J A C C : C A R D I O V A S C U L A R I M A G I N G V O L . 7 , N O . 3 , 2 0 1 4

ª 2 0 1 4 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 1 9 3 6 - 8 7 8 X / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c m g . 2 0 1 4 . 0 1 . 0 0 6

E D I T O R I A L C OMM E N T

Leaning Heavily on PET Myocardial Perfusion for Prognosis*

Frank A. Flachskampf, MD,y Vasken Dilsizian, MDz

Uppsala, Sweden; and Baltimore, Maryland

The prevalence of obesity has been steadilyincreasing in industrialized nations, reaching nearly30% of the population (1). Visceral adiposity hasbeen associated with early and accelerated progres-sion of atherosclerotic disease (2,3). Beyond playingan important role in the pathogenesis of metabolicdisorders that ultimately affect the coronary vessels,obesity is associated with increased risk formyocardial infarction, heart failure, and decreasedsurvival, predominantly because of increased car-diovascular morbidity and mortality (2).

See page 278

Recent data suggest that plasma proteins origi-nating from the adipose tissue, such as endo-cannabinoids, leptin, and adiponectin, play acentral role in the regulation and control of coro-nary circulatory function in obesity (4). A clinicalstudy investigating the relationship between bodyweight and coronary circulation in subjects withouttraditional coronary risk factors showed the pro-gression of endothelium-dependent coronary vas-omotion impairment in overweight subjects tototal vasodilatory capacity dysfunction in obeseindividuals (3). These findings imply that obesityis an independent mediator of coronary arterydisease rather than an epiphenomenon to othertraditional risk factors commonly associated withobesity, such as diabetes mellitus, hypertension,dyslipidemia, and insulin resistance (5). Noninva-sive diagnostic imaging studies that can identifycoronary circulatory and metabolic abnormalities

*Editorials published in JACC: Cardiovascular Imaging reflect the views

of the authors and do not necessarily represent the views of JACC:Cardiovascular Imaging or the American College of Cardiology.

From the yUppsala Universitet, Akademiska Sjukhuset, Uppsala, Sweden;

and the zUniversity of Maryland School of Medicine, Baltimore, Maryland.

Both authors have reported that they have no relationships relevant to the

contents of this paper to disclose.

early in the progression of cardiovascular diseasemay help improve patient outcome in suchpatients.

Myocardial blood flow assessed at rest and duringstress with positron emission tomographic (PET)imaging provides a noninvasive surrogate of coro-nary circulatory function (6). PET myocardialperfusion imaging uses radionuclides with half-livesthat are considerably shorter than those used insingle-photon emission computed tomography(SPECT), such as rubidium-82, with a 75-s half-life, resulting in lower radiation exposure than withSPECT (7). Moreover, PET images have betterspatial and temporal resolution compared withSPECT and allow reliable and accurate soft tissueattenuation correction (8). The latter is particularlyrelevant for patients who are obese or have largebody habitus. When combined with tracer kineticmodeling, PET imaging permits the noninvasiveassessment of the coronary circulatory function inabsolute (ml/min/g) terms (9,10).

In a recent multicenter observational registry,consisting of more than 7,000 patients with knownor suspected coronary artery disease, the extent andseverity of ischemia and scar as assessed byrubidium-82 vasodilator PET myocardial perfusionimaging provided significant incremental risk esti-mates of cardiac death and all-cause mortalitycompared with traditional coronary risk factors (11).The risk-adjusted hazard ratio of cardiac deathincreased with each 10% increment in stress myocar-dial perfusion defect frommild to moderate to severe,resulting in a nearly 5-fold higher hazard of cardiacdeath among patients with severe PET perfusiondefects compared with those with normal PET results(11). Certainly, there are inherent limitations in suchnonrandomized, observational multicenter registrystudies that use site-specific protocols on a variety ofPET cameras from different manufacturers, applyingdifferent attenuation correction methods and inter-pretation algorithms. Demographic information on

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 7 , N O . 3 , 2 0 1 4 Flachskampf and Dilsizian

M A R C H 2 0 1 4 : 2 8 8 – 9 1 Editorial Comment

289

patients was somewhat limited, because funda-mental information such as symptom status, leftventricular ejection fraction, and laboratory values,such as serum glucose, lipid profile, and renalfunction, were missing; hence, classic risk scoressuch as the Framingham score or the coronary riskscore of the European Society of Cardiology couldnot be calculated. However, the geographic diversityof the patients, protocols, and cameras may also beseen as a strength. Such registry data may moreclosely reflect real-world experience than a tightlymonitored randomized study.

