Small Absolute Differences Research Articles

Machine learning derived segmentation of phase velocity encoded cardiovascular magnetic resonance for fully automated aortic flow quantification

BackgroundPhase contrast (PC) cardiovascular magnetic resonance (CMR) is widely employed for flow quantification, but analysis typically requires time consuming manual segmentation which can require human correction. Advances in machine learning have markedly improved automated processing, but have yet to be applied to PC-CMR. This study tested a novel machine learning model for fully automated analysis of PC-CMR aortic flow.MethodsA machine learning model was designed to track aortic valve borders based on neural network approaches. The model was trained in a derivation cohort encompassing 150 patients who underwent clinical PC-CMR then compared to manual and commercially-available automated segmentation in a prospective validation cohort. Further validation testing was performed in an external cohort acquired from a different site/CMR vendor.ResultsAmong 190 coronary artery disease patients prospectively undergoing CMR on commercial scanners (84% 1.5T, 16% 3T), machine learning segmentation was uniformly successful, requiring no human intervention: Segmentation time was < 0.01 min/case (1.2 min for entire dataset); manual segmentation required 3.96 ± 0.36 min/case (12.5 h for entire dataset). Correlations between machine learning and manual segmentation-derived flow approached unity (r = 0.99, p < 0.001). Machine learning yielded smaller absolute differences with manual segmentation than did commercial automation (1.85 ± 1.80 vs. 3.33 ± 3.18 mL, p < 0.01): Nearly all (98%) of cases differed by ≤5 mL between machine learning and manual methods. Among patients without advanced mitral regurgitation, machine learning correlated well (r = 0.63, p < 0.001) and yielded small differences with cine-CMR stroke volume (∆ 1.3 ± 17.7 mL, p = 0.36). Among advanced mitral regurgitation patients, machine learning yielded lower stroke volume than did volumetric cine-CMR (∆ 12.6 ± 20.9 mL, p = 0.005), further supporting validity of this method. Among the external validation cohort (n = 80) acquired using a different CMR vendor, the algorithm yielded equivalently small differences (∆ 1.39 ± 1.77 mL, p = 0.4) and high correlations (r = 0.99, p < 0.001) with manual segmentation, including similar results in 20 patients with bicuspid or stenotic aortic valve pathology (∆ 1.71 ± 2.25 mL, p = 0.25).ConclusionFully automated machine learning PC-CMR segmentation performs robustly for aortic flow quantification - yielding rapid segmentation, small differences with manual segmentation, and identification of differential forward/left ventricular volumetric stroke volume in context of concomitant mitral regurgitation. Findings support use of machine learning for analysis of large scale CMR datasets.

The impact of using donor sperm in assisted reproductive technology cycles on perinatal outcomes

To assess the impact of using donor sperm in assisted reproductive technology (ART) cycles on perinatal outcomes. Historical cohort study. US national database from the Society of Assisted Reproductive Technology Clinic Outcome Reporting System (SART CORS) from 2012 to2013. Patients undergoing the first fresh autologous ART cycle using either donor or partner sperm. None. Miscarriage, preterm birth, low birthweight rates. A total of 134,710 fresh autologous ART cycles were evaluated from the SART CORS database. Following exclusion criteria and after restricting to the first cycle, 2,123 donor sperm and 42,799 partner sperm ART cycles were included in the final analyses. After adjusting for all confounding variables (including maternal age, race, body mass index, smoking status, gravidity, history of preterm birth, highest follicle stimulating hormone count, blastocyst transfer percentage, total embryo transferred, and etiology of infertility), no statistically significant differences in miscarriage rates, preterm births, very preterm births, low birthweight, and very low birthweight were observed. Birthweight was significantly lower in the partner sperm group than in the donor sperm group (3,292 ± 601 and 3,233 ± 592g in donor and partner sperm groups, respectively, adjusted P value 0.003); however, this small absolute difference (adjusted effect estimate 42g, 95% CI 14.7-70.9) does not carry clinical significance. The use of donor sperm in fresh autologous ART cycles was not associated with increased miscarriage, preterm births, or low birthweights, as compared to cycles using partner sperm. This information can be used in patient counseling to reassure patients using donor sperm in ART cycles.

Kinematic alignment more closely restores the groove location and the sulcus angle of the native trochlea than mechanical alignment: implications for prosthetic design.

