Original article

Prognostic value of fatty acid imaging in patients

with angina pectoris without prior myocardial infarction:
comparison with stress thallium imaging
Takayuki Matsuki1, Nagara Tamaki2, Tomoaki Nakata3, Atsushi Doi1, Hiroshi Takahashi1, Michihiro Iwata1,
Takashi Sakamoto1, Kazuaki Yamauchi1, Masaru Shimazaki1, Koichi Morita2, Kazuaki Shimamoto3
1 Departmentof Cardiology, Shin-Nittetsu Muroran General Hospital, Muroran, Japan
2 Departmentof Nuclear Medicine, Hokkaido University Graduate School of Medicine, Sapporo, Japan
3 Second Department of Internal Medicine, School of Medicine, Sapporo Medical University, Sapporo, Japan

Received: 20 November 2003 / Accepted: 16 March 2004 / Published online: 24 July 2004
Springer-Verlag 2004

Abstract. A fatty acid analogue, 123I-labelled -methyl
iodophenyl pentadecanoic acid (BMIPP), has been used
to identify ischaemic and metabolically impaired myo-
cardium. However, the prognostic value of BMIPP imag-
ing, particularly in relation to stress myocardial perfu-
sion imaging, remains unclear. Data from 167 consecu-
tive patients with angina pectoris but without prior myo-
cardial infarction (MI) who had undergone both BMIPP
and stress 201Tl (sTL) imaging were analysed. Tracer up-
take was graded using a 13-segment, 4-point scoring
model. Patients were followed up for 48 months with
primary end points (cardiac death, non-fatal MI) as hard
cardiac events and with secondary end points (late revas-
cularisation, recurrent angina and heart failure) as soft
events. For overall cardiac events (5 hard and 29 soft
events), Kaplan-Meier analysis revealed significantly
lower event rates in subgroups with normal BMIPP up-
take, a summed difference score of sTL (SDS) of <3 or
absence of diabetes mellitus when compared to each
counterpart. Multivariate Coxs analysis revealed re-
duced BMIPP uptake, SDS 3, diabetes and reduced
ejection fraction to be significant predictors. Negative
predictive values of normal BMIPP and SDS <3 for all
events were 91% and 84%, respectively. No hard event
occurred in 66 patients with normal BMIPP uptake,
whereas two patients with SDS <3 but impaired BMIPP
uptake had hard events. In conclusion, normal BMIPP
imaging is an excellent prognostic sign, independently of
stress myocardial perfusion imaging, in patients with an-
gina pectoris without prior MI.
Nagara Tamaki ()
Department of Nuclear Medicine,
Hokkaido University Graduate School of Medicine,
Kita 15 Nishi 7, Kita-Ku, Sapporo, 060-8638, Japan
e-mail: natamaki@med.hokudai.ac.jp
Keywords: Angina Prognosis Fatty acid Imaging
Eur J Nucl Med Mol Imaging (2004) 31:15851591
DOI 10.1007/s00259-004-1551-8
Introduction
Stress myocardial perfusion imaging is a useful tech-
nique for determining the prognosis of patients with
stable coronary heart disease or post-infarct patients
[14] because it permits precise evaluation of the sever-
ity and extent of inducible myocardial ischaemia, thus
enabling identification of patients at high risk and those
at low risk [1]. However, this procedure is not sufficient
to predict acute ischaemic events. Our previous studies
[5, 6] revealed that decreased uptake of 123I-labelled
-methyl iodophenyl pentadecanoic acid (BMIPP) rela-
tive to 201Tl uptake (i.e. perfusionBMIPP mismatch
pattern) could be a valuable predictor of future cardiac
events in comparison with the resting perfusion defect
alone in patients who have survived acute myocardial
infarction. Although impaired myocardial uptake of
fatty acids in the resting condition has been observed in
myocardium with reversible ischaemia but no previous
myocardial infarction [7], it is still unclear whether im-
paired myocardial fatty acid metabolism is closely asso-
ciated with long-term prognosis, as has been observed
in patients who have survived acute myocardial infarc-
tion [5].
Therefore, the present study was designed to deter-
mine whether impaired uptake of long-chain fatty acids
has prognostic value in patients with angina pectoris
without prior myocardial infarction and to compare the
prognostic value of BMIPP imaging with that of stress
201Tl scintigraphy.

