The global impact of the coronavirus disease 2019 (COVID-19) is unquestionable. Concerning deaths, 68% were concentrated in 10 countries: Brazil, Egypt, India, Indonesia, Mexico, Peru, Russia, South Africa, Turkey, and the United States (1). The first COVID-19 case in Brazil was confirmed on February 26, 2020. The disease rapidly spread in the capital and countryside regions, and within a month, community transmission was documented in Brazilian cities. Bahia is Brazil’s fourth most populous State and the sixth state in cumulative deaths as of 2022, with the first case confirmed on March 6, 2020, through reverse transcription-quantitative polymerase chain reaction (RT-qPCR) (2–4).

By December 2022, Brazil had an incidence coefficient of 17,152, with an incidence rate of 11,738 per 100,000 inhabitants in Bahia. One of the main cities in the Southern Region of Bahia State, Ilhéus, has an incidence rate of 17,129 per 100,000 inhabitants, which is higher than that of Bahia. While Brazil’s lethality rate is 1.9%, Bahia’s rate is 1.8%, and Salvador, Vitória da Conquista, Feira de Santana, and Ilhéus have the highest number of deaths (2, 5).

Generally, the infection can manifest in a varied clinical spectrum ranging from asymptomatic to critical presentations. In addition to respiratory symptoms, severe cases may present with extrapulmonary complications or multiple organ failure, and early identification and treatment of at-risk patients are essential to prevent mortality (6–8). From an epidemiological perspective, a profile analysis of severe COVID-19 cases indicates that males have higher mortality rates than females do. Furthermore, comorbidities such as hypertension, diabetes, heart disease, malignancy, and immunodeficiency are more prevalent in individuals with severe COVID-19, irrespective of sex (9–13). However, in the Southern region of Bahia State, at the beginning of the pandemic, males with comorbidities were more likely to test positive for SARS-CoV-2 (14).

Early indicators of death in hospitalized patients guide clinical decision-making and include blood pressure, respiratory rate, D-dimer levels, international normalized ratio (INR), and Simplified Acute Physiology Score 3 (SAPS 3), which are predictors of in-hospital mortality (15–20). Multiple biomarkers are necessary to assess disease progression and an individual’s response to clinical interventions (21–23). Notably, well-established biomarkers include interleukin-6 (IL-6) and C-reactive protein levels (24–26).

Nonetheless, the profile of SARS-CoV-2 infection changes as new variants emerge, increasing the infection rate, mortality, and symptomatic profile (27–29). People in low-income countries, especially those with low socio-economic conditions, are more likely to test positive for SARS-CoV-2, with higher mortality rates (30). Accordingly, a study conducted in South America showed high seropositivity in individuals with low socio-economic status (31). Similarly, the unequal conditions of public health systems increase the infection rate and make early identification and treatment of at-risk patients difficult (32–34).

We aimed to characterize the clinicopathological profile of hospitalized patients with COVID-19 admitted to the intensive care unit (ICU) of a reference hospital for COVID-19 in the Southern Region of Bahia State, in Brazil, between April 2020 and July 2021. Additionally, we analyzed the data to identify the laboratory and clinical markers associated with death in patients admitted to the ICU.

Herein, we describe the epidemiological and clinical characteristics of COVID-19 patients admitted to the ICU of a hospital for COVID-19 treatment in the Southern Region of the Bahia State, Brazil. We also analyzed the factors associated with death. For example, we identified clinical parameters such as the use of mechanical ventilation, central venous catheters, arterial catheters, vasopressors, and bladder catheters related to the respiratory, cardiovascular, and urinary systems, which increased the odds of death in COVID-19 patients in intensive care.

The impact of the COVID-19 pandemic has not been homogeneous worldwide, with some countries being more affected and presenting different mortality rates (1). Social factors and precarious socio-economic conditions are drivers of increased infection and mortality rates (30–34, 39, 40). For example, the positivity of SARS-CoV-2 infection in cities in the Southern Region of Bahia State was negatively correlated with a low Human Development Index (HDI) (41) (Bahia State has a low HDI, and the average worker salary is less than US$ 600.00). Furthermore, it was also shown that individual and community risk factors for SARS-CoV-2 infection varied between the Bahia cities; for example, gender and age were not homogenous risk factors for SARS-CoV-2 infection between the 12 cities studied (42).

