External quality assurance for image grading in the Scottish Diabetic Retinopathy Screening Programme.

AIMS: To develop and evaluate an image grading external quality assurance system for the Scottish Diabetic Retinopathy Screening Programme. METHOD: A web-based image grading system was developed which closely matches the current Scottish national screening software. Two rounds of external quality assurance were run in autumn 2008 and spring 2010, each time using the same 100 images. Graders were compared with a consensus standard derived from the top-level graders' results. After the first round, the centre lead clinicians and top-level graders reviewed the results and drew up guidance notes for the second round. RESULTS: Grader sensitivities ranged from 60.0 to 100% (median 92.5%) in 2008, and from 62.5 to 100% (median 92.5%) in 2010. Specificities ranged from 34.0 to 98.0% (median 86%) in 2008, and 54.0 to 100% (median 88%) in 2010. There was no difference in sensitivity between grader levels, but first-level graders had a significantly lower specificity than level-two and level-three graders. In 2008, one centre had a lower sensitivity but higher specificity than the majority of centres. Following the feedback from the first round, overall agreement improved in 2010 and there were no longer any significant differences between centres. CONCLUSIONS: A useful educational tool has been developed for image grading external quality assurance.

Costs and consequences of automated algorithms versus manual grading for the detection of referable diabetic retinopathy.

AIMS: To assess the cost-effectiveness of an improved automated grading algorithm for diabetic retinopathy against a previously described algorithm, and in comparison with manual grading. METHODS: Efficacy of the alternative algorithms was assessed using a reference graded set of images from three screening centres in Scotland (1253 cases with observable/referable retinopathy and 6333 individuals with mild or no retinopathy). Screening outcomes and grading and diagnosis costs were modelled for a cohort of 180 000 people, with prevalence of referable retinopathy at 4%. Algorithm (b), which combines image quality assessment with detection algorithms for microaneurysms (MA), blot haemorrhages and exudates, was compared with a simpler algorithm (a) (using image quality assessment and MA/dot haemorrhage (DH) detection), and the current practice of manual grading. RESULTS: Compared with algorithm (a), algorithm (b) would identify an additional 113 cases of referable retinopathy for an incremental cost of pound 68 per additional case. Compared with manual grading, automated grading would be expected to identify between 54 and 123 fewer referable cases, for a grading cost saving between pound 3834 and pound 1727 per case missed. Extrapolation modelling over a 20-year time horizon suggests manual grading would cost between pound 25,676 and pound 267,115 per additional quality adjusted life year gained. CONCLUSIONS: Algorithm (b) is more cost-effective than the algorithm based on quality assessment and MA/DH detection. With respect to the value of introducing automated detection systems into screening programmes, automated grading operates within the recommended national standards in Scotland and is likely to be considered a cost-effective alternative to manual disease/no disease grading.

Cost-effectiveness of implementing automated grading within the national screening programme for diabetic retinopathy in Scotland.

AIMS: National screening programmes for diabetic retinopathy using digital photography and multi-level manual grading systems are currently being implemented in the UK. Here, we assess the cost-effectiveness of replacing first level manual grading in the National Screening Programme in Scotland with an automated system developed to assess image quality and detect the presence of any retinopathy. METHODS: A decision tree model was developed and populated using sensitivity/specificity and cost data based on a study of 6722 patients in the Grampian region. Costs to the NHS, and the number of appropriate screening outcomes and true referable cases detected in 1 year were assessed. RESULTS: For the diabetic population of Scotland (approximately 160,000), with prevalence of referable retinopathy at 4% (6400 true cases), the automated strategy would be expected to identify 5560 cases (86.9%) and the manual strategy 5610 cases (87.7%). However, the automated system led to savings in grading and quality assurance costs to the NHS of 201,600 pounds per year. The additional cost per additional referable case detected (manual vs automated) totalled 4088 pounds and the additional cost per additional appropriate screening outcome (manual vs automated) was 1990 pounds. CONCLUSIONS: Given that automated grading is less costly and of similar effectiveness, it is likely to be considered a cost-effective alternative to manual grading.

