Milton Henriques Guimarães-Júnior

Cirrhosis has been pointed out as a clinical entity that leads to worse clinical prognosis in COVID-19 patients. However, this concept is controversial in the literature. We aimed to evaluate clinical outcomes by comparing patients with cirrhosis to those without cirrhosis in a Brazilian cohort. Data from 20,164 COVID-19 inpatients were collected from 41 hospitals in Brazil between March to September 2020 and March 2021 to August 2022. We compared 117 patients with cirrhosis to 632 matched controls. A propensity score model was used to adjust for potential confounding variables, incorporating some predictors: age, sex at birth, number of comorbidities, hospital of admission, whether it was an in-hospital clinical manifestation of COVID-19, and admission year. Closeness was defined as being within 0.16 standard deviations of the logit of the propensity score. The median age was 61 (IQR 50–70) years old, and 63.4% were men. There were no significant differences in the self-reported symptoms. Patients with cirrhosis had lower median hemoglobin levels (10.8 vs. 13.1 g/dl), lower platelets (127,000 vs. 200,000 cells/mm3), and leukocyte counts, as well as lower median C-reactive protein (63.0 vs. 76.0 p = 0.044) when compared to controls. They also had higher mortality compared to matched controls (51.3% vs. 21.7%, p < 0.001). They also had higher frequencies of admission in an intensive care unit (51.3% vs. 38.0%, p = 0.007), invasive mechanical ventilation (43.9% vs. 26.6%, p < 0.001), dialysis (17.9% vs. 11.1%, p = 0.038), septic shock (23.9% vs. 14.9%; p = 0.015) and institution of palliative care (19.7% vs. 7.4%; p < 0.001). This study has shown that COVID-19 inpatients with cirrhosis had significantly higher incidence of severe outcomes, as well as higher frequency of institution of palliative care when compared to matched controls. Our findings underscore the need for these patients to receive particular attention from healthcare teams and allocated resources.

The Covid-19 pandemic caused a negative impact on other infectious diseases control, prevention, and treatment. Consequently, low and middle-income countries suffer from other endemic diseases, such as tuberculosis. This study was designed to compare Covid-19 manifestations and outcomes between patients with previously treated tuberculosis and controls without this condition. We performed a matched case-control study drawn from the Brazilian Covid-19 Registry data, including in-hospital patients aged 18 and over with laboratory-confirmed Covid-19 from March 1, 2020, to March 31, 2022. Cases were patients with a past history of tuberculosis. Controls were Covid-19 patients without a tuberculosis history. Patients were matched by hospital, sex, presence of HIV, and number of comorbidities, with a 1:4 ratio. Of 13,636 patients with laboratory-confirmed diagnoses of Covid-19 enrolled in this study, 80 had a history of tuberculosis. Statistical differences in history of chronic pulmonary obstructive disease (15% vs. 3.2%), psychiatric disease (10% vs. 3.5%,), chronic kidney disease (11.2% vs. 2.8%), and solid-organ transplantation; (5% vs. 0.9%, p < 0.05 for all) were higher in patients with a past history of tuberculosis. Prior use of inhalatory medications (5% vs. 0.6%,), oral corticoids (8.8% vs. 1.9%), immunosuppressants (8.8% vs. 1.9%,) and the use of illicit drugs were more common in the case group (6.2% vs. 0.3% p < 0.05for all). There were no significant differences in in-hospital mortality, mechanical ventilation, need for dialysis, and ICU admission. Patients with a history of tuberculosis infection presented a higher frequency of use of illicit drugs, chronic pulmonary obstructive disease, psychiatric disease, chronic kidney disease, solid-organ transplantation, prior use of inhalatory medications, oral corticoids, and immunosuppressants. The outcomes were similar between cases and controls.

