We selected the study participants among 206 patients with community-acquired pneumonia (CAP) and sepsis who were admitted to the ICU of the People s Hospital of Qiandongnan Miao and Dong Autonomous Prefecture from January 1, 2016 to April 10, 2022. We excluded 22 patients with cardiac insufficiency, 18 with lung malignancy, and 10 with chest trauma; thus, 156 patients were enrolled in this study. They comprised 108 men and 48 women, aged 56-80 years (mean 69.74 ± 6.72 years); there were 76 cases of ARDS and 80 of non-ARDS.

The criteria were as follows: A, community onset; B, pneumonia-related clinical manifestations-(1) recent coughing, expectoration, or aggravation of original respiratory disease symptoms; (2) fever; (3) signs of pulmonary consolidation or audible moist wheezing; (4) peripheral blood leukocytes >10 × 10⁹/L or <4 × 10⁹/L; C, chest imaging showed new patchy infiltrates, consolidation of the lobe or segment, and ground-glass or interstitial changes. We were able to confirm the clinical diagnosis of CAP if we found any of criteria A-C and had excluded the following: pulmonary tuberculosis; pulmonary tumor; noninfectious pulmonary interstitial disease; pulmonary edema; atelectasis; pulmonary embolism; pulmonary eosinophilic infiltration; and pulmonary vasculitis8.

We followed the definition and diagnostic criteria of sepsis 3.0 jointly issued by the American Society of Critical Care Medicine and European Society of Critical Care Medicine9.

The criteria were as follows: A, onset time within 1 week after known predisposing factors, new respiratory symptoms, or exacerbation of original symptoms; B, increased opacity of both lungs in chest imaging that could not be satisfactorily explained by pleural effusion, atelectasis, or nodules; C, pulmonary edema with respiratory failure that could not be accounted for by cardiac failure or fluid overload; D, hypoxemia with an OI ≤300 mmHg6.

We treated all patients with antibiotics according to standard practice; they received expectorants, anti-asthmatics, nutritional support, and mechanical ventilation as necessary. We undertook the timing and mode of ARDS mechanical ventilation according to guidelines for such ventilation 6,8.

We excluded patients with complicated cardiac insufficiency, pulmonary poisoning, lung malignancy, and pulmonary trauma. We also excluded patients with incomplete data.

Patients were assessed for ARDS within 24 hours of ICU admission; we divided them into ARDS and non-ARDS groups. After patient admission to hospital, 2 ml of arterial blood was withdrawn under anoxic conditions; we immediately undertook blood gas analysis (REDU ABL800 blood gas analyzer, Denmark, according to manufacturer instructions) to determine P(Aa)O₂ and OI. We compared the two groups with respect to gender, age, and P(Aa)O₂ and OI levels. We examined the correlation between P(Aa)O₂ and ARDS. To analyze the diagnostic ability of P(Aa)O₂ for ARDS, we drew the receiver operating characteristic (ROC) curve.

We observed no significant differences in gender, age, P(Aa)O₂, and OI between the two groups (χ²/t = 0.681, 0.461; P >0.05). P(Aa)O₂ in the ARDS group was greater than in the non-ARDS group (t = 8.875; P <0.001; OI in the ARDS group was smaller than in the non-ARDS group (t = –6.956; P <0.001. The differences were statistically significant (Table 1). Our ROC curve analysis showed that the area under the curve for the diagnostic ability of P(Aa)O₂ for ARDS was 0.931 (0.873–0.988). The cutoff value was 214.70 mmHg, the sensitivity 89.50%, and the specificity 85.00% (Figure 1).

Serious lung inflammation in sepsis patients may directly lead to pulmonary injury. It can arise as an inflammatory response to pulmonary capillary endothelial cells and alveolar epithelial cells damaged by a significant increase in alveolar capillary wall permeability; it may spread extensively in the lung and produce pulmonary interstitial edema 1011121314. The result can be impaired oxygen exchange, lung compliance, and significantly reduced pulmonary function 15. ARDS is clinically characterized by stubborn hypoxemia. ARDS is a common cause of death among ICU patients with severe pulmonary infection; the fatality rate is high and treatment is difficult16. Reportedly, most patients who recover suffer severe sequelae 17,18, which makes it difficult for them to lead normal lives. For ARDS patients with severe pulmonary infection, timely effective preventive measures, appropriate determination of ARDS status, and well-timed effective intervention are key to successful treatment 3. ARDS patients have an imbalance between ventilation and blood flow ratio, resulting in a decreased respiratory quotient; simultaneously, there is reduced arterial oxygen partial pressure, leading to an increase in P(Aa)O₂ 19,20. Studies have shown that P(Aa)O₂ can accurately reflect ARDS patients intrapulmonary shunt and diffuse dysfunction; P(Aa)O₂ can precisely evaluate pulmonary gas exchange capacity, which correlates well with pulmonary oxygen level and function 7,21. Thus, monitoring P(Aa)O₂ can help evaluate dynamic changes in pulmonary lesions in ARDS patients, which is advantageous in individual management of ARDS.

As a common diagnostic and risk stratification indicator for ARDS, OI provides a reference for medical treatment 22. In the present study, P(Aa)O₂ correlated negatively with OI (r = -0.555, P <0.001), indicating that P(Aa)O₂ has diagnostic ability for ARDS. This investigation showed that when P(Aa)O₂ was greater than 214.70 mmHg, the sensitivity and specificity of P(Aa)O₂ were 89.50% and 85.00%, respectively, in diagnosing ARDS; this finding underlines the clinical value of P(Aa)O₂ in such diagnosis. P(Aa)O₂ is appropriate owing to the pathology of ARDS, which is an oxygen exchange disorder caused by inflammation: P(Aa)O₂ has high sensitivity in reflecting pulmonary oxygen exchange. Some studies have reported the significance of P(Aa)O₂ for ARDS. Wang and Associates suggested that P(Aa)O₂ can provide a clear early warning for ARDS in patients with severe pneumonia 23,24. P(Aa)O₂ can be used for the stratified diagnosis and guiding treatment of ARDS patients 25. Yang et al. suggested that P(Aa)O₂ may play an important role in indicating the severity of ARDS in patients with pulmonary infection; it can also be used to assess the risk of fatality 26. P(Aa)O₂ may be employed as an important reference indicator for judging clinical efficacy. Zhang et al. found that P(Aa)O₂ dynamic monitoring was effective in making a dynamic assessment for pulmonary disease and could be used to guide clinical treatment 27. Yu et al. 20 and Shen et al. 28 suggested that P(Aa)O₂ can guide the withdrawal of mechanical ventilation in ARDS patients. The clinical application of P(Aa)O₂ for ARDS patients was confirmed in the present study.

Some limitations of this study deserve mention. We had relatively few cases, and we selected only sepsis patients from one central ICU; their conditions were quite severe, so there could have been some significant deviation in the cutoff value we determined. We measured P(Aa)O₂ directly by blood gas analysis with arterial blood, which was simple, convenient, and repeatable. However, whether P(Aa)O₂ is as convenient a measure as OI requires further prospective clinical studies regarding the ability to assess ARDS and its practicability.