A retrospective study in Brazil using population-based registers demonstrated that individuals hospitalized for less than 4 days presented high odds of death (OR 2.07, 95% CI 2.05–2.10). Moreover, the odds of death were five times higher than for individuals requiring ICU admission (OR 5.19, 95% CI 5.14–5.24) (43). Notably, in our study, 35.32% of the COVID-19 patients in the ICU died. An in-hospital mortality rate of 37% for COVID-19 was reported in Brazil, and the mortality rate increased with advanced age, low education level, comorbidities, and in individuals of black/brown self-reported race (44).

Notably, determining the clinical-epidemiological and laboratory profiles of COVID-19 patients can provide valuable information for a multidisciplinary healthcare team for more assertive clinical management, better resource allocation, and improved survival of patients admitted with COVID-19 in the ICU (44, 45). In this study, when multifactorial variables were correlated using regression analysis, the male sex had a higher chance of death, consistent with previous studies (46, 47). Furthermore, male-specific variables such as hypogonadism and low testosterone levels have been linked to the development of comorbidities that increase mortality from COVID-19, including type 2 diabetes, obesity, and cardiovascular disease (46). Additionally, evidence suggests that unbalanced testosterone levels may facilitate infection and disease progression in men because of their impact on the expression of the SARS-CoV-2 receptor, angiotensin-converting enzyme-2, and major fusogenic transmembrane serine protease 2 under regular transcription by androgens (47, 48).

The average age of the ICU patients in our study was 64 years. We observed that older patients, especially those aged >66 years, were more likely to die from COVID-19. Comparing patients from wards and ICU, Pereira and coauthors showed that mortality rates increased with advanced age, according to sex, ethnic/racial background, and vaccination status (43). Moreover, previous studies have indicated that individuals aged 65 years and older have a higher risk of death from COVID-19 (49, 50). This may be attributed to an age-related decline in innate immunity and immunosenescence. Accordingly, a study conducted in 2020 in the Southern Region of Bahia State with hospitalized patients showed a higher frequency of COVID-19 among patients of advanced age (51). Furthermore, in severe cases of COVID-19, hematological changes in peripheral leukocytes reflect a compromised immune response during SARS-CoV-2 infection. These changes are early indicators of fatal outcomes and are crucial for maintaining immune homeostasis during viral infections (52).

In addition, the current study also suggests a high WBC count >10.03 cells x 1,000/mm³ as a predictive death parameter, which was higher in patients with death outcomes than in discharged patients. These data are consistent with those of previous studies (53) and a meta-analysis examining the relationship among WBC count, COVID-19 severity, and mortality (54). The meta-analysis reported a WBC count of 0.41 × 10⁹ /L for patients with moderate COVID-19, while the count increased significantly to 4.15 × 10⁹ /L in patients who died (55). Another meta-analysis showed that the WBC and neutrophil counts decreased significantly in patients with mild COVID-19. However, similar to the results of the present study, higher counts were observed in severe COVID-19 (56).

Although we have shown that clinical parameters such as the use of mechanical ventilation, central venous catheters, arterial catheters, vasopressors, and bladder catheters increased the odds of death in COVID-19 patients, we also analyzed laboratory markers, including arterial lactate, serum creatine, urea, and serum urea nitrogen, which were higher in patients who died than in those discharged from the ICU. Investigators have suggested that determining changes in lactate levels can provide insights into COVID-19 pathophysiology and multisystem interactions (57). Furthermore, oxygen deprivation in tissues leads to lactate overproduction because pyruvate cannot be oxidized in the Krebs cycle. Predisposing factors for lactic acidosis, including diabetes and acute respiratory distress syndrome are common in hospitalized COVID-19 patients. In addition, COVID-19-related damage to alveolar cells may contribute to increased lactic acid (22, 58, 59).

