Correlating drug prescriptions with prognosis in severe COVID-19: first step towards resource management.

BACKGROUND: Optimal COVID-19 management is still undefined. In this complicated scenario, the construction of a computational model capable of extracting information from electronic medical records, correlating signs, symptoms and medical prescriptions, could improve patient management/prognosis. METHODS: The aim of this study is to investigate the correlation between drug prescriptions and outcome in patients with COVID-19. We extracted data from 3674 medical records of hospitalized patients: drug prescriptions, outcome, and demographics. The outcome evaluated was hospital outcome. We applied correlation analysis using a Logistic Regression algorithm for machine learning with Lasso and Matthews correlation coefficient. RESULTS: We found correlations between drugs and patient outcomes (death/discharged alive). Anticoagulants, used very frequently during all phases of the disease, were associated with good prognosis only after the first week of symptoms. Antibiotics very frequently prescribed, especially early, were not correlated with outcome, suggesting that bacterial infections may not be important in determining prognosis. There were no differences between age groups. CONCLUSIONS: In conclusion, we achieved an important result in the area of Artificial Intelligence, as we were able to establish a correlation between concrete variables in a real and extremely complex environment of clinical data from COVID-19. Our results are an initial and promising contribution in decision-making and real-time environments to support resource management and forecasting prognosis of patients with COVID-19.

Predicting the outcome for COVID-19 patients by applying time series classification to electronic health records.

BACKGROUND: COVID-19 caused more than 622 thousand deaths in Brazil. The infection can be asymptomatic and cause mild symptoms, but it also can evolve into a severe disease and lead to death. It is difficult to predict which patients will develop severe disease. There are, in the literature, machine learning models capable of assisting diagnose and predicting outcomes for several diseases, but usually these models require laboratory tests and/or imaging. METHODS: We conducted a observational cohort study that evaluated vital signs and measurements from patients who were admitted to Hospital das Clínicas (São Paulo, Brazil) between March 2020 and October 2021 due to COVID-19. The data was then represented as univariate and multivariate time series, that were used to train and test machine learning models capable of predicting a patient's outcome. RESULTS: Time series-based machine learning models are capable of predicting a COVID-19 patient's outcome with up to 96% general accuracy and 81% accuracy considering only the first hospitalization day. The models can reach up to 99% sensitivity (discharge prediction) and up to 91% specificity (death prediction). CONCLUSIONS: Results indicate that time series-based machine learning models combined with easily obtainable data can predict COVID-19 outcomes and support clinical decisions. With further research, these models can potentially help doctors diagnose other diseases.

The role of PNPLA3 and TM6SF2 polymorphisms on liver fibrosis and metabolic abnormalities in Brazilian patients with chronic hepatitis C.

BACKGROUND: Despite the growing body of knowledge about TM6SF2 and PNPLA3 polymorphisms in non-alcoholic fatty liver disease, their influence in the spectrum of HCV liver disease is not yet fully defined. Besides that, admixed populations, such as Brazilians, were not included in most of the studies. METHODS: This cross-sectional study enrolled 365 treatment-naïve patients with HCV and 134 healthy individuals. TM6SF2 (rs58542926 c.499C > T) and PNPLA3 (rs738409 c.444C > G) polymorphisms were evaluated regarding their association with clinical and laboratory data, histological liver steatosis and fibrosis, and with components of the metabolic syndrome. RESULTS: In HCV subjects, the frequencies of TM6SF2 CC and CT + TT were 89% and 11%, while PNPLA3 frequencies of CC and CG + GG were 51.4% and 48.6%. In the univariate logistic regression analysis, the TM6SF2 CT + TT genotype in HCV was associated with significant liver fibrosis (p = 0.047; OR 1.953; 95% CI 1.009-3.788). In comparison to the CT + TT genotype, the TM6SF2 CC genotype in HCV was associated with older age (p = 0.002), higher frequency of arterial hypertension (p = 0.032), obesity (p = 0.030), metabolic syndrome (p = 0.014) and lower total cholesterol levels (p = 0.036). The PNPLA3 GG subjects had lower body mass index than CG/ CC individuals (p = 0.047). None of the polymorphisms, or their combinations, was independently associated with hepatic steatosis or fibrosis. On the other hand, older age, lower serum levels of total cholesterol, and higher serum levels of alanine aminotransferase and alkaline phosphatase were associated with liver fibrosis in the multivariate logistic regression analysis. CONCLUSION: In this evaluation of an admixed HCV population, neither TM6SF2 nor PNPLA3 polymorphisms were independently associated with hepatic steatosis or fibrosis. Other factors seem more influential than these specific polymorphisms in isolation. More studies are warranted to clarify the role of the TM6SF2 and PNPLA3 polymorphisms in Brazilians with HCV.

