splenic switch-off

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

Purpose:

To investigate the pharmacology and potential clinical utility of splenic switch-off to identify understress in adenosine perfusion cardiac magnetic resonance (MR) imaging.

Materials and Methods:

Splenic switch-off was assessed in perfusion cardiac MR examinations from 100 patients (mean age, 62 years [age range, 18–87 years]) by using three stress agents (adenosine, dobutamine, and regadenoson) in three different institutions, with appropriate ethical permissions. In addition, 100 negative adenosine images from the Clinical Evaluation of MR Imaging in Coronary Heart Disease (CE-MARC) trial (35 false and 65 true negative; mean age, 59 years [age range, 40–73 years]) were assessed to ascertain the clinical utility of the sign to detect likely pharmacologic understress. Differences in splenic perfusion were compared by using Wilcoxon signed rank or Wilcoxon rank sum tests, and true-negative and falsenegative findings in CE-MARC groups were compared by using the Fisher exact test.

Results:

The spleen was visible in 99% (198 of 200) of examinations and interobserver agreement in the visual grading of splenic switch-off was excellent (k = 0.92). Visually, splenic switch-off occurred in 90% of adenosine studies, but never in dobutamine or regadenoson studies. Semiquantitative assessments supported these observations: peak signal intensity was 78% less with adenosine than at rest (P , .001), but unchanged with regadenoson (4% reduction; P = .08). Calculated peak splenic divided by myocardial signal intensity (peak splenic/myocardial signal intensity) differed between stress agents (adenosine median, 0.34; dobutamine median, 1.34; regadenoson median, 1.13; P , .001). Failed splenic switch-off was significantly more common in CE-MARC patients with falsenegative findings than with true-negative findings (34% vs 9%, P , .005).

Conclusion:

Failed splenic switch-off with adenosine is a new, simple observation that identifies understressed patients who are at risk for false-negative findings on perfusion MR images. These data suggest that almost 10% of all patients may be understressed, and that repeat examination of individuals with failed splenic switch-off may significantly improve test sensitivity.

1

From the Heart Hospital Imaging Centre, University College London, 16-18 Westmoreland St, London W1G 8PH, England (C.M., A.S.H., G.C., J.C.M.); Multidisciplinary Cardiovascular Research Centre and Leeds Institute of Genetics, Health and Therapeutics, University of Leeds, Leeds, England (D.P.R., S.P., J.P.G.); Department of Medicine (T.C.W., E.B.S.) and UPMC Cardiovascular Magnetic Resonance Center (E.B.S.), University of Pittsburgh School of Medicine, Pittsburgh, Pa; NIHR Cardiovascular Biomedical Research Unit, Barts Health NHS Trust and Queen Mary University of London, London, England (S.E.P.); and Wessex Cardiothoracic Unit, Southampton University Hospitals NHS Trust, Southampton, England (C.P.). Received September 10, 2014; revision requested November 10; revision received December 30; accepted January 29, 2015; final version accepted February 5. Address correspondence to C.M. (e-mail: [email protected]). RSNA, 2015

q

Radiology: Volume 000: Number 0—   2015 n radiology.rsna.org

Imaging

Charlotte Manisty, PhD, MRCP David P. Ripley, MRCP Anna S. Herrey, MD, PhD, MRCP Gabriella Captur, MD, MRCP, MSc Timothy C. Wong, MD, MS Steffen E. Petersen, MD, DPhil Sven Plein, MBChB, PhD Charles Peebles, MRCP, FRCR Erik B. Schelbert, MD, MS John P. Greenwood, MBChB, PhD James C. Moon, MD, MRCP

Original Research n Cardiac

Splenic Switch-off: A Tool to Assess Stress Adequacy in Adenosine Perfusion Cardiac MR Imaging1

RSNA, 2015

q

Online supplemental material is available for this article. 1

CARDIAC IMAGING: Assessment of Stress Adequacy in Adenosine Perfusion Cardiac MR Imaging

F

unctional imaging to detect myocardial ischemia requires patients to be adequately stressed, but there is a lack of objective indicators of the stress response to adenosine. Myocardial stress perfusion cardiovascular magnetic resonance (MR) imaging is a safe, accurate, and reproducible technique with recent single- and multicenter studies demonstrating higher sensitivity than single photon emission computed tomographic (SPECT) imaging (1,2). However, rates of false-negative findings between 5% and 10% are reported when compared with quantitative coronary angiographic imaging with or without invasive functional testing (1–3). It is thought that up to 50% of these examinations with falsenegative findings result from patients who receive insufficient pharmacologic stress to unmask perfusion deficits (4) either because of drug interactions, such as caffeine antagonism, or because of administration errors. Unlike with other functional imaging tests for ischemia, where the stress agent is administered until patients reach a prespecified physiologic target, adenosine is administered according to a fixed