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
1586
Materials and methods
Patient population
Based on the following entry and exclusion criteria, we prospec-
tively registered 196 consecutive patients with known or suspected
coronary artery disease who had visited as an outpatient or who
had been admitted to Shin-Nittetsu Muroran General Hospital be-
tween January 1995 and December 2000. Criteria for admission
were recurrent pain, ECG changes or cardiac marker abnormalities
suggesting acute coronary syndrome. Clinical data concerning car-
diac history and physical examination were obtained. There were
no Q waves in 12-lead ECG in any of the patients, and rest echo-
cardiography was performed in all patients to exclude prior myo-
cardial infarction. The entry criteria were: (1) known or suspected
coronary artery disease without a history of previous myocardial
infarction, (2) performance of myocardial fatty acid metabolism
imaging using iodine-123-labelled BMIPP and stress 201Tl scintig-
raphy within 6 weeks of each other, (3) follow-up for at least
1 year or until cardiac death or non-cardiac death and (4) provision
of informed consent. The exclusion criteria were: (1) follow-up pe-
riod of less than 1 year, (2) presence of critical conditions such as
cardiogenic shock or uncontrollable heart failure that would hinder
stress testing, (3) presence of organic cardiac disorders other than
coronary artery disease, such as congenital, valvular or myocardial
heart disease (cardiomyopathy) and (4) complete left bundle branch
block or a history of permanent pacemaker implantation.
The 196 patients included 103 men and 93 women, and their
mean age was 65.310.5 years (range 3984 years). The mean left
ventricular ejection fraction (LVEF) was 68.5%8.2%. There were
90 patients with effort-induced angina pectoris, 78 with vasospas-
tic angina and 28 with unstable angina. The diagnosis of effort-in-
duced and vasospastic angina was based on the history of chest
symptoms, the results of stress or rest electrocardiography and/or
the results of coronary angiography. The diagnosis of vasospastic
angina was made by the detection of ST segment elevation with
pain using a computerised 24-h ECG or 12-lead ECG recording in
patients with normal coronary arteries, and, if needed, provocative
tests for coronary spasm were performed. The diagnosis of unsta-
ble angina was based on the criteria of Braunwald [8]. Patients
with unstable angina were included only if there was an intention
to treat medically based on a low-risk stress test.
Diabetes was defined as chronic use of insulin or oral hypogly-
caemic agents. Hypertension was defined as systemic blood pres-
sure of 140/90 mmHg on repeated measurements or the chronic
use of antihypertensive drugs. Hyperlipidaemia was defined as a
total cholesterol level of >220 mg/dl or chronic use of a cholester-
ol-lowering agent. The study protocol was approved by the Ethical
Committee of Shin-Nittetsu Muroran General Hospital and the
experiments complied with the current laws of Japan.
Patient follow-up protocol
Patients were followed up with hard cardiac events (cardiac death,
non-fatal myocardial infarction) as primary end points and with soft
events (late revascularisation, re-admission due to recurrent angina
and heart failure) as secondary end points. Elective coronary revas-
cularisation performed within 3 months after radionuclide studies or
due to restenosis within 6 months after the first coronary angioplas-
ty was, however, excluded from cardiac events analysis in order to
reduce selection bias and the influence of scintigraphic findings on
the event rate of coronary revascularisation. Survival status was de-
European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
termined by telephone contact with patients, their relatives or the re-
ferring physician and/or from patient medical records. The cause of
death was categorised as cardiac death or non-cardiac death. Cardi-
ac death was defined as death attributable to congestive heart fail-
ure, myocardial infarction or cardiac arrest, or sudden death.
Myocardial fatty acid imaging
In the fasting and resting condition, 111 MBq of BMIPP (provided
by Nihon Medi-Physics Co., Nishinomiya, Japan) was intra-
venously injected, and then tomographic data acquisition was
started 30 min after injection using a digital gamma camera and a
low-energy all-purpose collimator (Shimadzu 510R, Shimadzu
Co., Kyoto, Japan). Data were collected from 32 directions with a
6 step between a 45 left posterior oblique view and a 45 right
anterior oblique view for 20 s per direction. Data were stored in a
6464 matrix on an on-line nuclear medicine data processor (Scin-
tipack 2000, Shimadzu Co., Kyoto. Japan). Tomographic images
along the vertical long, horizontal long and short axes were recon-
structed by a back-projection method with a Shepp and Logans
filter. Scatter or attenuation correction was not performed.
Stress myocardial perfusion imaging
One hundred and sixty-seven patients underwent either exercise
(n=143) or pharmacological stress (n=24) tomographic imaging
with 201Tl on a different day to the day on which BMIPP imaging
was performed. After overnight fasting, patients underwent a tread-
mill stress test using a standardised multistage exercise protocol.
At peak exercise, 74111 MBq 201Tl was intravenously injected
and patients continued to exercise for an additional 60 s. When a
diagnostic exercise stress test could not be performed, pharmaco-
logical stress imaging was performed using an intravenous injec-
tion of dipyridamole (0.14 mg/kg per minute for 4 min) [9] in 17
patients or ATP (0.15 mg/kg per minute for 5 min) [10] in seven
patients. Stress and 4-h delayed images were obtained. The imag-
ing protocol, the system and reconstruction protocols (including fil-
ters) were the same as those used for myocardial BMIPP imaging.
Image interpretation
Regional uptake of BMIPP and the perfusion tracer was graded
visually using a 13-segment, 4-point scoring system (0, normal
uptake; 1, mildly reduced; 2, severely reduced; 3, absence) in a
blinded manner by three nuclear cardiologists without knowledge
of clinical data. Differences in opinion were resolved by consen-
sus. Segments with a score of 1 or more on rest BMIPP images
were considered abnormal. The summed stress score (SSS) and
summed rest score (SRS) were calculated by adding the 13-seg-
ment scores in the stress and resting states, respectively. The
stress-rest summed difference score (SDS), which represents the
amount of stress-induced ischaemia, was also calculated.
Echocardiography
Two-dimensional echocardiography was performed using 3.5-MHz
and 2.5-MHz transducers and a commercially available scanner
(Toshiba SSH-160A system, Tokyo, Japan) for calculation of
LVEF. Echocardiographic studies were performed by specialists in
echocardiography within a month after radionuclide imaging.
 1587
Table 1. Clinical characteristics and backgrounds of the 167 pa- Table 2. Comparison of clinical parameters in groups with and
tients whose data were used for further analysis and of the 29 pa- without cardiac events
tients whose data were not used for further analysis
Parameter Cardiac event No cardiac P value
Variable 167 patients 29 patients P value group event group
analysed (%) excluded (%) (n=34) (n=133)

