Metabolic remodeling in heart failure

Arteriovenous balance studies

Arteriovenous (A-V) balance studies involve the technique that was pioneered by Bing and colleagues. In this technique, catheters are placed in the coronary arteries and in the coronary sinus. These catheters are then used to directly sample the blood in these locations for measurement of plasma substrates, such as free fatty acids and glucose. The theory is that by measuring how much of each substrate the heart is presented with through the coronary arteries and by subtracting how much of each substrate passes through to the coronary sinus, one can calculate how much of the substrate is taken up and metabolized by the heart. Surprisingly, an early study of myocardial metabolism in patients with HF using this technique did not show much of a perturbation in substrate use. Specifically, a study by Blain et al. in 1956 showed that the failing heart under fasting conditions derived about 58% of its energy requirements from fatty acids, 30% from carbohydrates, 7% from ketone bodies, and 5% from amino acids, results that are similar to the normal human heart. There are several possible explanations for this lack of a difference. Early A-V balance studies did not use concomitant nonradiolabeled tracer infusions (e.g., C-13 palmitate). Using these tracer substrates allows for the distinction between exogenous substrate use and endogenous substrate production. Another possible explanation is that this early study of myocardial metabolism in HF did not include patients with concomitant obesity or diabetes, which would affect plasma substrate levels and hence myocardial metabolism. Indeed, work by Danforth et al. suggested that different etiologies of HF may affect myocardial metabolism differently, an important concept that will be explained later in this chapter. An advantage of A-V balance studies is that direct measurements are made in vivo . A significant disadvantage of this technique is its invasive nature because it requires arterial access and cannulation of the heart’s vessels. Although the absolute risk for a significant complication of A-V balance studies is small, it exists. A-V balance studies are also impractical for longitudinal follow-up in response to pharmacologic or other forms (e.g., bariatric surgery) of interventions.

Magnetic resonance techniques

Magnetic resonance (MR) techniques are predominantly used in the research setting rather than in clinical practice. There are two predominant MR-based techniques: ¹ H-MR spectroscopy and the new MR with hyperpolarization and dynamic nuclear polarization (DNP). ¹ H-MR specroscopy is used to quantify myocardial fatty acid/triglyceride deposits in the myocardium rather than flux through fatty acid metabolic pathways. In this technique, a region of interest (ROI) is placed on the interventricular septum, so as to avoid contamination of the signal from pericardial/epicardial fat. Because the human heart is made mostly of water, the water peak on the spectra is very large compared with the spectra from fat. Nevertheless, despite the relatively small size of the triglyceride peaks, there is good agreement between ¹ H-MR spectroscopy and direct biochemical measurement of lipids in the myocardium. Thus ¹ H-MR spectroscopy can be used to evaluate the effect of risk factors for HF, such as obesity and type 2 diabetes, which are associated with steatosis. This technique may also be used to evaluate the effect of different therapies on cardiac steatosis.

MR with hyperpolarization and DNP is a novel technique for measurement of visualization of biochemical pathways such as the Krebs cycle, when ¹³ C-labeled tracers are used. Hyperpolarization helps improve the signal-to-noise ratio. DNP requires very low temperatures and a high Tesla magnetic field. DNP aligns most nuclear spins, which increases the MR signal. A full review of this new technique is beyond the scope of this chapter but is discussed in detail by Schroeder et al. It should be noted that this technique has only recently been shown to be feasible for evaluating human myocardial metabolism. An advantage of these noninvasive MR-based techniques is the lack of exposure to ionizing radiation. Of course, there are potential downsides with MR techniques, including the high cost, the necessity of a large MR scanner, and the inability to scan patients with certain implanted electronic devices, such as non-MR-compatible implanted cardiac defibrillators (ICDs), which are common in patients with HF with reduced ejection fraction (HFrEF).

