Author + information
- Received June 20, 2012
- Accepted June 26, 2012
- Published online October 1, 2012.
- Venkatesh L. Murthy, MD, PhD⁎,†,
- Masanao Naya, MD, PhD‡,
- Courtney R. Foster, BS§,
- Jon Hainer, BS‡,
- Mariya Gaber, ALM‡,
- Sharmila Dorbala, MD, MPH⁎,†,‡,
- David M. Charytan, MD, MSc§,
- Ron Blankstein, MD⁎,† and
- Marcelo F. Di Carli, MD⁎,†,‡,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Marcelo F. Di Carli, Brigham and Women's Hospital, ASB-L1 037C, 75 Francis Street, Boston, Massachusetts 02115
Objectives This study sought to evaluate whether impaired vasodilator function, an early manifestation of coronary artery disease, which precedes angiographic stenosis, accounts for increased risk among patients with moderate to severe renal dysfunction.
Background Patients with renal dysfunction are at increased risk of adverse cardiac outcomes, even in the absence of overt myocardial ischemia or infarction.
Methods We included 866 consecutive patients with moderate to severe renal dysfunction referred for rest and stress myocardial perfusion positron emission tomography and followed them for a median of 1.28 years (interquartile range: 0.64 to 2.34). Regional myocardial perfusion abnormalities were assessed by semiquantitative visual analysis of positron emission tomography images. Rest and stress myocardial blood flow were calculated using factor analysis and a 2-compartment kinetic model; they were also used to compute coronary flow reserve (stress/rest myocardial blood flow). The primary endpoint was cardiac death.
Results Overall, 3-year cardiac mortality was 16.2%. After adjusting for clinical risk, left ventricular ejection fraction, as well as the magnitude of scar and/or ischemia, coronary flow reserve below the median (<1.5) was associated with a 2.1-fold increase in the risk of cardiac death (95% confidence interval [CI]: 1.3 to 3.5, p = 0.004). Incorporation of coronary flow reserve into cardiac death risk assessment models resulted in an increase in the C-index from 0.75 to 0.77 (p = 0.05) and in a net reclassification improvement of 0.142 (95% CI: 0.076 to 0.219). Among patients at intermediate risk based on all data other than coronary flow reserve, the net reclassification improvement was 0.489 (95% CI: 0.192 to 0.836). Corresponding improvements in risk assessment for mortality from any cause were also demonstrated.
Conclusions The presence of coronary vascular dysfunction in patients with moderate to severe renal dysfunction, as assessed by positron emission tomography, is a powerful, independent predictor of cardiac mortality and provides meaningful incremental risk stratification over conventional markers of clinical risk.
Cardiovascular disease is the leading cause of mortality among patients with moderate to severe renal dysfunction (1). In selected high-risk patients, early referral to cardiac catheterization and coronary revascularization may improve outcomes (2). However, acute deterioration of renal function following diagnostic coronary angiography occurs in approximately 10% of patients (3) and up to 30% of patients after percutaneous coronary intervention (4). Contrast medium–induced renal dysfunction after coronary procedures carries poor prognosis (4,5), especially if dialysis becomes necessary (5). In addition, coronary revascularization procedures in patients with renal impairment are associated with markedly higher risks of both fatal and nonfatal adverse outcomes (6,7). Consequently, careful selection of high-risk patients for referral to coronary angiography and revascularization is of paramount importance.
Unfortunately, traditional approaches for cardiac risk assessment, including stress imaging, have been unable to accurately identify low-risk individuals in this patient subgroup (8). This may be related, in part, to the fact that noninvasive imaging methods are insensitive for detecting the presence of diffuse atherosclerosis and its impact on coronary epicardial and microcirculatory function and myocardial ischemia. The latter may help explain the increase in biomarkers of myocardial injury including serum N-terminal pro B-type natriuretic peptide and troponin T seen among patients with renal dysfunction and their effects on prognosis (9). This issue is of relevance because both microvascular dysfunction and myocardial ischemia may be amenable to treatment.
This study was designed to test the hypothesis that in patients with moderate to severe renal dysfunction, coronary vasodilator dysfunction, as measured by positron emission tomography (PET), is prevalent and helps explain the observed excess risk of cardiac mortality in this population.