Advances in Knowledge nn Splenic switch-off (the visual attenuation of splenic perfusion during adenosine stress perfusion MR compared with rest) is a potential simple indicator of stress adequacy. nn The spleen is visible in 99% of routinely acquired studies and splenic switch-off can be reliably visually determined in 90% of studies.

protocol (5) with a heart rate rise of 10 beats per minute and/or symptomatic response to adenosine interpreted as signs of stress adequacy. However, this approach is not guaranteed to induce coronary hyperemia and the markers of stress response are unpredictable and subjective (6–8). During reporting of clinical adenosine perfusion examinations we noted marked attenuation of splenic perfusion in the stress images compared with the corresponding rest images (in this article, we refer to the visual attenuation of splenic perfusion during adenosine stress perfusion MR compared with rest as splenic switch-off). The short-axis image orientation, in which most myocardial perfusion MR images are acquired, typically includes the spleen, which allows for simultaneous assessment of myocardial and splenic blood flow. We hypothesized that visual observation of splenic perfusion during adenosine infusion might help to gauge the adequacy of adenosine stress; failure of splenic switch-off would alert the operator and/or reporting physician to potentially understressed patients. Accordingly, we formed a multicenter collaboration to investigate the potential clinical utility of splenic switch-off (and the failed splenic switch-off) to determine stress

Implications for Patient Care nn Current practice (at least in three three high-volume centers) may result in up to one in 11 participants who undergoes adenosine stress perfusion MR imaging to be inadequately stressed.

nn In cardiac MR perfusion images with proven false-negative results, failed splenic switch-off was four times more common than in perfusion scans with true negative results, which shows that it is a valid marker of understress.

nn Cardiac MR readers should look for failed splenic switch-off routinely during perfusion MR imaging because, when present, it identifies participants at risk of false-negative results, who should be considered for repeat imaging.

2

nn Splenic switch-off is specific to adenosine (not present with dobutamine or regadenoson).

nn Observation of failed splenic switch-off is a simple and quick marker of stress inadequacy and is incremental over hemodynamic response.

Manisty et al

adequacy in myocardial adenosine perfusion MR imaging, with the ultimate goal to improve test sensitivity. This study investigates the pharmacology and potential clinical utility of splenic switch-off to identify understress in adenosine perfusion cardiac magnetic resonance imaging.

Materials and Methods Study Data A retrospective observational study of splenic perfusion in stress myocardial perfusion MR examinations was performed by using three different pharmacologic stressors acquired in four separate institutions in a total of 200 patients (133 men; mean age, 61.6 years [age range, 18–87 years]; 67 women; mean age, 59.2 years [age range, 30–81 years]). At each of the four institutions, appropriate ethics and permissions from the institutional review board were in place or not needed for the retrospective analysis of anonymized data or for the Clinical Evaluation of MR Imaging in Coronary Heart Disease (CE-MARC) trial, from which these data were collected. The observation cohort consisted of 50 consecutive adenosine perfusion

Published online before print 10.1148/radiol.2015142059 Content codes: Radiology 2015; 000:1–9 Abbreviations: CE-MARC = Clinical Evaluation of MR Imaging in Coronary Heart Disease IQR = interquartile range Author contributions: Guarantors of integrity of entire study, C.M., S.P., C.P., J.C.M.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, C.M., D.P.R., G.C., T.C.W., C.P., J.C.M.; clinical studies, C.M., D.P.R., C.P., E.B.S., J.P.G., J.C.M.; experimental studies, C.M., J.C.M.; statistical analysis, C.M., D.P.R., S.E.P., E.B.S., J.C.M.; and manuscript editing, all authors Conflicts of interest are listed at the end of this article.

radiology.rsna.org n Radiology: Volume 000: Number 0—   2015


examinations with no exclusions acquired at the Heart Hospital (University College London, London, England) to assess whether splenic tissue is consistently visible, whether splenic switch-off consistently occurs with adenosine, and to determine interobserver agreement in grading splenic switch-off. Two comparison cohorts that consisted of 25 consecutive dobutamine perfusion examinations acquired at the University of Southampton Hospital (Hampshire, England) and 25 regadenoson perfusion MR examinations acquired at University of Pittsburgh Medical Center (Pittsburgh, Pa) were used to assess whether splenic switch-off is adenosine-specific or a generic response to pharmacologic stress. A validation cohort consisted of 100 adenosine perfusion images that were negative for inducible perfusion defects and acquired as part of the CE-MARC trial (1) in Leeds, England, to evaluate the clinical use of splenic switch-off as an indicator of adenosine stress adequacy. This cohort included all 35 false-negative studies and 65 randomly selected true-negative studies from the CE-MARC trial. Previous analysis had suggested that over half of those with false-negative findings at perfusion MR examinations had inadequate pharmacologic stress (4); therefore, a higher rate of failed splenic switch-off in participants with false-negative results compared with true-negative results would indicate potential clinical utility of the observation to detect pharmacologic understress.