Radionuclide techniques

Radionuclide techniques are noninvasive. Unlike MR imaging (MRI), they are not contraindicated in patients with implanted electronic devices such as ICDs and are used for measurement of substrate flux (as opposed to deposition).

Single photon emission computed tomography (SPECT) imaging can be used to evaluate myocardial glucose and fatty acid metabolism. There are several advantages to SPECT imaging, including the general availability of the technology, the simultaneous measurement of cardiac function, and the long half-life of SPECT radiotracers. This long half-life is especially important because it allows SPECT radiotracers to be delivered from an off-site central radiopharmacy rather than requiring that each institution have an on-site cyclotron or generator for radiotracer production. The primary disadvantage of SPECT imaging of myocardial metabolism is that it is only a visual and/or semiquantitative analysis of relative radiotracer uptake in the myocardium. Further limitations of SPECT imaging technology may be seen in its imperfect photon attenuation correction and relatively poor spatial and temporal resolution. That said, SPECT can be used to trace myocardial metabolism.

Table 19.1 shows the primary radiopharmaceuticals used in combination with radionuclide imaging to trace myocardial metabolism. ¹²³ I-β-methyl-P-iodophenylpentadecanoic acid (BMIPP), which is a branched-chain analog of 15-( p -iodophenyl)-pentadecanoic acid (IPPA), is often used in combination with SPECT to estimate fatty acid flux in the myocardium. BMIPP leads to better-quality images than IPPA because the turnover of the latter is so fast. In contrast, the branched chain of BMIPP prevents its β-oxidation and thus increases shunting of the radiotracer into the myocardial triglyceride pools, increasing retention of BMIPP in the heart. This enhances the SPECT images of fatty acid storage (but does not allow for measurement of fatty acid oxidation kinetics). Although glucose metabolism kinetics are not typically measured using SPECT, if used in combination with a specific collimator or detection scheme, the radiolabeled tracer ¹⁸ fluorodeoxyglucose (FDG) can be used to measure glucose uptake. Positron emission tomography (PET), however, is used more often than SPECT for measurement of myocardial glucose metabolism.

PET has several advantages over SPECT for myocardial metabolism measurement. First, PET can be used to quantify myocardial metabolism processes. Another significant advantage of PET over SPECT is that positron-emitting radionuclides of key elements such as carbon, nitrogen, and oxygen can be used to create tracers that nearly exactly mimic the kinetics of the nonradiolabeled substrates of interest (see Table 19.1 and Fig. 19.1 ). For example, ¹¹ C-palmitate is much closer in structure to palmitate that is not radiolabeled and so ¹¹ C-palmitate can be expected to more closely trace all aspects of fatty acid metabolism than a tracer such as BMIPP, which has a branched side chain that is not a part of the endogenous, unlabeled palmitate. ¹¹ C can be also be used to label acetate, glucose, and lactate to trace the uptake and metabolism of these substrates. ¹⁵ O can be used to label acetate to measure total oxygen consumption (MVO ₂ ), and ¹⁸ F can also be used in place of a hydroxyl group of a substrate/substrate analog of interest without a major alteration of the structure. ¹⁸ F is the positron-emitting radionuclide in FDG and in several tracers of fatty acid metabolism (see Table 19.1 ). The main disadvantages of PET are its expense and the complexity of kinetic modeling for quantification of substrate metabolism and of radiotracer generation. Several of the radiotracers (e.g., ¹¹ C-acetate) require an on-site cyclotron for generation, thus limiting their widespread use/distribution.

