[Clinical characteristics and prognostic factors of adult patients with acute respiratory failure due to influenza infection].

OBJECTIVE

To investigate the clinical characteristics and prognostic predictors of adult patients with acute respiratory failure due to influenza infection.

METHODS

A retrospective analysis was performed on adult patients with acute respiratory failure due to confirmed influenza infection admitted to intensive care units (ICU) of the First Affiliated Hospital of Zhengzhou University and the Sixth People's Hospital of Zhengzhou between January 2018 and January 2020. The subjects were divided into survival and death groups according to whether the patients died before discharge. Demographic and clinical data including underlying conditions, laboratory variables, therapy and prognostic factors of hospital mortality between the two groups were analyzed. The risk factors of mortality were evaluated by univariate and multivariate Logistic regression analysis. Then, the correlation between lymphocyte (LYM) count and LYM subsets were analyzed. The survival rates of different acute physiologic and chronic health evaluation II (APACHE II) and LYM level subgroups were compared.

RESULTS

A total of 104 patients were enrolled. Among them, 67 cases (64.4%) had underlying conditions, 91.3% of the patients (95 cases) were infected by influenza A virus, and the hospital mortality rate was 39.4% (41 cases). Compared with survival group, the patients of death group had higher respiratory rate (times/min: 26.0±5.6 vs. 23.7±5.0), APACHE II score (18.20±4.88 vs. 12.35±4.58), procalcitonin [PCT (μg/L): 0.82 (0.23, 4.63) vs. 0.39 (0.11, 0.92)], higher percentage of cardiovascular disease [24.4% (10/41) vs. 7.9% (5/63)] and invasive mechanical ventilation [63.4% (26/41) vs. 17.5% (11/63), all P < 0.01], but had lower oxygenation index [PaO/FiO (mmHg, 1 mmHg = 0.133 kPa): 131.8±34.5 vs. 181.7±31.6] at ICU admission, LYM (×10/L: 0.53±0.40 vs. 0.92±0.44), hemoglobin [Hb (g/L): 105.66±28.17 vs. 118.29±28.29], platelet count [PLT (×10/L): 135.12±85.40 vs. 199.81±110.11], T lymphocyte count [cells/μL: 181 (131, 275) vs. 319 (238, 528)], CD4 count [cells/μL: 110 (71, 161) vs. 190 (120, 311)] and CD8 count [cells/μL: 71 (33, 100) vs. 121 (81, 188), all P < 0.01]. Patients of death group also had a shorter length of hospital stay [days: 7.0 (4.0, 11.0) vs. 12.0 (8.0, 20.0), P < 0.01]. Univariate analysis showed that APACHE II score [odds ratio (OR) = 1.207, 95% confidence interval (95%CI) was 1.094-1.332, P < 0.001], LYM (OR = 0.070, 95%CI was 0.018-0.271, P < 0.001), Hb (OR = 0.984, 95%CI was 0.970-0.999, P = 0.031), PLT (OR = 0.992, 95%CI was 0.987-0.997, P = 0.003), T lymphocyte count (OR = 0.996, 95%CI was 0.993-0.998, P = 0.001) and PaO/FiO (OR = 0.955, 95%CI was 0.938-0.972, P < 0.001) were the risk factors for the prognosis of influenza patients with acute respiratory failure. Further multivariate Logistic analysis also showed that APACHE II score (OR = 1.195, 95%CI was 1.041-1.372, P = 0.011), LYM (OR = 0.063, 95%CI was 0.011-0.369, P = 0.002) and PaO/FiO (OR = 0.953, 95%CI was 0.933-0.973, P < 0.001) were the predictors of mortality. Moreover, patients with peripheral blood LYM < 0.65×10/L or APACHE II score > 14 had a higher risk of poor outcome. There were significantly positive correlation between LYM and LYM subsets (T lymphocyte, CD4 and CD8 lymphocyte, r value was 0.593, 0.563, and 0.500, respectively, all P < 0.001).

CONCLUSIONS

Influenza patients with acute respiratory failure were critically ill and had a high mortality rate. APACHE II score, PaO/FiO and LYM at ICU admission were independent risk factors affecting the prognosis of patients.