Hepatitis C Virus positivity prediction from serum samples using NIRS and L1-penalized classification.

AIMS: The hepatitis C virus (HCV) has developed a strategy to coexist with its host resulting in varying degrees of tissue and cell damage, which generate different pathological phenotypes, such as varying degrees of fibrosis, cirrhosis, and hepatocellular carcinoma (HCC). However, there is no integrated information that can predict the evolutionary course of the infection. We propose to combine Near-infrared spectroscopy (NIRS) and machine learning techniques to provide a predictive model. In this work, we propose to discriminate HCV positivity in biobank patient serum samples. METHODS: 126 serum samples from 38 HCV patients in different stages of the disease were obtained from the Biobank of Hospital Universitario Fundación Alcorcon. NIRS spectrum was captured by a FT-NIRS Spectrum 100 (Perkin Elmer) device in reflectance mode. For each patient, the HCV positivity was identified (PCR) and labeled as detectable =1 and undetectable =0. We propose an L1-penalized logistic regression model to classify each spectrum as positive (1) or negative (0) for HCV presence (x). The regularization parameter is selected using 5- fold cross-validation. The penalized model will induce sparsity in the solution so that only a few relevant wavelengths will be different from zero. RESULTS: L1-penalized logistic regression model provided 167 wavelengths different from zero. The accuracy on an independent test set was 0.78. CONCLUSIONS: We present a straightforward promising approach to detect HCV positivity from patient serum samples combining NIRS and machine learning techniques. This result is encouraging to predict HCV progression, among other applications. Clinical relevance- We presented a simple while promising approach to use machine learning and NIRS to analyze viral presence on sample serums.

Predictive modeling of hypophosphatasia based on a case series of adult patients with persistent hypophosphatasemia.

Approximately half of individuals with hypophosphatasemia (low levels of serum alkaline phosphatase) have hypophosphatasia, a rare genetic disease in which patients may have stress fractures, bone and joint pain, or premature tooth loss. We developed a predictive model based on specific biomarkers of this disease to better diagnose this condition. INTRODUCTION: Hypophosphatasemia is a condition in which low levels of alkaline phosphatase (ALP) are detected in the serum. Some individuals presenting with this condition may have a rare genetic disease called hypophosphatasia (HPP), which involves mineralization of the bone and teeth. Lack of awareness of HPP and its nonspecific symptoms make this genetic disease difficult to diagnose. We developed a predictive model based on biomarkers of HPP such as ALP and pyridoxal 5'-phosphate (PLP), because clinical manifestations sometimes are not recognized as symptoms of HPP. METHODS: We assessed 325,000 ALP results between 2010 and 2015 to identify individuals suspected of having HPP. We performed univariate and multivariate analyses to characterize the relationship between hypophosphatasemia and HPP. Using several machine learning algorithms, we developed several models based on biomarkers and compared their performance to determine the best model. RESULTS: The final cohort included 45 patients who underwent a genetic test. Half (23 patients) showed a mutation of the ALPL gene that encodes the tissue-nonspecific ALP enzyme. ALP (odds ratio [OR] 0.61, 95% confidence interval [CI] 0.3-0.8, p = 0.01) and PLP (OR 1.06, 95% CI 1.01-1.15, p = 0.04) were the only variables significantly associated with the presence of HPP. Support vector machines and logistic regression were the machine learning algorithms that provided the best predictive models in terms of classification (area under the curve 0.936 and 0.844, respectively). CONCLUSIONS: Given the high probability of a misdiagnosis, its nonspecific symptoms, and a lack of awareness of serum ALP levels, it is difficult to make a clinical diagnosis of HPP. Predictive models based on biomarkers are necessary to achieve a proper diagnosis. Our proposed machine learning approaches achieved reasonable performance compared to traditional statistical methods used in biomedicine, increasing the likelihood of properly diagnosing such a rare disease as HPP.

[Prevalence and clinical features of non-alcoholic steatohepatitis in a hypertensive population]. / Prevalencia y características clínicas de la esteatohepatitis no alcohólica en una población con hipertensión arterial esencial.