The efficacy of automated "disease/no disease" grading for diabetic retinopathy in a systematic screening programme.

AIM: To assess the efficacy of automated "disease/no disease" grading for diabetic retinopathy within a systematic screening programme. METHODS: Anonymised images were obtained from consecutive patients attending a regional primary care based diabetic retinopathy screening programme. A training set of 1067 images was used to develop automated grading algorithms. The final software was tested using a separate set of 14 406 images from 6722 patients. The sensitivity and specificity of manual and automated systems operating as "disease/no disease" graders (detecting poor quality images and any diabetic retinopathy) were determined relative to a clinical reference standard. RESULTS: The reference standard classified 8.2% of the patients as having ungradeable images (technical failures) and 62.5% as having no retinopathy. Detection of technical failures or any retinopathy was achieved by manual grading with 86.5% sensitivity (95% confidence interval 85.1 to 87.8) and 95.3% specificity (94.6 to 95.9) and by automated grading with 90.5% sensitivity (89.3 to 91.6) and 67.4% specificity (66.0 to 68.8). Manual and automated grading detected 99.1% and 97.9%, respectively, of patients with referable or observable retinopathy/maculopathy. Manual and automated grading detected 95.7% and 99.8%, respectively, of technical failures. CONCLUSION: Automated "disease/no disease" grading of diabetic retinopathy could safely reduce the burden of grading in diabetic retinopathy screening programmes.

Differences in myocardial velocity gradient measured throughout the cardiac cycle in patients with hypertrophic cardiomyopathy, athletes and patients with left ventricular hypertrophy due to hypertension.

OBJECTIVES: We sought to compare the myocardial velocity gradient (MVG) measured across the left ventricular (LV) posterior wall during the cardiac cycle between patients with hypertrophic cardiomyopathy (HCM), athletes and patients with LV hypertrophy due to systemic hypertension and to determine whether it might be used to discriminate these groups. BACKGROUND: The MVG is a new ultrasound variable, based on the color Doppler technique, that quantifies the spatial distribution of transmyocardial velocities. METHODS: A cohort of 158 subjects was subdivided by age into two groups: Group I (mean [+/-SD] 30 +/- 7 years) and Group II (58 +/- 8 years). Within each group there were three categories of subjects: Group Ia consisted of patients with HCM (n = 25), Group Ib consisted of athletes (n = 21), and Group Ic consisted of normal subjects; Group IIa consisted of patients with HCM (n = 19), Group IIb consisted of hypertensive patients (n = 27), and Group IIc consisted of normal subjects (n = 33). RESULTS: The MVG (mean [+/-SD] s-1) measured in systole was lower (p < 0.01) in patients with HCM (Group Ia 3.2 +/- 1.1; Group IIa 2.9 +/- 1.2) compared with athletes (Group Ib 4.6 +/- 1.1), hypertensive patients (Group IIb 4.2 +/- 1.8) and normal subjects (Group Ic 4.4 +/- 0.8; Group IIc 4.8 +/- 0.8). In early diastole, the MVG was lower (p < 0.05) in patients with HCM (Group Ia 3.7 +/- 1.5; Group IIa 2.6 +/- 0.9) than in athletes (Group Ib 9.9 +/- 1.9) and normal subjects (Group Ic 9.2 +/- 2.0; Group IIc 3.6 +/- 1.5), but not hypertensive patients (Group IIb 3.3 +/- 1.3). In late diastole, the MVG in patients with HCM (Group Ia 1.3 +/- 0.8; Group IIa 1.4 +/- 0.8) was lower (p < 0.01) than that in hypertensive patients (Group IIb 4.3 +/- 1.7) and normal subjects (Group IIc 3.8 +/- 0.9). An MVG < or = 7 s-1, as a single diagnostic approach, differentiated accurately (0.96 positive and 0.94 negative predictive value) between patients with HCM and athletes when the measurements were taken during early diastole. CONCLUSIONS: In both age groups, the MVG was lower in both systole and diastole in patients with HCM than in athletes, hypertensive patients or normal subjects. The MVG measured in early diastole in a group of subjects 18 to 45 years old would appear to be an accurate variable used to discriminate between HCM and hypertrophy in athletes.