All patients with moderate to severe renal dysfunction (estimated glomerular filtration rate [eGFR] ≤60 ml/min/1.73 m2) referred for rest/stress cardiac PET at the Brigham and Women's Hospital (Boston, Massachusetts) between January 1, 2006, and June 30, 2010, were included in this study, excluding those whose images were missing or uninterpretable due to poor image quality. In cases of repeat PET scans during the study period, only the earliest evaluable study was included. Demographic factors and key elements of the patients' history including risk factors and medication use were ascertained at the time of the study by patient interview and review of medical records. eGFR was calculated using the Modification of Diet in Renal Disease formula (10) based on a mean creatinine in the 90 days preceding the imaging study. Patients being treated with chronic renal replacement therapy, determined by patient interview and billing codes (11), were assumed to have eGFR of 1 ml/min/1.73 m2. The study was approved by the Partners Healthcare Institutional Review Board and conducted in accordance with institutional guidelines.
Patients were studied using a whole body PET-computed tomography scanner (Discovery RX or STE LightSpeed 64, GE Healthcare, Milwaukee, Wisconsin) after an overnight fast. Patients refrained from caffeine and methylxanthine containing substances and drugs for 24 h prior to their scans. Myocardial blood flow (MBF) was measured during rest and peak stress using rubidium-82 as a perfusion tracer, as described previously (12). Briefly, after transmission imaging and beginning with the intravenous bolus administration of rubidium-82 (1,480 to 2,200 MBq), list mode images were acquired for 7 min. Then, a standard intravenous infusion of dipyridamole, adenosine, regadenoson, or dobutamine was given. At peak stress, a second dose of rubidium-82 was injected and images were recorded in the same manner. The average radiation exposure per study was 4.6 mSv (13). Heart rate, blood pressure, and 12-lead electrocardiogram were recorded at baseline and every minute during and after pharmacological stress.
Rubidium-82 undergoes very rapid radioactive decay, with a physical half-life of 75 s, producing trace quantities of stable, nonradioactive krypton-82 gas, which is passively exhaled by the lungs. Neither renal nor hepatic excretion contributes meaningfully to rubidium-82 elimination. As a result, administered tracer doses and analytic methods do not require adjustment for renal function.
Semiquantitative Analysis of Myocardial Perfusion
Semiquantitative 17-segment visual interpretation of the gated myocardial perfusion images was performed by experienced observers using a standard 5-point scoring system (14). Summed rest and stress scores were calculated as the sum of individual segmental scores on the respective images, and their difference was recorded as summed difference score. These were converted to percentages of left ventricular myocardium by dividing by the maximum score (i.e., 68).
Left ventricular systolic function
Rest and stress left ventricular ejection fraction (LVEF) were calculated from gated myocardial perfusion images using commercially available software. LVEF reserve was considered present when LVEF increased from rest to stress.
Quantitative MBF and flow reserve
Absolute MBF (in ml/g/min) was computed from the dynamic rest and stress images using commercially available software (Corridor4DM, INVIA Medical Imaging Solutions, Ann Arbor, Michigan) and previously validated methods (15). Automated factor analysis was used to generate blood pool (arterial input function) and tissue time-activity curves. Regional and global rest and peak stress MBF were calculated by fitting the rubidium-82 time-activity curves to a 2-compartment tracer kinetic model as described previously (15). Per-patient global coronary flow reserve (CFR) was calculated as the ratio of absolute MBF at stress over rest for the entire left ventricle. Quantitation of MBF was performed by 4 operators. The intraclass correlation coefficient (16) for CFR among these 4 readers was 0.94 (95% confidence interval [CI]: 0.88 to 0.98), indicating excellent reproducibility.
Assessment of outcomes
The primary outcome was death from any cardiac cause. Patients who died from noncardiac causes were censored. Mortality from any cause was used as a secondary endpoint. Vital status of all patients was ascertained by integrating data from the Social Security Death Index, the National Death Index, and the Partners Healthcare Research Patient Data Registry. Cause of death was determined by blinded adjudication of hospital records and death certificates. Early revascularization (within 90 days) was ascertained from the Partners Healthcare Research Patient Data Registry and hospital records.
Statistical significance was assessed using Wilcoxon, Fisher exact, and chi-square tests for continuous, dichotomous, and categorical variables, respectively. Two-sided p values <0.05 were considered significant. All statistical analyses were performed using SAS (version 9.3, SAS Institute Inc., Cary, North Carolina).
The Cox proportional hazards model was used to assess the impact of CFR on cardiac mortality after controlling for the effects of critical covariates. A series of models were developed starting with the Duke Clinical Score, an index of coronary artery disease (CAD) likelihood and prognosis based on clinical covariates (17). Rest LVEF, combined extent and severity of scar and ischemia, stress-induced LVEF augmentation (LVEF reserve), and CFR (as a continuous variable or dichotomized at the median) were then sequentially incorporated into the model. To investigate the effects of absolute peak stress MBF, we generated an additional model containing absolute stress MBF instead of CFR. The models were examined for the validity of the proportional hazards assumption and additive value, taking care to avoid overfitting. Survival was plotted using direct adjusted survival probabilities (18) from the Cox survival model.