Age 66.610.2 63.511.7 0.162 Age 65.210.3 65.710.3 0.802

Gender (male) 90 (53.9%) 13 (44.8%) 0.483 Gender (male) 23/34 (67.6%) 68/133 (51.1%) 0.125
Effort-induced angina 78 (46.7%) 12 (41.1%) 0.742 UAP 6/34 (17.6%) 13/133 (10.0%) 0.323
Vasospastic angina 70 (41.9%) 8 (27.6%) 0.211
Coronary risk factors
Unstable angina pectoris 19 (11.4%) 9 (31.0%) 0.012
Smoking 12/34 (35.3%) 40/133 (30.1%) 0.705
Coronary risk factors Diabetes mellitus 13/34 (38.2%) 11/133 (8.3%) 0.001
Smoking 52 (31.1%) 9 (31.0%) 1.000 Hypertension 20/34 (58.8%) 60/133 (45.1%) 0.217
Diabetes mellitus 24 (14.4%) 2 (6.9%) 0.424 Hyperlipidaemia 10/34 (29.4%) 39/133 (29.3%) 0.841
Hypertension 80 (47.9%) 17 (58.6%) 0.249
Medications
Hyperlipidaemia 49 (29.3%) 8 (27.6%) 1.000
ACE inhibitors 5/34 (14.7%) 18/133 (13.5%) 0.919
Medications Beta blockers 6/34 (17.6%) 19/133 (14.3%) 0.826
ACE inhibitors 23 (13.8%) 2 (6.9%) 0.470 Ca antagonists 18/34 (52.9%) 65/133 (48.9%) 0.817
Beta blockers 25 (15.0%) 6 (20.7%) 0.615 LVEF 65.2%10.4% 69.4%6.9% 0.033
Ca channel antagonists 83 (49.7%) 16 (55.2%) 0.732
Scintigraphic parameters
LVEF 68.5%7.9% 68.8%10.0% 0.433
Positive BMIPP 28/34 (82.4%) 73/133 (54.9%) 0.006
Scintigraphic parameters SDS >3 13/34 (38.0%) 21/133 (15.8%) 0.040
BMIPP-SRS 2.23.3 2.22.7 0.495 BMIPP SRS 4.04.7 1.82.7 0.011
sTL-SSS 2.63.1 3.34.1 0.195 sTL SSS 4.15.1 2.22.2 0.045
sTL-SRS 1.72.3 1.92.8 0.318 sTL SRS 2.33.3 1.61.9 0.246
sTL-SDS 0.92.3 1.52.7 0.090 sTL SDS 1.83.3 0.61.9 0.059

