niska RW, Xu J. National Hospital Ambulatory Medical Care Survey: 2005 emergency department summary. Adv Data. 2007;(386):1–32. [PubMed] [Google Scholar]
- 3).Kapoor WN. Current evaluation and management of syncope. Circulation. 2002;106 (13):1606–1609. doi: 10.1161/01.CIR.0000031168.96232.BA [DOI] [PubMed] [Google Scholar]
- 4).Soteriades ES, Evans JC, Larson MG, Chen MH, Chen L, Benjamin EJ, et al. Incidence and prognosis of syncope. N Engl J Med. 2002; 347(12):878–885. doi: 10.1056/NEJMoa012407 [DOI] [PubMed] [Google Scholar]
- 5).Mitrani RD, Hendel RC. The appropriateness of an ischemia evaluation for syncope. Circ Cardiovasc Imaging. 2013;6(3):358–359. doi: 10.1161/CIRCIMAGING.113.000301 [DOI] [PubMed] [Google Scholar]
- 6).Sun BC, Emond JA, Comargo CA Jr. Direct medical costs of syncope-related hospitalizations in the United States. Am J Cardiol. 2005;95(1):668–671. doi: 10.1016/j.amjcard.2004.11.013 [DOI] [PubMed] [Google Scholar]
- 7).Strickberger SA, Benson DW, Biaggioni I, Callans DJ, Cohen MI, Ellenbogen KA, et al. American Heart Association Councils on Clinical Cardiology, Cardiovascular Nursing, Cardiovascular Disease in the Young, and Stroke; Quality of Care and Outcomes Research Interdisciplinary Working Group; American College of Cardiology Foundation; Heart Rhythm Society. AHA/ACCF scientific statement on the evaluation of syncope: from the American Heart Association Councils on Clinical Cardiology, Cardiovascular Nursing, Cardiovascular Disease in the Young, and Stroke, and the Quality of Care and Outcomes Research Interdisciplinary Working Group; and the American College of Cardiology Foundation In Collaboration With the Heart Rhythm Society. J Am Coll Cardiol. 2006;47(2):473–484. doi: 10.1016/j.jacc.2005.12.019 [DOI] [PubMed] [Google Scholar]
- 8).Ward RP, Al-Mallah MH, Grossman GB, Hansen CL, Hendel RC, Kerwin TC, et al. , American Society of Nuclear Cardiology. American Society of Nuclear Cardiology Review of the ACCF/ASNC Appropriateness Criteria for Single-Photon Emission Computed Tomography Myocardial Perfusion Imaging (SPECT MPI). J Nucl Cardiol. 2007;14(6): e26–38. doi: 10.1016/j.nuclcard.2007.10.001 [DOI] [PubMed] [Google Scholar]
- 9).Hendel RC, Berman DS, Di Carli MF, Heidenreich PA, Henkin RE, Pellikka PA, Pohost GM, et al. , American College of Cardiology Foundation Appropriate Use Criteria Task Force, American Society of Nuclear Cardiology, American College of Radiology, American Heart Association, American Society of Echocardiology, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, Society of Nuclear Medicine. ACCF/ASNC/ACR/AHA/ASE/SCCT/SCMR/SNM 2009 Appropriate Use Criteria for Cardiac Radionuclide Imaging. J Am Coll Cardiol. 2009;53(23):2201–2229. doi: 10.1016/j.jacc.2009.02.013. [DOI] [PubMed] [Google Scholar]
- 10).Wolk MJ, Bailey SR, Doherty JU, Douglas PS, Hendel RC, Kramer CM, et al. , American College of Cardiology Foundation Appropriate Use Criteria Task Force. ACCF/AHA/ASE/ASNC/HFSA/HRS/SCAI/SCCT/SCMR/STS 2013 Multimodality Appropriate Use Criteria for the Detection and Risk Assessment of Stable Ischemic Heart Disease. J Am Coll Cardiol. 2014;63(4):380–406. doi: 10.1016/j.jacc.2013.11.009 [DOI] [PubMed] [Google Scholar]
- 11).AlJaroudi WA, Alraies MC, Wazni O, Cerqueira MD, Jaber WA. Yield and diagnostic value of stress myocardial perfusion imaging in patients without known coronary artery disease presenting with syncope. Circ Cardiovasc Imaging. 2013;6(3):384–91. doi: 10.1161/CIRCIMAGING.112.000051 [DOI] [PubMed] [Google Scholar]