Adenosine Observation Cohort Splenic perfusion was assessed retrospectively in 50 adenosine myocardial perfusion MR images acquired for routine clinical purposes, (38 men; mean age, 63 years [age range, 38–83 years]; 12 women; mean age, 59 years [age range, 43–70 years]). Examinations were performed by using a 1.5-T MR imager (Avanto; Siemens Healthcare, Erlangen, Germany) with adenosine administered at 140 (mg ∙ kg21)/min

for 4 minutes (with the last minute at 170 [mg ∙ kg21]/min if the patient had no symptoms or no heart rate response) and the contrast agent gadoterate meglumine (Dotarem; Guerbet, Villepinte, France) administered intravenously at 0.05 mmol/kg. First-pass perfusion imaging was performed every cardiac cycle by using a T1-weighted saturation recovery gradient-echo sequence with fast low-angle shot readout with stress and after 10 minutes of recovery (ie, at rest).

Comparison Cohort 1: Dobutamine Dobutamine myocardial perfusion MR images were acquired for clinical purposes in 25 patients (16 men; mean age, 66 years [age range, 48–76 years]; nine women; mean age, 63 years [age range, 30–72 years]). Examinations were performed according to standard protocols for regional wall motion abnormalities (9) with additional perfusion images acquired at peak stress (rest perfusion images were not acquired). Images were acquired by using a 1.5-T clinical MR imager (Avanto; Siemens Healthcare) with dobutamine initially infused at 10 (mg ∙ kg21)/min and increasing in increments up to a maximum of 40 (mg ∙ kg21)/min until the target heart rate was achieved or a recognized indication to stop was reached. Contrast agent dose was 0.15 mmol/kg of dimeglumine gadobenate. The threesection myocardial perfusion MR image was typically acquired over two heartbeats because of the high heart rates with dobutamine. Comparison Cohort 2: Regadenoson Splenic perfusion was retrospectively assessed after administration of regadenoson (0.4-mg bolus; Lexiscan, Astellas, Northbrook, Ill) followed by aminophylline (100 mg) in 25 examinations (nine men; mean age, 57 years [age range, 18–80 years]; 16 women; mean age, 59 years [age range, 38–81 years]) acquired for routine clinical purposes by using a 1.5-T MR imager (Espree; Siemens Healthcare). First-pass perfusion used steady-state free-precession readout


Manisty et al

after a T1-weighted saturation recovery preparation with a gadoteridol contrast agent (0.05 mmol/kg). Rest perfusion images were acquired after a 10-minute delay.

Validation Cohort: CE-MARC Adenosine Data The methods used in CE-MARC were previously published (1,10,11). In summary, consecutive patients suspected of having angina pectoris were screened and enrolled if they had at least one major cardiovascular risk factor and a cardiologist judged them to have likely stable angina that required investigation (39% prevalence of significant coronary disease in this population). All patients were scheduled by protocol to undergo multiparametric cardiac MR (including adenosine stress perfusion) and SPECT (test order randomized) imaging, followed by invasive coronary angiography within 4 weeks. All cardiac MR examinations were performed on a 1.5-T MR system (1.5 T Intera CV Machine; Philips Healthcare, Best, the Netherlands), and first-pass perfusion was compared after a 4-minute infusion of adenosine at 140 (mg ∙ kg21)/min and then at rest by using 0.05 mmol/kg dimeglumine gadopentetate contrast agent (Magnevist; Bayer Healthcare, Wayne, NJ). Myocardial MR perfusion images were compared with the reference standard of invasive quantitative coronary angiography to define significant coronary artery disease. Images were categorized as false negatives if the myocardial MR perfusion image was considered to be normal but quantitative coronary angiography depicted significant coronary lesions; this categorization did not include the other components of the multiparametric cardiac MR analysis performed in the original study. In this current study we used false-negative and true-negative results as they were originally reported, with no additional analysis. We used all 35 examinations with false negative results from CE-MARC, and added 65 true-negative examinations (70 men; mean age, 60 years [age range, 37–73 years]; 30 women; 3


mean age, 57 years [age range, 42–72 years]). We then anonymized and randomly coded the cohort of 100 examinations so that readers were blinded to all clinical and invasive data at the time of grading splenic perfusion.

