Alzheimer’s disease (AD) is the most common cause of dementia in patients over 65 years of age. It causes approximately 50–60 % of all dementias, followed by dementia with Lewy bodies (DLB) and frontotemporal dementia (FTD) [14]. Approximately 27 million individuals are diagnosed with AD worldwide, a number that is estimated to quadruple by 2050, meaning 1 in 85 people will be affected [15, 16].

Alzheimer’s disease was first described by Alois Alzheimer in 1906 as an unusual disease of the cerebral cortex in a 51 year old woman named Auguste Deter. Her symptomatology included presenile dementia with memory loss, disorientation, hallucinations, and ultimately death by the age of 55. The autopsy showed classic neuropathological changes as senile plaques and neurofibrillary tangles. He also described a granulovascular degeneration and amyloid angiopathy. The diagnosis of Alzheimer’s disease has traditionally been through the NINCDS/ADRDA criteria [17]. In these clinical criteria, dementia is established by clinical examination and documented by the mini-mental test or Blessed Dementia Scale and confirmed by a neuropsychological examination. Alzheimer’s dementia requires cognitive deficits in two or more areas, with progressive worsening of memory and other cognitive function. There should be no disturbances in consciousness. The age of onset is typically between the ages of 40 and 90 years, most often after 65 years. In addition, the absence of systemic disorders or other brain diseases are required as these can confound the diagnosis of Alzheimer’s disease.

The two basic types of Alzheimer’s disease are familial and sporadic. Familial AD (FAD) is a rare form of AD, affecting less than 10 % of AD patients. All FAD is early onset, meaning the disease develops before age 65. Apolipoprotein E (APOE) epsilon4 gene dose (i.e., the number of epsilon4 alleles in a person’s APOE genotype) is associated with a higher risk of AD and a younger age at dementia onset [18], and correlates with reduced regional hypometabolism in brains of patients with AD. In addition, advanced age, prior head trauma, low educational levels, and gender, with female greater than male predominance, have been associated with an increased risk for Alzheimer’s disease.

Previous guidelines for the detection of AD included meeting the Diagnostic and Statistical Manual of Mental Disorders (fourth edition) criteria for dementia, which required an episodic memory disorder and impairment in >1 cognitive domain that interfered with daily life activity or social function. Beyond that, a diagnosis of “probable AD” was essentially a diagnosis of exclusion [16, 19, 20]. The Diagnostic and Statistical Manual of Mental Disorders (fifth edition), released in May 2013, replaces the term dementia with major neurocognitive disorder and mild neurocognitive disorder. For a diagnosis of AD, it requires memory problems and impairment in at least one other cognitive domain interfering with functional independence. For those with Mild Neurocognitive Disorder, AD diagnosis can only be made if the subject additionally tests positive for a mutation in an autosomal dominant AD gene or certain biomarkers. However, for the first time in 27 years, the National Institute on Aging and the Alzheimer’s Association established new guidelines for the diagnosis and treatment of dementia. The new guidelines examine the biological changes underlying symptoms of dementia, whereas previous diagnosis relied primarily on neuropsychological evaluation, clinical assessment, and evaluation of family history or reports. Albert et al. [21] described three distinct phases for AD: presymptomatic, mildly symptomatic but predementia, and dementia caused by AD. Furthermore, in conjunction with clinical evaluation, three biomarkers—cerebrospinal fluid (CSF), magnetic resonance imaging (MRI) volume, and PET—were identified to be useful for diagnosis. The presymptomatic stage marks a preclinical form of AD when biomarker changes indicate the presence of an early stage of a dementia process, but memory loss or other behavioral symptoms are not yet noticeable. At this stage, individuals could potentially benefit from vaccination approaches or other preventive strategies. The mildly symptomatic phase is marked by noticeable cognitive changes that do not necessarily interfere with daily life. The last stage is distinguished by definite memory, cognitive, and behavioral deficits that disrupt daily tasks.

