Introduction

The flow of blood to an organ is a fundamental physiological factor affecting tissue health, growth, and repair. Blood flow and volume are perturbed in many disease conditions, most notably in vascular disease and in tumors. The ability to determine non-invasively blood flow and blood volume using imaging methods therefore has important diagnostic and therapeutic implications. Since the early days of radiological imaging, scientists and physicians have been searching for methods that can accurately and non-invasively depict the major blood vessels of the body, and measure blood flow in tissue. For instance, X-ray projection imaging of blood vessels (angiography) was first demonstrated in 1927 by Moniz [1], using iodinated contrast agents injected intravascularly, while early measurements of tissue blood flow were based on the inhalation of freely diffusible tracers (e.g., nitrous oxide [N2O] [2], or radioactive xenon or krypton [3]). Subsequently, stable (i.e., non-radioactive) xenon was used in conjunction with X-ray computed tomography (CT) to image cerebral blood flow (CBF) [4], while other methods such as single-photon emission CT (SPECT) [5, 6] and positron emission tomography (PET) [7, 8] imaging using a variety of radiotracers also became available. More recently, dynamic CT perfusion imaging using bolus injection of iodinated contrast agents has been growing in popularity [9], particularly as fast multi-slice CT scanners have become widely available.

Blood flow is one of the most fundamental physiological parameters. Maintenance of adequate blood flow is vital for the health of biological tissue. The growth and function of many organ systems are linked tightly to their blood supply. In addition, many disease processes are associated with either increases or decreases in flow compared with normal values. The development and validation of non-invasive tools for the measurement of flow have been longstanding goals, both in biomedical research and in clinical practice.

Traditionally, the imaging of flow, or perfusion, has been accomplished using either nuclear medicine-based techniques involving radioactive isotopes, or X-ray computed tomography (CT) methods using radio-opaque contrast agents. However, soon after the introduction of magnetic resonance imaging (MRI) for anatomical imaging, research began on techniques for depicting flow. Since then, progress has been rapid, not least because MR methods have the advantage of not involving radiation, and in the case of arterial spin labeling-based techniques, are completely non-invasive. This makes them particularly appealing for use in a wide range of populations, including children and normal subjects. In addition, MR perfusion can be combined with the armamentarium of other structural, vascular, physiological, metabolic, and functional techniques available with MR to provide a comprehensive, “one-stop” examination for the patient.

Introduction

Nuclear magnetic resonance (NMR) spectroscopy was demonstrated for the first time in bulk matter in 1945 when Bloch and Purcell independently demonstrated that a strong magnetic field induced splitting of the nuclear spin energy levels, resulting in a detectable resonance phenomenon.[1,2] The method was originally of interest only to physicists for the measurement of the so-called gyromagnetic ratios (γ) of different nuclei, but applications of NMR to chemistry became apparent after the discovery of chemical shift and spin–spin coupling effects in 1950 and 1951 respectively.[3,4] High-resolution liquid state NMR spectra contain fine structure because the resonance frequency of each molecule is influenced by both neighboring nuclei (coupling) and the chemical environment (shift), which allows information on the structure of the molecule to be deduced. Hence, NMR spectroscopy rapidly became an important, and widely used, technique for chemical analysis and structure elucidation of chemical and biological molecules.

Major technical advances in the 1960s included the introduction of superconducting magnets (1965), which were very stable and allowed higher field strengths to be attained than with conventional electromagnets, and in 1966 the use of the Fourier transform (FT) for signal processing. In FT spectroscopy, the sample is subjected to periodic radiofrequency transmitter pulses followed by collection of the signal as a function of time (i.e., a time-domain signal), and the frequency-domain spectrum is calculated by FT. Use of FT NMR provides increased sensitivity compared with previous (so-called “continuous-wave”) techniques, and also led to the development of a huge variety of pulsed NMR methods, including multidimensional NMR techniques.

Introduction

As discussed in the previous chapter, a stroke is the rapidly developing loss of brain functions owing to a disturbance in the vessels supplying blood to the brain. This can be caused by ischemia (lack of blood supply) following thrombosis or embolism, or a hemorrhage. Acute stroke is a medical emergency, and imaging plays an important role in confirming (or otherwise) the clinical diagnosis of stroke, categorizing it as either ischemic or hemorrhagic, and identifying the underlying cause. Increasingly, imaging is also being used to guide therapeutic interventions and monitor their success. While traditionally X-ray computed tomography (CT) has been the imaging modality of choice (primarily because of its speed and widespread availability), MRI has been increasingly used because of its excellent contrast, high sensitivity and possibilities for multimodal acquisition (e.g., diffusion, perfusion, MR angiography, etc.). These topics are discussed in the following chapters.

Proton magnetic resonance spectroscopy (MRS) of the human brain was first demonstrated in the mid 1980s,[1–4] and shortly thereafter the first reports of its application to the study of human stroke appeared.[5,6] Although there have been reports of 31P,[7] 23Na,[8] and 13C [9] spectroscopy in human stroke, the majority of studies to date have utilized the proton nucleus, both because of its high sensitivity and the fact that proton MRS can be readily combined with conventional MRI without hardware modifications. Although the proton MRS studies performed in the early 1990s appeared to offer promise for diagnostic value in acute stroke, this modality has had relatively little impact for several reasons. The most important reason is the technical difficulty of performing it in a timely fashion in this patient population; secondly, much of the required clinical information is available from other (easier to perform) sequences, such as diffusion-weighted imaging (DWI), perfusion-weighted imaging (PWI), and T2-weighted MRI. Nevertheless, it is important to be aware of the spectroscopic correlates of acute and chronic infarction as, on occasion, MRS may be helpful in these clinical contexts.