Ferumoxytol-enhanced magnetic resonance imaging methodology and normal values at 1.5 and 3T.

BACKGROUND: Ultrasmall superparamagnetic particles of iron oxide (USPIO)-enhanced magnetic resonance imaging (MRI) can detect tissue-resident macrophage activity and identify cellular inflammation. Clinical studies using this technique are now emerging. We aimed to report a range of normal R2* values at 1.5 and 3 T in the myocardium and other tissues following ferumoxytol administration, outline the methodology used and suggest solutions to commonly encountered analysis problems. METHODS: Twenty volunteers were recruited: 10 imaged each at 1.5 T and 3 T. T2* and late gadolinium enhanced (LGE) MRI was conducted at baseline with further T2* imaging conducted approximately 24 h after USPIO infusion (ferumoxytol, 4 mg/kg). Regions of interest were selected in the myocardium and compared to other tissues. RESULTS: Following administration, USPIO was detected by changes in R2* from baseline (1/T2*) at 24 h in myocardium, skeletal muscle, kidney, liver, spleen and blood at 1.5 T, and myocardium, kidney, liver, spleen, blood and bone at 3 T (p < 0.05 for all). Myocardial changes in R2* due to USPIO were 26.5 ± 7.3 s-1 at 1.5 T, and 37.2 ± 9.6 s-1 at 3 T (p < 0.0001 for both). Tissues showing greatest ferumoxytol enhancement were the reticuloendothelial system: the liver, spleen and bone marrow (216.3 ± 32.6 s-1, 336.3 ± 60.3 s-1, 69.9 ± 79.9 s-1; p < 0.0001, p < 0.0001, p = ns respectively at 1.5 T, and 275.6 ± 69.9 s-1, 463.9 ± 136.7 s-1, 417.9 ± 370.3 s-1; p < 0.0001, p < 0.0001, p < 0.01 respectively at 3 T). CONCLUSION: Ferumoxytol-enhanced MRI is feasible at both 1.5 T and 3 T. Careful data selection and dose administration, along with refinements to echo-time acquisition, post-processing and analysis techniques are essential to ensure reliable and robust quantification of tissue enhancement. TRIAL REGISTRATION: ClinicalTrials.gov Identifier - NCT02319278 . Registered 03.12.2014.

A novel deep learning method for large-scale analysis of bone marrow adiposity using UK Biobank Dixon MRI data.

Background: Bone marrow adipose tissue (BMAT) represents > 10% fat mass in healthy humans and can be measured by magnetic resonance imaging (MRI) as the bone marrow fat fraction (BMFF). Human MRI studies have identified several diseases associated with BMFF but have been relatively small scale. Population-scale studies therefore have huge potential to reveal BMAT's true clinical relevance. The UK Biobank (UKBB) is undertaking MRI of 100,000 participants, providing the ideal opportunity for such advances. Objective: To establish deep learning for high-throughput multi-site BMFF analysis from UKBB MRI data. Materials and methods: We studied males and females aged 60-69. Bone marrow (BM) segmentation was automated using a new lightweight attention-based 3D U-Net convolutional neural network that improved segmentation of small structures from large volumetric data. Using manual segmentations from 61-64 subjects, the models were trained to segment four BM regions of interest: the spine (thoracic and lumbar vertebrae), femoral head, total hip and femoral diaphysis. Models were tested using a further 10-12 datasets per region and validated using datasets from 729 UKBB participants. BMFF was then quantified and pathophysiological characteristics assessed, including site- and sex-dependent differences and the relationships with age, BMI, bone mineral density, peripheral adiposity, and osteoporosis. Results: Model accuracy matched or exceeded that for conventional U-Nets, yielding Dice scores of 91.2% (spine), 94.5% (femoral head), 91.2% (total hip) and 86.6% (femoral diaphysis). One case of severe scoliosis prevented segmentation of the spine, while one case of Non-Hodgkin Lymphoma prevented segmentation of the spine, femoral head and total hip because of T2 signal depletion; however, successful segmentation was not disrupted by any other pathophysiological variables. The resulting BMFF measurements confirmed expected relationships between BMFF and age, sex and bone density, and identified new site- and sex-specific characteristics. Conclusions: We have established a new deep learning method for accurate segmentation of small structures from large volumetric data, allowing high-throughput multi-site BMFF measurement in the UKBB. Our findings reveal new pathophysiological insights, highlighting the potential of BMFF as a novel clinical biomarker. Applying our method across the full UKBB cohort will help to reveal the impact of BMAT on human health and disease.