Values are shown as mean SD Values are shown as mean SD

ACE, Angiotensin converting enzyme; Ca, Calcium; LVEF, left ven- ACE, Angiotensin converting enzyme; Ca, Calcium; SSS, summed
tricular ejection fraction; SSS, summed stress score; SRS, summed stress score; SRS, summed rest score; SDS, summed difference
rest score; SDS, summed difference score; sTL, stress 201Tl imaging score; sTL, stress 201Tl imaging; UAP, unstable angina

Statistical analysis imaging or within 6 months after a previous interventional

Data were analysed using an SAS statistical program package
(SAS Institute, Inc., NC, USA, 1994). Data were presented as
mean values SD, and follow-up intervals were expressed as me-
dians and ranges. Differences in mean values between two groups
and in the prevalence of variables were compared using an un-
paired t test and a 22 2 test, respectively. Cardiac event-free
curves were created by the Kaplan-Meier method and compared
using the log-rank test. Following univariate analysis, multivariate
analysis using Coxs proportional hazard model was performed.
A P value <0.05 was considered statistically significant.
Results
One hundred and ninety-six patients were followed up,
and the mean follow-up period was 4819 months (range
185 months). During follow-up, seven patients were lost.
Therefore, the follow-up rate was 96.4% (189/196). Data
from two patients for whom the cause of death could not
be determined and from three patients who had non-cardi-
ac death due to malignancy were excluded from further
analysis. Coronary revascularisations that had been per-
formed in 17 patients within 3 months after radionuclide
treatment due to restenosis were excluded from cardiac
events for prognostic analysis. Therefore, prognostic anal-
ysis was performed using data from 167 patients. Table 1
shows the clinical characteristics and backgrounds of the
167 patients whose data were used for prognostic analysis
and of the 29 patients whose data were not used for prog-
nostic analysis. There was no significant difference be-
tween any parameters in these two groups of patients
except for the prevalence of unstable angina. The mean
age of the 167 patients (90 men and 77 women) was
66.610.2 years (range 3984 years), and the mean LVEF
was 68.5%7.9%. The 167 patients included 78 with
effort-induced angina pectoris, 70 with vasospastic angina
and 19 with unstable angina (Table 1).
During follow-up, 34 cardiac events were document-
ed in 34 patients (five hard events and 29 soft events)
and used for further prognostic analysis. There was one
sudden cardiac death and one cardiac death due to con-
gestive heart failure or shock. Two patients were admit-
ted due to congestive heart failure, three due to non-fatal
acute myocardial infarction, ten because of the need for
revascularisation at a late stage, and 17 because of recur-
rent angina pectoris. Table 2 shows clinical parameters

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
1588
Fig. 1. Event-free curve based
on Kaplan-Meier analysis indi-
cating that the subgroups with
negative BMIPP imaging (left
upper), summed difference
score (SDS) of stress 201Tl
scintigraphy of <3 (right
upper) and absence of diabetes
mellitus (DM) (left lower)
showed significantly higher
cardiac event-free rates