To assess for biases introduced by early revascularization, analyses were repeated censoring all patients who underwent early revascularization (19). In an exploratory analysis, we considered the effect of any revascularization, including those >90 days after the PET scan, as a time-dependent covariate.
Assessment of incremental value
Incremental prognostic value of CFR was assessed with the likelihood ratio test to determine the improvement in prediction power of each sequential Cox model. The C-index was calculated for each model (20) with comparisons using the method of Antolini et al. (21). The potential impact of CFR on risk stratification was assessed by net reclassification improvement (NRI) (22) at 2 years using threshold annual rates of cardiac mortality of 2% and 4%. These thresholds were selected to be slightly higher than American College of Cardiology/American Heart Association guidelines for management of chronic stable angina (23) because of the higher periprocedural morbidity and mortality of coronary angiography and revascularization in patients with moderate to severe renal impairment (3–7). Reclassification metrics using guideline-derived (23) 1% and 3% thresholds were also computed as a secondary analysis.
A total of 866 consecutive patients met inclusion and exclusion criteria during the study period and were followed for a median of 1.28 years (interquartile range: 0.64 to 2.34 years). Baseline characteristics are given in Table 1. The most common indications for testing were evaluation for chest pain, dyspnea, or their combination. Approximately one-half of all studies were normal by semiquantitative visual analysis.
Mortality from any cause occurred in 155 (17.9%) patients, of which 88 (56.8%) were due to cardiac causes (Table 2,Online Appendix). Three-year cardiac mortality was 16.2%. Compared with patients without cardiac death, those who experienced cardiac death were older, more likely to be men, have diabetes, had a lower body mass index, more likely to be referred for dyspnea evaluation, have prior CAD, lower rest LVEF, and larger abnormalities on PET scans (Table 1).
The annualized rate of cardiac death increased with increasing extent and severity of perfusion abnormalities (Fig. 1A) and, importantly, was 2.7% per year among patients with a visually normal PET scan. Furthermore, in each category of abnormality on PET scanning (combined ischemia and scar extent), an impaired CFR identified higher risk subgroups, including among those with visually normal scans (Fig. 1B). Likewise, in each category of LVEF, a higher CFR was associated with a decrease in the risk of cardiac mortality (Fig. 1C).
Univariate predictors of cardiac mortality
CFR values below the median were associated with a 3.3-fold increased risk of cardiac death. Other significant predictors of increased risk included age, male sex, diabetes, and prior CAD (Fig. 2,Online Appendix). Somewhat surprisingly, chest pain as a reason for testing was associated with a decreased risk, possibly reflecting confounding. In addition, dyspnea, a decrease in rest LVEF, as well as increasing burden of scar, ischemia, or their combination on semiquantitative visual analysis, were all significantly associated with increased risk.
Multivariable survival analysis and incremental prognostic value
A series of multivariable models were then constructed to assess the incremental value of CFR after adjustment for critical covariates known to be associated with increased risk of cardiac mortality (Table 2). Addition of CFR to a model including the Duke clinical risk score, early revascularization, eGFR, rest LVEF, LVEF reserve, and the total burden of ischemia and scar was associated with a significant increase in global chi-square and decrease in Akaike Information Criterion, indicating improved model fit, improved model calibration, and a borderline significant increase in the C-index from 0.75 to 0.77 (p = 0.05). Dichotomization of CFR at the median value led to better model fit than as a continuous variable. Compared with those with CFR ≥1.5, the fully adjusted hazard ratio for cardiac death was 2.1 (95% CI: 1.3 to 3.5, p = 0.0004) for those with CFR <1.5 (Fig. 2).
Addition of CFR estimates to the model resulted in the reclassification of 14%, 36%, and 12% of patients at low (<2% annualized cardiac mortality), intermediate (2% to 4% annualized cardiac mortality), and high cardiac risk (≥4% annualized cardiac mortality), respectively, based on the pre-CFR model (Table 2, Model 5; Fig. 3). The benefit of CFR on risk reclassification was greatest among patients with an intermediate pre-CFR risk, in whom addition of CFR to risk estimation downgraded risk in 21% (0% annualized cardiac mortality) and upgraded it in 15% (9.8% annualized cardiac mortality).
