134. Nuclear and Multimodality Imaging: ...

CardioNerd Amit Goyal is joined by Dr. Erika Hutt (Cleveland Clinic general cardiology fellow), Dr. Aldo Schenone (Brigham and Women’s advanced cardiovascular imaging fellow), and Dr. Wael Jaber (Cleveland Clinic cardiovascular imaging staff and co-founder of Cardiac Imaging Agora) to discuss nuclear and complimentary multimodality cardiovascular imaging for the evaluation of cardiac sarcoidosis. Show notes created by Dr. Hussain Khalid (University of Florida general cardiology fellow and CardioNerds Academy fellow in House Thomas). To learn more about multimodality cardiovascular imaging, check out Cardiac Imaging Agora!

Cardiac sarcoidosis is a leading cause of morbidity and mortality for patients with sarcoidosis. A high index of suspicion is needed for the diagnosis as it is often recognized late or unrecognized. It is difficult to diagnose given the focal nature of the cardiac involvement limiting the utility of biopsy and the available clinical criteria have limited diagnostic accuracy. Multimodality imaging plays a large role in the diagnosis and management of patients with cardiac sarcoidosis with the different imaging modalities offering complimentary information and functions. 

Collect free CME/MOC credit just for enjoying this episode! 

Quoatables

“It’s not important for you to love the Soviet Union. It’s important for the Soviet Union to love you back [Stalin regarding the famous dissonant Russian poet Anna Akhmatova]. When we talk about PET, you love PET, but the PET has to love you back, and it has to love you back in a way where you have to know how to approach this test. With, first, some humility about its limitations: 1) inflammation is universal…and 2) the prep is extremely important.”  — 11:25

“A test without a good preparation is a preparation to fail.” –15:30

“Sarcoidosis is kind of the tuberculosis that we have in medicine—it can present as anything.” –36:40

Pearls

1. Cardiac Magnetic Resonance Imaging (Cardiac MRI) and/or 18-Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) are complimentary tests in the evaluation of cardiac sarcoidosis. Both tests look for scarring and inflammation. Cardiac MRI is a good initial test due to its high negative predictive value (i.e. absence of LGE makes cardiac sarcoidosis less likely) but not great for following a cardiac sarcoidosis patient’s response to therapy. Cardiac FDG-PET is great to follow a patient’s response to therapy especially in patients with intracardiac devices such as a pacemaker.

1. 18-fluorodeoxyglucose (FDG) is a glucose analog and just like glucose, is transported into the cell by transporters. Once in the cell, it is phosphorylated, like glucose is, by hexokinase in preparation for use in glycolysis. Unlike glucose, however, it does not proceed to be metabolized any further in the glycolysis pathway and remains trapped in the cell. In the inflammatory cells within sarcoid granulomas, glycolysis is significantly increased to fuel the large energy requirement. Thus, these inflammatory cells (i.e. macrophages) can take up large amounts of FDG.

1. When planning to obtain a cardiac FDG-PET for evaluation of cardiac sarcoidosis, patient preparation is key! There are several available dietary protocols to accomplish the goal of switching the patient’s metabolism to be reliant on fatty acids instead of glucose as an energy source. One such protocol used by the discussants in the episode is prolonged fasting (10-12 hours) prior to the study preceded by two meals that are high in fat and proteins and low in carbohydrates—a ketogenic diet. By having the patient eat this diet, we are trying to switch the metabolism because there is no ability or no offer of glucose for the body to use as an energy source! After we have switched the body’s metabolism to purely fat, when we inject the patient with FDG, hopefully most of the myocardium not affected by inflammatory cells within a granuloma will not have any uptake!

1. Why do we start with resting perfusion images in the imaging portion of the cardiac FDG-PET protocol for cardiac sarcoidosis? Resting perfusion images allow us to identify any perfusion defects at baseline. These images can be compared to the FDG images to see if there is match or mismatch in areas of abnormalities. Resting perfusion images also allow us to assess LV and RV function. Resting perfusion images in conjunction with FDG images can also allow us to monitor the patient’s response to treatment by demonstrating return to normal myocardium from active sarcoid granuloma after treatment or by showing the progression to development of scar.

