Previous chapters have demonstrated the important role of myocardial perfusion imaging (MPI) for the diagnosis and risk stratification of coronary artery disease (CAD) in the general population. There is also robust literature validating the use of MPI in different patient cohorts. This chapter will describe the value of MPI for the assessment of CAD and risk stratification in special populations consisting of diabetic patients, women, patients with chronic renal disease, the elderly, and patients with acute chest pain syndromes.

In 2014, the global prevalence of diabetes was estimated to be 9% among adults aged 18+ years by the World Health Organization (WHO).¹ In the United States, the Centers for Disease Control (CDC) estimates those with diabetes is rising, with recently 29 million people (9.3% prevalence) suffering from diabetes.² This increase in prevalence of diabetes mirrors the obesity epidemic in the United States with 33% of the population now classified as being obese.³ The overall prevalence of CAD has been estimated to be as high as 55% in diabetic patients compared with 4% in the general population. In a landmark Finnish observational trial, the authors elegantly demonstrated the similar survival (15.4% vs. 15.9%) and rate of myocardial infarction (MI) (20.2% vs. 18.8%) among diabetic patients without prior MI and nondiabetic patients with prior MI, respectively, leading to the designation of diabetes as CAD equivalent by the Adult Treatment Panel-III (ATP-III) guidelines.^4,5 There are also data demonstrating that these patients have a higher incidence of more severe disease and are at greater risk of developing acute coronary syndromes.⁶ From these data it is clear that diabetic patients are at high risk for coronary events. Nuclear cardiology imaging offers potential for early and accurate identification of such patients.

The Role of Stress Myocardial Perfusion Imaging in Symptomatic Diabetic Patients

Stress MPI in symptomatic diabetic patients can be very beneficial in providing the appropriate diagnosis and risk stratification. Several studies demonstrate similar sensitivity (80–86%) and specificity (79–87%) of single-photon emission computed tomography (SPECT) MPI in diabetic patients as compared to nondiabetic patients.^7,8 In the largest study to date, Kang et al.⁹ evaluated retrospectively 138 diabetics and 188 nondiabetics with suspected CAD who had SPECT imaging and coronary angiography within 6 months. The diagnostic accuracy was similar between diabetic and nondiabetic patients (p = nonsignificant), respectively. The normalcy rate for low-likelihood patients was 89% in diabetics and 90% in nondiabetics (p = NS). These data indicate a strong diagnostic role for MPI in diabetic patients which is not different from a nondiabetic cohort.

Risk Stratification in Diabetic Patients with Stress Myocardial Perfusion Imaging

Several studies have reported the value of SPECT MPI for the risk stratification of diabetic patients. In a single-center prospective study of 1080 diabetic and 5130 nondiabetic patients, Kang et al.⁹ evaluated risk stratification according to defect size and extent (Fig. 19-1). As in studies in the general population, they found the greater the defect extent, the greater the risk for coronary events (cardiac death and nonfatal MI). Diabetic patients had more perfusion defects and higher event rates (nonfatal MI, cardiac death, and revascularization) during the follow-up period. The authors’ conclusion was that exercise and adenosine stress myocardial perfusion SPECT add incremental prognostic value for patients with diabetes.

Wiersma et al.^10,11 performed a subanalysis of the prematurely terminated MERIDIAN trial (medical therapy vs. invasive therapy in diabetic patients with mild angina and reversible perfusion defects) to examine the prognostic value of SPECT MPI in diabetics with stable angina and documented ischemia on MPI. In multivariate analysis, severe myocardial ischemia (summed difference score >8) (p = 0.001) and the use of insulin (p = 0.02) were independent predictors of cardiac death/nonfatal MI.

Data from a multicenter analysis also corroborate the above-mentioned findings. Using data from six centers, Giri et al.¹² examined whether the nuclear perfusion study could contribute to risk stratification beyond the presence of diabetes using the incremental chi-square approach (Fig. 19-2). Indeed, for both the prediction of cardiac death alone and as a combined end point with nonfatal MI, the perfusion imaging data were significantly better predictors of cardiac events than the presence of diabetes in combination with clinical risk factors.