Visual Analysis of Splenic Perfusion Splenic perfusion was graded by using a simple visual comparison of the splenic tissue contrast enhancement on whichever of the three stress perfusion short-axis sections in which the spleen was seen best, with both the rest perfusion images of the spleen and the myocardium. Perfusion was graded as either switched off (ie, clearly visually lower splenic enhancement on stress than rest images) or as failed switch-off (ie, visually similar splenic enhancement at rest and stress). When only stress images were available (ie, in the dobutamine comparison cohort), splenic perfusion was compared with myocardium. Images were assessed independently by two observers from different institutions (C.M. and D.P.R., each level-3 accredited with .4 years of experience with cardiac MR imaging) who were blinded to clinical data, and we assessed interobserver agreement of grading. Semiquantitative Analysis of Splenic Perfusion Semiquantitative analysis of splenic perfusion was performed by using signal intensity curves, and analyzed by using standard proprietary software (OsiriX; OsiriX Foundation, Geneva, Switzerland). Regions of interest in the myocardium and spleen were drawn on the stress perfusion images of each patient and were then copied onto the corresponding rest perfusion images. Tissue perfusion (ie, peak signal intensity) was estimated by subtracting the maximal signal intensity after injection of contrast agent from the baseline signal intensity obtained before administration of the contrast agent. For graphs, consecutive heart beats were determined from the Digital Imaging and Communications in Medicine headers. 4

To allow statistical comparison of data acquired in different centers, we calculated the percentage reduction in the signal intensity values from resting to stress states. Signal intensity ratios were then calculated by comparing splenic signal intensity at stress with that in another tissue (myocardium). We calculated the splenic-to-myocardial signal intensity ratio at stress and rest, and we then calculated the stress-to-rest signal intensity ratios for both myocardial and splenic tissue.

Validation of Splenic Switch-off as an Indicator of Stress Adequacy in the CEMARC Dataset We graded the hemodynamic response to adenosine in the CE-MARC validation cohort as normal (rise in heart rate of .10 beats per minute) or reduced (12). Studies were then categorized by using three dichotomous variables: true-negative or false-negative examination results (as per original CE-MARC report); splenic switch-off or failed splenic switch-off; and presence or absence of hemodynamic response. We then assessed whether failed splenic switch-off was more common in the false-negative examination group versus true-negative examination group of the CE-MARC cohort, and also whether splenic switch-off follows hemodynamic response and whether it could aid identification of understressed patients who are at risk for false-negative examination results. Statistical Analyses Interobserver agreement on classifi cation of splenic perfusion was measured by using Cohen k for two observers. The distribution of the quantitative signal intensity data was first assessed for normality by using the Shapiro-Wilk test, quantile-quantile plots, and density plots. Because the adenosine stress data were not parametrically distributed, analysis of differences between semiquantitative measures of perfusion with different pharmacologic agents was performed by using the Wilcoxon rank sum test

Manisty et al

with continuity correction, and within-patient differences in perfusion between stress and rest were compared by using the Wilcoxon signed rank test. We compared signal intensities in the true-negative and false-negative examination groups by using the Fisher exact test. Data are expressed as the median and interquartile range (IQR) unless otherwise stated, and P values less than .05 indicate statistical significance.

Results Assessment of Splenic Switch-off in Myocardial Perfusion MR Examinations Adequate splenic tissue for analysis was seen in 99% of the myocardial perfusion examinations (49 of 50 adenosine observation cohort examinations, 24 of 25 dobutamine examinations, all 25 regadenoson examinations, and all 100 of the adenosine validation cohort). The spleen was typically not visible with cardiac heterotopia, for example when the heart lies in a coronal plane (apex toward the anterior axillary line), or where there had been splenectomy. Interobserver agreement for grading splenic switch-off was excellent (Cohen k = 0.92), with concordance in 191 of the 198 examinations with visible splenic tissue. Visual Assessment of Splenic Perfusion with Different Pharmacologic Stress Agents Splenic switch-off with adenosine was present in 90% of the adenosine perfusion MR studies in the observation cohort (44 of 49 studies where splenic tissue was visible) (Fig 1 and Movie [online]). Splenic switch-off did not occur with regadenoson or dobutamine in any of the studies. Semiquantitative Measurements of Splenic Perfusion with Different Pharmacologic Stress Agents Semiquantitative measurements of splenic perfusion by using signal intensity corroborated the results of visual grading. Peak splenic signal intensity



1.13 [IQR 0.96–1.44]; P , .001; Fig 4), which confirms that splenic switch-off was not the result of generalized hypoperfusion with adenosine.

