[Effect of hypercapnia on the clinical prognosis and severity of infection in patients with severe community-acquired pneumonia]

Objective: To investigate the effect of hypercapnia at admission on the clinical prognosis and the severity of infection in patients with severe community-acquired pneumonia (SCAP).

Methods: The clinical data of 219 SCAP patients admitted to the department of emergency and critical care medicine of Guangdong Provincial People's Hospital from December 2017 to November 2019 were retrospectively analyzed. Based on the partial pressure of arterial carbon dioxide (PaCO₂) within 1 day after admission, the patients were divided into hypocapnia group [HO group, PaCO₂ < 35 mmHg (1 mmHg = 0.133 kPa)], normal carbonation group (NC group, PaCO₂ 35-45 mmHg) and hypercapnia group (HC group, PaCO₂ > 45 mmHg). The clinical parameters of patients, such as gender, age, underlying diseases, white blood cell (WBC), procalcitonin (PCT), C-reactive protein (CRP), interleukin-6 (IL-6), pH value and lactate (Lac) within 1 day after admission were reviewed. The oxygenation index (PaO₂/FiO₂), pneumonia severity index (PSI) score and acute physiology and chronic health evaluation II (APACHE II) score were evaluated. The change tendencies of each index on day 1, day 3, and day 5 after admission were observed subsequently. Meanwhile, the rate of invasive mechanical ventilation (IMV), length of hospital stays and 28-day mortality among three groups were compared. Kaplan-Meier survival analysis was performed to assess the 28-day cumulative survival rate of patients with SCAP among three groups. Multivariate Logistic regression analysis was used to screen the risk factors of IMV and 28-day death in patients with SCAP.

Results: Compared with the HO group (n = 68) and NC group (n = 72), the HC group (n = 79) had higher proportion of preexisting comorbid chronic obstructive pulmonary disease (COPD) and PSI score, lower PCT, CRP, IL-6, and pH values. Compared with the HO group and NC group, there were smaller improvement trends on the levels of WBC, PCT, CRP, IL-6, PaO₂/FiO₂ and Lac at day 3 and day 5 as compared with day 1 in the HC group. On the 5th day after admission, the levels of WBC, PCT, CRP, IL-6, and Lac in the HC group were significantly higher than those in the HO group and NC group [WBC (×10⁹/L): 18.33±1.44 vs. 10.89±2.37, 11.15±1.74; PCT (μg/L): 5.04±1.18 vs. 3.46±0.87, 3.58±0.83; CRP (mg/L): 78.43±7.17 vs. 54.24±4.97, 57.93±5.39; IL-6 (ng/L): 75.35±11.92 vs. 60.11±10.27, 57.88±12.34; Lac (mmol/L): 4.36±1.24 vs. 0.78±0.39, 0.86±0.64; all P < 0.01], and the lowest in PaO₂/FiO₂ was found in the HC group as compared with the HO and NC groups (mmHg: 171.31±6.73 vs. 226.68±7.36, 225.93±6.92, both P < 0.01). Compared with the HO group and NC group, the HC group had highest proportion of IMV (29.1% vs. 22.1%, 22.2%, both P < 0.01) and 28-day mortality (26.6% vs. 13.2%, 13.9%, both P < 0.01). Even when the patients with COPD were excluded from the analysis, the differences persisted among the groups. Kaplan-Meier survival analysis suggested that HC group had a higher 28-day cumulative survival rate as compared with the HO and NC groups (Log-Rank test: χ₁² = 4.976, P₁ = 0.026; χ₂² = 4.629, P₂ = 0.031). Multivariate Logistic regression analysis showed that IL-6, PSI score and hypercapnia within 1 day and PCT on the 5th day after admission were the independent risk factors of requiring IMV and 28-day death in patients with SCAP [odds ratio (OR) were 0.325, 1.229, 1.396, 1.313, respectively, all P < 0.01]. Even when patients with COPD were excluded from the analysis, the above results had not been changed.

Conclusions: Hypercapnia at admission was associated with higher proportion of IMV and 28-day mortality in patients with SCAP, which may be related to its early suppression of inflammation and then increment of infection.