Previous studies suggested that diabetic women have a greater risk of adverse outcomes at any level of stress perfusion defects.¹³ Recently, Santos et al.¹⁴ reported a large retrospective study of more than 4600 diabetic patients to examine whether the gender differences in cardiovascular event rates persisted in the modern era of aggressive treatment of CAD. This large study demonstrated that size and severity of stress perfusion abnormalities’ predicted outcomes equally in both genders. It is also appreciated that patients undergoing pharmacologic as opposed to exercise myocardial perfusion SPECT are at greater risk of cardiac events in diabetic patients.¹⁵ Other aspects of diabetics contributing to cardiac risk were evaluated by Barmpouletos et al.¹⁶ They demonstrated that diabetic patients with moderate to severe perfusion abnormalities are at greater risk for adverse cardiac events with long-standing diabetes (>10 years) and/or receiving insulin therapy.

Various studies have shown a higher rate of cardiac events following a normal MPI in diabetic patients.^9,12 A recent meta-analysis¹⁷ also confirmed these findings revealing an annualized event rate with normal MPI of 1.60%, higher than in the general population (<1%), leading the authors to define diabetic patients with negative stress MPI results as a “relatively low-risk” cohort but not truly at low risk. This study, however, did not account for stress type (pharmacologic or exercise) in either cohort.

Is this higher cardiac risk in diabetic patients inherent to the disease? This question was addressed by a large study (n > 3000) by Ghatak et al.¹⁸ to examine how the type of stress SPECT MPI (exercise vs. pharmacologic) impacts outcomes in diabetic patients.¹⁸ They demonstrated that diabetic patients who were able to undergo exercise stress SPECT MPI had significantly lower annualized cardiac event rates compared to diabetic patients undergoing pharmacologic stress SPECT MPI across all perfusion categories. In the normal perfusion category, the subgroup of diabetic patients able to exercise had a low cardiac event rate (<1%) similar to nondiabetic population (Fig. 19-3). This finding suggests that previous studies may have been reflective of a high percentage of pharmacologic stress tests in diabetic patents.

Ventricular function assessment also adds additional prognostic value in diabetic patients. Acampa et al.¹⁹ studied the role of gated SPECT in 520 diabetic and nondiabetic patients with baseline normal MPI and showed that the highest probability of cardiac events and the major risk acceleration was observed in diabetic patients with poststress LVEF <45%. In diabetic patients with poststress LVEF <45%, the time to achieve a risk level of events >3% was earlier at 12 months than nondiabetics (Figs. 19-4 and 19-5).

In summary, MPI provides considerable value in diagnosis and risk stratification of the diabetic patient. The literature suggests that diabetic patients are at greater risk for CAD-related events than nondiabetic patients. The duration of diabetes and insulin use (perhaps an indicator of severity of the disease) are also predictors of greater risk in the same categories of perfusion abnormalities. However, the Ghatak study demonstrates that diabetic patients with good exercise capacity have more favorable outcomes similar to nondiabetic patients. Thus, the form of stress a diabetic patient can perform needs to be considered in management decisions.

Role of Stress Myocardial Perfusion Imaging in Asymptomatic Diabetic Patients

In the last decade, several studies have emerged demonstrating the varying rates of cardiovascular events in asymptomatic diabetics, perhaps reflecting the heterogeneity of risk in this group of patients. The screening of truly asymptomatic patients using MPI remains controversial and is not recommended by professional guidelines. Contemporary evidence regarding this topic is reviewed below.

