Age-specific percentile-based prostate-specific antigen cutoff values predict the risk of prostate cancer: A single hospital observation.

Background: Testing for prostate-specific antigen (PSA) is often recommended for men with a potential risk of prostate cancer (PCa) before requiring advanced examination. However, the best PSA cutoff value remains controversial. Object: We compared the predictive performance of age-specific percentile-based PSA thresholds with a conventional cutoff of >4 ng/mL for the risk of PCa. Methods: We included men who received PSA measurements between 2003 and 2017 in a medical center in Taiwan. Logistic regression modeling was used to assess the association between age-specific percentile-based PSA thresholds and PCa risk in age subgroups. We further applied C-statistic and decision curve analysis to compare the predictive performance of age-specific percentile-based PSA with that of a conventional cutoff PSA. Results: We identified 626 patients with PCa and 40 836 patients without PCa. The slope of PSA in patients >60-year-old was almost 3 times that of those <60-year-old (0.713 vs 0.259). The risk effect sizes of the 75th percentile PSA cutoff (<60-year-old: 2.19; 60-70-year-old: 4.36; >70-year-old: 5.84 ng/mL) were comparable to those observed based on the conventional cutoff in all age groups. However, the discrimination performance of the 75th percentile PSA cutoff was better than that of the conventional cutoff among patients aged <60-year-old (C-statistic, 0.783 vs. 0.729, p < 0.05). The 75th percentile cutoffs also correctly identified an additional 2 patients with PCa for every 100 patients with PSA screening at the threshold probability of 20%. Conclusions: Our data support the use of the 75th percentile PSA cutoff to facilitate individualized risk assessment, particularly for patients aged <60-year-old.

Cardiothoracic ratio values and trajectories are associated with risk of requiring dialysis and mortality in chronic kidney disease.

BACKGROUND: The prognostic role of the cardiothoracic ratio (CTR) in chronic kidney disease (CKD) remains undetermined. METHODS: We conducted a retrospective cohort study of 3117 patients with CKD aged 18-89 years who participated in an Advanced CKD Care Program in Taiwan between 2003 and 2017 with a median follow up of 1.3(0.7-2.5) and 3.3(1.8-5.3) (IQR) years for outcome of end-stage renal disease (ESRD) and overall death, respectively. We developed a machine learning (ML)-based algorithm to calculate the baseline and serial CTRs, which were then used to classify patients into trajectory groups based on latent class mixed modelling. Association and discrimination were evaluated using multivariable Cox proportional hazards regression analyses and C-statistics, respectively. RESULTS: The median (interquartile range) age of 3117 patients is 69.5 (59.2-77.4) years. We create 3 CTR trajectory groups (low [30.1%], medium [48.1%], and high [21.8%]) for the 2474 patients with at least 2 CTR measurements. The adjusted hazard ratios for ESRD, cardiovascular mortality, and all-cause mortality in patients with baseline CTRs ≥0.57 (vs CTRs <0.47) are 1.35 (95% confidence interval, 1.06-1.72), 2.89 (1.78-4.71), and 1.50 (1.22-1.83), respectively. Similarly, greater effect sizes, particularly for cardiovascular mortality, are observed for high (vs low) CTR trajectories. Compared with a reference model, one with CTR as a continuous variable yields significantly higher C-statistics of 0.719 (vs 0.698, P = 0.04) for cardiovascular mortality and 0.697 (vs 0.693, P < 0.001) for all-cause mortality. CONCLUSIONS: Our findings support the real-world prognostic value of the CTR, as calculated by a ML annotation tool, in CKD. Our research presents a methodological foundation for using machine learning to improve cardioprotection among patients with CKD.

An enlarged heart occurs during various medical conditions and can result in early death. However, it is unclear whether this is also the case in patients with chronic kidney disease (CKD). Although the size of the heart can be measured on chest X-rays, this process is time consuming. We used artificial intelligence to quantify the heart size of 3117 CKD patients based on their chest X-rays within hours. We found that CKD patients with an enlarged heart were more likely to develop end-stage kidney disease or die. This could improve monitoring of CKD patients with an enlarged heart and improve their care.

Association of poorly controlled HbA1c with increased risk of progression to end-stage kidney disease and all-cause mortality in patients with diabetes and chronic kidney disease.

Glycosylated hemoglobin (HbA1c) targets for patients with chronic kidney disease (CKD) and type 2 diabetes remain controversial. To evaluate whether baseline HbA1c and HbA1c trajectories are associated with the risk of end-stage kidney disease (ESKD) and all-cause mortality, we recruited adult patients with CKD and type 2 diabetes from a "Pre-ESKD Program" at a medical center in Taiwan from 2003 to 2017. Group-based trajectory modeling was performed to identify distinct patient groups that contained patients with similar longitudinal HbA1c patterns. Cox proportional hazard models were used to estimate hazard ratios (HRs) of ESKD and mortality associated with baseline HbA1c levels and HbA1c trajectories. In the analysis related to baseline HbA1c (n = 4543), the adjusted HRs [95% confidence interval (CI)] of all-cause mortality were 1.06 (0.95-1.18) and 1.25 (95% CI, 1.07-1.46) in patients with an HbA1c level of 7%-9% (53-75 mmol/mol) and >9% (>75 mmol/mol), respectively, as compared with those with an HbA1c level < 7% (<53 mmol/mol). In the trajectory analysis (n = 2692), three distinct longitudinal HbA1c trajectories were identified: nearly optimal (55.9%), moderate to stable (34.2%), and poor control (9.9%). Compared with the "nearly optimal" HbA1c trajectory group, the "moderate-to-stable" group did not have significantly higher mortality, but the "poorly controlled" group had 35% higher risk of mortality (adjusted HR = 1.35, 95% CI = 1.06-1.71). Neither baseline levels of HbA1c nor trajectories were associated with ESKD risk. In conclusion, in patients with CKD and type 2 diabetes, poor glycemic control was associated with an elevated risk of mortality but not associated with a risk of progression to ESKD.