Figure 1

Figure 1: Splenic perfusion MR images in three individual participants at rest and during stress with the three pharmacological stress agents. Perfusion is standard saturation recovery prepared spoiled gradient echo (fast low-angle shot) single-shot perfusion sequence with factor-2 parallel imaging and an inversion time of 140 msec for adenosine and dobutamine stress imaging with a steady state readout for regadenoson stress imaging. Adenosine perfusion MR images in a 58-year-old man show, A, rest perfusion and, B, adenosine stress. Regadenoson perfusion MR images in a 58-year-old man show, C, rest perfusion and, D, regadenoson stress. Dobutamine perfusion MR images in a 59-year-old man show, E, dobutamine stress perfusion. The spleen is visually darker (switched off) with adenosine stress compared with rest. There is no splenic switch-off with either regadenoson or dobutamine stress.

markedly decreased with adenosine compared with rest, both in the observation cohort (median reduction in signal intensity from resting levels with adenosine, 78% [IQR, 59%–91%]; P , .001) and in the validation cohort (median reduction in signal intensity from resting levels with adenosine, 71% [IQR, 38%–90%]; P , .001. This was a different response to with regadenoson, where the median splenic signal intensity with stress was unchanged to at rest, 4% reduction [IQR, 23%– 14%]; P = .22; Fig 2).

Manisty et al

Splenic stress-to-rest signal intensity ratios were significantly lower in the adenosine than in the regadenoson perfusion examinations (median signal intensity ratios, adenosine vs regadenoson, respectively, 0.21 [IQR, 0.09–0.41] vs 0.96 [IQR, 0.86–1.03]; P , .001; Fig 3). Splenic-to-myocardial signal intensity ratios were significantly lower with adenosine stress (median, 0.24 [IQR, 0.11–0.41]) than with both dobutamine (median, 1.34 [IQR, 0.89–1.49]; P , .001) and regadenoson stress (median,


Clinical Utility of Failed Splenic Switch-off as a Marker of Adenosine Understress The frequency of failed splenic switchoff was almost four times higher in examinations with known false-negative perfusion results than in examinations with true-negative results in the CEMARC adenosine validation cohort (34% vs 9%, respectively; P , .005; Fig 5). Suboptimal hemodynamic response (increase in heart rate of ,10 beats per minute) showed a trend toward higher frequency in examinations with false-negative results versus true-negative results (34% vs 17%, respectively), but the difference between groups was not statistically significant (P = .08). Overall, 22% of participants with false-negative results and 5% of participants with true-negative perfusion results had neither splenic switchoff nor a hemodynamic response (Fig 5), and 80% of patients had concordant hemodynamic and splenic responses to adenosine. However, eight of 35 (22%) patients in the group with false-negative perfusion results had failed splenic switch-off despite a normal hemodynamic response to adenosine, which suggests that splenic switch-off provides incremental information to aid the interpretation of negative adenosine perfusion MR examination results. Discussion We identified and described splenic switch-off, a new sign with a clear potential clinical application for assessment of stress adequacy during adenosine myocardial stress perfusion MR imaging. Splenic switch-off was observed in over 90% of adenosine perfusion examinations, but it is not a generic response; pharmacologic stress perfusion sequences acquired after administration of dobutamine and regadenoson showed no attenuation of splenic blood flow. Failed splenic switch-off occurred in more than a third of proven false-negative 5


Figure 2

Manisty et al

Figure 3

Figure 3: Graph shows stress-to-rest tissue perfusion ratio in myocardium and spleen with adenosine and regadenoson. Splenic switch-off with adenosine stress resulted in a markedly lower stress-to-rest tissue perfusion ratio in the spleen compared with the myocardium. With regadenoson the stress and rest ratios were similar. Data are means 6 standard error. NS = nonsignificant.

Figure 2: Graphs show example time-signal intensity curves for blood, spleen, and myocardium at rest and with stress from adenosine (A, rest and, B, stress), regadenoson (C, rest and, D, stress), and dobutamine (E, stress) in example patients, adjusted for baseline signal intensities in arbitrary units (au). Splenic perfusion is higher than myocardial perfusion in all scenarios except adenosine stress, where splenic switch-off occurs. Arrows highlight splenic perfusion response to pharmacologic stress agent. Time is shown in seconds.

adenosine perfusion examinations: almost four times more commonly than in examinations with true-negative results. Splenic switch-off is a simple and quick marker of stress adequacy, with no additional measures required for acquisition or detection. We hypothesize that a repeat perfusion cardiac MR examination in those patients with failed splenic switch-off by using higher dose adenosine (thereby ensuring that adequate stress was achieved second time) would reduce false-negative findings on myocardial perfusion MR examinations by up to a third. There may also be a nonnegligible rate 6

of understress in the general population with failed splenic switch-off in approximately one in 11 true-negative adenosine perfusion MR examinations. There is a growing body of data that supports the use of myocardial perfusion MR imaging in the diagnosis of coronary artery disease, in terms of diagnostic performance (1,2), safety (10,12,13), and cost effectiveness (14). Worldwide, adenosine is the most widely used pharmacologic stressor for perfusion MR imaging, although the test sensitivity remained relatively static during the past decade. It is speculated that many false-negative