Kinematic alignment (KA) and mechanical alignment (MA) position the prosthetic trochlea that guides patellar tracking differently. The present study determined whether KA or MA more closely restores the groove location and sulcus angle of the prosthetic trochlea to the native trochlea for three femoral component designs. Ten 3D femur-cartilage models were created by combining computer tomographic (CT) and laser scans of native human cadaveric femurs. Three femoral component designs were positioned using KA and MA. Measurements of the prosthetic and native trochlea were made along the arc length of the native trochlear groove. The alignment technique with the smaller absolute difference between prosthetic and native for the medial-lateral and radial locations of the groove and sulcus angle of the trochlea more closely restored the native trochlea. For three femoral component designs, KA more closely restored to native the mean medial-lateral location (p = 0.0033 to < 0.0001) and mean radial location (p = 0.0150 to < 0.0001) than MA. For two femoral component designs, KA more closely restored to native the mean sulcus angle (p = 0.0326 to 0.0006) than MA. However, the differences in the mean sulcus angles between KA and MA were less than 2° for all three designs. KA more closely restored the native trochlea, which explains why the reported risk of patellofemoral complications for KA is not higher than MA according to five randomized clinical trials. Small design modifications of the medial-lateral and radial locations and sulcus angle are strategies for restoring the native trochlea. Such modifications might further reduce the risk of patellofemoral complications. II.

Minimum sample size for developing a multivariable prediction model: PART II - binary and time-to-event outcomes.

When designing a study to develop a new prediction model with binary or time‐to‐event outcomes, researchers should ensure their sample size is adequate in terms of the number of participants (n) and outcome events (E) relative to the number of predictor parameters (p) considered for inclusion. We propose that the minimum values of n and E (and subsequently the minimum number of events per predictor parameter, EPP) should be calculated to meet the following three criteria: (i) small optimism in predictor effect estimates as defined by a global shrinkage factor of 0.9, (ii) small absolute difference of 0.05 in the model's apparent and adjusted Nagelkerke's , and (iii) precise estimation of the overall risk in the population. Criteria (i) and (ii) aim to reduce overfitting conditional on a chosen p, and require prespecification of the model's anticipated Cox‐Snell , which we show can be obtained from previous studies. The values of n and E that meet all three criteria provides the minimum sample size required for model development. Upon application of our approach, a new diagnostic model for Chagas disease requires an EPP of at least 4.8 and a new prognostic model for recurrent venous thromboembolism requires an EPP of at least 23. This reinforces why rules of thumb (eg, 10 EPP) should be avoided. Researchers might additionally ensure the sample size gives precise estimates of key predictor effects; this is especially important when key categorical predictors have few events in some categories, as this may substantially increase the numbers required.

Minimum sample size for developing a multivariable prediction model: Part I-Continuous outcomes.

In the medical literature, hundreds of prediction models are being developed to predict health outcomes in individuals. For continuous outcomes, typically a linear regression model is developed to predict an individual's outcome value conditional on values of multiple predictors (covariates). To improve model development and reduce the potential for overfitting, a suitable sample size is required in terms of the number of subjects (n) relative to the number of predictor parameters (p) for potential inclusion. We propose that the minimum value of n should meet the following four key criteria: (i) small optimism in predictor effect estimates as defined by a global shrinkage factor of ≥0.9; (ii) small absolute difference of ≤ 0.05 in the apparent and adjusted R2 ; (iii) precise estimation (a margin of error ≤ 10% of the true value) of the model's residual standard deviation; and similarly, (iv) precise estimation of the mean predicted outcome value (model intercept). The criteria require prespecification of the user's chosen p and the model's anticipated R2 as informed by previous studies. The value of n that meets all four criteria provides the minimum sample size required for model development. In an applied example, a new model to predict lung function in African-American women using 25 predictor parameters requires at least 918 subjects to meet all criteria, corresponding to at least 36.7 subjects per predictor parameter. Even larger sample sizes may be needed to additionally ensure precise estimates of key predictor effects, especially when important categorical predictors have low prevalence in certain categories.

Increased risks for random errors are common in outcomes graded as high certainty of evidence

ObjectivesThe aim of article was to assess the risk for random errors in outcomes graded as high certainty of evidence (CoE). Study Design and SettingWe randomly selected 100 Cochrane reviews with dichotomous outcomes rated as high CoE using Grading of Recommendations Assessment, Development, and Evaluation. To detect increased risks for random errors, two investigators independently conducted trial sequential analysis using conventional thresholds for type I (α = 0.05) and type II (β = 0.10) errors. We dually regraded all outcomes with increased risks for random errors and conducted multivariate logistic regression analyses to determine predictors of increased risks for random errors. ResultsOverall, 38% (95% confidence interval: 28–47%) of high CoE outcomes had increased risks for random errors. Outcomes assessing harms were more frequently affected than outcomes assessing benefits (47% vs. 12%). Regrading of outcomes with increased random errors showed that 74% should have been downgraded based on current guidance. Regression analyses rendered small absolute risk differences (P = 0.009) and low number of events (P = 0.001) as significant predictors of increased risks for random errors. ConclusionDecisionmakers need to be aware that outcomes rated as high CoE often have increased risks for false-positive or false-negative findings.