In the entire cohort, 19.5% of patients were reclassified (7.7% upward and 11.8% downward) into more accurate risk categories. NRI was 0.142 (95% CI: 0.076 to 0.219) across clinical risk categories of <2, 2 to 4, and >4% annual rate of cardiac death (Table 2, Online Appendix). The effect of reclassification was most pronounced among those patients classified as low (NRI = 0.914, 95% CI: 0.817 to 1.11) or intermediate risk without CFR (NRI = 0.489, 95% CI: 0.192 to 0.836). However, significant reclassification was also seen among patients classified as high risk without CFR (NRI = 0.145, 95% CI: 0.106 to 0.185). Using thresholds of <1, 1 to 3, and >3% annual rate of cardiac death, the NRI for all patients was 0.098 (95% CI: 0.055 to 0.148). The continuous NRI, which measures discriminatory potential, was 0.390 (95% CI: 0.131 to 0.639).
Analyses were repeated using mortality from any cause as a secondary outcome and the results were similar. After correction for clinical risk, left ventricular systolic function, extent of ischemia and scar, and stress-induced LVEF augmentation, CFR remained a significant predictor of mortality with CFR <1.5 associated with a hazard ratio of 1.9 (95% CI: 1.3 to 2.8, p = 0.0004). Addition of CFR was associated with favorable risk reclassification for all-cause mortality (continuous NRI = 0.461; 95% CI: 0.257 to 0.658). Using risk thresholds of 4% and 8% per year (double those for cardiac mortality), the NRI was 0.129 (95% CI: 0.044 to 0.222).
The principal finding of this study is that the severity of coronary vascular dysfunction, as assessed by PET, is an independent predictor of cardiac death in patients with moderate or severe renal impairment. We observed that the failure of MBF to increase adequately on demand identified patients with renal impairment who experienced a significantly higher rate of cardiac mortality (10.7% vs. 3.2% per year in those with relatively preserved coronary vasodilator reserve, p < 0.0001). Importantly, identification of coronary vasodilator dysfunction improved risk stratification beyond comprehensive clinical assessment, left ventricular systolic function, and semiquantitative measures of myocardial ischemia and scar. Indeed, a quantitative estimate of coronary vasodilator reserve in this cohort was able to improve risk stratification in more than one-third of patients with intermediate risk, appropriately downgrading risk in 15% and upgrading it in 21% of patients.
Prior studies have shown that coronary vasodilator function assessment improves prognostic assessment (24). This study demonstrates the benefits of improved risk stratification by quantitative measures of coronary vascular dysfunction also apply to patients with moderate to severe renal dysfunction, who are among the highest risk cohorts for CAD complications. Although future studies will be required to determine how CFR metrics should best be incorporated into treatment strategies, more aggressive medical treatment of patients with visually normal perfusion but impaired flow reserve could potentially improve outcomes. Similarly, it is possible that avoidance of angiography and revascularization in persons with myocardial scar and/or ischemia but with preserved CFR may decrease nephrotoxicity without compromising safety.
Noninvasive measures of coronary vasodilator reserve integrate the hemodynamic effects of focal epicardial coronary stenoses, the fluid dynamic effects of diffuse atherosclerosis, and the presence of coronary microvascular dysfunction. Thus, the close relationship between the blunting of the increase in MBF with stress and prognosis could be due to any or all of these. Patients with renal impairment may be more likely to have advanced multivessel epicardial coronary disease (25) and more rapid progression of disease (26), both of which may contribute to adverse prognosis (27). Prior investigations have also suggested that renal disease is associated with abnormal coronary vasodilator function (28), which may result from multiple mechanisms (29), including decreased capillary density (30) leading to microvascular dysfunction as well as vascular remodeling in epicardial arteries (31). Our demonstration of increased cardiac mortality in patients who failed to augment MBF in response to stress in the absence of overt evidence of myocardial ischemia (5.9% vs. 1.1% per year for those with relatively preserved flow reserve, p = 0.002) provides new evidence that microvascular abnormalities play a role in the increased cardiovascular risk of patients with renal impairment. This is supported by prior studies showing that angiographic measures of coronary disease severity could not fully explain the increased risk observed in patients with renal dysfunction (32). Consequently, it is likely that either diffuse atherosclerosis or microvascular dysfunction or both together account for at least part of the increased risk observed in patients with poor coronary vasodilator function.
The current study is a single-center, nonrandomized, observational study and carries all of the inherent limitations of that study design. As such, it is likely that some amount of residual confounding remains, despite careful adjustment for clinically relevant covariates. On the other hand, compared with data derived from patients selectively enrolled in a randomized trial, these data, with very limited exclusion criteria, may be more representative of patients seen in routine clinical practice. Although the Modification of Diet in Renal Disease formula for eGFR has been extensively validated, it represents an estimate of renal function at a single time point. However, for 99% of patients in this study, ≥2 creatinine values were averaged, reducing the impact of fluctuations.