Myocardial oxygen consumption

As previously mentioned, MVO ₂ can be quantified in vivo using ¹⁵ O-oxygen or ¹¹ C-acetate. The former has a very short half-life, which lends itself to multiple measurements over a short time period; however, imaging with ¹⁵ O-oxygen requires an additional tracer for myocardial blood flow measurement for the quantification of MVO ₂ . ¹¹ C-acetate, on the other hand, has around a 20-minute half-life, like all ¹¹ C-acetate compounds. ¹¹ C-acetate is primarily converted to ¹¹ C-acetyl-coA and is metabolized via Krebs cycle. Because the Krebs cycle and oxidative phosphorylation in the mitochondria are closely linked, ¹¹ C-acetate kinetics reflect overall MVO ₂ . The quantification of MVO ₂ using ¹¹ C-acetate necessitates compartmental modeling or exponential curve fitting. Especially relevant to HF, when there is low cardiac output compartmental modeling, correction for ¹¹ CO ₂ in the blood is preferred. This is due to spillover of ¹¹ C activity from the lungs and due to dispersion of the input function.

Carbohydrate metabolism

PET is also used to quantify glucose and lactate metabolism pathways. Measurement of carbohydrate metabolism is especially relevant to HF for several reasons. First, in nonischemic cardiomyopathy and HFrEF, there is a general metabolic shift toward relatively more glucose use. Second, full oxidation of glucose is more oxygen efficient than fatty acid oxidation, which is a theoretical advantage in ischemic cardiomyopathy. Third, identification of viable, glucose-metabolizing myocardium using PET in ischemic cardiomyopathy can alter patient management by helping select those who may benefit from revascularization. The two most commonly used radiotracers of carbohydrate metabolism are ¹⁸ FDG and ¹¹ C-glucose. ¹⁸ FDG is commonly available, generator-made, and has been used as a radiopharmaceutical for more than 35 years. As discussed in Chapter 20 , ¹⁸ FDG is extremely clinically useful for detecting myocardial viability and predicting improved left ventricular (LV) function after revascularization, which is especially relevant to disease management of ischemic cardiomyopathy. ¹⁸ FDG is taken up by cardiomyocytes by facilitated transport. The radiotracer is then phosphorylated by hexokinase and trapped in the cell. Thus, ¹⁸ FDG cannot be used to study the intramyocellular fate of glucose (e.g., glucose oxidation or glycogen formation), but ¹⁸ FDG can be used to assess myocardial glucose uptake and overall flux through glycolysis. To quantify myocardial glucose uptake using ¹⁸ FDG, a correction factor, called the lumped constant , must be used to correct for differences between endogenous glucose uptake and the tracer ¹⁸ FDG. Unfortunately, the value of the lumped constant may change depending on the substrate and hormonal conditions, which can affect the accuracy of glucose uptake using ¹⁸ FDG.

Tracing myocardial glucose metabolism with ¹¹ C-glucose and PET has some advantages over the more commonly used ¹⁸ FDG. First, there is no need for a lumped constant because ¹¹ C-glucose closely mimics nonradiolabeled endogenous glucose. Another major advantage is that different intramyocellular fates of glucose (e.g., glucose oxidation, lactate production, glycogen deposition) may be determined using ¹¹ C-glucose tracer kinetics combined with kinetic compartmental modeling. Evaluating the intramyocellular metabolic fate of glucose is especially relevant to myocardial energetics in HF because full glucose oxidation yields around 36 ATP equivalents but glycogen deposition costs around 2 ATP. A significant disadvantage of using ¹¹ C-glucose is its higher cost compared with ¹⁸ FDG. The former is also more complicated to synthesize and requires a cyclotron. Moreover, the input function of the kinetic modeling using ¹¹ C-glucose requires correction for ¹¹ CO ₂ and ¹¹ C-lactate levels in the input function.

As previously mentioned, ¹¹ C-glucose may be used to measure lactate production, but lactate use by the heart may be measured using the novel PET tracer L-3- ¹¹ C-lactate. The heart is a particularly avid user of lactate when it is available, which is typically during heavy workloads or ischemia. The use of this tracer of lactate metabolism, like ¹¹ C-glucose, requires complex kinetic modeling in conjunction with the PET kinetics. Although lactate metabolism is particularly pertinent to ischemia and hence ischemic cardiomyopathy, this tracer has not been used yet in this condition.