RESUMO
OBJECTIVE: To explore the value of bedside lung ultrasound in the early diagnosis and severity assessment of ventilator-associated pneumonia (VAP). METHODS: A prospective observational study was conducted in 60 patients with VAP (VAP group) and 62 patients without VAP (control group) who were admitted to department of intensive care unit of General Hospital of Ningxia Medical University from September 2018 to July 2020. The gender, age and underlying diseases of non-VAP group were matched with VAP group. The general clinical data such as gender, age, underlying diseases, department source of the patient, acute physiology and chronic health evaluation II (APACHE II) score, sequential organ failure assessment (SOFA) score were recorded. The body temperature, white blood cell count (WBC), procalcitonin (PCT), oxygenation index (PaO2/FiO2), alveolar artery oxygen differential pressure (PA-aDO2) were recorded. During mechanical ventilation, the patient's body temperature, WBC, sputum characteristics, and the change of the lung ultrasound were dynamically observed. With or without dynamic air bronchogram, lung ultrasound was considered to be positive as long as there were small subpleural consolidation or tissue-like sign. Ventilator-associated pneumonia lung ultrasound score (VPLUS) and lung ultrasound score (LUSS) were performed, and chest CT scan was completed on the same day. Use positive chest CT scan as the standard to evaluate the diagnostic efficacy of lung ultrasound, VPLUS score, and the combination of the two with PCT for VAP. LUSS was used to assess the severity of disease in patients with VAP. The correlation between LUSS and PaO2/FiO2, PA-aDO2, APACHE II score and SOFA score were analyzed. RESULTS: (1) General information: compared with non-VAP group, VAP group had more emergency surgery patients [51.7% (31/60) vs. 33.9% (21/62), P = 0.047], APACHE II score and SOFA score were significantly higher (APACHE II score: 15.4±5.7 vs. 13.4±3.4, P = 0.021; SOFA score: 8.8±4.2 vs. 6.3±3.3, P < 0.001), body temperature tended to rise (centigrade: 38.3±0.8 vs. 38.0±0.9, P = 0.054), more patients had airway purulent secretions [65.0% (39/60) vs. 41.9% (26/62), P = 0.011], and mechanical ventilation time and length of ICU stay were longer [mechanical ventilation time (days): 10.5 (6.6, 15.0) vs. 4.3 (3.0, 6.0), P < 0.001; length of ICU stay (days): 14.8 (9.0, 18.0) vs. 6.0 (4.0, 9.1), P < 0.001], 28-day mortality rate was higher [31.7% (19/60) vs. 9.7% (6/62), P = 0.003]. (2) Diagnostic efficacy evaluation: when lung ultrasound was positive, VPLUS ≥ 3 and PCT > 0.5 µg/L were used separately for the diagnosis of VAP, the sensitivity was 73.3%, 75.0%, 61.7%, respectively; the specificity was 80.6%, 58.1% and 59.7%, respectively; the 95% confidence interval (95%CI) was 0.685-0.842, 0.574-0.748, 0.514-0.694, respectively,all P < 0.05, positive lung ultrasound had good sensitivity and specificity. When positive lung ultrasound or VPLUS ≥ 3 were combined with PCT > 0.5 µg/L for tandem test, the specificity of VAP diagnosis was increased to 95.2% and 83.9%, respectively; but the specificity of VAP diagnosis of positive lung ultrasound combined with PCT > 0.5 µg/L was higher than VPLUS ≥ 3 combined with PCT > 0.5 µg/L (95.2% vs. 83.9%, P < 0.05). (3) Correlation analysis: LUSS showed a significant positive correlation with APACHE II and SOFA score (r values were 0.407, 0.399, P values were 0.001, 0.002, respectively), LUSS had no relation with PaO2/FiO2 and PA-aDO2 (r values were 0.189, -0.064, P values were 0.629, 0.149, respectively). CONCLUSIONS: Lung ultrasound can early detect VAP , and its diagnostic specificity is significantly improved when combined with PCT > 0.5 µg/L. LUSS is closely related to the severity of disease in VAP patients, therefore, lung ultrasound may be an effective method for early diagnosis and efficacy evaluation of VAP patients.