Neck muscle cross-sectional area, brain volume and cognition in healthy older men: a cohort study.

BACKGROUND: Two important consequences of the normal ageing process are sarcopenia (the age-related loss of muscle mass and function) and age-related cognitive decline. Existing data support positive relationships between muscle function, cognition and brain structure. However, studies investigating these relationships at older ages are lacking and rarely include a measure of muscle size. Here we test whether neck muscle size is positively associated with cognition and brain structure in older men. METHODS: We studied 51 healthy older men with mean age 73.8 (sd 1.5) years. Neck muscle cross-sectional area (CSA) was measured from T1-weighted MR-brain scans using a validated technique. We measured multiple cognitive domains including verbal and visuospatial memory, executive functioning and estimated prior cognitive ability. Whole brain, ventricular, hippocampal and cerebellar volumes were measured with MRI. General linear models (ANCOVA) were performed. RESULTS: Larger neck muscle CSA was associated with less whole brain atrophy (t = 2.86, p = 0.01, partial eta squared 17%). Neck muscle CSA was not associated with other neuroimaging variables or current cognitive ability. Smaller neck muscle CSA was unexpectedly associated with higher prior cognition (t = -2.12, p < 0.05, partial eta squared 10%). CONCLUSIONS: In healthy older men, preservation of whole brain volume (i.e. less atrophy) is associated with larger muscle size. Longitudinal ageing studies are now required to investigate these relationships further.

The magnetic resonance imaging subset of the mngu0 articulatory corpus.

This paper announces the availability of the magnetic resonance imaging (MRI) subset of the mngu0 corpus, a collection of articulatory speech data from one speaker containing different modalities. This subset comprises volumetric MRI scans of the speaker's vocal tract during sustained production of vowels and consonants, as well as dynamic mid-sagittal scans of repetitive consonant-vowel (CV) syllable production. For reference, high-quality acoustic recordings of the speech material are also available. The raw data are made freely available for research purposes.

Ferumoxytol-enhanced magnetic resonance imaging in acute myocarditis.

OBJECTIVES: Ultrasmall superparamagnetic particles of iron oxide (USPIO)-enhanced MRI can detect tissue-resident macrophage activity and identify cellular inflammation within tissues. We hypothesised that USPIO-enhanced MRI would provide a non-invasive imaging technique that would improve the diagnosis and management of patients with acute myocarditis. METHODS: Ten volunteers and 14 patients with suspected acute myocarditis underwent T2, T2* and late gadolinium enhancement (LGE) 3T MRI, with further T2* imaging at 24 hours after USPIO (ferumoxytol, 4 mg/kg) infusion, at baseline and 3 months. Myocardial oedema and USPIO enhancement were determined within areas of LGE as well as throughout the myocardium. RESULTS: Myocarditis was confirmed in nine of the 14 suspected cases of myocarditis. There was greater myocardial oedema in regions of LGE in patients with myocarditis when compared with healthy volunteer myocardium (T2 value, 57.1±5.3 vs 46.7±1.6 ms, p<0.0001). There was no demonstrable difference in USPIO enhancement between patients and volunteers even within regions displaying LGE (change in R2*, 35.0±15.0 vs 37.2±9.6 s-1, p>0.05). Imaging after 3 months in patients with myocarditis revealed a reduction in volume of LGE, a reduction in oedema measures within regions displaying LGE and improvement in ejection fraction (mean -19.7 mL, 95% CI (-0.5 to -40.0)), -5.8 ms (-0.9 to -10.7) and +6% (0.5% to 11.5%), respectively, p<0.05 for all). CONCLUSION: In patients with acute myocarditis, USPIO-enhanced MRI does not provide additional clinically relevant information to LGE and T2 mapping MRI. This suggests that tissue-resident macrophages do not provide a substantial contribution to the myocardial inflammation in this condition.Clinical trial registration NCT02319278; Results.

18F-Sodium Fluoride Uptake in Abdominal Aortic Aneurysms: The SoFIA3 Study.