Background: Proper analysis and interpretation of health care data can significantly improve patient outcomes by enhancing services and revealing the impacts of new technologies and treatments. Understanding the substantial impact of temporal shifts in these data is crucial. For example, COVID-19 vaccination initially lowered the mean age of at-risk patients and later changed the characteristics of those who died. This highlights the importance of understanding these shifts for assessing factors that affect patient outcomes. Objective: This study aims to propose detection, initial characterization, and semantic characterization (DIS), a new methodology for analyzing changes in health outcomes and variables over time while discovering contextual changes for outcomes in large volumes of data. Methods: The DIS methodology involves 3 steps: detection, initial characterization, and semantic characterization. Detection uses metrics such as Jensen-Shannon divergence to identify significant data drifts. Initial characterization offers a global analysis of changes in data distribution and predictive feature significance over time. Semantic characterization uses natural language processing–inspired techniques to understand the local context of these changes, helping identify factors driving changes in patient outcomes. By integrating the outcomes from these 3 steps, our results can identify specific factors (eg, interventions and modifications in health care practices) that drive changes in patient outcomes. DIS was applied to the Brazilian COVID-19 Registry and the Medical Information Mart for Intensive Care, version IV (MIMIC-IV) data sets. Results: Our approach allowed us to (1) identify drifts effectively, especially using metrics such as the Jensen-Shannon divergence, and (2) uncover reasons for the decline in overall mortality in both the COVID-19 and MIMIC-IV data sets, as well as changes in the cooccurrence between different diseases and this particular outcome. Factors such as vaccination during the COVID-19 pandemic and reduced iatrogenic events and cancer-related deaths in MIMIC-IV were highlighted. The methodology also pinpointed shifts in patient demographics and disease patterns, providing insights into the evolving health care landscape during the study period. Conclusions: We developed a novel methodology combining machine learning and natural language processing techniques to detect, characterize, and understand temporal shifts in health care data. This understanding can enhance predictive algorithms, improve patient outcomes, and optimize health care resource allocation, ultimately improving the effectiveness of machine learning predictive algorithms applied to health care data. Our methodology can be applied to a variety of scenarios beyond those discussed in this paper.

Background: Although dementia has emerged as an important risk factor for severe SARS-CoV-2 infection, results on COVID-19-related complications and mortality are not consistent. We examined the clinical presentations and outcomes of COVID-19 in a multicentre cohort of in-hospital patients, comparing those with and without dementia. Methods: This retrospective observational study comprises COVID-19 laboratory-confirmed patients aged ≥ 60 years admitted to 38 hospitals from 19 cities in Brazil. Data were obtained from electronic hospital records. A propensity score analysis was used to match patients with and without dementia (up to 3:1) according to age, sex, comorbidities, year, and hospital of admission. Our primary outcome was in-hospital mortality. We also assessed admission to the intensive care unit (ICU), invasive mechanical ventilation (IMV), kidney replacement therapy (KRT), sepsis, nosocomial infection, and thromboembolic events. Results: Among 1,556 patients included in the study, 405 (4.5%) had a diagnosis of dementia and 1,151 were matched controls. When compared to matched controls, patients with dementia had a lower frequency of dyspnoea, cough, myalgia, headache, ageusia, and anosmia; and higher frequency of fever and delirium. They also had a lower frequency of ICU admission (32.7% vs. 47.1%, p < 0.001) and shorter ICU length of stay (7 vs. 9 days, p < 0.026), and a lower frequency of sepsis (17% vs. 24%, p = 0.005), KRT (6.4% vs. 13%, p < 0.001), and IVM (4.6% vs. 9.8%, p = 0.002). There were no differences in hospital mortality between groups. Conclusion: Clinical manifestations of COVID-19 differ between older inpatients with and without dementia. We observed that dementia alone could not explain the higher short-term mortality following severe COVID-19. Therefore, clinicians should consider other risk factors such as acute morbidity severity and baseline frailty when evaluating the prognosis of older adults with dementia hospitalised with COVID-19.