- 12).Shen WK, Sheldon RS, Benditt DG, Cohen MI, Forman DE, Goldberger ZD, et al. 2017 ACC/AHA/HRS guideline for the evaluation and management of patients with syncope: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. Circulation. 2017;136(5):e25–e59. doi: 10.1161/CIR.0000000000000498 [DOI] [PubMed] [Google Scholar]
- 13).Brignole M, Moya A, de Lange FJ, Deharo JC, Elliott PM, Fanciulli A, et al. 2018 ESC Guidelines for the Diagnosis and Management of Syncope. Eur Heart J. 2018;39(21):1883–1948. doi:10/1093/eurheartj/ehy037. [DOI] [PubMed] [Google Scholar]
- 14).Diamond GA, Forrester JS. Analysis of Probability as an Aid in the Clinical Diagnosis of Coronary Artery Disease. N Engl J Med. 1979;300(24):1350–1358. doi: 10.1056/nejm197906143002402 [DOI] [PubMed] [Google Scholar]
- 15).Genders TS, Steyerberg EW, Alkadhi H, Leschka S, Desbiolles L, Nieman K, et al. , CAD Consortorium. CAD Consortium. A Clinical prediction rule for the diagnosis of coronary artery disease: validation, updating, and extension. Eur Heart J. 2011;32(11):1316–30. doi: 10.1093/eurheartj/ehr014 [DOI] [PubMed] [Google Scholar]
- 16).Henzlova MJ, Duvall WL, Einstein AJ, Travin MI, Verberne HJ. ASNC Imaging Guidelines for SPECT Nuclear Cardiology Procedures: Stress, Protocols, and Tracers. J Nucl Cardiol. 2016;23(3):606–639. doi: 10.1007/s12350-015-0387-x [DOI] [PubMed] [Google Scholar]
- 17).Dilsizian V, Bacharach SL, Beanlands RS, Bergmann SR, Gropler RJ, Knuuti J, et al. ASNC Imaging Guidelines/SNMMI Procedure Standard for Positron Emission Tomography (PET) Nuclear Cardiology Procedures. J Nucl Cardiol. 2016;23(5):1187–1226. doi: 10.1007/s12350-016-0522-3 [DOI] [PubMed] [Google Scholar]
- 18).Depuey EG, Mahmarian JJ, Miller TD, Einstein AJ, Hansen CL, Holly TA, et al. Patient-Centered Imaging. J Nucl Cardiol. 2012;19(2):185–215. doi: 10.1007/s12350-012-9523-z. [DOI] [PubMed] [Google Scholar]
- 19).Yoshida K, Mullani N, Gould KL> Coronary flow and flow reserve by PET simplified for clinical applications using rubidium-82 or nitrogen-13-ammonia. J Nucl Med 1996;37:1701–1712. [PubMed] [Google Scholar]
- 20).Nesterov SV, Deshayes E, Sciagra R, Settimo L, Declerck JM, Pan X-B, et al. Quantification of myocardial blood flow in absolute terms using (82)Rb PET imaging: the RUBY-10 study. JACC Cardiovasc Imaging. 2014;7:1119–1127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21).Dorbala S, Hachamovitch R, Curillova Z, Thomas D, Vangala D, Kwong RY, et al. Incremental prognostic value of gated Rb-82 positron emission tomography myocardial perfusion imaging over clinical variables and rest LVEF. JACC Cardiovasc Imaging. 2009;2(7):846–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22).Sharir T, Germano G, Kavanagh PB, Lai S, Cohen I, Lewin HC, et al. Incremental prognostic value of post-stress left ventricular ejection fraction and volume by gated myocardial perfusion single photon emission computed tomography. Circulation. 1999;100(10):1035–42. [DOI] [PubMed] [Google Scholar]
- 23).Patel KK, Spertus JA, Chan PS, Sperry BW, Al Badarin F, Kennedy KF, et al. Myocardial blood flow reserve assessed by positron emission tomography myocardial perfusion imaging identifies patients with a survival benefit from early revascularization. Eur Heart J. 2019; ehz389, 10.1093/eurheartj/ehz389 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1: Yield of single photon emission computed tomography myocardial perfusion imaging in patients with syncope, stratified by pre-test risk probability for coronary artery disease