Table 3. Significant predictors of overall cardiac events in univari- significantly lower than that of the non-cardiac event
ate analysis group (P=0.033). SRS of BMIPP (p=0.011) and SSS of
sTL (p=0.045) in the cardiac event group were signifi-
Variable P value cantly higher than those in the non-cardiac event group.
Gender (male) 0.081
Kaplan-Meier analysis revealed that the subgroups
Unstable angina 0.177 without diabetes mellitus (P=0.000), SDS <3 (P=0.006)
or negative BMIPP imaging (P=0.002) had significantly
Coronary risk factors higher event-free rates (Table 3, Fig. 1). When adjusted
Smoking 0.483 for diabetes, lower LVEF, reduced BMIPP uptake and
Diabetes mellitus 0.000 SDS 3 using multivariate analysis with Coxs propor-
Hypertension 0.115 tional hazard model, diabetes (adjusted hazard ratio,
Hyperlipidaemia 0.956
2.95; 95% CI, 1.356.47; P=0.006), lower LVEF (adjust-
Medications (during follow-up) ed hazard ratio, 25.56; 95% CI, 5.00130.82; P=0.000),
ACE inhibitors 0.807 reduced BMIPP uptake (adjusted hazard ratio, 4.05; 95%
Beta blockers 0.481 CI, 1.4311.43; P=0.007) and SDS 3 (adjusted hazard
Ca antagonists 0.722 ratio, 2.21; 95% CI, 1.064.61; P=0.031) were identified
Scintigraphic parameters as significant predictors of overall cardiac events (Fig. 1,
Table 4).
Positive BMIPP 0.002
SDS >3 0.006
When five hard cardiac events alone were considered,
none of the patients with normal BMIPP uptake had hard
ACE, Angiotensin converting enzyme; Ca, calcium; SDS, summed cardiac events. Two patients with SDS <3 but impaired
difference score of stress 201Tl scintigraphy BMIPP uptake, however, had hard events (Fig. 2); the
negative predictive values for hard cardiac events were
100% (66/66) for BMIPP imaging and 99% (131/133)
in the group of patients with cardiac events and the for SDS. The negative predictive values of BMIPP imag-
group without cardiac events. The cardiac event group ing and SDS for overall cardiac events were 91% and
had significantly higher prevalences of diabetes mellitus 84%, respectively.
(P=0.001) and SDS >3 (P=0.040) than the non-cardiac In addition, 11 out of 13 events (in cases of late revas-
event group. The LVEF of the cardiac event group was cularisation and non-fatal myocardial infarction using

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
 1589

fatty acid was observed in 101 (60.5%) who had histo-

ries of chest symptoms but were stable during imaging
protocols, suggesting that stable, chronic or repeated
ischaemia could be responsible for impairment of cardi-
ac fatty acid metabolism. These findings also suggest
that altered myocardial fatty acid metabolism following
a transient ischaemic event persists to some extent, even
though a persistent perfusion defect or irreversible myo-
cardial injury is not induced.
This possibility is supported by experimental findings
[15] and our clinical findings in patients with acute coro-
Fig. 2. Hard events in 167 patients with angina pectoris during a nary syndrome [16, 17]. These studies have demonstrated
mean follow-up period of 4819 months. In patients with normal
delayed recovery of myocardial fatty acid uptake com-
BMIPP imaging, no hard events occurred during the follow-up
period. In contrast, hard cardiac events occurred in two patients
pared with the recovery of flow or perfusion tracer uptake
(1.5%) whose summed differential score (SDS) of stress 201Tl following acute coronary occlusion and reperfusion. It is,
scintigraphy was less than 3 however, not clear how long metabolic alterations in-
duced by reversible ischaemia last if myocardial isch-
aemia is completely abolished without irreversible myo-
Table 4. Multivariate Coxs proportional hazard regression for cardial injury. Ischaemia-related impairment of myocar-
predicting all cardiac events
dial BMIPP uptake has been observed more often (with
Item Hazard 95% P value
an incidence of 6179%) in patients with unstable angina
ratio Confidential [12, 18], coronary spasm [19] or acute chest pain [13].
interval Thus, it is thought that it takes several days to months to
recover metabolically, depending on the severity of myo-
DM 2.95 1.356.47 0.006 cardial ischaemia and the induced intracellular metabolic
LVEF 25.56 5.00130.82 0.000 injury.
Reduced BMIPP uptake 4.05 1.4311.43 0.007 In addition, patients in the present study who had re-
SDS >3 2.21 1.064.61 0.031 duced BMIPP uptake that was thought to have been
caused by a mechanism different from that addressed by
DM, Diabetes mellitus; LVEF, left ventricular ejection fraction;
ischaemic memory imaging, such as prior myocardial in-
SDS, summed difference score of stress 201Tl scintigraphy
farction with necrosis, primary myocardial disease and
valvular heart disease, were excluded.
coronary angiography to clarify coronary artery lesions)
were related to the area of reduced uptake of BMIPP.
Comparison of prognostic values
of stress 201Tl scintigraphy and BMIPP
Discussion
Stress-induced myocardial perfusion abnormality as-
This is the first study to demonstrate that myocardial sessed by radionuclide imaging has a strong prognostic
fatty acid imaging has a significant prognostic value, in- value for patients with coronary artery disease [14]. It
dependently of stress myocardial perfusion imaging, for is also well known that normal stress myocardial perfu-
symptomatic patients with known or suspected coronary sion imaging indicates a better prognosis [1]. Similarly,
artery disease but no history of prior myocardial infarc- the present results showed that SDS of stress 201Tl scin-
tion. tigraphy and fatty acid imaging could be important pre-
dictors of overall cardiac events in univariate and multi-
variate analyses using Coxs proportional hazard model.
Impaired myocardial uptake of long-chain fatty acids This study also revealed that BMIPP imaging has excel-
lent negative predictive values that are basically compa-
Persistent metabolic alterations are often seen following rable to that of SDS of stress 201Tl scintigraphy for hard
myocardial ischaemia after recovery of myocardial blood events (100% vs 99%) and better than that of SDS of
flow. Thus, prior ischaemic insult may be identified as stress 201Tl scintigraphy for overall events (91% vs
areas of altered metabolism despite normal myocardial 84%). These findings suggest that resting myocardial
perfusion (so-called ischaemic memory imaging) [11]. fatty acid imaging can provide intracellular information
Several studies have shown impaired BMIPP uptake on ischaemia-related myocardial injury that is distinct
in patients with stable coronary artery disease without from the information obtained from assessment of epi-
previous myocardial infarction [1214]. Among the 167 cardial coronary flow reserve by stress myocardial perfu-
patients in this study, impaired myocardial uptake of sion imaging.