1. The hallmark for detecting cardiac sarcoidosis with cardiac MRI is late gadolinium enhancement (LGE) in the mid-wall and subepicardial regions. Gadolinium is an extracellular contrast agent that washes out slowly from areas of inflammation or scar (both processes with result in an expansion of the extracellular space). Because of this it is important to look at the distribution pattern of the LGE (i.e. a subendocardial enhancement in a coronary distribution is more suggestive of scar from a prior myocardial infarction in this area). Some features of LGE that favor the diagnosis of cardiac sarcoidosis include: 1) multifocal involvement and 2) involvement of the basal anteroseptum and inferoseptum with contiguous spread into the right ventricle.

Show Notes

1. What is the typical patient population with cardiac sarcoidosis and how does it present?

• Sarcoidosis is a multisystem disorder of unclear etiology characterized by the formation of noncaseating granulomas in multiple organs with an annual incidence of 5 to 40 cases per 100 000 persons in the USA and Europe. It has a 3-fold higher risk in blacks than in whites and it is more common in females. In the USA most disease occurs between ages 25-45, however in Europe and Japan there is a second peak in women older than 70 years old.

• Around 5% of patients with pulmonary/systemic sarcoidosis have symptomatic cardiac involvement and autopsy studies have shown that there is cardiac involvement in 25% of patients. More recent advanced cardiac imaging studies in patients with known extracardiac sarcoidosis suggests asymptomatic cardiac involvement may be present in ~40% of patients. It is difficult to detect given its patchy focal distribution in the heart such that cardiac biopsy has a sensitivity of only 20-30% (this yield may improve with image guidance from cardiac MRI or FDG PET or guidance from electroanatomic voltage mapping). 

• We may suspect cardiac sarcoidosis in patients presenting with new and unexplained atrioventricular (AV) block, atrial or ventricular arrhythmias, or left ventricular dysfunction—especially in patients with a history of prior non-cardiac sarcoidosis. Palpitations, presyncope, syncope or other nonspecific symptoms may also be the initial presentation.
  • As with our patient in the episode, we should suspect cardiac sarcoidosis and infra-hisian disease in a young previously healthy and active patient who is presenting with evidence of cardiac conduction system disease (i.e. AV block, bundle branch block [BBB], etc.)!

2. Using the patient from the episode as an example (previously healthy presenting with syncope, BBB, frequent non-sustained tachycardia on telemetry monitoring, and intermittent complete heart block, with bilateral hilar fullness on chest x-ray), how should we proceed with evaluation for patients with suspected cardiac sarcoidosis?

• To continue our evaluation, we could obtain an echocardiogram to look for wall motion abnormalities and myocardial thinning in the basal areas—findings that can be seen with cardiac sarcoidosis. Even if this is evident, we should still rule out coronary artery disease (CAD) in these patients. In a patient with low risk factors for CAD, we can pursue a coronary computed tomography angiogram (CCTA) and if there is no evidence of obstructive CAD, then we start thinking about an inflammatory process with resultant scar causing the patient’s echocardiography findings. If we have an elderly patient or a patient with significant risk factors for obstructive CAD, then we can obtain a coronary angiogram for evaluation. The bottom line is: common things being common—we don’t want to miss obstructive CAD!
  • Abnormal findings on echocardiography in patients with cardiac sarcoidosis include wall motion abnormalities, diastolic dysfunction, and changes in the left ventricular geometry including: 1) abnormal myocardial wall thickness in a noncoronary distribution (possibly caused by sarcoid granulomas), 2) myocardial wall thinning as a result of a later stage of the same process, 3) left ventricular dilation, and 4) left ventricular systolic dysfunction.
  • Sensitivity, specificity, and negative predictive value of echocardiography for diagnosis of cardiac sarcoidosis are low.
  • There are recent studies that have been published about the value of strain imaging with echocardiography in the diagnosis and assessment of response to therapy of patients with cardiac sarcoidosis, however this is still an evolving area or research.
• After ruling out obstructive CAD we can proceed to Advanced Cardiac Imaging modalities for further evaluation: Cardiac Magnetic Resonance Imaging (Cardiac MRI) and/or 18-Fluorodeoxyglucose Positron Emission Tomography (FDG-PET).  These are complimentary tests. Both tests look for scarring and inflammation. Cardiac MRI is a good initial test but not great for following a cardiac sarcoidosis patient’s response to therapy. Cardiac FDG-PET is great to follow a patient’s response to therapy especially in patients with intracardiac devices  such as pacemaker or ICD. These patients can also have a cardiac MRI done but the diagnostic yield becomes more challenging. Let’s discuss the details and characteristic findings of Cardiac FDG-PET and Cardiac MRI in the next sections, and then circle back and continue our discussion of some comparisons between the two modalities in the evaluation and management of cardiac sarcoidosis.

