Screening Process
A total of 3405 hospitalized patients with RNA tests positive for influenza and 1313 eligible adult patients with Flu-p were included in the final analysis, including 1191 immunocompetent patients and 122 IC patients (Fig.1). The three most common immunocompromising conditions were immunosuppressive therapy (57/122, 46.7%), active malignancy (25/122, 20.5%) and organ transplantation (17/122, 13.9%) (Supplementary Material 4).
Comparison of clinical characteristics and outcomes between IC and non-IC patients with Flu-p
Compared with non-IC patients, IC patients were younger (median: 45.0 years vs. 61.0 years, p < 0.001), and were admitted to hospitals earlier after illness onset (median: 2.0 days vs. 4.0 days, p < 0.001). Chest pain (31.1% vs. 17.2 %, p < 0.001) and pleural effusion (47.5% vs. 32.0%, p < 0.001) at admission were present significantly more often in IC patients, while fever (63.9% vs. 76.6%, p < 0.001) was present significantly less often. Baseline lymphocyte counts were significantly lower in IC patients than in non-IC patients [ (0.9 ± 0.6) vs. (1.2 ± 0.8), p < 0.001]. Other baseline vital signs, symptoms, and laboratory and radiologic findings were similar between patients in the two groups (Table 1).
Compared to non-IC patients, IC patients (43.4% vs. 34.0%, p = 0.037) were more commonly coinfected with other pathogens at admission. Incidence of gram-negative bacterium (69.8% vs. 48.4%, p = 0.003), fungus (11.3% vs. 0.5%, p < 0.001) and Cytomegalovirus (CMV) (3.8% vs. 0.0% , p < 0.001) was higher in IC patients, while the proportion of gram-positive bacterium (37.7% vs. 55.1%, p = 0.017) was lower (Supplementary Material 5).
All of the Flu-p patients were administered antibiotics during their hospital stays. Compared to non-IC patients, less IC patients were administrated with systemic corticosteroids at admission (7.1% vs 26.2%, p < 0.001); more IC patients received noninvasive (48.8% vs. 25.7%, p < 0.001) and invasive (30.3% vs. 17.7%, p = 0.001) ventilation. Additionally, respiratory failure (41.0% vs. 22.9%, p < 0.001), heart failure (34.4% vs. 23.9%, p = 0.011), nosocomial pneumonia (17.2% vs. 7.4%, p < 0.001), septic shock (19.7% vs. 8.0%, p < 0.001) and nosocomial blood stream infections (7.4% vs. 0.8%, p < 0.001) were more frequent in IC patients. The duration between admission to clinical stability (median: 14.0 days vs. 3.0 days, p < 0.001), as well as the length of hospital stays (median: 17.0 days vs. 10.0 days, p < 0.001) were significantly longer for IC patients than for non-IC patients. Finally, more IC patients were admitted to the ICU (48.4% vs. 22.4%, p < 0.001), and the all-cause 30-day mortality rate (37.7% vs. 20.3%, p < 0.001) was higher in IC patients (Table 2).
Impact of immunocompromised status on clinical outcomes for Flu-p patients
After controlling for age, sex, duration between illness onset and admission, influenza virus type, comorbidities, pregnancy, obesity, smoking history, early NAI therapy, administration of systemic corticosteroids, antibiotics use, and coinfection with other community-acquired pathogens, IC status was associated with increased risks for invasive ventilation [odds ratio (OR): 2.475, 95% confidence interval (IC): 1.511-4.053, p < 0.001] , ICU admission (O:R 3.247, 95% CI: 2.064-5.106, p < 0.001) and 30-day mortality (OR: 3.206, 95% CI: 1.926-5.335, p < 0.001) for patients with Flu-p (Table 2).
Baseline predictors for 30-day mortality in IC patients with Flu-p
Results from univariate analysis revealed that respiratory rates ≥ 30 breaths/min at admission, and decreased baseline blood lymphocytes, blood albumin, and PaO2/FiO2 were associated with an increased risk for 30-day mortality in IC patients with Flu-p (Supplementary Material 6).
A multivariate backward stepwise logistic regression suggested that lymphocyte counts (OR: 0.993, 95% CI: 0.990-0.996, p < 0.001), coinfection (OR: 5.450, 95% CI: 1.638-18.167, p = 0.006), early NAI therapy (OR: 0.401, 95% CI: 0.127-0.878, p = 0.001), and systemic corticosteroid use at admission (OR: 6.414, 95% CI: 1.348-30.512, p = 0.020) were independent predictors for 30-day mortality in IC patients with Flu-p (Table 3).
Association of coinfected pathogens with baseline lymphocytes and 30-day mortality in IC patients with Flu-p
The baseline lymphocyte count levels for IC patients coinfected with gram-positive bacterium [(1.2 ± 0.8) ×109/L] was significantly higher than those for IC patients coinfected with gram-negative bacterium [(0.5 ± 0.3) ×109/L, p < 0.001], gram-negative bacterium/fungus or CMV [(0.3 ± 0.2)×109/L, p < 0.001] (Fig. 2A).
The survival rates of IC patients coinfected with gram-negative bacterium (80.0% vs. 25.0%, p < 0.001) and gram-negative bacterium/fungus or CMV (83.3% vs. 25.0%, p < 0.001) were lower than those of IC patients coinfected with gram-positive bacterium (25.0%) during the first 30 days after admission (Fig. 2B).
Baseline lymphocytes as predictors for 30-day mortality in IC patients with Flu-p
The ROC determined that the optimal cutoff of baseline lymphocyte counts was 0.6×109/L, which reached an AUROC of 0.825 (95% CI: 0.744 - 0.887), with sensitivity of 97.8% and specificity of 73.7% (Fig. 3A).
The Kaplan-Meier curves showed that the 30-day mortality of IC patients with baseline lymphocyte counts ≤ 0.6×109/L was higher than that of patients with lymphocyte counts > 0.6×109/L (69.2% vs. 1.8%, log-rank test, p < 0.001) (Fig. 3B).