Diagnosis of vasa previa with ultrasound and color flow Doppler: a case report.

Ultrasound and Doppler imaging have proven to be valuable antepartum diagnostic tools for the early recognition of vasa previa. Recent reports have demonstrated that successful detection and timely management can subsequently reduce the risk of a life-threatening fetal hemorrhage. We present the fourth reported case of vasa previa detected antenatally by ultrasound. While conscientious attention to the risk factors and a high index of clinical suspicion remain important considerations, we believe that ultrasound and color Doppler flow analysis represent a highly sensitive diagnostic tool for those patients at risk. In the past, antenatal diagnosis has relied on insensitive and inconclusive means, often only after signs of fetal compromise have become apparent. Sonography has been demonstrated to offer reduced fetal mortality and therefore must be considered the new standard of care for management of suspected vasa previa.

Age-related transmural peak mean velocities and peak velocity gradients by Doppler myocardial imaging in normal subjects.

Doppler myocardial imaging is a new cardiac ultrasound technique based on the principles of colour Doppler imaging which can determine myocardial velocities by detecting the changes of phase-shift of the ultrasound signal returning directly from the myocardium. To determine the normal range of transmural velocities in healthy hearts a prospective study was carried out involving 42 normal subjects (age from 21 to 78, mean 47 +/- 16 years). Using M-mode Doppler myocardial imaging the peak values of the mean velocity and velocity gradient across the left ventricular posterior wall were measured during standardized phases of the cardiac cycle. Peak mean velocities had the following values during the cardiac cycle: isovolumic contraction - 1.3 +/- 1.2 cm. s-1, early ventricular ejection 4.2 +/- 1.2 cm. s-1, late ventricular ejection 1.8 +/- 1.1 cm. s-1, isovolumic relaxation -2.0 +/- 0.8 cm. s-1, rapid ventricular filling -6.6 +/- 2.2 cm. s-1, atrial contraction -2.8 +/- 1.8 cm. s-1, atrial relaxation 1.2 +/- 1.1 cm. s-1. Peak velocity gradients were: isovolumic contraction 1.3 +/- 1.9 s-1, early ventricular contraction 4.7 +/- 1.9 s-1, late ventricular contraction 1.1 +/- 1.0 s-1, isovolumic relaxation -0.6 +/- 0.5 s-1, rapid ventricular filling 6.1 +/- 3.4 s-1, atrial contraction 2.6 +/- 1.7 s-1, atrial relaxation 0.0 +/- 0.3 s-1. Linear regression analysis showed that with the increase of age, peak velocity gradient decreases during rapid ventricular filling (r = 0.83; P < 0.0001) and increases during atrial contraction (r = 0.86; P < 0.0001) while peak mean velocity increases only during atrial contraction (r = 0.80, P < 0.0001). Thus, there was no correlation between increasing age and systolic peak mean velocity and peak velocity gradient but both diastolic filling phases rapid ventricular filling and atrial contraction demonstrated age-related changes. In summary, this study has determined the age-related range of normal transmural myocardial velocities within the left ventricular posterior wall in healthy hearts during the cardiac cycle. We conclude that these measurements of peak mean velocities and peak velocity gradients, should form the baseline for subsequent Doppler myocardial imaging clinical studies on myocardial diseases processes.

Verification of cardiac doppler tissue images using grey-scale M-mode images.

Doppler tissue imaging (DTI) is a new ultrasonic technique that can be used to give quantified information about the motion of the myocardium. DTI M-mode recordings can be generated that display cardiac wall motion in great detail. In order to verify the motion as displayed in these images comparisons were made with simultaneously obtained grey-scale M-mode recordings. After capturing 135 simultaneous DTI and grey-scale M-mode recordings, those were selected in which wall motion could be accurately assessed from the grey-scale recording. Comparison with the DTI images shows: (1) that DTI accurately displays the direction of wall motion; and (2) that DTI displays whether the wall is thickening or thinning as a velocity distribution across the heart wall. This information is more reliably displayed and easier to interpret in the DTI M-mode recordings than in the grey-scale M-mode recordings.