3. Why is patient preparation prior to a cardiac FDG-PET study for evaluation of cardiac sarcoidosis so important? How do we utilize cardiac FDG-PET in the diagnosis of cardiac sarcoidosis?

• There are three main components to the cardiac FDG-PET study for evaluation of cardiac sarcoidosis: 1) Patient preparation with high-fat, high-protein, low-carbohydrate meals followed by prolonged fasting for 10-12 hours prior to the study in an effort to switch the body’s metabolism to using primarily fatty acids as a source of energy, 2) A resting perfusion imaging component using either Rubidium-82 or N-13 Ammonia, and finally 3) the FDG-imaging component. After obtaining resting perfusion images, the patient may or may not be given intravenous heparin (for lipolysis—discussed further below) and then an intravenous injection of the FDG. They are given around a 90 minute period to allow  for FDG uptake during which they are asked to rest quietly (to minimize skeletal muscle uptake of FDG!). After this, images of the heart looking for FDG uptake are obtained. Following this, whole-body imaging from the base of the skull to the mid-thigh is done to help identify extracardiac sites of sarcoidosis.
• What is the philosophy behind the required strict patient preparation prior to a cardiac FDG-PET study for evaluation of cardiac sarcoidosis? FDG is a glucose analog and just like glucose, is transported into the cell by transporters. Once in the cell, it is phosphorylated, like glucose is, by hexokinase in preparation for use in glycolysis. Unlike glucose, however, it does not proceed to be metabolized any further in the glycolysis pathway and remains trapped in the cell. In the inflammatory cells within sarcoid granulomas, glycolysis is significantly increased to fuel the large energy requirement. Thus, these inflammatory cells (i.e. macrophages) can take up large amounts of FDG.
  • The heart primarily functions with fatty acid uptake, however, it does have some glucose uptake as well—especially those areas that are ischemic or hibernating. We want FDG to only be taken up by inflammatory cells within a potential granuloma and minimize the uptake of FDG by other myocardial cells. Preparation is key!
  • There are several available dietary protocols to accomplish this goal. One such protocol used by the discussants in the episode is prolonged fasting (10-12 hours) prior to the study preceded by two meals that are high in fat and proteins and low in carbohydrates—a ketogenic diet. By having the patient eat this diet, we are trying to switch the metabolism because there is no ability or no offer of glucose for the body to use as an energy source! After we have switched the body’s metabolism to purely fat, when we inject the patient with FDG, hopefully most of the myocardium not affected by inflammatory cells within a granuloma will not have any uptake!
    • For patient’s receiving intravenous infusions during a hospital admission, it is important to avoid including dextrose in the formulations!
  • Along with the above diet, we also don’t want the patient engaging in vigorous physical activity on the day of or day prior to the test as there will then be musculoskeletal uptake of FDG rather than myocardial uptake. We try to avoid simultaneously doing stress imaging during the same encounter as sarcoid imaging because we don’t want to induce ischemia that could potentially induce FDG uptake by the ischemic myocardium. Exercise or pharmacologic stress can induce myocardial ischemia which shifts the metabolism from fatty acids to glucose. Despite this change happening in minutes after the onset of ischemia, it can persist for a long time, so that if you inject FDG after this, that area of ischemic myocardium will light up. This is known as “ischemic memory!”
    • If a simultaneous stress testing evaluation is needed in addition to evaluation for cardiac sarcoidosis, then you can do the rest perfusion imaging and FDG portion on Day 1 and the stress perfusion imaging portion on Day 2.
  • As mentioned above, some centers use intravenous heparin prior to the FDG imaging portion with the thought being that heparin induces lipolysis and can increase the offer of fatty acids to the myocardium (every effort to shift the metabolism!)
  • Frequently on cardiac FDG-PET imaging, patients have failed to suppress the myocardium uptake of glucose due to suboptimal compliance. A study by Dr. Erika Hutt (a discussant on the episode!) and Dr. Paul Cremer is being done with the use of a ketone shake for breakfast and lunch on the day before the study followed by fasting after lunch until the study the following day and is finding that in patients who failed to suppress the myocardium with a standard diet, these patients were able to suppress the myocardium with ketone shake.