CSF is a clear fluid similar to blood plasma. The intracranial and spinal cord structures float in CSF and are protected from jolts and blows. Principally, the choroid plexus in the lateral, third, and fourth ventricles produces the major portion of CSF. Normally, between 125 and 150 mL of CSF is circulating within the ventricles and subarachnoid space at any given time. Approximately 600 mL of CSF is produced daily. The CSF normally drains from the lateral ventricles sequentially through the interventricular foramen of Monro, the third ventricle, and the cerebral aqueduct of Sylvius into the fourth ventricle and then leaves the ventricular system through the median foramen of Magendie and two lateral foramina of Luschka. Here, the CSF enters the subarachnoid space. Along the base of the brain, this space extends into a number of lakes called cisterns. The CSF is absorbed through the pacchionian granulations of the pia-arachnoid villi into the superior sagittal sinus.

The term hydrocephalus generally refers to those conditions that produce an imbalance between the rate of production and absorption of the cerebrospinal fluid, leading to dilatation of the ventricular system. Hydrocephalus normally occurs as a result of obstruction to the flow and absorption of CSF, although there are rare cases of choroid plexus papillomas causing hydrocephalus by the overproduction of CSF.

Hydrocephalus is traditionally classified as communicating and noncommunicating, based on whether ventricular obstruction is present. In the former, the ventricular system continues to communicate with the subarachnoid spaces outside the brain through the fourth ventricular foramina of Luschka and Magendie.

Noncommunicating hydrocephalus correspondingly refers to the presence of occlusion within the ventricular system. Hydrocephalus may be either congenital or acquired. Arnold–Chiari malformation, Dandy–Walker malformations, and aqueductal stenosis/atresia are common causes of the congenital variety. In the acquired type, many pathologic conditions, including inflammatory, infectious, traumatic, and neoplastic disorders, can cause hydrocephalus [27].

Noncommunicating hydrocephalus can be the result of intraventricular mass, aqueductal obstruction, or fourth ventricular obstruction. Communicating hydrocephalus, on the other hand, results from meningitis, meningeal carcinomatosis, or cerebral dural sinus thrombosis, or it is idiopathic in elderly patients. Normal-pressure hydrocephalus (NPH) is a communicating hydrocephalus of particular interest to nuclear medicine professionals since radionuclide cisternography is useful in its diagnosis and management. In NPH, the usual flow of CSF is impaired somewhere in the intracranial subarachnoid space, resulting in a reversal of CSF flow and dilatation of the lateral ventricles. There is free communication between the ventricular system and the subarachnoid pathways and no elevation of CSF pressure. Clinically, the entity presents as dementia, gait disturbances, and fecal and urinary incontinence. Most commonly, this condition results from subarachnoid hemorrhage or meningoencephalitis.

Leaking of CSF may be etiologically classified into:

To understand the uptake mechanism of 99mTc-hexamethylpropyleneamine oxime (99mTc-HMPAO), a three-compartmental analysis model can be used for analysis [47]. In this model, the first compartment is the lipophilic tracer in the blood pool of the brain, but outside of the blood–brain barrier. The second compartment is consisting of the lipophilic tracer inside of the blood–brain barrier. The third compartment is the hydrophilic form of the tracer that is retained in the brain. Transport from the first compartment to the second compartment represents efflux of lipophilic tracer from the blood compartment to the brain compartment. Back-exchange from the third compartment to the second compartment represents back-diffusion of the lipophilic form of the tracer and is essentially equal to zero since the tracer is irreversibly trapped (by intracellular reaction with glutathione) in the brain. Figure 18.6 shows a normal brain SPECT scan after injection of 20 mCi (740 MBq) IV of 99mTc-HMPAO and acquired on a triple-head Picker Prism (Picker International, Cleveland, OH). There is noted to be excellent uptake of this tracer in the gray matter of the brain, and there is clear distinction of small brain structures.