BACKGROUND: Fluorine-18-sodium fluoride (18F-NaF) uptake is a marker of active vascular calcification associated with high-risk atherosclerotic plaque. OBJECTIVES: In patients with abdominal aortic aneurysm (AAA), the authors assessed whether 18F-NaF positron emission tomography (PET) and computed tomography (CT) predicts AAA growth and clinical outcomes. METHODS: In prospective case-control (n = 20 per group) and longitudinal cohort (n = 72) studies, patients with AAA (aortic diameter >40 mm) and control subjects (aortic diameter <30 mm) underwent abdominal ultrasound, 18F-NaF PET-CT, CT angiography, and calcium scoring. Clinical endpoints were aneurysm expansion and the composite of AAA repair or rupture. RESULTS: Fluorine-18-NaF uptake was increased in AAA compared with nonaneurysmal regions within the same aorta (p = 0.004) and aortas of control subjects (p = 0.023). Histology and micro-PET-CT demonstrated that 18F-NaF uptake localized to areas of aneurysm disease and active calcification. In 72 patients within the longitudinal cohort study (mean age 73 ± 7 years, 85% men, baseline aneurysm diameter 48.8 ± 7.7 mm), there were 19 aneurysm repairs (26.4%) and 3 ruptures (4.2%) after 510 ± 196 days. Aneurysms in the highest tertile of 18F-NaF uptake expanded 2.5× more rapidly than those in the lowest tertile (3.10 [interquartile range (IQR): 2.34 to 5.92 mm/year] vs. 1.24 [IQR: 0.52 to 2.92 mm/year]; p = 0.008) and were nearly 3× as likely to experience AAA repair or rupture (15.3% vs. 5.6%; log-rank p = 0.043). CONCLUSIONS: Fluorine-18-NaF PET-CT is a novel and promising approach to the identification of disease activity in patients with AAA and is an additive predictor of aneurysm growth and future clinical events. (Sodium Fluoride Imaging of Abdominal Aortic Aneurysms [SoFIA3]; NCT02229006; Magnetic Resonance Imaging [MRI] for Abdominal Aortic Aneurysms to Predict Rupture or Surgery: The MA3RS Trial; ISRCTN76413758).

Exploring the Biological and Mechanical Properties of Abdominal Aortic Aneurysms Using USPIO MRI and Peak Tissue Stress: A Combined Clinical and Finite Element Study.

Inflammation detected through the uptake of ultrasmall superparamagnetic particles of iron oxide (USPIO) on magnetic resonance imaging (MRI) and finite element (FE) modelling of tissue stress both hold potential in the assessment of abdominal aortic aneurysm (AAA) rupture risk. This study aimed to examine the spatial relationship between these two biomarkers. Patients (n = 50) > 40 years with AAA maximum diameters > = 40 mm underwent USPIO-enhanced MRI and computed tomography angiogram (CTA). USPIO uptake was compared with wall stress predictions from CTA-based patient-specific FE models of each aneurysm. Elevated stress was commonly observed in areas vulnerable to rupture (e.g. posterior wall and shoulder). Only 16% of aneurysms exhibited co-localisation of elevated stress and mural USPIO enhancement. Globally, no correlation was observed between stress and other measures of USPIO uptake (i.e. mean or peak). It is suggested that cellular inflammation and stress may represent different but complimentary aspects of AAA disease progression.

Ferumoxytol-enhanced magnetic resonance imaging assessing inflammation after myocardial infarction.