perfusion MR examination results may be because patients are insufficiently stressed by the adenosine to unmask perfusion defects, and currently it is difficult to determine stress adequacy. In addition to its actions as a coronary vasodilator, adenosine causes systemic vasodilatation and reflex sympathetic activation, which cause a mild reduction in systemic blood pressure and a mild increase in heart rate—it is these small hemodynamic changes that are monitored to determine whether patients are sufficiently stressed. Studies show that around one in six patients have a blunted hemodynamic response to adenosine (12) because of physiologic, pathologic (15), or pharmacologic differences (16,17). Furthermore, peripheral heart rate and mean arterial pressure changes are known to be poor predictors of myocardial blood flow and coronary vascular resistance changes (8), and the blunted hemodynamic response itself may be a diagnostic and prognostic marker (6). Therefore, it may be misleading to determine



Figure 4

Figure 4: Graph shows splenic perfusion relative to myocardial perfusion with stress induced by using the three different pharmacologic stressors. Splenic perfusion was markedly attenuated with adenosine stress compared with myocardial perfusion, which was a different response to both regadenoson and dobutamine stress. Data are means 6 standard error.

adenosine stress adequacy from hemodynamic responses alone. One particular concern with adenosine and regadenoson stress is caffeine. Caffeine is a competitive inhibitor of adenosine, and therefore caffeine consumption can increase the risk of false-negative findings (18). Although patients are told to avoid caffeine for 12–24 hours before the examination, compliance is inconsistent and difficult to verify (19). The biologic half-life of caffeine also varies widely with genetic differences in metabolism (20) and drug interactions; fluvoxamine can prolong half-life 10-fold, to 56 hours (21). Without a reliable marker of stress adequacy, the potential presence of caffeine may undermine confidence in negative test results. Adenosine visually attenuates splenic blood flow on perfusion MR images, an observation that is supported by data from animal studies that measure regional blood flow during adenosine infusion (22–24). Regarding, adenosine

Manisty et al

Figure 5

Figure 5: Flowchart shows results of assessment of splenic switch-off and hemodynamic response to adenosine in the validation cohort of 100 participants from the CE-MARC trial (35 participants with falsenegative perfusion MR examination results, and 65 participants with true-negative examination results). Significantly more patients had failed splenic switch-off in the group with false-negative examination results compared with the group with true negative examination results. CMR = cardiac MR imaging.

mediates its effects on splenic perfusion via A1 and/or A2B adenosine receptors (25,26) to maintain circulatory volume in conditions of shock (25), whereas coronary vasodilatation is mediated via the adenosine A2A receptor. This explains why regadenoson, the selective coronary A2A receptor agonist, had no discernable effect on splenic perfusion during stress perfusion MR imaging in this study. Failed splenic switch-off may be a better marker than poor heart rate response to detect inadequate pharmacologic stress with adenosine; although rates of both markers were similar in the group with false-negative examination results (Fig 5; 34%), rates of absent heart rate response were twice as high as failed splenic switch-off in the group with true-negative examination results (absent heart rate response vs failed splenic switch-off, 17% vs 9%, respectively). Although failed splenic switch-off with adenosine can be indicative of understressed patients, once the contrast has been administered for image acquisition, immediate repetition of the stress perfusion examination with


adenosine does not ensure success. One future strategy may be to directly measure real-time splenic (or even left main stem coronary arterial) flow during adenosine stress before contrast agent administration, which would ensure that perfusion images are acquired only when an adequate stress response is achieved. In the meantime, our clinical practice for all examinations is a quick visual assessment of splenic perfusion when images are acquired (a visual comparison with the myocardium) and reported (a comparison of stress vs rest splenic perfusion, before focusing on the heart). If failure of splenic switch-off is supported by a lack of symptomatic and hemodynamic response, we either administer the adenosine at an increased dose ensuring that the intravenous access is adequate (ie, failed stress followed by repeated stress), or recall the patient for repeat examination by using high-dose adenosine, emphasizing the importance of caffeine abstention on another visit. This study has limitations. Our study used retrospective data with all images acquired for clinical purposes 7