The second tracer commonly used in brain SPECT to measure regional cerebral perfusion is 99mTc-ECD [48]. This radiopharmaceutical is lipophilic, similar to 99mTc-HMPAO, and rapidly traverses the endothelium and capillary membranes into the brain cells [49]. However, in the third compartment irreversible trapping mechanism of this tracer differs from 99mTc-HMPAO, since 99mTc-ECD is enzymatically metabolized to a polar complex, which is trapped in the brain. This tracer has been reported to demonstrate less nonspecific scalp and facial tissue background activity compared with 99mTc-HMPAO. Figure 18.7 shows a normal 99mTc-ECD scan brain SPECT scan after injection of 20 mCi (740 MBq) IV. However, it has been reported that there are differences in regional uptake of these tracers, predominantly in the thalamus and the cerebellum.

This difference in distribution is illustrated by scans from a 42-year-old female normal subject who received both tracers separated by a 48-h time period (Fig. 18.8).

A recent advancement in brain SPECT imaging has been the development of special software primarily used on the Picker Prism triple-head camera system which allows dynamic scan acquisition (10 s per scan for 7 min). This acquisition results in a total of 42 scans of temporally separated individual tomographic image data sets. This allows calculation of tomographically displayed absolute quantification of regional cerebral perfusion (rCBF) in milliliters per 100 g of tissue per minute [50]. This is possible since the clearance of ¹³³Xe is linearly proportional to the rCBF, and unlike tracers such as 99mTc-HMPAO or 99mTc-ECD, ¹³³Xe does not underestimate rCBF due to the limitations on extraction fractions at high cerebral blood flow rates. In our experience, we have found that the 133Xe clearance technique is more sensitive to changes in blood flow during Diamox augmentation of rCBF in the evaluation of hemodynamically significant vascular stenosis and in cases of cortical blood flow changes in brain activation studies. One of the major limitations of 133Xe SPECT is the relatively low energy of the emission photon resulting in a significant attenuation and loss of spatial resolution of the central structures of the brain. In addition, due to the rapid SPECT acquisition necessary to obtain accurate clearance on a pixel-by-pixel basis, the count rate is low, requiring use of a 64 × 64 matrix, resulting in reduction in spatial resolution throughout the scan.

Dynamic ¹³³Xe SPECT is excellent in assessing large territorial vascular abnormalities due to its ease of quantitation and its ability to detect large major vessel territorial reductions in regional cerebral perfusion. Its utility is exemplified in a study designed to measure cerebrovascular perfusion reserve with rest/stress SPECT brain scans in patients with cerebrovascular disease (CVD) and suffering from TIA undergoing evaluation for extracranial/intracranial (EC/IC) arterial anastomosis or superficial temporal artery (STA)/middle cerebral artery (STA/MCA) bypass to assess and to specifically identify the presence or the absence of a vascular reserve constraint, which has been previously documented to be of value using increased oxygen extraction fraction PET [51]. Figure 18.9 illustrates the imaging results of a 62-year-old female with TIA. The angiogram showed the presence of a 99 % left ICA stenosis. The ¹³³Xe SPECT study confirms the presence of hemodynamic vascular reserve constraint due to a high-grade stenosis of the left internal carotid artery.

In a study on nine CVD patients, 99mTc-HMPAO SPECT was found to be a more specific indicator of hemodynamic constraint since it was asymmetric only in cases of severe ischemia [52]. 133Xe SPECT detected asymmetries, even in the mild ischemic group, and therefore is a more sensitive detector of vascular disease. Because ¹³³Xe SPECT measures absolute rCBF (ml/100 g/min), it has one major advantage (i.e., to measure absolute perfusion) and thus can establish parameters of significant ischemia independent from semiquantitative asymmetry values.