Association of Cataract Surgical Outcomes With Late Surgeon Career Stages

Evidence suggests that the quality of some aspects of care provided by physicians may decrease during their late career stage. However, to our knowledge, data regarding the association of advancing surgeon career phase with cataract surgical outcomes have been lacking. To investigate whether an increase in cataract surgical adverse events occurs during later surgeon career stages. This population-based study of 499 650 cataract operations performed in Ontario, Canada, between January 1, 2009, and December 31, 2013, investigated the association between late surgeon career stage and the risk of surgical adverse events. Linked health care databases were used to study cataract surgical complications while controlling for patient-, surgeon-, and institution-level covariates. All ophthalmologists who performed cataract surgery in Ontario within the study period were included in the analysis. Isolated cataract surgery performed by surgeons at early, mid, and late career stages. Four serious adverse events were evaluated: dropped lens fragments, posterior capsule rupture, suspected endophthalmitis, and retinal detachment. Of 416 502 participants, 244 670 (58.7%) were women, 90 429 (21.7%) were age 66 to 70 years, 111 530 (26.8%) were age 71 to 75 years, 90 809 (21.8%) were age 76 to 80 years, and 123 734 (29.7%) were 81 years or older. Late-career surgeons performed 143 108 of 499 650 cataract operations (28.6%) during the study period. Late surgeon career stage was not associated with an increased overall risk of surgical adverse events (odds ratio [OR] vs midcareer, 1.06; 95% CI, 0.85-1.32). In a sensitivity analysis with surgeon volume removed from the model, late career stage was still not associated with overall adverse surgical events (OR, 1.10; 95% CI, 0.88-1.38). Among individual complications, late surgeon career stage was associated with an increased risk of dropped lens fragment (OR, 2.30; 95% CI, 1.50-3.54) and suspected endophthalmitis (OR, 1.41; 95% CI, 1.01-1.98). These corresponded with small absolute risk differences of 0.11% (95% CI, 0.085%-0.130%) and 0.045% (95% CI, 0.028%-0.063%) for dropped lens fragment and suspected endophthalmitis, respectively. These findings suggest that later-career surgeons are performing a substantial proportion of cataract operations with overall low surgical adverse event rates. Future studies might extend evaluations to the frequency of secondary surgical interventions as additional measures of surgical care quality.

Association of Risk for Venous Thromboembolism With Use of Low-Dose Extended- and Continuous-Cycle Combined Oral Contraceptives

Continuous/extended cyclic estrogen use (84/7 or 365/0 days cycles) in combined oral contraceptives (COCs) could potentially expose women to an increased cumulative dose of estrogen, compared with traditional cyclic regimens (21/7 days cycle), and may increase the risk for venous thromboembolism (VTE). To determine, while holding the progestogen type constant, whether the risk for VTE is higher with use of continuous/extended COCs than with cyclic COCs among women who initiated a COC containing ethinyl estradiol and levonorgestrel. Incident user retrospective cohort study of primarily commercially insured US population identified from the Sentinel Distributed Database. Participants were women aged 18 to 50 years at the time of initiating a study COC between May 2007 and September 2015. Using a propensity score approach and Cox proportional hazards regression models, we estimated the hazard ratios of VTE overall and separately by ethinyl estradiol dose and age groups. Initiation of continuous/extended or traditional cyclic COCs containing ethinyl estradiol or levonorgestrel of any dose. First VTE hospitalization that occurred during the study follow-up, identified by an inpatient International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code of 415.1, 415.1x, 453, 453.x, or 453.xx. We identified 210 691 initiators of continuous/extended COCs (mean [SD] age, 30.4 [8.6] years) and 522 316 initiators of cyclic COCs (mean [SD] age, 28.8 [8.3] years), with a mean of 0.7 person-years at risk among continuous/extended and cyclic users. Baseline cardiovascular and metabolic conditions (7.2% vs 4.7%), gynecological conditions (39.7% vs 32.3%), and health services utilization were slightly higher among continuous/extended cyclic than cyclic COC users. Propensity score matching decreased the hazard ratio estimates from 1.84 (95% CI, 1.53-2.21) to 1.32 (95% CI, 1.07-1.64) for continuous/extended use compared with cyclic COC use. The absolute risk difference (0.27 per 1000 persons) and the incidence rate difference (0.35 cases per 1000 person-years [1.44 vs 1.09 cases per 1000 person-years]) between the 2 propensity score-matched cohorts remained low, which may not translate into a clinically significant risk differences between cyclic and noncyclic estrogen use. Holding the progestogen type constant (levonorgestrel), we observed a slightly elevated VTE risk in association with continuous/extended COC use when compared with cyclic COC use. However, due to the small absolute risk difference and potential residual confounding, our findings did not show strong evidence supporting a VTE risk difference between continuous/extended and cyclic COC use.