or as part of the CE-MARC clinical trial and was not designed to assess the performance of perfusion MR imaging for the detection of coronary artery disease. Instead, the aim was to assess the feasibility, pharmacology, and clinical usefulness of splenic switch-off as an indicator of inadequate stress in adenosine perfusion examinations. There may be other mechanisms for failed splenic switchoff of which we are unaware, such as receptor variability, and similarly, because we have no measure of stress adequacy with adenosine, we cannot definitively assess whether patients are truly stressed. The scenario in which hemodynamic and splenic switch-off are discordant merits further study. For simplicity, semiquantitative measurements of perfusion were acquired instead of formal quantification of splenic and myocardial blood flow. Objective quantification was performed only to corroborate the straightforward visual assessments of splenic perfusion, and therefore the lack of calibrated measurements should not detract from the results. Signal intensity ratios, albeit unconventional, were used to compare images acquired in the different centers because of the use of different protocols, gadolinium-based compounds, administration protocols, and equipment. The CE-MARC trial used quantitative coronary angiographic imaging as the gold standard with which the perfusion MR imaging was compared, however, invasive coronary fractional flow reserve measurements are the contemporary gold standard. Although recent studies (10,27,28) found that myocardial perfusion MR imaging performs well when compared with fractional flow reserve, the CE-MARC study may have over-estimated the rates of false-negative findings on perfusion MR images because some of the stenoses greater than 70% may not be functionally obstructive. Splenic switch-off with adenosine is a simple visual marker of stress adequacy during myocardial perfusion MR imaging. Failure of splenic switch-off is an indicator of inadequate pharmacologic stress that has the potential 8

to identify patients at risk for falsenegative examination results. Repeat examinations of individuals with failure of splenic switch-off could reduce falsenegative perfusion MR examination results by a third, but it may be that up to one in 11 of all adenosine perfusion patients are inadequately stressed. Disclosures of Conflicts of Interest: C.M. disclosed no relevant relationships. D.P.R. disclosed no relevant relationships. A.S.H. disclosed no relevant relationships. G.C. disclosed no relevant relationships. T.C.W. Financial activities related to the present article: disclosed no relevant relationships. Financial activities not related to the present article: author disclosed money paid to author’s institution for grant from the American Heart Association and Children’s Cardiomyopathy Foundation; author received money for travel expenses from Gilead Sciences. Other relationships: disclosed no relevant relationships. S.E.P. Financial activities related to the present article: disclosed no relevant relationships. Financial activities not related to the present article: author received personal fees from Circle Cardiovascular Imaging. Other relationships: disclosed no relevant relationships. S.P. disclosed no relevant relationships. C.P. disclosed no relevant relationships. E.B.S. disclosed no relevant relationships. J.P.G. disclosed no relevant relationships. J.C.M. disclosed no relevant relationships.

References 1. Greenwood JP, Maredia N, Younger JF, et al. Cardiovascular magnetic resonance and single-photon emission computed tomography for diagnosis of coronary heart disease (CEMARC): a prospective trial. Lancet 2012; 379(9814):453–460. 2. Schwitter J, Wacker CM, Wilke N, et al. MRIMPACT II: Magnetic Resonance Imaging for Myocardial Perfusion Assessment in Coronary artery disease Trial: perfusion-cardiac magnetic resonance vs. single-photon emission computed tomography for the detection of coronary artery disease: a comparative multicentre, multivendor trial. Eur Heart J 2013;34(10):775–781. 3. Hamon M, Fau G, Née G, Ehtisham J, Morello R, Hamon M. Meta-analysis of the diagnostic performance of stress perfusion cardiovascular magnetic resonance for detection of coronary artery disease. J Cardiovasc Magn Reson 2010;12(1):29. 4. Plein S, Kidambi A, Sourbron S, et al. Associated factors for a false negative cardiovascular magnetic resonance perfusion study: a CE-MARC substudy. J Cardiovasc Magn Reson 2013;15(Suppl 1):P214.