Dynamic 15O-water PET scans with arterial sampling provide the ability for the quantitative assessment of regional cerebral perfusion. Each emission scan is short, approximately 3 min in duration. Data are typically analyzed using a 1-tissue-compartment model to obtain K1 (mL/min/mL), and quantification is expressed in standard units of ml/100 g/min [53].

The PET scan shown in Fig. 18.10a was acquired on a Siemens/CTI ECAT HR + scanner in 3D imaging mode (63 parallel planes); axial field-of-view, 15.2 cm; in-plane resolution, 4.1-mm full width at half maximum; and slice width, 2.0 mm. The scanner gantry is equipped with a Neuro-insert (CTI PET Systems, Knoxville, TN) to reduce the contribution of scattered photon events. PET data was reconstructed using filtered back-projection (Fourier rebinning and 2D back-projection with Hann filter: kernel FWHM = 3 mm). Data was corrected for photon attenuation, scatter, and radioactive decay. A windowed transmission scan (10–15 min) was obtained for attenuation correction using rotating ⁶⁸Ge/⁶⁸Ga rods, and a model-based correction was applied to account for the 3D scatter fraction. The final reconstructed PET image resolution was about 6 mm (transverse and axial planes). PET–MRI fusion images shown in Fig. 18.10b are obtained from the MRI data that were transferred to the PET facility and co-registered to the dynamic 15O-water PET scans by automated image registration (AIR) software [54].

²⁰¹Tl in the form of thallous chloride is a cyclotron produced radiopharmaceutical shown to have affinity for brain tumors as early as the 1970s [55]. Although more commonly used as a myocardial perfusion imaging agent, thallium has high sensitivity for detection of new, recurrent, or residual viable tumor, which is difficult to differentiate from postradiation necrosis and edema on CT or MRI.

Thallium decays by electron capture with a half-life of 73 h and emits photons with a range of 0.78–167.4 keV [56]. The useful energy for imaging is at 80 keV corresponding to mercury x-rays when thallium decays to stable Hg-201. In its intravenous form, ²⁰¹Tl is supplied in isotonic solution at pH 4.5–7.0 and contains NaCl for isotonicity and 0.9 % benzyl alcohol as a bactericidal agent [57]. The uptake of thallium in normal tissues has been hypothesized to act as a potassium analog. Both elements belong to group IIIA of the periodic table. The distribution and retention of ²⁰¹Tl in the normal brain and tumors is an active process related to blood flow, loss of integrity of the blood–brain barrier, tumor cell viability, tumor type, tumor cell membrane function, and the Na⁺–K⁺ ATPase pump activity.

Normal brain tissues show minimal to no uptake of ²⁰¹Tl. The normal physiological distribution in the head and neck region includes the scalp, lacrimal gland, nasopharyngeal area, salivary gland, and the pituitary gland. There is also minimal thallium uptake in the choroid plexus. Thallium is normally taken up by regions of the brain which do not have a blood–brain barrier (BBB) such as the pituitary gland and pineal gland and minimally taken up by the choroid plexus (Fig. 18.11). The evaluation of viable tumor can be performed with great accuracy using brain SPECT or brain PET imaging. Brain SPECT imaging employs the tracer ²⁰¹Tl to detect new, residual, or recurrent viable tumor due to the fact that there is transport of ²⁰¹Tl across the breakdown in the blood–brain barrier and uptake of ²⁰¹Tl into regions of hypermetabolism. ²⁰¹Tl is postulated to represent “potassium analog” with affinity for the sodium–potassium ATPase enzyme [58]. Thallium accumulates in the residual or recurrent viable tumor cells in proportion to malignant grade and total viable tumor bulk.