Thursday, September 27, 2018 1:05 PM–2:05 PM Lumbar Spine Surgery: What You Need to Know: 113. Routine microscope use does not increase operative time or affect short term outcomes in single-level lumbar microdecompression

BACKGROUND CONTEXT Lumbar microdecompression is one of the most common procedures in orthopedic spine surgery with an overall low complication and high satisfaction rate. The procedure is typically performed with visualization aids, either loupes or an operating room microscope. Several studies to date have investigated differences between these two magnification tools, with conflicting results. Some authors argue in favor of the microscope, citing improved visualization, better ergonomics and improved patient outcomes, while others argue that its use makes no clinical difference. These studies consist primarily of small patient cohorts numbering 50-100 subjects. PURPOSE The purpose of this study is to determine how the routine use of an operating room microscope during single level lumbar microdecompression impacts intraoperative and short-term surgical outcomes through an examination of the largest retrospective cohort to date. STUDY DESIGN/SETTING Retrospective review of American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) database. PATIENT SAMPLE A total of 43,948 patients who underwent single level lumbar microdecompression from 2005 to 2015. OUTCOME MEASURES Assessment of surgical benchmarks of operative time and length of hospital stay as well as differences in early (30 day) complication rates. The specific complications evaluated included the need for blood transfusion, return to the operating room, pulmonary embolism, deep venous thrombosis, sepsis, cerebrovascular accident, cardiac arrest, ventilator time greater than 48 hours, urinary tract infection, pneumonia, myocardial infarction, surgical site infections, and patient mortality. METHODS Using the ACS-NSQIP database, we identified all patients who underwent single level lumbar microdecompression (CPT 63030). They were divided into two cohorts, those cases that involved the use of an operating room microscope (CPT 69990), and those that did not. We compared intraoperative and postoperative procedural data as well as the incidences of local and systemic complications. RESULTS The ACS-NSQIP database identified 43,948 patients who underwent single level lumbar microdecompression from 2005 to 2015. Of that patient cohort 4,654 cases included the use of an operating room microscope. There were no clinically significant demographic differences between patient cohorts. We identified clinically important statistical differences in length of stay (1.37 vs. 1.75 days, p CONCLUSIONS The routine use of an operating room microscope in single level lumbar microdecompression does not affect short-term patient outcomes. Importantly, there is no difference in mean operative time or infection rate with the addition of a microscope. There is a statistically significant trend toward decreased length of stay and fewer blood transfusions in the microscope cohort; however, in the setting of a large database study and small absolute differences, the clinical implications of these findings are difficult to interpret. Based on these results, we conclude that the decision to use an operating room microscope in single level lumbar microdecompression should be left to the treating surgeon and based upon considerations of improvements in visualization, surgeon ergonomics and ability to educate surgical assistants. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.

Prediction Models for Clinical Outcome After a Carotid Revascularization Procedure.

Background and Purpose- Prediction models may help physicians to stratify patients with high and low risk for periprocedural complications or long-term stroke risk after carotid artery stenting or carotid endarterectomy. We aimed to evaluate external performance of previously published prediction models for short- and long-term outcome after carotid revascularization in patients with symptomatic carotid artery stenosis. Methods- From a literature review, we selected all prediction models that used only readily available patient characteristics known before procedure initiation. Follow-up data from 2184 carotid artery stenting and 2261 carotid endarterectomy patients from 4 randomized trials (EVA-3S [Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis], SPACE [Stent-Protected Angioplasty Versus Carotid Endarterectomy], ICSS [International Carotid Stenting Study], and CREST [Carotid Revascularization Endarterectomy Versus Stenting Trial]) were used to validate 23 short-term outcome models to estimate stroke or death risk ≤30 days after the procedure and the original outcome measure for which the model was developed. Additionally, we validated 7 long-term outcome models for the original outcome measure. Predictive performance of the models was assessed with C statistics and calibration plots. Results- Stroke or death ≤30 days after the procedure occurred in 158 (7.2%) patients after carotid artery stenting and in 84 (3.7%) patients after carotid endarterectomy. Most models for short-term outcome after carotid artery stenting (n=4) or carotid endarterectomy (n=19) had poor discriminative performance (C statistics ranging from 0.49-0.64) and poor calibration with small absolute risk differences between the lowest and highest risk groups and overestimation of risk in the highest risk groups. Long-term outcome models (n=7) had a slightly better performance with C statistics ranging from 0.59 to 0.67 and reasonable calibration. Conclusions- Current models did not reliably predict outcome after carotid revascularization in a trial population of patients with symptomatic carotid stenosis. In particular, prediction of short-term outcome seemed to be difficult. Further external validation of existing prediction models or development of new prediction models is needed before such models can be used to support treatment decisions in individual patients.