Application of machine learning methods for the prediction of true fasting status in patients performing blood tests.

The fasting blood glucose (FBG) values extracted from electronic medical records (EMR) are assumed valid in existing research, which may cause diagnostic bias due to misclassification of fasting status. We proposed a machine learning (ML) algorithm to predict the fasting status of blood samples. This cross-sectional study was conducted using the EMR of a medical center from 2003 to 2018 and a total of 2,196,833 ontological FBGs from the outpatient service were enrolled. The theoretical true fasting status are identified by comparing the values of ontological FBG with average glucose levels derived from concomitant tested HbA1c based on multi-criteria. In addition to multiple logistic regression, we extracted 67 features to predict the fasting status by eXtreme Gradient Boosting (XGBoost). The discrimination and calibration of the prediction models were also assessed. Real-world performance was gauged by the prevalence of ineffective glucose measurement (IGM). Of the 784,340 ontologically labeled fasting samples, 77.1% were considered theoretical FBGs. The median (IQR) glucose and HbA1c level of ontological and theoretical fasting samples in patients without diabetes mellitus (DM) were 94.0 (87.0, 102.0) mg/dL and 5.6 (5.4, 5.9)%, and 92.0 (86.0, 99.0) mg/dL and 5.6 (5.4, 5.9)%, respectively. The XGBoost showed comparable calibration and AUROC of 0.887 than that of 0.868 in multiple logistic regression in the parsimonious approach and identified important predictors of glucose level, home-to-hospital distance, age, and concomitantly serum creatinine and lipid testing. The prevalence of IGM dropped from 27.8% based on ontological FBGs to 0.48% by using algorithm-verified FBGs. The proposed ML algorithm or multiple logistic regression model aids in verification of the fasting status.

Insomnia and Poor Sleep in CKD: A Systematic Review and Meta-analysis.

Rationale & Objective: Poor sleep quality and insomnia are pervasive among patients with advanced chronic kidney disease (CKD); however, these health issues have not been systematically evaluated. Study Design: Systematic review and meta-analysis. Setting & Study Populations: Adult patients with CKD not receiving kidney replacement therapy (KRT), as well as adults receiving KRT, including hemodialysis, peritoneal dialysis, and kidney transplantation. Selection Criteria for Studies: A systematic literature search using PubMed, Embase, and PsycNET, was conducted for articles published between January 1, 1990, and September 28, 2018. Data Extraction: Data on the prevalences of poor sleep quality and insomnia in patients with CKD, including those receiving and not receiving KRT, were extracted. Analytical Approach: Pooled prevalences were estimated using a random-effects meta-analysis and were stratified according to age, CKD stage, World Health Organization region, risk of bias, Pittsburgh Sleep Quality Index score, and the different criteria for insomnia that were used at diagnosis. Results: Of 3,708 articles, 93 were selected, and significant methodological heterogeneity was present. The pooled prevalences of poor sleep quality for CKD without KRT, hemodialysis, peritoneal dialysis, and kidney transplantation were 59% (95% CI, 44%-73%), 68% (95% CI, 64%-73%), 67% (95% CI, 44%-86%), and 46% (95% CI, 34%-59%), respectively. The corresponding prevalences of insomnia were 48% (95% CI, 30%-67%), 46% (95% CI, 39%-54%), 61% (95% CI, 41%-79%), and 26% (95% CI, 9%-49%), respectively. Insomnia was significantly more prevalent among patients aged 51-60 years and those aged >60 years than among those aged <50 years. The prevalence of insomnia in the European region was the lowest of all World Health Organization regions. Limitations: High interstudy heterogeneity. Conclusions: Approximately half of the patients with advanced CKD had poor sleep quality or insomnia, and the prevalence was even higher among those who received KRT. Kidney transplantation may reduce the burden of poor sleep quality and insomnia.

Electronic medical record-based deep data cleaning and phenotyping improve the diagnostic validity and mortality assessment of infective endocarditis: medical big data initiative of CMUH.

BACKGROUND: International Classification of Diseases (ICD) code-based claims databases are often used to study infective endocarditis (IE). However, the quality of ICD coding can influence the reliability of IE research. The impact of complementing the ICD-only approach with data extracted from electronic medical records (EMRs) has yet to be explored. METHODS: We selected the information of adult patients with discharge ICD codes for IE (ICD-9: 421, 112.81, 036.42, 098.84, 115.04, 115.14, 115.94, 424.9; ICD-10: I33, I38, I39) during 2005-2016 in China Medical University Hospital. Data extraction was conducted on the basis of the modified Duke criteria to establish a reference group comprising patients with definite or possible IE. Clinical characteristics and in-hospital mortality were compared between ICD-identified and Duke-confirmed cases. The positive predictive value (PPV) was used to quantify the IE identification performance of various phenotyping algorithms. RESULTS: A total of 593 patients with discharge ICD codes for IE were identified, only 56.7% met the modified Duke criteria. The crude in-hospital mortality for Duke-confirmed and Duke-rejected IE were 24.4% and 8.2%, respectively. The adjusted in-hospital mortality for ICD-identified IE was lower than that for Duke-confirmed IE by a difference of 5.1%. The best PPV was achieved (0.90, 95% CI 0.86-0.93) when major components of the Duke criteria (positive blood culture and vegetation) were integrated with ICD codes. CONCLUSION: Integrating EMR data can considerably improve the accuracy of ICD-only approaches in phenotyping IE, which can improve the validity of EMR-based studies and their applications, including real-time surveillance and clinical decision support.