Tc-99m Hexakis-2-methoxy-2-isobutyl isonitrile is a monovalent cation complex formed by a central technetium atom surrounded by six 2-methoxy-2-isobutyl isonitrile groups. This compound is also used extensively in myocardial perfusion imaging. The normal brain tissue shows minimal uptake of 99mTc-sestamibi. The normal physiological distribution in the head and neck region is similar to that of thallium and includes the scalp, nasopharyngeal area, salivary gland, and pituitary gland. There is notable significant choroid plexus uptake, much greater when compared to ²⁰¹Tl. The choroid plexus uptake may be due in part from the pertechnetate in the solution that is known to be actively taken up and secreted by the cells. This may account for secretion into the CSF and the presence of activity in the 4th ventricle which is visible on careful scrutiny of MIBI brain SPECT images (Fig. 18.12). It is postulated that after crossing the cell membrane MIBI is taken by the mitochondria in relation to negative electric potential. Normal myocardial uptake of MIBI depends on blood flow and the uptake of the mitochondria in metabolically active tissue. In brain tumors, the mechanism of tumor uptake is also thought to be dependent on mitochondrial activity and the presence of P-glycoprotein [59].

A 36-year-old normal female volunteer underwent a fully dynamic 18F-FDG PET with arterial sampling at rest as shown in Fig. 18.13. ¹⁸F-FDG PET scan image slice thickness = 2.0 mm and reconstructed in-plane image resolution = 4 mm FWHM. The 18F-FDG PET data is acquired over 90 min (34 frames) with arterial blood sampling throughout the scan period. The method of quantitative assessment of 18F-FDG PET has been well established [60–64].

18F-FDG PET has led to a more widespread capability in allowing evaluation of cerebral neoplasms as well as other diseases of the brain which were previously imaged using SPECT radiopharmaceuticals [65, 66]. In addition, due to the relatively long half-life of 18F (109 min), it can be transported regionally (within approximately 2–4-h travel time from a cyclotron production facility) enabling a centrally located production facility to supply several camera sites.

A PET amino acid isotope, l-[methyl-¹¹C] methionine (11C-MET) [67] has a relatively short half-life of 20 min. The tracer’s use requires a nearby cyclotron. A study by Hustinx et al. explains its potential role in differentiating tumor recurrence from radiation necrosis [68]. The extent of tracer uptake is greater than the degree of contrast enhancement indicative of better delineation of tumor margins [69].

The tracer uptake has been shown to correlate with prognosis and survival in low-grade gliomas [70, 71], where the uptake is increased in the absence of BBB breakdown which is a significant advantage over CT, conventional MRI, and ^18F-FDG PET [72, 73]. In high-grade gliomas, ¹¹C-MET uptake is greater than in low-grade tumors [74–76] establishing its potential for use in monitoring anaplastic transformation.

In a study of 21 patients with brain metastases status post stereotactic radiosurgery, the tracer accurately identified 7 of 9 recurrences and 10 of 12 radiation injuries [77]. A combined ¹⁸F-FDG and ¹¹C-MET study for stereotactic biopsy of 32 unresectable glioma patients demonstrated that ¹¹C-MET generates a more sensitive signal, making it a potential single-tracer PET agent for neurosurgical intervention of gliomas [78].

A study by Ullrich et al. shows that increased ¹¹C-MET uptake during tumor growth parallels an upregulation of angiogenic markers such as vascular endothelial growth factor (VEGF) [79]. A study by Yamane et al. talks about the clinical impact of ¹¹C-MET and that the addition of ¹¹C-MET PET changed patient management [80].

O-(2-[¹⁸F] fluoroethyl)-l-tyrosine (18F-FET) is a PET tracer studied for its potential role in the differentiation of radiation necrosis and residual tumor. Studies have shown the absence of ¹⁸F-FET uptake in a case of radiation necrosis [81], but further systematic studies are necessary to confirm this finding. In contrast to ¹⁸F-FDG, ¹⁸F-FET uptake was absent from macrophages, a common inflammatory mediator [82]. In another study, the ratio of ¹⁸F-FET uptake in radiation necrosis to that in the normal cortex was much lower than the corresponding ratios for ¹⁸F-FDG and ¹⁸F-choline suggestive of its potential for differentiating radiation necrosis from tumor recurrence [83].