At present, cardiac PET is only available at a relatively small number of institutions compared with other stress testing modalities such as single-photon emission computed tomography and echocardiography. However, with increased PET scanner availability and emerging longer half-life tracers for myocardial perfusion imaging available for unit dose distribution (33), accessibility of cardiac PET is likely to continue to improve (34).
Other methods of stress imaging are not able to routinely quantify myocardial perfusion. Advances in single-photon emission computed tomography imaging technology might enable this in the near future. Routine stress echocardiography can detect impaired myocardial perfusion once severe enough to result in overt systolic dysfunction, quantification of subclinical abnormalities in myocardial perfusion requires more advanced methods than are available at most sites. One approach, Doppler interrogation of coronary flow velocities, typically in the left anterior descending artery, has been demonstrated to identify patients at risk of future coronary events (35). This method requires excellent acoustic windows, which are often not available due to body habitus or lung disease. Furthermore, this technique can only evaluate a subset of the coronary tree and may thus underestimate disease burden. Myocardial contrast echocardiography may increase the proportion of evaluable myocardium, but lacks regulatory approval in the United States.
In summary, among patients with moderate to severe renal dysfunction, noninvasive assessment of coronary vasodilator function provides incremental risk stratification beyond routine measures of clinical risk, including estimates of left ventricular systolic function and the extent and severity of myocardial ischemia and scar, and results in a meaningful risk reclassification of 1 in 5 patients with known or suspected CAD. These findings have potentially important implications for optimal identification of high-risk individuals and selection of more aggressive management strategies, especially given the markedly higher rates of morbidity and mortality associated with cardiac catheterization and revascularization in patients with renal impairment (6,7).
For supplemental figures and tables, please see the online version of this paper.
The study was funded in part by grants from the United States National Institutes of Health (RC1 HL101060-01 and T32 HL094301-01A1). Dr. Murthy owns equity in General Electric. Dr. Dorbala has received research grants from Astellas and Bracco Diagnostic. Dr. Charytan was supported by a Carl S. Gottschalk award from the American Society of Nephrology. He has consulted for Medtronic, Affymax, and Tengion. Dr. Di Carli receives research grant support from Toshiba Medical Systems. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- coronary artery disease
- coronary flow reserve
- confidence interval
- left ventricular ejection fraction
- myocardial blood flow
- net reclassification improvement
- positron emission tomography
- Received June 20, 2012.
- Accepted June 26, 2012.
- American College of Cardiology Foundation
- U.S. Renal Data System
- Charytan D.M.,
- Wallentin L.,
- Lagerqvist B.,
- et al.
- Rihal C.S.,
- Textor S.C.,
- Grill D.E.,
- et al.
- Gruberg L.,
- Mintz G.S.,
- Mehran R.,
- et al.
- Cooper W.A.,
- O'Brien S.M.,
- Thourani V.H.,
- et al.
- Al-Mallah M.H.,
- Hachamovitch R.,
- Dorbala S.,
- Di Carli M.F.
- McMurray J.J.V.,
- Uno H.,
- Jarolim P.,
- et al.
- El Fakhri G.,
- Sitek A.,
- Guerin B.,
- Kijewski M.F.,
- Di Carli M.F.,
- Moore S.C.
- Senthamizhchelvan S.,
- Bravo P.E.,
- Lodge M.A.,
- Merrill J.,
- Bengel F.M.,
- Sgouros G.
- Cerqueira M.D.,
- Weissman N.J.,
- Dilsizian V.,
- et al.,
- American Heart Association Writing Group on Myocardial Segmentation and Registration for Cardiac Imaging
- El Fakhri G.,
- Kardan A.,
- Sitek A.,
- et al.
- Nieto F.J.,
- Coresh J.
- Hachamovitch R.,
- Di Carli M.F.
- Gibbons R.J.,
- Chatterjee K.,
- Daley J.,
- et al.
- Murthy V.L.,
- Naya M.,
- Foster C.R.,
- et al.
- Joki N.,
- Hase H.,
- Nakamura R.,
- Yamaguchi T.
- Gradaus F.,
- Ivens K.,
- Peters A.J.,
- et al.
- Amann K.,
- Ritz E.
- Amann K.,
- Breitbach M.,
- Ritz E.,
- Mall G.
- Schwarz U.,
- Buzello M.,
- Ritz E.,
- et al.
- Maddahi J.,
- Czernin J.,
- Lazewatsky J.,
- et al.
- Cortigiani L.,
- Rigo F.,
- Galderisi M.,
- et al.