    • Additionally, it can be challenging for vegans and vegetarians to follow the high protein component of the diet given the lack of meat in their normal diet. For these patients it is more difficult to adhere to the diet (typically some formulation of tofu or oil can be recommended), but ketone shake may also be helpful in this situation.
• Why do we start with resting perfusion images in the imaging portion of the cardiac FDG-PET protocol for cardiac sarcoidosis? Resting perfusion images allow us to identify any perfusion defects at baseline. These images can be compared to the FDG images to see if there is match or mismatch in areas of abnormalities. This is particularly important because despite the patient’s and our best efforts with the patient preparation, we may not be successful in suppressing the myocardium. In these situations we may FDG-uptake in normal myocardium. Resting perfusion imaging can help clarify this situation—if there is a matching perfusion defect, then this likely represents an area of inflammation. If there is no matching perfusion defect, this FDG avid area may represent an area of inflammation in cardiac sarcoidosis in the early stages of the disease when scar has not yet developed or it may represent FDG uptake by normal myocardium due to our failure to suppress the myocardium. Resting perfusion images also allow us to assess LV and RV function. Resting perfusion images in conjunction with FDG images can also allow us to monitor the patient’s response to treatment by demonstrating return to normal myocardium from active sarcoid granuloma after treatment or by showing the progression to development of scar.
  • When analyzing the cardiac FDG-PET cardiac sarcoidosis study, there are a few potential scenarios in addition to the scenario mentioned above (normal resting perfusion and FDG uptake) :
    • Normal resting perfusion and no FDG uptake
      • Normal Study
    • Resting perfusion defect and focal FDG uptake in the same area
      • Possible early stage of cardiac sarcoidosis with the perfusion defects either representative of: 1) compression of the microvasculature by inflammation or 2) scar related to fibrosis
      • This can also be seen in viability studies which suggest that there is an area of viable myocardium in a myocardial region with scar secondary to coronary artery disease. However, the patient preparation for a viability study with FDG PET and a cardiac sarcoidosis study with FDG PET are very different, so this possibility is less likely. Importantly, we should still be mindful of this possibility and the possibility of obstructive coronary artery disease.
    • Resting perfusion defect and no FDG uptake
      • Representative of scare without any active  inflammation. Notably, this does not rule out cardiac sarcoidosis. This pattern can be seen in later stages of the disease during which a larger burden of scar has developed compared to the amount of living cells, and thus, there happens to be no active inflammation (macrophage uptake of glucose results in areas of FDG uptake).
• FDG-PET has both diagnostic and prognostic uses in cardiac sarcoidosis. A prior meta-analysis by Youssef et al. demonstrated a sensitivity and specificity of FDG-PET in identifying cardiac sarcoidosis of 89% and 78% respectively (the lower specificity may have been impacted by the low sensitivity of the available diagnostic criteria which are used as gold standards for reference in these studies). An abnormal FDG-PET with myocardial perfusion defect and focal inflammation had a 4x increased risk of ventricular tachycardia or death irrespective of JMHW criteria and left ventricular ejection fraction. Among patients with active inflammation, those with RV involvement had the highest event rates.
• Whole-body imaging is particularly important when there is a high clinical suspicion of cardiac sarcoidosis. In these cases, if we see extra-cardiac involvement (i.e. Hilar lymphadenopathy that lights up or mesenteric lymphadenopathy that lights up) and we have evidence of conduction system abnormalities, then we know we likely have cardiac involvement even if we don’t see it on images. These extra-cardiac areas of FDG-uptake can also be areas of the body more amenable to biopsy than the heart. Finally, whole-body imaging can help with assessing the need and/or benefit of immunosuppressive therapy.