OBJECTIVE: Non-alcoholic fatty liver is a chronic liver disease in which fat is deposited in the liver, causing an inflammation called non-alcoholic steatohepatitis (NASH), and fibrosis. NASH is associated with metabolic syndrome (MS) and other cardiovascular risk factors. The aim of this study was to analyse the epidemiological features of NASH within a hypertensive population, with a high prevalence of MS, and to determine the features related to NASH. METHODS: The computerised records were collected from 3,473 patients from Mostoles University Hospital's Hypertension Unit in order to perform a retrospective, cross-sectional study. NASH was considered as ultrasound-detected fatty liver disease along with serum levels of alanine aminotransferase or aspartate aminotransferase 1.5 times above normal values, having ruled out other causes of liver disease: alcohol abuse, autoimmune hepatitis, drug toxicity, virus and hemochromatosis. A univariate, multivariate, and ANOVA analysis were performed to assess the effect of the studied features on the response of interest. RESULTS: The cohort included 2,242 patients (51.3% men). NASH was present in 255 patients (11.4%) of whom 71% were men. MS was detected in 52.6% of patients (69.4% in the NASH group, and 50.5% in the non-NASH group, P=.001). Prevalence of type 2 diabetes mellitus was 11.5% (16.5% in the NASH group, and 10.8% in the non-NASH group, P=.01). In a multivariate analysis, waist circumference, MS, body mass index, type 2 diabetes mellitus, age, fasting serum insulin, and serum ferritin were associated with NASH. ANOVA revealed that NASH and transaminases were also significantly associated with components of metabolic syndrome. CONCLUSIONS: In the population studied, MS, type 2 diabetes mellitus, and several components of MS were independently associated with NASH. Therefore, NASH can be considered as the liver manifestation of MS in patients with arterial hypertension.

Cystatin C as a predictor of cardiovascular outcomes in a hypertensive population.

Calculating the estimated glomerular filtration rate (eGFR) using creatinine-based equations may underestimate cardiovascular risk. Cystatin C-based eGFR may be a stronger prognostic biomarker than creatinine-based eGFR when assessing cardiovascular outcomes and mortality. Our aim was to determine whether levels of serum cystatin C, as an estimator of GFR, had a higher predictive value than creatinine-based eGFR for incident cardiovascular disease among hypertensive patients. We retrospectively analyzed the records of 2016 hypertensive patients from the Hypertension Unit at Mostoles University Hospital between 2006 and 2016. We calculated the eGFR using 3 CKD-EPI equations. The outcomes we included in our study were cardiovascular death, non-cardiovascular death, stroke, incident heart failure, and myocardial infarction. We used the Cox regression hazard model to estimate the hazard ratio. Our analysis found that, in terms of cardiovascular morbidity and mortality, both cystatin C-based eGFR (HR 2.88, 95% CI 1.86-4.47, P<0.0001) showed a higher prognostic value than creatinine-based eGFR (HR 2.83, 95% CI 1.73-4.63, P<0.0001). In terms of all-cause mortality, cystatin C-based eGFR (HR 4.24, 95% CI 2.38-7.53, P<0.0001) showed a higher prognostic value than creatinine-based eGFR (HR 2.77, 95% CI 1.43-5.36, P<0.0001). Our findings suggest that both cystatin C-based eGFRs may be stronger predictors of all-cause mortality and cardiovascular events in our hypertensive cohort when compared to creatinine-based eGFR, and may improve the risk assessment in certain populations.

[Uricemia and metabolic syndrome in patients with arterial hypertension]. / Uricemia y síndrome metabólico en pacientes con hipertensión arterial.

OBJECTIVE: Serum urate levels have been associated with metabolic syndrome (MS). However, the relationship between these two variables in patients with essential arterial hypertension has not been studied. PATIENTS AND METHODS: A Cross-sectional study in 592 patients with essential hypertension. The MS was defined according to the ATP-III criteria. We excluded patients with hypouricemic treatment. RESULTS: The prevalence of MS was 52% (95% CI, 48-56%) and there was a graded increase with increasing serum urate (uricemia ≤ 4.7 mg/dl, 36%; uricemia ≥ 6.8 mg/dl, 70%, P < 0.001). Hypertensive patients with MS showed a higher mean uricemia than those without this comorbidity (6.1 ± 1.5 mg/dl versus 5.4 ± 1.3 mg/dl, P < 0.0001). The prevalence of hyperuricemia (men, > 7.0 mg/dL; women, > 6.0 mg/dL) in hypertensive patients without diuretic treatment, was 24% (in those with MS 40% versus 11% without MS). In multivariate analysis, triglycerides (OR = 1.008, CI 95%: 1.004-1.012, P < 0.001) and body mass index (BMI) (OR = 1.118, CI 95%: 1.059-1.181, P < 0.001), were independent predictors of serum uric acid levels. CONCLUSIONS: In patients with essential hypertension, about half have MS and one out of four has hyperuricemia. The most important determinant of hyperuricemia is BMI.