The Molecular Characterization of Hepatitis A Virus Strains Circulating during Hepatitis A Outbreaks in São Paulo, Brazil, from September 2017 to May 2019.

Outbreaks of hepatitis A may occur in countries of medium and high socioeconomic levels in which the population generally exhibits an increased susceptibility in young adults to this infection if they are not vaccinated against the hepatitis A virus (HAV). In Europe, an outbreak involved approximately 22 European countries with 4475 cases reported from 2016 to 2018; most of them were men who have sex with men (MSM). This outbreak expanded to North and South America, including Brazil, particularly in São Paulo city with 1547 reported cases from 2016 to 2019. In the present study, we characterized the HAV strains involved in the acute hepatitis A cases identified in the reference centers of São Paulo city during this outbreak. A total of 51 cases with positive anti-HAV IgM were included, 80.4% male, 68.6% of them between 20 and 40 years old and 41.7% MSM. HAV RNA was detected in 92% (47/51) of the cases. Subgenotype IA of HAV was identified and most of the strains were closely related to that isolated in outbreaks that occurred in different European countries in 2016. These results showed the epidemiological relation between these outbreaks and reinforce the need to implement vaccination against hepatitis A for the adult population, particularly for a population with a high-risk behavior.

Tegumentary leishmaniasis mimicking visceralization in a cirrhotic patient: atypical cutaneous lesions and local immunological features.

Tegumentary leishmaniasis (TL) diagnosis is challenging due to the lack of a gold standard diagnostic tool. The diagnosis is significantly harder in regions where visceral leishmaniasis (VL) is also prevalent since immunological tests may present cross-reactivity. A cirrhotic patient from an endemic Brazilian region for TL and VL presented with atypical cutaneous lesions, a usual clinico-laboratory feature of VL (including a positive rk39 test result), but he was diagnosed with TL histopathologically; VL was ruled out by necropsy. Physicians working in co-prevalent areas should be aware of atypical features, unusual clinical course, and unexpected laboratory findings of leishmaniasis.

Tegumentary leishmaniasis mimicking visceralization in a cirrhotic patient: atypical cutaneous lesions and local immunological features

Abstract Tegumentary leishmaniasis (TL) diagnosis is challenging due to the lack of a gold standard diagnostic tool. The diagnosis is significantly harder in regions where visceral leishmaniasis (VL) is also prevalent since immunological tests may present cross-reactivity. A cirrhotic patient from an endemic Brazilian region for TL and VL presented with atypical cutaneous lesions, a usual clinico-laboratory feature of VL (including a positive rk39 test result), but he was diagnosed with TL histopathologically; VL was ruled out by necropsy. Physicians working in co-prevalent areas should be aware of atypical features, unusual clinical course, and unexpected laboratory findings of leishmaniasis.

Validation of PNPLA3 polymorphisms as risk factor for NAFLD and liver fibrosis in an admixed population.

INTRODUCTION AND AIM: Studies have shown that two polymorphisms were associated with steatosis and progression of non-alcoholic fatty liver disease (NAFLD) in different populations: the Patatin-like Phospholipase Domain Containing 3 (PNPLA3) and Transmembrane 6 Superfamily Member 2 (TM6SF2). However, the frequency and significance of these polymorphisms in an admixed population, i.e., Brazilian, is unknown. Therefore, we aimed to evaluate them in healthy subjects in comparison to patients with NAFLD. MATERIAL AND METHODS: This was a multicenter cross-sectional study in 248 patients with biopsy-proven NAFLD and in 134 healthy controls from two tertiary centers in Brazil. PNPLA3 (rs738409 c.444C>G) and TM6SF2 (rs58542926 c.449C>T) polymorphisms were evaluated. RESULTS: In controls, the frequencies of PNPLA3 CC and CG+GG were 49.25% and 50.74%, respectively; in NAFLD patients, this was 31.05% and 68.88% (p=0.0044, 95% CI 1.037-2.977). PNPLA3 GG subjects had an increased risk (3.29-fold) of having NAFLD when compared to CC subjects (p=0.0044, 95% CI 1.504-7.225). In patients with nonalcoholic steatohepatitis (NASH), PNPLA3 GG compared to CC was associated with higher AST levels [38.4±25.3 versus 36.7±40.1IU/L, p=0.0395)] and with the presence of liver fibrosis (≥F2 fibrosis, p=0.0272). TM6SF2 polymorphisms were not in Hardy-Weinberg equilibrium in our NAFLD group precluding further analysis. CONCLUSION: We demonstrated for the first time that PNPLA3 CG+GG increase the risk of NAFLD among Brazilian subjects. Moreover, PNPLA3 GG was associated with liver enzyme elevation and fibrosis in NASH patients.