4.  What are the characteristic findings of cardiac sarcoidosis on cardiac MRI?

• In addition to morphological changes of the ventricle as seen with echocardiography (described above), the hallmark for detecting cardiac sarcoidosis with cardiac MRI is late gadolinium enhancement (LGE) in the mid-wall and subepicardial regions.
  • Gadolinium is an extracellular contrast agent that washes out slowly from areas of inflammation or scar (both processes with result in an expansion of the extracellular space). Because of this it is important to look at the distribution pattern of the LGE (i.e. a subendocardial enhancement in a coronary distribution is more suggestive of scar from a prior myocardial infarction in this area). Even in non-infarct distributions, however, LGE is nonspecific and can be seen in myocarditis or idiopathic cardiomyopathy.
    • Some features of LGE that favor the diagnosis of cardiac sarcoidosis include: 1) multifocal involvement and 2) involvement of the basal anteroseptum and inferoseptum with contiguous spread into the right ventricle.
  • Increased T2-weighted signal (a marker of increased water content which can be seen in areas of inflammation) can also suggest inflammation secondary to cardiac sarcoidosis.
  • A prior study by Smedema et al. described a cohort of patients with biopsy proven pulmonary sarcoidosis and evaluated the diagnostic accuracy of cardiac MRI for diagnosing cardiac sarcoidosis in the patients within this cohort who met Japanese Ministry of Health and Welfare (JMHW) diagnostic criteria. The study suggested that the sensitivity and specificity of cardiac MRI for diagnosing cardiac sarcoidosis is 100% and 78% respectively (the lower specificity, as for FDG-PET, may be attributed to the low sensitivity of the JMHW criteria).
  • Just as with FDG-PET, Cardiac MRI also has prognostic value in regards to patients with cardiac sarcoidosis. A study by Greulich et al. showed that in patients with systemic sarcoidosis, LGE had a hazard ratio for death or aborted sudden cardiac death of 31.6. 

5. How should we use cardiac MRI and cardiac FDG-PET in the evaluation of suspected cardiac sarcoidosis?

• In studies comparing cardiac FDG-PET and cardiac MRI in the diagnosis of cardiac sarcoidosis, there are certain situations where FDG-PET has been observed  to be a more sensitive test: 1) patients with new-onset AV block, 2) patients that are more likely to recover from AV block after steroid therapy, and 3) identification of inflammation in cardiac sarcoidosis (as compared to T2-weighted imaging with cardiac MRI). However, cardiac MRI may be more sensitive that FDG-PET for patients who are already on steroid therapy in the diagnosis of cardiac sarcoidosis. In general, FDG-PET and Cardiac MRI provide complimentary information which can help bolster clinical decision making for a disease that is difficult to diagnose, help risk stratify patients for immunosuppressive therapies and ICD placement purposes, and help monitoring for treatment response.
• Because of the high negative predictive value of cardiac MRI for diagnosis of cardiac sarcoidosis, a proposed strategy by Blankstein et al. for patients with suspected cardiac sarcoidosis it to first obtain cardiac MRI as long as there are no contraindications to the study. If this is negative (i.e. if there is absence of LGE), FDG PET can be considered if high clinical suspicion remains. If the cardiac MRI is positive or inconclusive, FDG PET should be obtained for complimentary information for diagnosis, risk stratification, and monitoring response to therapy (data is emerging on risk stratification for sudden cardiac death benefits of FDG-PET however as of the writing of the HRS recommendations in 2014, the writing group felt there was insufficient data to include a formal recommendation on this in their Expert Consensus Statement).
• Osborne et al. showed that a decrease in intensity (assessed by standardized uptake value max [SUV_max]) or extent (volume of inflammation above a prespecified SUV threshold—usually 2.7 g/mL) of inflammation on cardiac FDG-PET was associated with improvement in left ventricular ejection fraction. Thus, this can be used to monitor treatment efficacy and can help with the decision of timing of tapering of steroid therapy and with the decision of whether a different immunosuppressive regimen is needed (if an inadequate response is observed).
• It is also important to consider the expertise and familiarity with different imaging modalities available at the institutions when deciding between imaging modalities.

6. What diagnostic criteria do we have to establish a probable diagnosis of cardiac sarcoidosis?

• The Japanese Ministry of Health and Welfare (JMHW) criteria was used originally for research purposes and then incorporated into clinical practice. It was created in 1993 and modified in 2007 with the addition of some advanced imaging diagnostic criteria. Now, more commonly, the Heart Rhythm Society (HRS) criteria which was more recently introduced and based on more contemporary evidence is being used by most practitioners. Similar to the JMHW criteria, the HRS criteria includes two pathways for the diagnosis of cardiac sarcoidosis: 1) a histological diagnosis, and 2) a clinical diagnosis
  • For histological diagnosis, cardiac sarcoidosis “is diagnosed in the presence of non-caseating granuloma on histological examination of myocardial tissue with no alternative cause identified (including negative organismal stains if applicable).”
  • For clinical diagnosis, information from invasive and non-invasive studies are used. You need to have extracardiac histological diagnosis, no other explanation for the patient’s clinical presentation, and one or more of the following:
    • Cardiomyopathy or heart block that response to steroids or immunosuppressants
    • Unexplained reduction in the LVEF <40%
    • Unexplained high-degree AV block
    • Unexplained sustained ventricular tachycardia
    • Imaging findings suggestive of sarcoid (Patchy uptake of FDG on dedicated Cardiac PET in a pattern consistent with cardiac sarcoidosis or late gadolinium enhancement on Cardiac MRI in a pattern consistent with cardiac sarcoidosis)
    • Positive gallium uptake (in a pattern consistent with cardiac sarcoidosis)

