Variability in PET image quality and quantification measured with a permanently filled 68Ge-phantom: a multi-center study

BACKGROUND

This study evaluated, as a snapshot, the variability in quantification and image quality (IQ) of the clinically utilized PET [¹⁸F]FDG whole-body protocols in Finland using a NEMA/IEC IQ phantom permanently filled with ⁶⁸Ge.

METHODS

The phantom was imaged on 14 PET-CT scanners, including a variety of models from two major vendors. The variability of the recovery coefficients (RC_max, RC_mean and RC_peak) of the hot spheres as well as percent background variability (PBV), coefficient of variation of the background (COV_BG) and accuracy of corrections (AOC) were studied using images from clinical and standardized protocols with 20 repeated measurements. The ranges of the RCs were also compared to the limits of the EARL ¹⁸F standards 2 accreditation (EARL2). The impact of image noise on these parameters was studied using averaged images (AVIs).

RESULTS

The largest variability in RC values of the routine protocols was found for the RC_max with a range of 68% and with 10% intra-scanner variability, decreasing to 36% when excluding protocols with suspected cross-calibration failure or without point-spread-function (PSF) correction. The RC ranges of individual hot spheres in routine or standardized protocols or AVIs fulfilled the EARL2 ranges with two minor exceptions, but fulfilling the exact EARL2 limits for all hot spheres was variable. RC_peak was less dependent on averaging and reconstruction parameters than RC_max and RC_mean. The PBV, COV_BG and AOC varied between 2.3-11.8%, 9.6-17.8% and 4.8-32.0%, respectively, for the routine protocols. The RC ranges, PBV and COV_BG were decreased when using AVIs. With AOC, when excluding routine protocols without PSF correction, the maximum value dropped to 15.5%.

CONCLUSION

The maximum variability of the RC values for the [¹⁸F]FDG whole-body protocols was about 60%. The RC ranges of properly cross-calibrated scanners with PSF correction fitted to the EARL2 RC ranges for individual sphere sizes, but fulfilling the exact RC limits would have needed further optimization. RC_peak was the most robust RC measure. Besides COV_BG, also RCs and PVB were sensitive to image noise.

Decreased Gray-White Matter Contrast of [11C]-PiB Uptake in Cognitively Unimpaired Subjects with Severe Obstructive Sleep Apnea

BACKGROUND

Very recently, cognitively normal, middle-aged adults with severe obstructive sleep apnea (OSA) were shown to have regional cortical amyloid-β deposits. In the normal brain, amyloid tracer (e.g., [11C]-PiB) uptake is observed in white matter (WM) but not in cortical gray matter (GM), resulting in clear GM-WM contrast. There are no reports on possible changes in this contrast in severe OSA.

OBJECTIVES

Evaluate changes in the global [11C]-PiB GM-WM contrast and study if factors reflecting clinical and imaging characteristics are associated with them.

DESIGN AND SETTING

Cross-sectional imaging study.

PARTICIPANTS

19 cognitively intact middle-aged (mean 44 years) patients with severe OSA (Apnea-Hypopnea Index >30/h), carefully selected to exclude any other possible factors that could alter brain health.

MEASUREMENTS

Detailed neuroimaging (amyloid PET, MRI). Signs of possible alterations in amyloid tracer GM-WM contrast and kinetics were studied with static and dynamic [11C]-PiB PET and WM structures with detailed 3.0T MRI.

RESULTS

Static [11C]-PiB PET uptake showed significantly decreased GM-WM contrast in 5 out of 19 patients. This was already clearly seen in visual evaluation and also detected quantitatively using retention indexes. Dynamic imaging revealed decreased contrast due to alterations in trace accumulation in the late phase of [11C]-PiB kinetics. Decreased GM-WM contrast in the late phase was global in nature. MRI revealed no corresponding alterations in WM structures. Importantly, decreased GM-WM contrast was associated with smoking (p = 0.007) and higher Apnea-Hypopnea Index (p = 0.001).

CONCLUSIONS

Severe OSA was associated with decreased GM-WM contrast in amyloid tracer uptake, with significant correlation with clinical parameters of smoking and AHI. The results support and further extend the current understanding of the deleterious effect of severe OSA on proper amyloid clearance, possibly reflecting dysfunction of the brain glymphatic system.

Inclusion of Regional Body Composition Parameters Improves Bone Mineral Density Cross-Calibration Between GE Lunar Prodigy and iDXA Densitometers

Since 1989, the Kuopio Osteoporosis Risk Factor and Prevention (OSTPRE, n = 14220) Study has followed long-term changes of bone mineral density (BMD) and body composition in women with GE Lunar devices. During the course of OSTPRE, the dual-energy X-ray absorptiometry device had to be replaced by a newer model. Then, it was essential to determine whether systematic measurement differences in BMD and body composition will occur. As a part of the OSTPRE study, BMD was measured in 54 women, whereas body composition was determined in 55 women, aged 27-71, by using both the GE Healthcare Lunar Prodigy and iDXA narrow-angle fan beam densitometers during the same visit. The total body fat mass (FM) and lean body mass (LBM) results of these scanners showed a high linear correlation (r = 0.981-0.994, p < 0.0001). However, the mean total body FM and LBM values measured by iDXA were on average 2.3% (0.5 kg, 95% confidence interval: 0.3-0.7 kg) higher and 0.8% (0.3 kg, 95% confidence interval: 0.1-0.6 kg) lower, respectively, than those measured by Prodigy. Inclusion of local soft tissue measurements (total body LBM, legs/android FM) improved the agreement of total body, total hip, and lumbar spine BMD values between the devices but not femoral neck BMD agreement. Equations, based on linear regression analyses, were derived to minimize differences between the instruments. Then, the differences in BMD and body composition measurements were negligible between Prodigy and iDXA. Using correction equations enables an objective comparison of longitudinal BMD and body composition measurements.