In this issue of iJACC, Chow et al. (12) furtherextend the prognostic role of rubidium-82 vasodilatorPETmyocardial perfusion imaging by stratifying theaforementioned patient population by body massindex (BMI) and thus providing a look at the per-formance of PET imaging in a population known tobe difficult to assess by conventional imaging studies.The detrimental effects of soft tissue attenuation,which tends to degrade image quality and increaseinterpretive errors, have long been recognized withSPECT (13). Although more recent SPECT cam-eras include hardware and software solutions forattenuation correction, they remain nonideal and lessrobust than attenuation correction algorithmsapplied with PET imaging. Beyond differences insoft tissue attenuation correction, PET imagingprovides higher myocardial count density at a shorterimage acquisition time than SPECT and less scatterof activity from adjacent subdiaphragmatic visceralstructures into the myocardial region, allowing a finerreslicing of the left ventricular myocardium.

As for echocardiography, despite steady im-provements in transducer technology and increasinguse of left heart contrast, obese patients remain achallenge for (stress) echocardiographers: “Imagequality is subject to the body habitus of patients, andultrasound imaging in general may yet become acasualty of the obesity epidemic” (14). Obesity alsoreduces image quality and/or necessitates higherradiation doses when applying cardiac computedtomography (15,16). Cardiac magnetic resonance isthe only imaging technique largely immune to im-age deterioration due to obesity.

Remarkably, in the study by Chow et al. (12), only22% of patients had normal BMIs (<25 kg/m2),underlining the fact that obesity really is the “newnormal.”Another full 22% qualified as severely obese(BMI $35 kg/m2). The important findings of thestudy are as follows:

1. Normal PET results conferred a low risk forcardiac death (<0.5%/year) during the median

follow-up period of 2.2 years, regardless ofBMI.

2. Added to clinical variables, PET data (thesummed stress score) predicted cardiac oroverall mortality with similar power in all BMIcategories.

3. Within all BMI categories, there was a gradedrelation between the extent of stress perfusiondefects on PET imaging and annual cardiacdeath rate. This relation held up even inseverely obese patients (BMI $35 kg/m2),among whom those with the most severeperfusion defects ($20% of the myocardium)had an annual cardiac death rate of 4.8%compared with 0.1% in those with normalPET results.

4. Added to clinical variables, PET data assignedpatients a more precise risk for future death,measured by the “net reclassification improve-ment” (NRI), and did so equally well in all BMIcategories.

Reclassification of risk is a relatively novelconcept, which determines how many patients areshifted to more appropriate risk categories after theevaluation of new disease markers (17,18). In thepresent study, reclassification was quantified as“category-free NRI” (19,20). The category-freeNRI, different from the conventional “categorical”NRI, does not depend on the number and distri-bution of levels of cardiovascular risk into whichpatients are categorized. Note, however, that thecategory-free NRI may be several-fold higher inabsolute value than the conventional “categorical”NRI (21) and, therefore, cannot be directlycompared with categorical NRIs for diagnosticmarkers or procedures reported in the publishedresearch. For example, in the first report from thePET registry (11), PET imaging had a categoricalNRI of 11.6% but a category-free NRI of 54% forthe prediction of cardiac death. In the present study(12), a category-free NRI of 46% (95% confidenceinterval: 31% to 61%) was calculated for the entirestudy group for which data on cardiac death wereavailable, and similar NRIs were calculated acrossthe different BMI groups.