In the last decade, studies on combined 18F-FET and MRI have shown improved identification of tumor tissue as compared to either modality alone [84, 85]. The specificity of distinguishing gliomas from normal tissue could be increased from 68 % with the use of MRI alone to 97 % with the use of MRI in conjunction with ¹⁸F-FET PET and MRI spectroscopy [86].

3,4-dihydroxy-6-¹⁸F-fluoro-l-phenylalanine (¹⁸F-FDOPA) is an amino acid tracer initially used for the evaluation of movement disorders [87–89] but recently being studied in the imaging of brain tumors. ¹⁸F-DOPA crosses the BBB in the normal brain via the neutral amino acid transporter [90, 91]. Although increased sensitivity and specificity of ¹⁸F-DOPA over ¹⁸F-FDG was shown, no correlation to tumor grade or contrast enhancement was observed [83]. The same study found a tumor-to-normal-brain ratio of less than 1.6 with ¹⁸F-DOPA in four cases of radiation necrosis. A recent study showed that correlation between tracer uptake and tumor proliferation was observed only in newly diagnosed gliomas and not in recurrent gliomas [92]. When compared to ¹¹C-MET PET, no significant difference in uptake was shown in either low- or high-grade tumors [93]. Larger series of radiation necrosis cases will be needed for confirmation of these findings, however.

¹⁸F-Fluoromisonidazole is a nitroimidazole derivative PET agent used to image hypoxia [94], a physiological marker for tumor progression and resistance to radiotherapy [95]. Its preferential uptake in high-grade rather than low-grade gliomas [96], a significant relationship with the upregulation of angiogenic markers such as VEGF-R1 [97], and correlation to progression and survival after radiotherapy [98] suggest its potential role in monitoring response to therapy targeting hypoxic tissue.

3′-deoxy-3′-18F-fluorothymidine (18F-FLT) has been used to indicate tumor proliferation in both preclinical and clinical studies [99, 100]. Transport of ¹⁸F-FLT is mediated by both passive diffusion and Na⁺-dependent carriers. The tracer is subsequently phosphorylated by thymidine kinase 1 (TK₁) into ¹⁸F-FLT monophosphate where TK₁ is a principal enzyme in the salvage pathway of DNA synthesis. Whereas the TK₁ activity is virtually absent in quiescent cells, its activity reaches the maximum in the late G₁ and S phases of the cell cycle in proliferating cells [101]. The phosphorylation of the tracer by TK₁ therefore makes ¹⁸F-FLT a good marker for tumor proliferation.

Imaging of brain tumor proliferative activity has been performed using semiquantitative measures of standard uptake values. ¹⁸F-FLT imaging can be correlated with stereotactic biopsies representing the Ki-67 proliferation index. Recurrent or residual viable tumor demonstrates increased quantitative ¹⁸F-FLT utilization and can provide a useful index to separate residual or recurrent viable tumor from radiation or chemotherapy necrosis. ¹⁸F-FLT is more specific for detection of viable tumor proliferation since the background activity in the normal brain is low (Fig. 18.14), unlike ¹⁸F-FDG which has a high normal brain background.

Figure 18.15 demonstrates 1-year follow-up gadolinium-enhanced MRI of patient in image Fig. 18.14.

Nuclear medicine techniques have been used for the past 55–60 years to investigate cerebrovascular diseases and stroke mainly through tomographic applications, such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT) [110]. It is ironic that tracers which directly measure regional cerebral perfusion have not assumed greater clinical application in the evaluation of cerebrovascular disease. This is partly due to the fact that, in many cases, the identification of a poststroke blood flow defect has not provided unequivocal useful clinical information beyond that offered by neurological exam combined with CT or MRI. Furthermore, the use of rCBF tracers in patients with TIA has had unclear clinical usefulness since a rest perfusion scan may not provide significant additional clinical information beyond the neurological examination.