European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
1590
Several studies have demonstrated that the degree of
impaired BMIPP uptake is correlated with the severity of
coronary artery stenosis [18, 20]. The greater the limita-
tion of coronary flow reserve and the frequency of isch-
aemic attack in the coronary artery with significant ste-
nosis, the greater is the possibility of decrease in fatty
acid utilisation. For this reason, impaired BMIPP uptake
might be correlated with poor prognosis. Moreover, cor-
onary spasm, which is sometimes the cause of sudden
death [21], is one of the major causes of chest pain.
BMIPP imaging also enables detection of coronary
spasm, which usually cannot be detected by stress myo-
cardial perfusion imaging, since altered fatty acid utilisa-
tion may be associated with repetitive ischaemic
episodes as mentioned above.
Therefore, cardiac fatty acid metabolic imaging may
improve the value of stress myocardial perfusion imag-
ing for prognostic assessment in patients with coronary
artery disease by overcoming the pitfalls of stress myo-
cardial perfusion imaging.
Study limitations
The major limitations of this study are the relatively
small size of the study population and the small number
of cardiac events during the follow-up. A multi-centre
study using a larger study population with a longer fol-
low-up period is required to establish the prognostic
value of myocardial fatty acid imaging in relation to
stress myocardial perfusion imaging for patients with
angina pectoris but no history of previous myocardial in-
farction. The fact that normal BMIPP findings indicated
better prognosis (excellent negative predictive value) in
this study seems important. It is, therefore, necessary to
determine the cost-effectiveness of this simple metabolic
imaging technique alone or in combination with stress
myocardial perfusion imaging for prognostic assessment.
Finally, further investigation is needed to determine
whether myocardial fatty acid imaging still has a signifi-
cant prognostic value when electrocardiogram gating is
applied to stress myocardial perfusion imaging, because
the gating approach improves the prognostic power of
stress myocardial perfusion imaging [22].
Conclusion
This study demonstrated that impaired myocardial fatty
acid uptake is closely related to future cardiac events in
patients with angina pectoris without prior myocardial
infarction, independently of stress myocardial perfusion
imaging.
Acknowledgements. We thank Dr. Takeshi Ashie, MD, Mr.
Kazumasa Tsukamoto and Mr. Masatsuyo Takano, RT, for their
cooperation and encouragement.
European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
References
1. Gibbons RJ, Hodge DO, Berman DS, et al. Long-term out-
come of patients with intermediate-risk exercise electrocardio-
grams who do not have myocardial perfusion defects on radio-
nuclide imaging. Circulation 1999;100:21405
2. Hachamovitch R, Berman DS, Kiat H, et al. Value of stress
myocardial perfusion single photon emission computed tomo-
graphy in patients with normal resting electrocardiograms: an
evaluation of incremental prognostic value and cost-effective-
ness. Circulation 2002;105:8239
3. Diaz LA, Brunken RC, Blackstone EH, et al. Independent con-
tribution of myocardial perfusion defects to exercise capacity
and heart rate recovery for prediction of all-cause mortality in
patients with known or suspected coronary heart disease. J Am
Coll Cardiol 2001;37:15584
4. Galassi AR, Azzarelli S, Tomaselli A, et al. Incremental prog-
nostic value of technetium-99m-tetrofosmin exercise myocar-
dial perfusion imaging for predicting outcomes in patients
with suspected or known coronary artery disease. Am J
Cardiol 2001;88:1016
5. Nakata T, Kobayashi T, Tamaki N, et al. Prognostic value of
impaired myocardial fatty acid uptake in patients with acute
myocardial infarction. Nucl Med Commun 2000;21:897
906
6. Tamaki N, Tadamura E, Kudoh T, et al. Prognostic value of
iodine-123 labelled BMIPP fatty acid analogue imaging in
patients with myocardial infarction. Eur J Nucl Med 1996;
23:2729
7. Nishimura S, Ohta Y. BMIPP in angina pectoris. Int J Card
Imaging 1999;15:359
8. Braunwald E. Unstable angina. A classification. Circulation
1989;80:4104
9. Gould KL, Westcott RJ, Albro PC, et al. Noninvasive assess-
ment of coronary stenoses by myocardial imaging during phar-
macologic coronary vasodilatation. II. Clinical methodology
and feasibility. Am J Cardiol 1978;41:27987
10. Watanabe K, Sekiya M, Ikeda S, et al. Comparison of adeno-
sine triphosphate and dipyridamole in diagnosis by thallium-
201 myocardial scintigraphy. J Nucl Med 1997;38:57781
11. Tamaki N, Kuge Y, Tsukamoto E. Clinical roles of perfusion
and metabolic imaging. J Cardiol 2001;37(Suppl 1):5764
12. Yamabe H, Fujiwara S, Rin K, et al. Resting 123I-BMIPP scin-
tigraphy for detection of organic coronary stenosis and thera-
peutic outcome in patients with chest pain. Ann Nucl Med
2000;14:18792
13. Kawai Y, Tsukamoto E, Nozaki Y, et al. Significance of
reduced uptake of iodinated fatty acid analogue for the evalua-
tion of patients with acute chest pain. J Am Coll Cardiol
2001;38:188894
14. Morimoto K, Tomoda H, Yoshitake M, et al. Prediction of
coronary artery lesions in unstable angina by iodine 123 beta-
methyl iodophenyl pentadecanoic acid (BMIPP), a fatty acid
analogue, single photon emission computed tomography at
rest. Angiology 1999;50:63948
15. Schwaiger M, Schelbert HR, Ellison D, et al. Sustained
regional abnormalities in cardiac metabolism after transient
ischemia in the chronic dog model. J Am Coll Cardiol
1985;6:33647
16. Hashimoto A, Nakata T, Tsuchihashi K, et al. Postischemic
functional recovery and BMIPP uptake after primary percuta-
neous transluminal coronary angioplasty in acute myocardial
infarction. Am J Cardiol 1996;77:2530