An interesting aspect of this study is the reap-pearance of the debated “obesity paradox” (22,23).In the cohort of subjects studied, overall and cardiacmortality were lower in patients with higher BMIs.This of course does not prove that obesity protectsagainst death. As the investigators point out,interaction with age (inversely associated with BMI)as well as with dyspnea symptoms and referral

Flachskampf and Dilsizian J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 7 , N O . 3 , 2 0 1 4

Editorial Comment M A R C H 2 0 1 4 : 2 8 8 – 9 1

290

likelihood (both positively associated with BMI)may be the reason for this finding, similar to thebetter known “smokers’ paradox.”The investigators conclude that myocardial

perfusion PET imaging is an attractive option forrisk stratification in all patients with suspected orknown coronary artery disease, including those withincreased BMIs. Is PET imaging therefore the nextwave of prognostic testing in the growing popula-tion of obese patients? Although there is littledoubt that myocardial perfusion PET imaging addsprognostic power to classic risk factors, there are nolarge head-to-head comparisons of the prognosticpower of PET imaging with other tests. In partic-ular, prognostic comparisons with current stressechocardiography or cardiac magnetic resonanceimaging are lacking. Furthermore, statistical mea-sures of predictive power, such as the ones used inthis study (Harrell’s c or the category-free NRI) arenot easy to translate into clinically meaningful in-formation or to compare between studies. Finally,as we have learned from new risk markers for cor-onary artery disease, heart failure, and other

conditions, adding prognostic power is not enoughfor a test to be clinically meaningful (24). Rather,information garnered from a diagnostic test needsto guide patient management decisions. Ultimately,such marker-based patient management decisionsshould improve patient outcome. This is a formi-dable threshold to widespread introduction of newdiagnostic techniques, which is similarly faced byother candidate tests, such as ultrasound measure-ment of intima-media thickness of the carotid arteryor coronary calcium scoring by computed tomog-raphy. For PET imaging in particular, availabilityand cost further limit its large-scale adoption at thispoint. Nonetheless, the study of Chow et al. (12)shows convincingly that PET imaging is a validprognostic instrument to consider for risk assess-ment of the large and growing group of patients withobesity.

Reprint requests and correspondence: Dr. Frank A.Flachskampf, Uppsala Universitet, Akademiska Sjukhu-set, Ingång 40, Plan 5, 751 85 Uppsala, Sweden. E-mail:frank.flachskampf@medsci.uu.se.

R E F E R E N C E S

1. Flegal KM, Carroll MD, Ogden CL,et al. Prevalence and trends in obesityamong US adults 1999-2008. JAMA2010;303:235–41.

2. Apovian CM, Gokce N. Obesity andcardiovascular disease. Circulation2012;125:1178–82.

3. Schindler TH, Cardenas J, Prior JO,et al. Relationship between increasingbody weight, insulin resistance,inflammation, adipocytokine leptin,and coronary circulatory function.J Am Coll Cardiol 2006;47:1188–95.

4. Quercioli A, Pataky Z, Montecucco F,et al. Coronary vasomotor control inobesity and morbid obesity: contrast-ing flow responses with endocannabi-noids, leptin, and inflammation. J AmColl Cardiol Img 2012;5:805–15.

5. Valenta I, Dilsizian V, Schelbert HR,Schindler TH. The influence ofinsulin-resistance, obesity and diabetesmellitus on vascular tone and myocar-dial blood flow. Curr Cardiol Rep2012;14:217–25.

6. Valenta I, Dilsizian V, Querciolo A,Ambrosio G, Wahl R, Schindler TH.Impact of obesity and bariatric surgeryon metabolism and coronary circula-tory function. Curr Cardiol Rep 2014;16:433.

7. Laskey WK, Feinendegen LE,Neumann RD, Dilsizian V. Low-levelionizing radiation from non-invasive

cardiac imaging: can we extrapolateestimated risks from epidemiologicdata to the clinical setting? J Am CollCardiol Img 2010;3:517–24.