OBJECTIVES: Macrophages play a central role in the cellular inflammatory response to myocardial infarction (MI) and predict subsequent clinical outcomes. We aimed to assess temporal changes in cellular inflammation and tissue oedema in patients with acute MI using ultrasmallsuperparamagnetic particles of iron oxide (USPIO)-enhanced MRI. METHODS: Thirty-one patients were recruited following acute MI and followed up for 3 months with repeated T2 and USPIO-enhanced T2*-mapping MRI. Regions of interest were categorised into infarct, peri-infarct and remote myocardial zones, and compared with control tissues. RESULTS: Following a single dose, USPIO enhancement was detected in the myocardium until 24 hours (p<0.0001). Histology confirmed colocalisation of iron and macrophages within the infarcted, but not the non-infarcted, myocardium. Following repeated doses, USPIO uptake in the infarct zone peaked at days 2-3, and greater USPIO uptake was detected in the infarct zone compared with remote myocardium until days 10-16 (p<0.05). In contrast, T2-defined myocardial oedema peaked at days 3-9 and remained increased in the infarct zone throughout the 3-month follow-up period (p<0.01). CONCLUSION: Myocardial macrophage activity can be detected using USPIO-enhanced MRI in the first 2 weeks following acute MI. This observed pattern of cellular inflammation is distinct, and provides complementary information to the more prolonged myocardial oedema detectable using T2 mapping. This imaging technique holds promise as a non-invasive method of assessing and monitoring myocardial cellular inflammation with potential application to diagnosis, risk stratification and assessment of novel anti-inflammatory therapeutic interventions. TRIAL REGISTRATION NUMBER: Trial registration number: 14663. Registered on UK Clinical Research Network (http://public.ukcrn.org.uk) and also ClinicalTrials.gov (https://clinicaltrials.gov/ct2/show/NCT02319278?term=DECIFER&rank=2).

Sertoli Cells Modulate Testicular Vascular Network Development, Structure, and Function to Influence Circulating Testosterone Concentrations in Adult Male Mice.

The testicular vasculature forms a complex network, providing oxygenation, micronutrients, and waste clearance from the testis. The vasculature is also instrumental to testis function because it is both the route by which gonadotropins are delivered to the testis and by which T is transported away to target organs. Whether Sertoli cells play a role in regulating the testicular vasculature in postnatal life has never been unequivocally demonstrated. In this study we used models of acute Sertoli cell ablation and acute germ cell ablation to address whether Sertoli cells actively influence vascular structure and function in the adult testis. Our findings suggest that Sertoli cells play a key role in supporting the structure of the testicular vasculature. Ablating Sertoli cells (and germ cells) or germ cells alone results in a similar reduction in testis size, yet only the specific loss of Sertoli cells leads to a reduction in total intratesticular vascular volume, the number of vascular branches, and the numbers of small microvessels; loss of germ cells alone has no effect on the testicular vasculature. These perturbations to the testicular vasculature leads to a reduction in fluid exchange between the vasculature and testicular interstitium, which reduces gonadotropin-stimulated circulating T concentrations, indicative of reduced Leydig cell stimulation and/or reduced secretion of T into the vasculature. These findings describe a new paradigm by which the transport of hormones and other factors into and out of the testis may be influenced by Sertoli cells and highlights these cells as potential targets for enhancing this endocrine relationship.

5α-reductase type 1 modulates insulin sensitivity in men.

CONTEXT: 5α-Reductase (5αR) types 1 and 2 catalyze the A-ring reduction of steroids, including androgens and glucocorticoids. 5α-R inhibitors lower dihydrotestosterone in benign prostatic hyperplasia; finasteride inhibits 5αR2, and dutasteride inhibits both 5αR2 and 5αR1. In rodents, loss of 5αR1 promotes fatty liver. OBJECTIVE: Our objective was to test the hypothesis that inhibition of 5αR1 causes metabolic dysfunction in humans. DESIGN, SETTING, AND PARTICIPANTS: This double-blind randomized controlled parallel group study at a clinical research facility included 46 men (20-85 years) studied before and after intervention. INTERVENTION: Oral dutasteride (0.5 mg daily; n = 16), finasteride (5 mg daily; n = 16), or control (tamsulosin; 0.4 mg daily; n = 14) was administered for 3 months. MAIN OUTCOME MEASURE: Glucose disposal was measured during a stepwise hyperinsulinemic-euglycemic clamp. Data are mean (SEM). RESULTS: Dutasteride and finasteride had similar effects on steroid profiles, with reduced urinary androgen and glucocorticoid metabolites and reduced circulating DHT but no change in plasma or salivary cortisol. Dutasteride, but not finasteride, reduced stimulation of glucose disposal by high-dose insulin (dutasteride by -5.7 [3.2] µmol/kg fat-free mass/min, versus finasteride +7.2 [3.0], and tamsulosin +7.0 [2.0]). Dutasteride also reduced suppression of nonesterified fatty acids by insulin and increased body fat (by 1.6% [0.6%]). Glucose production and glycerol turnover were unchanged. Consistent with metabolic effects of dutasteride being mediated in peripheral tissues, mRNA for 5αR1 but not 5αR2 was detected in human adipose tissue. CONCLUSION: Dual inhibition of 5αRs, but not inhibition of 5αR2 alone, modulates insulin sensitivity in human peripheral tissues rather than liver. This may have important implications for patients prescribed dutasteride for prostatic disease.