Due to altered dyspnea and extremely low oxygen saturation, individuals with impaired respiratory metabolism are at an exceptionally high risk of death. Specifically, changes in carbon dioxide levels trigger a hypoxic threshold, resulting in lung damage. Under normal hypoxic conditions, even a slight imbalance in PaCO₂ levels quickly evokes significant increases in ventilation per minute and brief respiratory alkalosis, which physiologically alters blood pH (37, 46, 47).

Regarding laboratory markers, we also observed an association between COVID-19 non-survivors and urea and serum creatinine levels. These biomarkers can help evaluate kidney injury, especially the acute forms that occur in 3–29% of COVID-19 patients. According to a study of 701 patients with COVID-19, both kidney injury and acute kidney injury increased the risk of death, with elevated serum creatine and urea nitrogen levels being predictive of mortality (60). Furthermore, we observed high levels of these biomarkers in COVID-19 patients who died. Data from 95 patients, of whom 25 were admitted to the ICU, showed a short-term increase in the urea and serum creatine ratios (OR, 1.72; 95% CI, 1.20–2.66), characterizing them as independent predictors of the prognosis of death.

Finally, we observed that the odds of death were five times higher for individuals requiring mechanical ventilation in the ICU, and patients with a higher FiO₂ were more likely to die of COVID-19 in the ICU. During the COVID-19 outbreak, ICU stay and mechanical ventilation devices have been associated with respiratory failure (61–66), and the unprecedented number of patients weaned from non-invasive ventilation proved to be highly challenging. Additionally, here, the bladder catheter and central venous catheter groups presented higher ORs for death, 79.3 and 45.12, respectively.

Invasive ventilation is an intricate procedure that requires skilled multidisciplinary teams and expensive equipment. The lack of trained professionals to administer and maintain the technique, the increased number of patients with respiratory injuries, and the shortage of materials increase the risk of infection during these procedures (67). Furthermore, the use of a bladder catheter increases the risk of catheter-associated urinary tract infection (68), and a central venous catheter is associated with mortality in chronic hemodialysis patients with COVID-19 in Brazil (69). Additionally, SAPS3, a scoring system widely used for predicting in-hospital mortality, was able to discriminate between survivors and non-survivors in our study (17–19), and the highest concentration of deaths due to COVID-19 was in ICU patients with SAPS3 > 51.5.

In summary, this study reported the clinical profile of a low-income population admitted to the COVID-19 ICU at a reference hospital in the Southern Region of Bahia State, Brazil. Our data demonstrate that the use of a catheter (central venous, arterial, or bladder) was the main factor associated with death in COVID-19 patients. Although platelet count was not associated with the death of patients in the ICU, leukocyte count and biochemical parameters were valuable indicators of death. The SAPS3 presented the highest sensitivity (77.9%) and specificity (63.1%) for discriminating between survivors and non-survivors, with an AUC of 0.79. Lastly, we suggest that multi-laboratory parameters can be used to predict patient prognosis and guide healthcare teams toward more assertive clinical management, better resource allocation, and improved survival of patients admitted to COVID-19 in the ICU.

We identified some factors (epidemiological and laboratory) associated with a higher chance of death among patients with COVID-19 treated in the ICU. For example, patients aged 65 years or older, those with a prolonged ICU stay, and those who required catheter use were more likely to die of COVID-19. Identifying predictors of death is important for choosing the best clinical management and therapeutic approaches to patients to avoid or minimize unfavorable outcomes. Moreover, it is important that epidemiological and clinical laboratory data are available for decision-making purposes. Thus, by knowing the predictors of worse prognosis and having these data, clinicians can act early and with scientific evidence.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author/s.

The studies involving humans were approved by the Human Beings Research Ethics Committee of the State University of Santa Cruz, Ilhéus, Brazil. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

LC, JG, CJ, and HP: conceptualization. MD’C, US, LC, and RF: methodology. RF, AC, CR, and AS: validation. MD’C and US: formal analysis. US, MD’C, and LC: data curation. MD’C, AC, CM, EG, SG, LM, and LC: original draft preparation. US, CJ, RR, RF, SG, AC, CR, and AS: writing review and editing. LC: supervision. All authors have read and agreed to the published version of the manuscript.