Doppler tissue imaging: myocardial wall motion velocities in normal subjects.

With a scanner modified for Doppler tissue imaging, mean myocardial velocities (MMV) across the myocardium were measured. The aim of this study was to determine the normal range of the maximum MMV in six standardized phases of the cardiac cycle. The MMV was defined as the average value of the myocardial velocity measured along each M-mode scan line throughout the thickness of the myocardium. The maximum MMV was defined as the maximum value of the MMV during the particular cardiac phase. Simultaneous gray-scale and Doppler tissue imaging M-mode images were taken of the interventricular septum and the left ventricular posterior wall from the parasternal long-axis and short-axis views in 15 normal volunteers (aged 21 to 47 years; mean 32 +/- 6 years). Each cardiac cycle was divided into six phases: atrial contraction, isovolumetric contraction, ventricular ejection, isovolumetric relaxation, rapid ventricular filling, and diastasis. Isovolumetric contraction, isovolumetric relaxation, and diastasis were subdivided into two parts a and b because of changes in the direction of the myocardial movement. For each volunteer, the mean and standard deviation of the maximum MMV were measured for each cardiac phase averaged from 12 cardiac cycles from both long-axis and short-axis views. Finally, the mean and standard deviation were taken for each cardiac phase from 180 cardiac cycles from 15 volunteers. We have found that specific cardiac phases show significant differences in the maximum MMV between the adjoining cardiac phases and significant differences also occur between the maximum MMV measured in the interventricular septum and the left ventricular posterior wall during the same cardiac phases. These normal values provide a standard against which future Doppler tissue imaging M-mode studies of abnormal left ventricular function might be compared.

Myocardial velocity gradients detected by Doppler imaging.

Using a scanner whose colour Doppler mode has been adapted to display tissue motion (instead of blood flow), velocity gradients have been detected across the myocardium. A velocity gradient is a gradual spatial change in the value of velocity estimates. Velocity gradients have potential for assessing regional myocardial contractility. 28 M-mode scans were performed on nine normal volunteers at different locations in the left-ventricle posterior wall. In each case simultaneous Doppler M-mode and pulse-echo M-mode images were obtained. Doppler velocity gradient (DVG) was calculated from Doppler M-mode images and rate of change of wall thickness (RCWT) was calculated from pulse-echo M-mode images. In all Doppler M-mode images statistically significant velocity gradients were observed. In all but one scan, cyclically consistent peaks in DVG occur relative to the electrocardiogram waveform. 99% of systolic and 89% of early diastolic peaks in RCWT have a corresponding peak in DVG. Velocity gradients are consistent with wall thickness changes, suggesting that they have potential for assessment of myocardial contractility.

Assessment of colour Doppler tissue imaging using test-phantoms.

An investigation has been carried out on the velocity resolution, spatial resolution and accuracy of Doppler images as part of a study into the Doppler display of cardiac tissue motion. Test-phantoms were designed to perform this work and images were captured on a computer. The characteristics of the phantom images and of the image capture process were studied. The smallest spatial detail that was observed in the Doppler image was 3 mm by 3 mm. Doppler receive gain and Doppler ensemble size both affected velocity resolution. Different target materials gave different measures for velocity resolution. This could be related to the different back-scatter intensities of the materials.

Comparison of humerus length with femur length in fetuses with Down syndrome.