Usefulness of cancer-free survival in estimating the lifetime attributable risk of cancer incidence from radiation exposure

Risk projection models estimating the lifetime cancer risk from radiation exposure are generally based on exposure dose, age at exposure, attained age, gender and study-population-specific factors such as baseline cancer risks and survival rates. Because such models have mostly been based on the Life Span Study cohort of Japanese atomic bomb survivors, the baseline risks and survival rates in the target population should be considered when applying the cancer risk. The survival function used in the risk projection models that are commonly used in the radiological protection field to estimate the cancer risk from medical or occupational exposure is based on all-cause mortality. Thus, it may not be accurate for estimating the lifetime risk of high-incidence but not life-threatening cancer with a long-term survival rate. Herein, we present the lifetime attributable risk (LAR) estimates of all solid cancers except thyroid cancer, thyroid cancer, and leukemia except chronic lymphocytic leukemia in South Korea for lifetime exposure to 1 mGy per year using the cancer-free survival function, as recently applied in the Fukushima health risk assessment by the World Health Organization. Compared with the estimates of LARs using an overall survival function solely based on all-cause mortality, the LARs of all solid cancers except thyroid cancer, and thyroid cancer evaluated using the cancer-free survival function, decreased by approximately 13% and 1% for men and 9% and 5% for women, respectively. The LAR of leukemia except chronic lymphocytic leukemia barely changed for either gender owing to the small absolute difference between its incidence and mortality. Given that many cancers have a high curative rate and low mortality rate, using a survival function solely based on all-cause mortality may cause an overestimation of the lifetime risk of cancer incidence. The lifetime fractional risk was robust against the choice of survival function.

Global burden of hypoglycaemia-related mortality in 109 countries, from 2000 to 2014: an analysis of death certificates

Aims/hypothesisIn the context of increasing prevalence of diabetes in elderly people with multimorbidity, intensive glucose control may increase the risk of severe hypoglycaemia, potentially leading to death. While rising trends of severe hypoglycaemia rates have been reported in some European, North American and Asian countries, the global burden of hypoglycaemia-related mortality is unknown. We aimed to investigate global differences and trends of hypoglycaemia-related mortality.MethodsWe used the WHO mortality database to extract information on death certificates reporting hypoglycaemia or diabetes as the underlying cause of death, and the United Nations demographic database to obtain data on mid-year population estimates from 2000 to 2014. We calculated crude and age-standardised proportions (defined as number of hypoglycaemia-related deaths divided by total number of deaths from diabetes [i.e. the sum of hypoglycaemia- and diabetes-related deaths]) and rates (hypoglycaemia-related deaths divided by mid-year population) of hypoglycaemia-related mortality and compared estimates across countries and over time.ResultsData for proportions were extracted from 109 countries (31 had data from all years analysed [2000–2014] available). Combining all countries, the age-standardised proportion of hypoglycaemia-related deaths was 4.49 (95% CI 4.44, 4.55) per 1000 total diabetes deaths. Compared with the overall mean, most Central American, South American and (mainly) Caribbean countries reported higher proportions (five more age-standardised hypoglycaemia-related deaths per 1000 total diabetes deaths in Chile, six in Uruguay, 11 in Belize and 22 in Aruba), as well as Japan (11 more age-standardised hypoglycaemia-related deaths per 1000 total diabetes deaths). In comparison, lower proportions were noted in most European countries, the USA, Canada, New Zealand and Australia. For countries with data available for all years analysed, trend analysis showed a 60% increase in hypoglycaemia-related deaths until 2010 and stable trends onwards. Rising trends were most evident for Argentina, Brazil, Chile, the USA and Japan. Data for rates were available for 105 countries (30 had data for all years analysed [2000–2014] available). Combining all countries, the age-standardised hypoglycaemia-related death rate was 0.79 (95% CI 0.77, 0.80) per 1 million person-years. Most Central American, South American and Caribbean countries similarly reported higher rates of hypoglycaemia-related death, whilst virtually all European countries, the USA, Canada, Japan, New Zealand and Australia reported lower rates compared with the overall mean. Age-standardised rates were very low for most countries (lower than five per 1 million person-years in 89.5% of countries), resulting in small absolute differences among countries. As noted with the proportions analysis, trend analysis showed an overall 60% increase in hypoglycaemia-related deaths until 2010 and stable rate trends onwards; rising rates were particularly evident for Brazil, Chile and the USA.Conclusions/interpretationMost countries in South America, Central America and the Caribbean showed the highest proportions of diabetes-related deaths attributable to hypoglycaemia and the highest rates of hypoglycaemia-related deaths. Between 2000 and 2014, rising trends were observed in Brazil, Chile and the USA for both rates and proportions of hypoglycaemia-related death, and in Argentina and Japan for proportions only. Further studies are required to unravel the contribution of clinical and socioeconomic factors, difference in diabetes prevalence and heterogeneity of death certification in determining lower rates and proportions of hypoglycaemia-related deaths in high-income countries in Europe, North America and Asia.Data availabilityData used for these analyses are available at https://doi.org/10.17632/ndp52fbz8r.1