Resumo Fundamento Apesar da ausência de evidência mostrando benefícios da hidroxicloroquina e da cloroquina combinadas ou não à azitromicina no tratamento da covid-19, esses medicamentos têm sido amplamente prescritos no Brasil. Objetivos Avaliar desfechos, incluindo moralidade hospitalar, alterações eletrocardiográficas, tempo de internação, admissão na unidade de terapia intensiva, e necessidade de diálise e de ventilação mecânica em pacientes hospitalizados com covid-19 que receberam cloroquina ou hidroxicloroquina, e comparar os desfechos entre aqueles pacientes e seus controles pareados. Métodos Estudo multicêntrico [...] Resumo Fundamento Apesar da ausência de evidência mostrando benefícios da hidroxicloroquina e da cloroquina combinadas ou não à azitromicina no tratamento da covid-19, esses medicamentos têm sido amplamente prescritos no Brasil. Objetivos Avaliar desfechos, incluindo moralidade hospitalar, alterações eletrocardiográficas, tempo de internação, admissão na unidade de terapia intensiva, e necessidade de diálise e de ventilação mecânica em pacientes hospitalizados com covid-19 que receberam cloroquina ou hidroxicloroquina, e comparar os desfechos entre aqueles pacientes e seus controles pareados. Métodos Estudo multicêntrico

Comparisons among the different vaccines against SARS-CoV-2 are important to understand which type of vaccine provides more protection. This study aimed to evaluate the real-life efficacy through symptomatic infection and the humoral response of six different vaccines against SARS-CoV-2—BNT162b2, mRNA-1273, ChAdOx1-S, CoronaVac, Ad26.COV2, and Ad5-nCoV. This multicentric observational longitudinal study involved hospitals from Mexico and Brazil in which volunteers who received complete vaccination schemes were followed for 210 days after the last dose. SARS-CoV-2 Spike 1–2 IgG levels were taken before receiving the first vaccine, 21 days after each dose, and the last sample at six months (+/−1 month) after the last dose. A total of 1132 individuals exposed to five COVID-19 waves were included. All vaccines induced humoral responses, and mRNA vaccines had the highest antibody levels during follow-up. At six months, there was a decline in the SARS-CoV-2 Spike 1–2 IgG antibody titers of 69.5% and 36.4% in subjects with negative and positive history of infection respectively. Infection before vaccination and after complete vaccination scheme correlated with higher antibody titers. The predictors of infection were vaccination with CoronaVac compared to BNT162b2 and ChAdOx1-S. In the presence of comorbidities such as diabetes, rheumatoid arthritis, or dyslipidemia, CoronaVac lowered the risk of infection.

The majority of early prediction scores and methods to predict COVID-19 mortality are bound by methodological flaws and technological limitations (e.g., the use of a single prediction model). Our aim is to provide a thorough comparative study that tackles those methodological issues, considering multiple techniques to build mortality prediction models, including modern machine learning (neural) algorithms and traditional statistical techniques, as well as meta-learning (ensemble) approaches. This study used a dataset from a multicenter cohort of 10,897 adult Brazilian COVID-19 patients, admitted from March/2020 to November/2021, including patients [median age 60 (interquartile range 48–71), 46% women]. We also proposed new original population-based meta-features that have not been devised in the literature. Stacking has shown to achieve the best results reported in the literature for the death prediction task, improving over previous state-of-the-art by more than 46% in Recall for predicting death, with AUROC 0.826 and MacroF1 of 65.4%. The newly proposed meta-features were highly discriminative of death, but fell short in producing large improvements in final prediction performance, demonstrating that we are possibly on the limits of the prediction capabilities that can be achieved with the current set of ML techniques and (meta-)features. Finally, we investigated how the trained models perform on different hospitals, showing that there are indeed large differences in classifier performance between different hospitals, further making the case that errors are produced by factors that cannot be modeled with the current predictors.