The Detection of Ischemia in Asymptomatic Diabetics (DIAD) study²⁰ was a randomized controlled trial that tested the hypothesis that systematic screening would identify higher-risk individuals and beneficially affect their risk of MI or cardiac death. The overall 5-year cardiac event rate was 2.9% (0.6% per year) and use of MPI screening had no discernable effect on subsequent cardiac events in asymptomatic diabetics.²¹ More recently, the BARDOT trial included 400 high-risk asymptomatic diabetic patients. The Patients included had evidence of end-organ damage or composite of age >55 years, diabetes duration >5 years, and at-least two cardiac risk factors of smoking, hypertension, hypercholesterolemia, positive family history of CAD in addition to diabetes. The rate of silent myocardial ischemia (SMI) in this study was 22% and “overt or silent CAD progression” (cardiac events + new ischemia/scar) was seven-fold higher in patients with an abnormal versus a normal MPS (35.6% vs. 4.6%), documenting the screening efficacy of MPS in these patients. Patients randomized to revascularization had similar rates of symptomatic CAD progression, but lower rates of asymptomatic CAD (more ischemia or new scar) progression (54.3% vs. 15.8%; p < 0.001).²² These data in a higher-risk population than the DIAD trial suggest some benefits to screening but this has not been adopted by professional organizations.

A new novel method to identify high-risk asymptomatic diabetics is to estimate coronary microvascular dysfunction (CMD) using coronary flow reserve (CFR) with positron emission tomography (PET). Murthy et al.²³ examined patients with and without diabetes using ⁸²Rb-PET MPI and found that addition of CFR to clinical and imaging risk models improved risk discrimination for both diabetics and nondiabetics. Diabetic patients without known CAD with impaired CFR experienced a rate of cardiac death comparable to that for nondiabetic patients with known CAD (2.8% per year vs. 2.0% per year; p = 0.33). Conversely, diabetics without known CAD and preserved CFR had very low annualized cardiac mortality, which was similar to patients without known CAD or diabetes mellitus and normal stress perfusion and systolic function (0.3% per year vs. 0.5% per year; p = 0.65) (Fig. 19-6).

These data suggest such a test might be useful in asymptomatic diabetic patients but further prospective studies are necessary.

Conclusions: Diabetic Patients

1. 
   

   MPI is an effective tool in the diagnosis and risk stratification of diabetic patients.

   
   
2. 
   

   Diabetic patients with longstanding and more severe disease are at greater risk for cardiac events.

   
   
3. 
   

   The ability to perform exercise in diabetic patients conveys a lower risk of cardiac events.

   
   
4. 
   

   Routine stress MPI is not recommended for asymptomatic diabetic patients.

   
   
5. 
   

   The role of cardiac PET with myocardial blood flow (MBF) assessment needs further assessment, but can identify an “at-risk” population.

While cardiovascular diseases (CVDs), including CAD and stroke, are the leading cause of death in both men and women in the United States, more women die from CVD than men.²⁴ In 2013, CVDs caused about one death every 80 seconds in women and were responsible for more number of deaths from cancer, chronic lower respiratory disease, and diabetes combined. Younger women (aged 35–45 years) for the first time in the last four decades had higher CAD death rates likely representing the influence of the obesity epidemic in younger women.^25,26 This section will review the data on women and heart disease and the value of MPI in assessment and prognosis.

Diagnostic Testing in Women

The diagnosis of CAD in women is challenging due to the atypical symptoms often experienced. This leads to less referral for evaluation in women with suspected CAD.^27,28 A recent meta-analysis revealed that women with MI had a lower likelihood of presenting with typical chest pain than men (odds ratio 0.63; 95% CI, 0.59–0.68) and were more likely than men to present with atypical symptoms.²⁹ Similar findings were also observed in a prospective cohort study which showed that despite chest pain being the most common presentation of acute coronary syndrome in both sexes, women presented more frequently without chest pain than men (19.0% vs. 13.7%; p = 0.03).³⁰