17. Nakata T, Hashimoto A, Kobayashi H, et al. Outcome signifi-
cance of thallium-201 and iodine-123-BMIPP perfusion-
metabolism mismatch in preinfarction angina. J Nucl Med
1998;39:14929
18. Tateno M, Tamaki N, Yukihiro M, et al. Assessment of fatty
acid uptake in ischemic heart disease without myocardial in-
farction. J Nucl Med 1996;37:19815
19. Nakajima K, Shimizu K, Taki J, et al. Utility of iodine-123-
BMIPP in the diagnosis and follow-up of vasospastic angina.
J Nucl Med 1995;36:193440
European Journal of Nuclear Medicine and Molecular Imaging Vol. 31, No. 12, December 2004
1591
20. Takeishi Y, Sukekawa H, Saito H, et al. Clinical significance of
decreased myocardial uptake of 123I-BMIPP in patients with sta-
ble effort angina pectoris. Nucl Med Commun 1995;16:10028
21. Nishizaki M, Arita M, Sakurada H, et al. Polymorphic ventric-
ular tachycardia in patients with vasospastic anginaclinical
and electrocardiographic characteristics and long-term out-
come. Jpn Circ J 2001;65:51925
22. Sharir T, Germano G, Kavanagh PB, et al. Incremental prog-
nostic value of post-stress left ventricular ejection fraction and
volume by gated myocardial perfusion single photon emission
computed tomography. Circulation 1999;100:103542
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.