Multi-site bone ultrasound measurements in elderly women with and without previous hip fractures

About 75% of patients suffering from osteoporosis are not diagnosed. This study describes a multi-site bone ultrasound method for osteoporosis diagnostics. In comparison with axial dual energy X-ray absorptiometry (DXA), the ultrasound method showed good diagnostic performance and could discriminate fracture subjects among elderly females.

INTRODUCTION

Axial DXA, the gold standard diagnostic method for osteoporosis, predicts fractures only moderately. At present, no reliable diagnostic methods are available at the primary health care level. Here, a multi-site ultrasound method is proposed for osteoporosis diagnostics.

METHODS

Thirty elderly women were examined using the ultrasound backscatter measurements in proximal femur, proximal radius, proximal and distal tibia in vivo. First, we predicted the areal bone mineral density (BMD) at femoral neck by ultrasound measurements in tibia combined with specific subject characteristics (density index, DI) and, second, we tested the ability of ultrasound backscatter measurements at proximal femur to discriminate between individuals with previously fractured hips from those without fractures. Areal BMD was determined by axial DXA.

RESULTS

Combined ultrasound parameters, cortical thickness at distal and proximal tibia, with age and weight of the subject, provided a significant estimate of BMD(neck) (r = 0.86, p < 0.001, n = 30). When inserted into FRAX (World Health Organization fracture risk assessment tool), the DI indicated the same treatment proposal as the BMD(neck) with 86% sensitivity and 100% specificity. The receiver operating characteristic analyses, with a combination of ultrasound parameters and patient characteristics, discriminated fracture subjects from the controls similarly as the model combining BMD(neck) and patient characteristics.

CONCLUSIONS

For the first time, ultrasound backscatter measurements of proximal femur were conducted in vivo. The results indicate that ultrasound parameters, combined with patient characteristics, may provide a means for osteoporosis diagnostics.

Structural parameters of normal and osteoporotic human trabecular bone are affected differently by microCT image resolution

This study employed microCT to investigate whether image resolution affects bone structural parameters differently in healthy normal and osteoporotic trabecular bone. With increasing image voxel size, the originally detected differences between sample groups diminished. The results suggest that structural differences may not be reliably detected with clinical scanners.

INTRODUCTION

Structural parameters of bone reflect its health status, but are highly dependent on the image resolution. We hypothesized that image resolution affects bone structural parameters differently in normal and osteoporotic trabecular bone.

METHODS

Human trabecular bone samples from the iliac crest and the knee were analyzed (normal n = 11, osteoporotic n = 15) using a high-resolution microCT (14 or 18 µm voxel sizes). Images were re-sampled to voxel sizes 1-16 times larger than the original image and thresholded with global or local adaptive algorithms. Absolute and normalized values of each structural parameter were calculated, and the effect of decreasing image resolution was compared between the normal and osteoporotic samples.

RESULTS

Normal and osteoporotic samples had different (p < 0.05) absolute bone volume fractions. However, the normalized values showed that the osteoporotic samples were more prone to errors (p < 0.05) with increased voxel size. The absolute values of trabecular number, trabecular separation, degree of anisotropy, and structure model index were different between the groups at the original voxel size (p < 0.05), but at voxel sizes between 60 and 110 µm, those differences were no longer significant.

CONCLUSIONS

The results suggest that structural differences between osteoporotic and normal trabecular bone may not be reliably detected with clinical CT scanners providing image voxel sizes above 100 µm.

Efficacy of Haemophilus influenzae type b polysaccharide-diphtheria toxoid conjugate vaccine in infancy

Haemophilus influenzae type b capsular polysaccharide-diphtheria toxoid conjugate vaccine has recently been shown to be capable of inducing antibodies to H. influenzae in infants. In an evaluation of its clinical efficacy, 60,000 children were enrolled in an open trial in Finland. Children born on odd-numbered days between October 1, 1985, and September 30, 1986, received the vaccine at 3, 4, 6, and 14 months; those born on even-numbered days served as controls. The geometric mean antibody titer measured in a cohort of 99 children rose from a prevaccination level of 0.08 microgram per milliliter at three months of age to 0.42 microgram per milliliter at seven months. Only minor adverse reactions were reported. Up to February 1987, two cases of invasive H. influenzae infection had occurred among the children who had received three doses of vaccine, whereas 12 cases had occurred among the controls (P = 0.0005 by Poisson one-tailed test). The rate of short-term (average follow-up time, five months) protection provided by this conjugate vaccine in infancy was thus 83 percent.