key: cord-1046072-d0pv1hcm authors: FERNANDES, A. T.; Rodrigues, E.; Araujo, E.; Formiga, M. F.; Sa Horan, P.; Ferreira, A. B.; Barbosa, H.; Barbosa, P. title: Risk factors and survival in patients with COVID-19 in northeastern Brazil. date: 2022-01-27 journal: nan DOI: 10.1101/2022.01.27.22269602 sha: 98d7f91cdfd61cb0192db540bb349642455fed2b doc_id: 1046072 cord_uid: d0pv1hcm BACKGROUND: Countries have focused research on developing strategies to fight COVID-19, prevent hospitalizations, and maintain economic activities. OBJECTIVE: This study aimed to establish a survival analysis and identify risk factors for patients with COVID-19 in a upper middle-income city in Brazil. METHODS: We performed a retrospective cohort study with 280 hospitalized patients with COVID-19. The eCOVID platform provided data used to monitor COVID-19 cases and help communication between professionals. RESULTS: Survival analysis showed that age [≥] 65 years was associated with decreased survival (54.8%). Females had lower survival rate than males (p=0.01). Regarding risk factors, urea concentration (p<0.001), hospital LOS (p=0.002), oxygen concentration (p=0.005), and age (p=0.02) were associated with death. CONCLUSION: Age, hospital LOS, high blood urea concentration, and low oxygen concentration were associated with death by COVID-19 in the studied population. These findings corroborate with studies conducted in research centers worldwide. A transmissible disease caused by the SARS-CoV-2 virus increased the number of severe acute respiratory syndrome (SARS) cases worldwide since December 2019. Although Efforts were made to determine prevalence and factors associated with severity of COVID-19 1, 2 . To date, more than 224 million people have been infected, and more than four million died. The United States of America and Brazil correspond to 71% of cumulative cases of COVID-19 in the Americas (more than 46 million cases) 3 . Many countries focused their research on defining and identifying strategies to fight the disease, prevent hospitalizations, and maintain economic activities [4] [5] [6] . One critical point of the disease is its form of presentation, varying from asymptomatic and very mild to critical symptoms. Symptoms may also persist even after the acute phase, and individuals who initially did not report or had mild symptoms may evolve to sudden health decline or death 4, 7, 8 . Identifying factors that predispose to high risk of hospitalization and death contributes to preventive measures, especially in developing countries facing difficulties in establishing early diagnosis 8 . Therefore, this study describes clinical and demographic characteristics, comorbidities, outcomes, survival and factors associated with mortality of hospitalized patients with COVID-19 in an upper-middle-income city of Brazil. . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. We included hospitalized patients of both sexes, aged 18 or over, diagnosed with COVID-19 using RT-PCR test (nasopharynx swab) 3 , with epidemiological history of COVID-19 and/or relevant clinical symptoms and serological parameters 9, 10 . Patients whose COVID-19 diagnosis was discarded or outcomes were not recorded were excluded. The eCOVID platform was developed to store patient data (e.g., symptoms, comorbidities, vital signs, and laboratory and imaging exams) and facilitate communication among professionals treating patients with COVID-19. The platform allows remote assistance/communication among professionals working in the public Brazilian Unified Health System and the private sector. It also provides a database for epidemiological studies in the city where the study was conducted. Variables and data collection . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint The following variables were included for data analysis: (1) profile of patients (age and sex); (2) diagnostic method (RT-PCR, immunoglobulin serology, or clinicalepidemiological); (3) symptoms presented on admission; (4) comorbidities; (5) vital signs; (6) laboratory and imaging exams; (7) risk stratification on admission; (8) hospital length of stay (LOS); (9) length of invasive mechanical ventilation (IMV); and (10) clinical outcomes (discharge or death). We obtained laboratory parameters, such as blood cell count and renal function (i.e., concentration of urea and creatinine), from medical records. Data from the first day (admission) to the last day of hospitalization (hospital discharge/death) were included for analysis. Vital signs, symptoms, and laboratory tests were considered if obtained on admission, while data regarding IMV, hospital LOS, and imaging tests were obtained when the outcome (discharge or death) was recorded. The multiple imputation approach (intention to treat analysis) 11 estimated plausible values and replaced missing data from laboratory tests. We computed five different imputations and generated five individual databases and five results. Then, results were grouped, and inferences were performed. The multiple imputation approach generated more accurate values than those generated by single imputation methods