Considerations in the determination of orientational order parameters from X-ray scattering experiments

ABSTRACTAn assessment of the data processing and analysis methods used to obtain the second- and fourth-rank orientational order parameters of liquid crystals from X-ray scattering experiments has been carried out, using experimental data from four extensively studied alkyl-cyanobiphenyls and calculated data generated from two general types of theoretical orientational distribution function. The application of a background subtraction and two different baseline correction methods to the scattering profiles is assessed, along with three different methods to analyse the processed data. The choice of baseline correction method is shown to have a significant effect: an offset to zero overestimates the order parameters from the experimental and calculated data sets, particularly for lower order parameters arising from broad distributions, whereas an offset to a value estimated from regions of low scattering intensity provides experimental values close to those reported from other experimental techniques. By contrast, the three different analysis methods are shown generally to result in relatively small absolute differences between the order parameters. We outline a straightforward general approach to experimental X-ray scattering data processing and analysis for uniaxial phases that results in order parameters that match well with those reported using other experimental techniques.

Vertebral Bone Mineral Density Measured by Quantitative Computed Tomography With and Without a Calibration Phantom: A Comparison Between 2 Different Software Solutions

Quantitative computed tomography (CT) can be used to quantify bone mineral density (BMD) in the spine from clinical CT scans. We aimed to determine agreement and precision of BMD measurements by 2 different methods: phantom-less internal tissue calibration and asynchronous phantom-based calibration in a cohort of patients with chronic kidney disease (CKD). Patients with CKD were recruited for CT angiography of the chest, abdomen, and pelvis. BMD was analyzed by 2 different software solutions using different calibration techniques; phantom-based by QCT Pro (Mindways Inc.) and phantom-less by Extended Brilliance Workspace (Philips Healthcare). Intraoperator reanalysis was performed on 53 patients (36%) for both methods. An interoperator reanalysis on 30 patients (20%) using the phantom-based method and 29 patients (19%) using the phantom-less method was made. XY- and Bland-Altman plots were used to evaluate method agreement. Phantom-based measured BMD was systematically higher than phantom-less measured BMD. Despite a small absolute difference of 3.3 mg/cm3 (CI: -0.2–6.9 mg/cm3) and a relative difference of 5.1% (CI: 2.2%–8.1%), interindividual differences were large, as seen by a wide prediction interval (PI: -47–40 mg/cm3). The Bland-Altman plot showed no systematic bias, apart from 5 outliers. Intraoperator variability was high for the phantom-less method (5.8%) compared to the phantom-based (0.8%) and the interoperator variability was also high for the phantom-less method (5.8%) compared to the phantom-based (1.8%). Despite high correlation between methods, the between-method difference on an individual level showed great variability. Our results suggest agreement between these 2 methods is insufficient to allow them to be used interchangeably in patients with CKD.

National Estimates of 30-Day Unplanned Readmissions of Patients on Maintenance Hemodialysis.

Patients on hemodialysis have high 30-day unplanned readmission rates. Using a national all-payer administrative database, we describe the epidemiology of 30-day unplanned readmissions in patients on hemodialysis, determine concordance of reasons for initial admission and readmission, and identify predictors for readmission. This is a retrospective cohort study using the Nationwide Readmission Database from the year 2013 to identify index admissions and readmission in patients with ESRD on hemodialysis. The Clinical Classification Software was used to categorize admission diagnosis into mutually exclusive clinically meaningful categories and determine concordance of reasons for admission on index hospitalizations and readmissions. Survey logistic regression was used to identify predictors of at least one readmission. During 2013, there were 87,302 (22%) index admissions with at least one 30-day unplanned readmission. Although patient and hospital characteristics were statistically different between those with and without readmissions, there were small absolute differences. The highest readmission rate was for acute myocardial infarction (25%), whereas the lowest readmission rate was for hypertension (20%). The primary reasons for initial hospitalization and subsequent 30-day readmission were discordant in 80% of admissions. Comorbidities that were associated with readmissions included depression (odds ratio, 1.10; 95% confidence interval [95% CI], 1.05 to 1.15; P<0.001), drug abuse (odds ratio, 1.41; 95% CI, 1.31 to 1.51; P<0.001), and discharge against medical advice (odds ratio, 1.57; 95% CI, 1.45 to 1.70; P<0.001). A group of high utilizers, which constituted 2% of the population, was responsible for 20% of all readmissions. In patients with ESRD on hemodialysis, nearly one quarter of admissions were followed by a 30-day unplanned readmission. Most readmissions were for primary diagnoses that were different from initial hospitalization. A small proportion of patients accounted for a disproportionate number of readmissions.