A recent report by FitzSimmons et al. demonstrated a greater frequency of upper- versus lower-extremity shortening in autopsies of second-trimester fetuses with trisomy 21. We undertook this study to determine whether this upper-limb shortening could be detected by prenatal ultrasonography in the second trimester and therefore identify fetuses at risk for trisomy 21. A retrospective review of all prenatal sonograms preceding genetic amniocentesis was conducted. Between 1987 and 1990 11 consecutive fetuses between 15 and 22 weeks' gestation with trisomy 21 were identified by genetic amniocentesis. Femur and humerus lengths were plotted on growth curves created from 1470 normal patients between 12 and 26 weeks. Gestational age was confirmed by last menstrual period and biparietal diameter. In fetuses with trisomy 21, seven of 11 humeri were less than 5th percentile, for a sensitivity of 64%, whereas only two of 11 femurs were less than 5th percentile, for a sensitivity of 18%. Biparietal diameter/femur length and biparietal diameter/humerus length ratios were also tested to predict Down syndrome. In only 2 of 11 cases was the biparietal diameter/femur length ratio greater than 95th percentile, whereas the biparietal diameter/humerus length ratio was greater than 95th percentile in 7 of 11. Since all seven were identified by shortened humerus alone, we conclude that humerus length versus gestational age is the simplest and most effective screen. The positive predictive value of an abnormally short humerus length in detecting Down syndrome was 6.8% in our population where the prevalence of Down syndrome was 1 of 173. The present study supports the observations of FitzSimmons et al. that shortened humerus length has a greater sensitivity than femur length in cases of trisomy 21. We conclude that in fetuses at risk for trisomy 21 humerus length should be determined, because it may, if shortened, aid in the prenatal diagnosis.

The relationship between umbilical artery Doppler velocimetry and fetal biometry.

The relationship between peak-systolic/end-diastolic ratio of the umbilical artery waveform and fetal biometry was studied in 127 uncomplicated pregnancies with established dates between 20 and 40 weeks' gestation. At each ultrasonographic examination fetal biometry included measurement of the biparietal diameter, head circumference, abdominal circumference, and femur length. The peak-systolic/end-diastolic ratio was measured by either a continuous or a pulsed-wave method. There were significant linear negative correlations between all the biometric parameters, as well as between the ultrasonographically estimated fetal weight and peak-systolic/end-diastolic ratio. Of the individual ultrasonographic parameters the femur length (for gestations less than 30 weeks) and the abdominal circumference (for gestations greater than or equal to 30 weeks) were found to be best correlated with the peak-systolic/end-diastolic ratio. Regression curves, including the 10th and the 90th percentile, were developed between each biometric parameter (biparietal diameter, head circumference, abdominal circumference, and femur length), as well as between estimated fetal weight and peak-systolic/end-diastolic ratio. The estimated fetal weight nomogram had the best sensitivity (48%) in predicting intrauterine growth retardation. These nomograms should prove most useful in assessing downstream placental vascular resistance in high-risk patients with unknown dates.

Abruptio placentae.

Despite our heightened awareness of placental abruption, this condition remains largely unpreventable. The incidence of placental abruption may actually be on the rise due to increasing use of cocaine and "crack" and a greater contribution from abdominal trauma. Whether recurrence of abruption associated with hypertension can be prevented with low-dose aspirin is yet to be determined. Unfortunately, although our recognition of this condition may have improved, placental abruption remains a high cause of perinatal morbidity and mortality.

Relationship between fetal biophysical activities and umbilical cord blood gas values.

In a prospective study of 62 patients undergoing cesarean section before the onset of labor a fetal biophysical profile assessment was performed within 3 hours before the cesarean section. The presence or absence of the individual fetal biophysical activities (fetal heart rate reactivity, fetal breathing movements, fetal body movements, and fetal tone) were correlated with umbilical cord blood gas and acid-base measurements (artery and vein). Fetuses with nonreactive nonstress test results or the absence of breathing had significantly lower cord artery pH, PO2 bicarbonate, and base excess measurements but not a significantly different PCO2 level as compared with fetuses that had these activities present. Fetuses with the absence of movements or tone had lower pH, PO2 bicarbonate, and base excess levels and higher PCO2 levels as compared with fetuses with the presence of movements or tone, respectively. These blood gas and acid-base differences were observed in both umbilical cord artery and vein. Subsequent analysis of the blood gas and acid-base measurements of the fetuses with compromised biophysical activities revealed that there are different levels of acidemia, hypoxemia, and hypercapnia at which the individual biophysical activities are compromised. These data suggest that the first manifestations of fetal hypoxemia and acidemia are nonreactive nonstress test results and loss of fetal breathing; in advanced acidemia, hypoxemia, and hypercapnia fetal movements and fetal tone are compromised.