Avaliação da sepse em terapia intensiva: análise dos dados clínico-laboratoriais e diagnósticos de enfermagem NANDA Internacional

Abstract

Sepsis remains one of the major challenges in Intensive Care Units, characterized by high morbidity and mortality and complex care demands. This study aimed to analyze clinical and laboratory data and NANDA-I nursing diagnoses recorded in the medical records of patients diagnosed with sepsis and septic shock in an adult Intensive Care Unit. This was a documentary, analytical, and retrospective study with a quantitative approach, conducted using medical records of patients admitted between 2022 and 2024 and approved by the Research Ethics Committee under protocol number 7,589,331. A total of 78 medical records were analyzed. The results showed a predominance of female patients and older adults (age range: 20 to 83 years), as well as significant hemodynamic instability, with frequent need for mechanical ventilation,

vasopressor therapy, and invasive devices. The main infectious foci were ventilator- associated pneumonia and catheter-associated bloodstream infection. Microbiological

analysis revealed a predominance of multidrug-resistant Gram-negative bacilli, including Pseudomonas aeruginosa, Klebsiella pneumoniae, and Acinetobacter baumannii, in addition to a relevant presence of Candida species. Antimicrobial therapy predominantly involved broad-spectrum agents, reflecting the resistance profile identified. The most frequent nursing diagnoses were Risk for Infection, Risk for Pressure Injury in Adults, Impaired Spontaneous Ventilation, and Impaired Self-Care Syndrome. After applying the False Discovery Rate adjustment, fourteen associations between nursing diagnoses and categorical variables remained statistically significant. Among continuous variables, a strong correlation was observed between urea and creatinine levels, and a moderate correlation between SOFA score and creatinine. For the nursing diagnosis Risk for Infection – Invasive Procedure, elevated bilirubin levels and the presence of septic shock were identified as strong predictors. Receiver operating characteristic curve analysis demonstrated that this diagnosis showed the highest accuracy among the tested models, indicating its potential as a useful predictive tool in clinical practice. It is concluded that sepsis in intensive care settings presents an etiology predominantly associated with the hospital environment and requires integrated action by the multidisciplinary team. The findings reinforce the strategic role of nurses in the Nursing Process, particularly in clinical surveillance, early risk identification, and diagnostic accuracy. Overall, the results highlight that certain nursing diagnoses have relevant predictive value, standing out as clinical judgments capable of anticipating clinical deterioration and supporting decision-making in patients with sepsis.