Manisty et al

5. Kramer CM, Barkhausen J, Flamm SD, Kim RJ, Nagel E; Society for Cardiovascular Magnetic Resonance Board of Trustees Task Force on Standardized Protocols. Standardized cardiovascular magnetic resonance (CMR) protocols 2013 update. J Cardiovasc Magn Reson 2013;15(1):91. 6. Hage FG, Dean P, Bhatia V, Iqbal F, Heo J, Iskandrian AE. The prognostic value of the heart rate response to adenosine in relation to diabetes mellitus and chronic kidney disease. Am Heart J 2011;162(2):356–362. 7. Norlén K. Central and regional haemodynamics during controlled hypotension produced by adenosine, sodium nitroprusside and nitroglycerin. Studies in the pig. Br J Anaesth 1988;61(2):186–193. 8. Mishra RK, Dorbala S, Logsetty G, et al. Quantitative relation between hemodynamic changes during intravenous adenosine infusion and the magnitude of coronary hyperemia: implications for myocardial perfusion imaging. J Am Coll Cardiol 2005;45(4):553– 558. 9. Nagel E, Lehmkuhl HB, Bocksch W, et al. Noninvasive diagnosis of ischemia-induced wall motion abnormalities with the use of high-dose dobutamine stress MRI: comparison with dobutamine stress echocardiography. Circulation 1999;99(6):763–770. 10. Bettencourt N, Chiribiri A, Schuster A, et al. Cardiac magnetic resonance myocardial perfusion imaging for detection of functionally significant obstructive coronary artery disease: a prospective study. Int J Cardiol 2013;168(2):765–773. 11. Greenwood JP, Maredia N, Radjenovic A, et al. Clinical evaluation of magnetic resonance imaging in coronary heart disease: the CEMARC study. Trials 2009;10:62. 12. Karamitsos TD, Ntusi NA, Francis JM, Holloway CJ, Myerson SG, Neubauer S. Feasibility and safety of high-dose adenosine perfusion cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2010;12:66. 13. Chen J, Einstein AJ, Fazel R, et al. Cumulative exposure to ionizing radiation from diagnostic and therapeutic cardiac imaging procedures: a population-based analysis. J Am Coll Cardiol 2010;56(9):702–711. 14. Walker S, Girardin F, McKenna C, et al. Cost-effectiveness of cardiovascular magnetic resonance in the diagnosis of coronary heart disease: an economic evaluation using data from the CE-MARC study. Heart 2013;99(12):873–881. 15. Triposkiadis F, Karayannis G, Giamouzis G, Skoularigis J, Louridas G, Butler J. The sympathetic nervous system in heart



failure physiology, pathophysiology, and clinical implications. J Am Coll Cardiol 2009;54(19):1747–1762. 16. Reyes E, Stirrup J, Roughton M, D’Souza S, Underwood SR, Anagnostopoulos C. Attenuation of adenosine-induced myocardial perfusion heterogeneity by atenolol and other cardioselective beta-adrenoceptor blockers: a crossover myocardial perfusion imaging study. J Nucl Med 2010;51(7):1036–1043. 17. Sharir T, Rabinowitz B, Livschitz S, et al. Underestimation of extent and severity of coronary artery disease by dipyridamole stress thallium-201 single-photon emission computed tomographic myocardial perfusion imaging in patients taking antianginal drugs. J Am Coll Cardiol 1998;31(7):1540– 1546. 18. Reyes E, Loong CY, Harbinson M, Dono van J, Anagnostopoulos C, Underwood SR. High-dose adenosine overcomes the attenuation of myocardial perfusion reserve caused by caffeine. J Am Coll Cardiol 2008; 52(24):2008–2016. 19. Banko LT, Haq SA, Rainaldi DA, et al. Incidence of caffeine in serum of patients un-

dergoing dipyridamole myocardial perfusion stress test by an intensive versus routine caffeine history screening. Am J Cardiol 2010; 105(10):1474–1479. 20. Cornelis MC, El-Sohemy A, Kabagambe EK, Campos H. Coffee, CYP1A2 genotype, and risk of myocardial infarction. JAMA 2006;295(10):1135–1141. 21. Culm-Merdek KE, von Moltke LL, Harmatz JS, Greenblatt DJ. Fluvoxamine impairs single-dose caffeine clearance without altering caffeine pharmacodynamics. Br J Clin Pharmacol 2005;60(5):486–493. 22. van Lambalgen AA, van Kraats AA, Mul der MF, van den Bos GC, Teerlink T, Thijs LG. Organ blood flow and distribution of cardiac output in dopexamine- or dobutamine-treated endotoxemic rats. J Crit Care 1993;8(2):117–127. 23. Rooney MW, Crystal GJ, Salem MR, Paulissian R. Influence of nifedipine on systemic and regional hemodynamics during adenosine-induced hypotension in dogs. Anesth Analg 1989;68(3):261–269.

Manisty et al

diating inhibition of intestinal secretion. Br J Pharmacol 1995;114(1):152–156. 25. Hamza SM, Kaufman S. Role of spleen in integrated control of splanchnic vascular tone: physiology and pathophysiology. Can J Physiol Pharmacol 2009;87(1):1–7. 26. Fozard JR, Milavec-Krizman M. Contrac tion of the rat isolated spleen mediated by adenosine A1 receptor activation. Br J Pharmacol 1993;109(4):1059–1063. 27. Chiribiri A, Hautvast GLTF, Lockie T, et al. Assessment of coronary artery stenosis severity and location: quantitative analysis of transmural perfusion gradients by high-resolution MRI versus FFR. JACC Cardiovasc Imaging 2013;6(5):600–609. 28. Costa MA, Shoemaker S, Futamatsu H, et al. Quantitative magnetic resonance perfusion imaging detects anatomic and physiologic coronary artery disease as measured by coronary angiography and fractional flow reserve. J Am Coll Cardiol 2007;50(6):514– 522.

24. Hancock DL, Coupar IM. Functional characterization of the adenosine receptor me-


9