Radiometric validation of atmospheric correction for MERIS in the Baltic Sea based on continuous observations from ships and AERONET-OC

The Baltic Sea is a semi-enclosed sea that is optically dominated by coloured dissolved organic material (CDOM) and has relatively low sun elevation which makes accurate ocean colour remote sensing challenging in these waters. The high absorption, low scattering properties of the Baltic Sea are representative of other optically similar water bodies including the Arctic Ocean, Black Sea, coastal regions adjacent to the CDOM-rich estuaries such as the Amazon, and highly absorbing lakes where radiometric validation is essential in order to develop accurate remote sensing algorithms. Previous studies in this region mainly focused on the validation and improvement of standard Chlorophyll-a (Chl a) and attenuation coefficient (kd) ocean colour products. The primary input to derive these is the water-leaving radiance (Lw) or remote sensing reflectance (Rrs) and it is therefore fundamental to obtain the most accurate Lw or Rrs before deriving higher level products. To this end, the retrieval accuracy of Rrs from Medium Resolution Imaging Spectrometer (MERIS) imagery using six atmospheric correction processors was assessed through above-water measurements at two sites of the Aerosol Robotic Network for Ocean Colour (AERONET-OC; 363 measurements) and a shipborne autonomous platform from which the highest number of measurements were obtained (4986 measurements). The six processors tested were the CoastColour processor (CC), the Case 2 Regional processor for lakes (C2R-Lakes), the Case 2 Regional CoastColour processor (C2R-CC), the FUB/WeW water processor (FUB), the MERIS ground segment processor (MEGS) and POLYMER. All processors except for CC had small average absolute percentage differences (ψ) in the wavelength range from 490nm to 709nm (ψ<40%), while other bands had larger differences with ψ>60%. Compared to in situ values, the Rrs(709)/Rrs(665) band ratio had ψ<30% for all processors. The most accurate Rrs in the 490 to 709nm domain was obtained from POLYMER with ψ<30% and coefficients of determination (R2)>0.6. Using a score system based on all statistical tests, POLYMER scored highest, while C2R-CC, C2R-Lakes and FUB had lower scores. This study represents the largest data base of in situ Rrs, the most comprehensive analysis of AC models for highly absorbing waters and for MERIS, conducted to date. The results have implications for the new generation of Copernicus Sentinel ocean colour satellites.

Incremental Sampling Methodology: Applications for Background Screening Assessments.

This article presents the findings from a numerical simulation study that was conducted to evaluate the performance of alternative statistical analysis methods for background screening assessments when data sets are generated with incremental sampling methods (ISMs). A wide range of background and site conditions are represented in order to test different ISM sampling designs. Both hypothesis tests and upper tolerance limit (UTL) screening methods were implemented following U.S. Environmental Protection Agency (USEPA) guidance for specifying error rates. The simulations show that hypothesis testing using two-sample t-tests can meet standard performance criteria under a wide range of conditions, even with relatively small sample sizes. Key factors that affect the performance include unequal population variances and small absolute differences in population means. UTL methods are generally not recommended due to conceptual limitations in the technique when applied to ISM data sets from single decision units and due to insufficient power given standard statistical sample sizes from ISM.

External Validation of the New International Association for the Study of Lung Cancer Tumor, Node, and Metastasis 8th Edition Staging System and Updated T Descriptors in Determining Prognosis for Patients With Non–Small Cell Lung Cancer Patients With N3 Disease

ObjectiveThe objective of this study is to externally validate the 8th Edition of the Tumor, Node, and Metastasis staging system and its updated T descriptors in patients with non–small cell lung cancer with N3 disease. MethodsData were extracted from the Surveillance, Epidemiology, and End Results database. Chi-square test, Kaplan–Meier method, and Cox regression models were used in SPSS 23.0 (IBM Corp, Armonk, NY). ResultsA total of 7732 patients with non–small cell lung cancer with T1-4N3M0 disease from 1988 to 2013 were identified. A total of 1410 patients (18.2%) had T1N3 disease, 2491 patients (32.2%) had T2N3 disease, 1563 patients (20.2%) had T3N3 disease, and 2268 patients (29.3%) had T4N3 disease. The 5-year overall survival (OS) for the entire cohort was 8.4%. There was a significant difference in OS concerning T stage (T1N3: 10.8% vs. T2N3: 8.3% vs. T3N3: 8.1% vs. T4N3: 7.3%; P < .001). When stratified by the median age of patients (66 years), a significant difference in OS by stage of disease (IIIB vs. IIIC) was still observed in both the younger (P < .001) and older (P < .001) patient populations. A significant difference in disease-specific survival (DSS) was observed by T stage (T1N3: 14.7% vs. T2N3: 11.6% vs. T3N3: 11.3% vs. T4N3: 9.7%; P < .001). On multivariate analysis, T stage, year of diagnosis, age, gender, histology, and receipt of radiotherapy remained independent prognostic factors for both OS and DSS. ConclusionsThe 8th Edition of the Tumor, Node, and Metastasis staging system significantly stratifies both overall and DSS between stages IIIB and IIIC among those with N3 disease. However, small absolute differences in 5-year outcomes between T stage may suggest limited clinical relevance.