and also helped to understand patterns of missing data from our data set. Categorical data were presented as relative frequency and numerical data as mean and standard error (SE). Patients were divided into survivors (patients discharged) and non-survivors (patients who died). Kolmogorov-Smirnov test assessed data normality, and the unpaired t-test compared data between groups. Chi-square test (Chi 2 ) compared comorbidities, signs, and symptoms between groups. Cox (proportional is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint hazards) regression investigated the effects of age (< 65 and ≥ 65 years) and sex on death, whereas survival analysis was performed using Kaplan-Meier method. We also developed logistic regression models using the progressive selection technique (i.e., forward selection), in which only significant variables (p ≤ 0.05) and predictive value remained in the model to identify those interfering with death. studied outcomes. The software Statistical Package for the Social Sciences version 22 (IBM Corp., USA) performed inferential analysis, and significance was set at p ≤ 0.05. Table 1 shows general and clinical characteristics, symptoms, and comorbidities presented by the studied population. Among 280 hospitalized patients with COVID-19 included in our study, 192 were discharged, and 88 died. Approximately 54% of patients were males with mean age of 62.42 years (SE: 1.45). The most common comorbidities were systemic arterial hypertension (SAH) (53%), diabetes mellitus (67%), and obesity (24%). Moreover, 10% of patients had other types of heart disease. Dyspnea was the most reported symptom (82%), followed by cough (59%), fever (57%), muscular pain (22%), headache (15%), and anosmia (11%). Furthermore, 27% of patients were critically ill, and 57% were moderately ill on admission. Age (p = 0.001), dyspnea (p = 0.003), and presence of other heart diseases (p = 0.005) were different between groups. SAH was the most reported comorbidity in the non-survivor group (53%). This group also presented a high prevalence of cough (58%) and fever (56%). We observed significant differences in SpO2 (p < 0.001), leukocyte count (p < 0.001), blood urea concentration (p < 0.001), anosmia (p = 0.019), number of associated is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint comorbidities (p = 0.007), hospital LOS (p < 0.001), length of IMV (p < 0.001), and time between symptom onset and hospitalization (p = 0.057). The overall model, including age and sex as predictors, significantly improved the fit compared with null model [χ2 (2) . The logistic regression model included all numerical variables assessed ( Table 2 ). Urea concentration (p < 0.001), hospital LOS (p = 0.002), SpO2 (p = 0.005), and age is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint (p = 0.02) were associated with death (Hosmer and Lemeshow test, p = 0.50). These variables were also different between groups. This study analyzed data from 280 hospitalized patients with COVID-19. Most common comorbidities were SAH, diabetes mellitus, and obesity, while most reported symptoms were dyspnea, cough, and fever. Age ≥ 65 years, low SpO2, high urea concentration, and prolonged hospital LOS were associated with increased risk of death in these patients. Moreover, females were associated with low survival rate when hospital LOS lasted more than eleven days. Hospitalization due to COVID-19 is linked to worsening symptoms and increased risk of developing SARS, which increased in Brazil between 2019 and 2020 12 . Thus, identifying clinical characteristics of patients who reach severe clinical conditions is a global concern. Casas-Rojo et al. 13 analyzed data from 15,111 hospitalized patients of 150 hospitals in Spain and found a high prevalence of males in the sample (more than 57% aged > 60 years). These authors also observed that SAH (50.9%), obesity (21.2%), and diabetes mellitus (19.4%) were the most prevalent comorbidities, while most reported signs and symptoms on admission were fever (63.4%), dry cough (58%), and dyspnea (57.6%). Results of our study were similar to those observed in the Spanish is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint males (82%) with SAH (49%). Furthermore, 89% of patients aged > 64 years needed IMV, increasing mortality rate and hospital LOS 12 . In our study, 28% of patients received IMV, and those who died needed this type of support for a longer period (4.38 days). We highlight that prolonged hospital LOS was associated with high risk of death. Although studies showed a high prevalence of SARS-CoV-2 or other types of coronavirus in males 15, 16 , survival rate dropped faster in females than males and presented a reduced percentage of survival. The survival analysis conducted by Salinas-Escudero et al. 17 with 133 Mexican patients showed high mortality in older females, especially aged > 75 years. Our data also showed higher mortality in females than males (32.3% vs. 30.7%) but without significant difference between groups (p= 0.77), probably due to the high rate of deaths not related to sex particularities. Studies showing