Adding to the challenge of diagnosing heart disease in women is the high prevalence of nonobstructive disease (40–60%). This finding was confirmed in a more recent analysis by Jespersen et al.,³¹ which showed a higher incidence of nonobstructive CAD in women (65% vs. 32%) as compared to men. In this study normal coronary arteries and nonobstructive CAD were associated with a 52% and 85% increased risk of MACE (cardiovascular mortality, hospitalization for MI, heart failure, or stroke) and with 29% and 52% increased risk of all-cause mortality, respectively. Similar findings were demonstrated in the 10-year follow-up of the Women’s Ischemia Syndrome Evaluation (WISE) study, which showed that two-thirds (62%) of women with angina had nonobstructive disease. The 10-year adverse outcome rates (cardiovascular death or MI rate) in the women with nonobstructive CAD were almost double (12.8% vs. 6.7%) than that observed in women with angiographically normal coronary arteries.³² These differences in the presentation and prognosis of CAD in women could be related to the increased incidence of CMD and different characteristics of plaque morphology in women presenting with acute coronary syndrome.³³ However, recent studies using optical coherence tomography (OCT) did not reveal a sex difference in the prevalence of plaque rupture or erosion.^34,35

Despite the above-mentioned challenges of diagnosing CAD in women, ETT still remains the most common screening test used to detect CAD. However, Table 19-1 summarizes the low sensitivity and specificity of this modality in women.^36–41

To address the concerns regarding the accuracy of ETT alone in women, the WOMEN (What is the Optimal Method for Ischemia Evaluation in Women) trial was undertaken to examine outcomes rather than diagnoses. In this trial, 824 symptomatic women with intermediate pretest probability of CAD were randomized to one of two diagnostic strategies: exercise tolerance testing (ETT) alone or exercise MPI. At 2 years, there was no difference in major adverse cardiac events (98.0% for ETT and 97.7% for MPI; p = 0.59) (Fig. 19-7).

The clinical implications of this trial’s results are noteworthy. For low- to intermediate-risk women capable of performing exercise, routine ETT without imaging appears a reasonable first-line test.⁴² Women with intermediate–high CAD risk may be referred for stress imaging because of a higher likelihood of CAD. Women at high CAD risk with stable symptoms may be referred for a stress imaging modality for assessment of their ischemic burden and to guide posttest and anti-ischemic therapeutic decision making.⁴³

Radionuclide Stress Imaging in Women

Robust data are available for the use of MPI for diagnosis as well as risk stratification of women with known or suspected CAD. The strength of MPI in the diagnostic accuracy of stress testing in women was demonstrated in a contemporary meta-analysis of 14 SPECT studies, which found a sensitivity of 81% and specificity of 78% in women with no known CAD.⁴⁴

The higher specificity of SPECT MPI than ETT alone is particularly useful in identifying the presence of ischemia in women with false-positive exercise tests. In a recent report of >4000 patients of which more than one-third were women, exercise MPI showed a net reclassification improvement of 36% over ETT. Thus, additional ischemic heart disease (IHD) risk information was available by the performance of MPI in one out of every three patients.⁴⁵ While radionuclide imaging has a higher sensitivity than exercise stress testing in the detection of single-vessel disease, the highest accuracy has been found in women with multiple-vessel disease compared to those with single-vessel disease.^46,47

It has also been questioned whether stress MPI is less accurate in women compared to men, similar to ETT. To address this issue, Iskandar et al.⁴⁸ performed a bivariate meta-analysis on 26 studies that met criteria. In contrast to ETT, SPECT imaging provided similar and high sensitivity and specificity for both genders, reassuring that SPECT is a reasonable diagnostic modality in women.

Pharmacologic stress is an important alternative for women with limited exercise capacity. Both dipyridamole and adenosine stress have been found to be comparable to exercise imaging in primarily male populations.⁴⁹ In a prospective study of 201 women, adenosine SPECT imaging had a 95% sensitivity, 66% specificity, and 85% accuracy in the detection of coronary stenosis greater than 70% regardless of presenting symptoms, prior history of MI, or pretest probability of coronary disease.⁵⁰

Data are also now available for use of SPECT MPI in women of different ethnicity.⁵¹ Cerci et al. studied 2225 Hispanic women in Brazil and showed that women with abnormal SPECT had three times higher event rate (13.1% vs. 4%) as compared to those with normal SPECT studies. Moreover, in the subgroup of patients with abnormal SPECT studies, further risk stratification could be performed using extent of perfusion abnormality and presence or absence of reversible ischemia.