Post-Discharge Antibiotic Therapy in Patients with Acute Bacterial Skin and Skin Structure Infections

Background There are limited published data regarding the post-discharge treatment epidemiology for most infections, including acute bacterial skin and skin structure infections (ABSSSIs).Methods This is a national descriptive study of Veterans Affairs medical center admissions with diagnosis codes for ABSSSIs between January 1, 2005 and September 30, 2015. Patients receiving vancomycin during this admission were selected for inclusion. Treatment approaches after hospital discharge, including oral antibiotics, as well as intravenous (IV) medications administered in an outpatient clinic were assessed. Differences between oral and IV groups were assessed with χ 2 or Wilcoxon tests as appropriate.ResultsOf the 118,135 ABSSSI admissions, 114,352 (96.8%) patients continued antibiotic therapy after discharge. Most patients (98.5%) continued on oral therapy (median length of stay 4 days vs 6 days in IV group; P < 0.05). The most common oral therapies after discharge were sulfamethoxazole/trimethoprim (n = 30,220, 26.8%) and amoxicillin clavulanate (n = 21,819, 19.4%). The most common IV antibiotics were vancomycin (n = 740, 57.5%) and ceftriaxone (n = 220, 17.1%). Significant differences in demographics and comorbidities were observed between the oral and IV groups; however, absolute differences were negligible in some cases (e.g., median age 61 in oral group and 62 in IV group). Cultures were taken in less than half of patients (37.9% oral, 49.7% IV; P < 0.05), of which most were Staphylococcus aureus (81.7% oral, 78.4% IV; P < 0.05), with higher methicillin resistance in the IV group (59.1% oral, 63.7% IV; P < 0.05). In unadjusted comparisons between the oral and IV groups, the following 30-day outcomes differed significantly (P < 0.05): emergency room visit (21.6% oral, 45.3% IV), readmission (13.8% oral, 25.5% IV), and reinfection (45.7% oral, 54.3% IV).Conclusion Most patients with ABSSSI admissions continued antibiotic therapy after discharge, with only 1.1% receiving IV antibiotics in an outpatient clinic after discharge. Demographic characteristics and comorbidity burden were similar between the oral and IV groups; however, small absolute differences were statistically significant as this was a large cohort.Disclosures A. Caffrey, Merck: Grant Investigator, Research grant. The Medicines Company: Grant Investigator, Research grant. Pfizer: Grant Investigator, Research grant. K. LaPlante, Merck: Grant Investigator, Grant recipient. Pfizer: Grant Investigator, Grant recipient. Cempra: Scientific Advisor, Consulting fee. The Medicines Company: Grant Investigator, Grant recipient. Allergan: Scientific Advisor, Consulting fee. Bard/ Davol: Scientific Advisor, Consulting fee. Ocean Spray: Grant Investigator and Scientific Advisor, Consulting fee and Grant recipient. Zavante: Scientific Advisor, Consulting fee. Achaogen: Scientific Advisor, Consulting fee

Effect of Self-Listening on Self-Evaluation Accuracy

Abstract The purpose of this study was to determine the effect of self-listening on self-evaluation accuracy. Twenty-six undergraduate musicians self-evaluated their individual performances of a musical excerpt. Volunteer participants were randomly assigned to the experimental group or the control group. The experimental group members listened to the audio playback of their self-recorded performance before self-evaluating, while control group members did not. Both groups read information about self-evaluation before scoring their performances. Participants’ self-evaluations were compared to the mean scores of a panel of expert judges to determine the accuracy of the self-evaluations. The panel of judges listened to the recordings of the individual performances before completing assessments. The participants and the judges used a researcher-modified version of the Woodwind Brass Solo Evaluation Form to assess the individual performances. Self-accuracy was calculated as the absolute difference between students’ self-scores and the mean of the three judges’ scores; a small absolute difference indicated high self-evaluation accuracy. For purposes of data analyses, the 2 groups (self-listening and nonlistening) served as the independent variable, and the total self-evaluation accuracy score served as the dependent variable. A one-way, between subjects ANOVA revealed no main effect for treatment group on total self-evaluation accuracy, F(1, 24) = 3.08, p = .09, η2 = .11.