7. Is there are any role for screening for cardiac sarcoidosis?

• The Heart Rhythm Society (HRS) has proposed an Expert Consensus recommendation for screening for cardiac sarcoidosis suggesting that all patients with biopsy-proven extracardiac sarcoidosis have an ECG done to screen for conduction system disease and asked about suggestive symptoms such as palpitations, presyncope, and syncope.  Echocardiography may be helpful (however, as discussed above, echocardiography is insensitive and a normal LVEF and normal wall motion does not exclude cardiac sarcoidosis). 
• In patients with biopsy-proven or clinically diagnosed extracardiac sarcoidosis who have signs or symptoms or possible cardiac involvement based on ECG, echocardiography, or symptoms further screening with cardiac MRI or FDG-PET may be useful.
• Screening with advanced cardiac imaging and high resolution CT chest may also be helpful in patients with no previous history of sarcoidosis but who have unexplained Mobitz II, third-degree AV block (sustained monomorphic ventricular tachycardia was also felt to be reasonable by a majority of the writing group for the HRS recommendations, however the vote did not reach a predefined threshold to become a formal recommendation).
  • If advanced imaging and high resolution CT chest are suggestive of cardiac sarcoidosis, extra-cardiac biopsy if feasible or guided endomyocardial biopsy can be pursued to confirm the diagnosis.
• Finally, there is noted overlap in the presentation of arrhythmogenic right ventricular cardiomyopathy and cardiac sarcoidosis. The literature is still emerging on the topic and thus, there are no formal guidelines for screening yet.

Guest Profiles

References and Links: Cardiac Sarcoidosis

1. Birnie DH, Sauer WH, Bogun F, et al. HRS expert consensus statement on the diagnosis and management of arrhythmias associated with cardiac sarcoidosis. Heart Rhythm. Jul 2014;11(7):1305-23. doi:10.1016/j.hrthm.2014.03.043

2. Blankstein R, Waller AH. Evaluation of Known or Suspected Cardiac Sarcoidosis. Circ Cardiovasc Imaging. Mar 2016;9(3):e000867. doi:10.1161/CIRCIMAGING.113.000867

3. Freeman AM, Curran-Everett D, Weinberger HD, et al. Predictors of cardiac sarcoidosis using commonly available cardiac studies. Am J Cardiol. Jul 2013;112(2):280-5. doi:10.1016/j.amjcard.2013.03.027

4. Greulich S, Deluigi CC, Gloekler S, et al. CMR imaging predicts death and other adverse events in suspected cardiac sarcoidosis. JACC Cardiovasc Imaging. Apr 2013;6(4):501-11. doi:10.1016/j.jcmg.2012.10.021

5. Iannuzzi MC, Fontana JR. Sarcoidosis: clinical presentation, immunopathogenesis, and therapeutics. JAMA. Jan 2011;305(4):391-9. doi:10.1001/jama.2011.10

6. Skali H, Schulman AR, Dorbala S. (18)F-FDG PET/CT for the assessment of myocardial sarcoidosis. Curr Cardiol Rep. May 2013;15(5):352. doi:10.1007/s11886-013-0352-8

7. Smedema JP, Snoep G, van Kroonenburgh MP, et al. Evaluation of the accuracy of gadolinium-enhanced cardiovascular magnetic resonance in the diagnosis of cardiac sarcoidosis. J Am Coll Cardiol. May 2005;45(10):1683-90. doi:10.1016/j.jacc.2005.01.047

8. Youssef G, Leung E, Mylonas I, et al. The use of 18F-FDG PET in the diagnosis of cardiac sarcoidosis: a systematic review and metaanalysis including the Ontario experience. J Nucl Med. Feb 2012;53(2):241-8. doi:10.2967/jnumed.111.090662