age as risk factor for worsening COVID-19 are frequent in the literature 13,14,18 . Li et al. 19 showed that age ≥ 65 years was associated with risk of severe disease and SAH. In the present study, survival rate in females aged ≥ 65 years declined to 50% when hospitalization was longer than eleven days. Also, mean age of patients who died was 70.87 years, 12 years more than those discharged. Although age as risk factor for developing COVID-19 is still controversial, its role in mortality is well established 20 . In this sense, are the aspects inherent to aging responsible for developing the disease, or is the presence of individual factors (e.g., immune response), comorbidities, and particular aspects of the elderly that worsen the response to viral infection? As already established, chronic diseases are common in the elderly, causing health 15 problems, especially during COVID-19 infection. Other factors, such as SpO2, hospital LOS, and blood urea concentration were associated with mortality in our sample. Blood urea concentration was also associated with COVID-19 severity in the study by Ok et al. 21 . These authors observed that patients is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint who progressed to severe disease had higher urea concentration than those with moderate disease. Moreover, urea/creatinine ratio, white cell count, C-reactive protein, monocyte/lymphocyte ratio, and neutrophil/lymphocyte ratio were considered predictive factors for disease severity. These findings reinforce that early assessment may identify patients with risk of disease worsening. Prolonged hospitalization, mainly in older individuals, was associated with mortality. Thai et al. 18 showed that patients who died had mean hospital LOS of eight days, whereas those who survived had hospital LOS of six days. Prolonged hospitalization causes damage in muscle strength, quality of life and functionality domains in older patients. 22 The negative impact on physical, cognitive, and social wellbeing in COVID-19 patients who recover from ICU admissions and prolonged LOS due to acute respiratory illnesses is already recognized. 23 Limited access of the population to COVID-19 diagnostic test and its acquisition and availability by the health system may be a potential limitation of this study. Specifically during hospital care, some patients were already outside the window of time to detect SARS-CoV-2 using RT-PCR. Therefore, diagnosis was performed by identifying immunoglobulins M and G and clinical and/or radiological evolution of the patient. Furthermore, the unfamiliarity of the population with signs and symptoms of COVID-19 may have delayed healthcare seeking. As a result, patients arrived at the hospital with advanced disease. Age, hospital LOS, high blood urea concentration, and low SpO2 are associated with mortality by COVID-19 in the evaluated population. Considering that several variables associated with increased mortality are assessed at hospital admission, this is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted January 27, 2022. ; https://doi.org/10.1101/2022.01.27.22269602 doi: medRxiv preprint COVID-19: immunopathology and its implications for therapy The novel coronavirus Disease-2019 (COVID-19): Mechanism of action, detection and recent therapeutic strategies WHO | World Health Organization The Natural History , Pathobiology , and Clinical Manifestations of SARS-CoV-2 Infections Clinical Characteristics of 138 Hospitalized Patients with 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China The 2019 -2020 Novel Coronavirus ( Severe Acute Respiratory Syndrome Coronavirus 2 ) Pandemic : A Joint American College of Academic International Medicine-World Academic Council of Emergency Medicine Multidisciplinary COVID-19 Working Group Consensus Pa Neurological Consequences of SARS-CoV-2 Infection and Concurrence of Treatment-Induced Neuropsychiatric Adverse Events in COVID-19 Patients : Navigating the Uncharted Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVIDresearch that is available on the COVID-19 resource centre -including this for unrestricted research re-use a Survival analysis of hospital length of stay of novel coronavirus (COVID-19) pneumonia patients in Sichuan, China. medRxiv Intention-to-Treat Analysis: Implications for Quantitative and Qualitative Research Orientações sobre Diagnóstico , Tratamento e Isolamento de Pacientes com COVID-19 Revista Clínica Española Características clínicas de los pacientes hospitalizados con COVID-19 en Espa ˜ Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region Prevalence of comorbidities in the Middle East respiratory syndrome coronavirus (MERS-CoV): a systematic review and meta-analysis Sex-based differences in susceptibility to SARS-CoV infection Factors associated with the duration of hospitalisation among COVID-19 patients in Vietnam : A survival analysis Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan A Comprehensive Review of Viral Characteristics, Transmission, Pathophysiology, Immune Response, and Management of SARS-CoV-2 and COVID-19 as a Basis for Controlling the Pandemic