8. Dilsizian V, Bacharach SL,Beanlands SR, et al. ASNC imagingguidelines for nuclear cardiology pro-cedures: PET myocardial perfusionand metabolism clinical imaging.J Nucl Card 2009;16:651.

9. Schindler TH, Schelbert HR,Quercioli A, Dilsizian V. CardiacPET imaging for the detection andmonitoring of coronary artery diseaseand microvascular health. J Am CollCardiol Img 2010;3:623–40.

10. Dilsizian V, Taillefer R. Journey inevolution of nuclear cardiology: willthere be another quantum leap withthe F-18 labeled myocardial perfusiontracers? J Am Coll Cardiol Img 2012;5:1269–84.

11. Dorbala S, Di Carli MF,Beanlands RS, et al. Prognostic valueof stress myocardial perfusion positronemission tomography: results from amulticenter observational registry.J Am Coll Cardiol 2013;61:176–84.

12. Chow BJW, Dorbala S, Di Carli MF,et al. Prognostic value of PETmyocardial perfusion imaging in obesepatients. J Am Coll Cardiol Img2014;7:278–87.

13. Duvall WL, Croft LB, Corriel, et al.SPECT myocardial perfusion imagingin morbidly obese patients: imagequality, hemodynamic response topharmacologic stress, and diagnosticand prognostic value. J Nucl Cardiol2006;13:202–9.

14. Marwick TH. The future of echocar-diography. Eur J Echocardiogr 2009;10:594–601.

15. Dewey M, Vavere AL, Arbab-Zadeh A, et al. Patient characteristicsas predictors of image quality anddiagnostic accuracy of MDCTcompared with conventional coronaryangiography for detecting coronaryartery stenoses: CORE-64 Multi-center International Trial. AJR Am JRoentgenol 2010;194:93–102.

16. Leschka S, Stinn B, Schmid F, et al.Dual source CT coronary angiographyin severely obese patients: trading offtemporal resolution and image noise.Invest Radiol 2009;44:720–7.

17. Pencina MJ, D’Agostino RB Sr.,D’Agostino RB Jr., Vasan RS. Evalu-ating the added predictive ability of anew marker: from area under the ROCcurve to reclassification and beyond.Stat Med 2008;27:157–72.

18. Cook NR, Ridker PM. Advances inmeasuring the effect of individualpredictors of cardiovascular risk: the

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 7 , N O . 3 , 2 0 1 4 Flachskampf and Dilsizian

M A R C H 2 0 1 4 : 2 8 8 – 9 1 Editorial Comment

291

role of reclassification measures. AnnIntern Med 2009;150:795–802.

19. Pencina MJ, D’Agostino RB Sr.,Steyerberg EW. Extensions of netreclassification improvement calcula-tions to measure usefulness of newbiomarkers. Stat Med 2011;30:11–21.

20. Cook NR, Paynter NP. Comments on“Extensions of net reclassificationimprovement calculations to measureusefulness of new biomarkers” by M. J.Pencina, R. B. D’Agostino, Sr. andE. W. Steyerberg. Stat Med 2012;31:93–5.

21. Mühlenbruch K, Heraclides A,Steyerberg EW, Joost HG, Boeing H,Schulze MB. Assessing improvementin disease prediction using net reclas-sification improvement: impact of riskcut-offs and number of risk categories.Eur J Epidemiol 2013;28:25–33.

22. Lavie CJ, Milani RV, Ventura HO.Obesity and cardiovascular disease:risk factor, paradox, and impact ofweight loss. J Am Coll Cardiol 2009;53:1925–32.

23. Tobias DK, Pan A, Jackson CL, et al.Body-mass index and mortality among

adults with incident type 2 diabetes.N Engl J Med 2014;370:233–44.

24. Dilsizian V, Chandrashekhar Y,Narula J. Introduction of new tests:low are the mountains, high the ex-pectations. J Am Coll Cardiol Img2010;3:117–9.

Key Words: cardiac death -

obesity - positron emissiontomography - prognosis.