Increased skeletal muscle 11ßHSD1 mRNA is associated with lower muscle strength in ageing.

BACKGROUND: Sarcopenia, the loss of muscle mass and function with age, is associated with increased morbidity and mortality. Current understanding of the underlying mechanisms is limited. Glucocorticoids (GC) in excess cause muscle weakness and atrophy. We hypothesized that GC may contribute to sarcopenia through elevated circulating levels or increased glucocorticoid receptor (GR) signaling by increased expression of either GR or the GC-amplifying enzyme 11 beta-hydroxysteroid dehydrogenase type 1 (11ßHSD1) in muscle. METHODS: There were 82 participants; group 1 comprised 33 older men (mean age 70.2 years, SD 4.4) and 19 younger men (22.2 years, 1.7) and group 2 comprised 16 older men (79.1 years, 3.4) and 14 older women (80.1 years, 3.7). We measured muscle strength, mid-thigh cross-sectional area, fasting morning plasma cortisol, quadriceps muscle GR and 11ßHSD1 mRNA, and urinary glucocorticoid metabolites. Data were analysed using multiple linear regression adjusting for age, gender and body size. RESULTS: Muscle strength and size were not associated with plasma cortisol, total urinary glucocorticoids or the ratio of urinary 5ß-tetrahydrocortisol +5α-tetrahydrocortisol to tetrahydrocortisone (an index of systemic 11ßHSD activity). Muscle strength was associated with 11ßHSD1 mRNA levels (ß -0.35, p = 0.04), but GR mRNA levels were not significantly associated with muscle strength or size. CONCLUSION: Although circulating levels of GC are not associated with muscle strength or size in either gender, increased cortisol generation within muscle by 11ßHSD1 may contribute to loss of muscle strength with age, a key component of sarcopenia. Inhibition of 11ßHSD1 may have therapeutic potential in sarcopenia.

Design and validation of a novel method to measure cross-sectional area of neck muscles included during routine MR brain volume imaging.

INTRODUCTION: Low muscle mass secondary to disease and ageing is an important cause of excess mortality and morbidity. Many studies include a MR brain scan but no peripheral measure of muscle mass. We developed a technique to measure posterior neck muscle cross-sectional area (CSA) on volumetric MR brain scans enabling brain and muscle size to be measured simultaneously. METHODS: We performed four studies to develop and test: feasibility, inter-rater reliability, repeatability and external validity. We used T1-weighted MR brain imaging from young and older subjects, obtained on different scanners, and collected mid-thigh MR data. RESULTS: After developing the technique and demonstrating feasibility, we tested it for inter-rater reliability in 40 subjects. Intraclass correlation coefficients (ICC) between raters were 0.99 (95% confidence intervals (CI) 0.98-1.00) for the combined group (trapezius, splenius and semispinalis), 0.92 (CI 0.85-0.96) for obliquus and 0.92 (CI 0.85-0.96) for sternocleidomastoid. The first unrotated principal component explained 72.2% of total neck muscle CSA variance and correlated positively with both right (r = 0.52, p = .001) and left (r = 0.50, p = .002) grip strength. The 14 subjects in the repeatability study had had two MR brain scans on three different scanners. The ICC for between scanner variation for total neck muscle CSA was high at 0.94 (CI 0.86-0.98). The ICCs for within scanner variations were also high, with values of 0.95 (CI 0.86-0.98), 0.97 (CI 0.92-0.99) and 0.96 (CI 0.86-0.99) for the three scanners. The external validity study found a correlation coefficient for total thigh CSA and total neck CSA of 0.88. DISCUSSION: We present a feasible, valid and reliable method for measuring neck muscle CSA on T1-weighted MR brain scans. Larger studies are needed to validate and apply our technique with subjects differing in age, ethnicity and geographical location.