Value of modified ROX index in predicting the prognosis of acute respiratory distress syndrome treated with high-flow nasal cannula oxygen therapy

Caixia ZHANG; Xinwei WANG; Xinnian LIU; Chuan DU

ABSTRACT

Objective:

To evaluate the value of the ROX index [blood oxygen saturation (SpO 2)/fraction of inspiration O 2 (FiO 2)/respiratory rate (RR)], ROX-heart rate (HR) index (ROX index/HR × 100), modified ROX (mROX) index [partial pressure of oxygen in the blood (PaO 2)/FiO 2/RR], and mROX-HR index (mROX index/HR × 100) in predicting prognosis for patients with acute respiratory distress syndrome (ARDS) treated with high-flow nasal cannula oxygen therapy (HFNC).

Methods:

The clinical data of 100 patients with ARDS who received HFNC between January 2018 and December 2022 at The Third People's Hospital of Hubei Province, Jianghan University, were retrospectively analyzed. These patients were divided into two groups based on whether HFNC treatment was successful or not a success group with 65 patients and a failure group with 35 patients. The differences in the ROX index, ROX-HR index, mROX index, and mROX-HR index in the observation group were observed at the designated time points 2, 12, and 24 hours after HFNC treatment. Receiver operating characteristic (ROC) curves were utilized to evaluate the value of ROX index, ROX-HR index, mROX index, and mROX-HR index in predicting the success or failure of HFNC treatment at 2, 12, and 24 hours. Cutoff values were determined.

Results:

There were no significant differences in age, gender, body mass index, Acute Physiology and Chronic Health Evaluation (APACHE II) score, Sequential Organ Failure Assessment score, or the proportions of underlying diseases and pulmonary causes between the success and failure groups (all P > 0.05). Furthermore, there were no significant differences in baseline HR, RR, FiO 2, SpO 2, partial pressure of carbon dioxide (PaCO 2), PaO 2, pH, lactate, oxygenation index, ROX index, mROX index, ROX-HR index, or mROX-HR index between the two groups (all P > 0.05). The ROX index in the success group at 2, 12, and 24 hours after HFNC treatment was 6.86 ± 1.09, 6.31 ± 1.61, and 8.24 ± 2.29, respectively. These values were significantly higher than those in the failure group (6.36 ± 0.67, 5.65 ± 1.44, and 5.41 ± 0.84) at the corresponding time points ( F = 5.97, 4.04, 49.40, all P < 0.05). At 2, 12, and 24 hours after HFNC treatment, the mROX index in the success group was 5.94 ± 1.28, 5.74 ± 1.23, and 8.51 ± 2.64, respectively. These values were significantly higher than those in the failure group (5.26 ± 0.74, 4.80 ± 0.97, 4.81 ± 1.17) at the corresponding time points ( F = 8.23, 15.38, 61.79, all P < 0.05). At 2, 12, and 24 hours after HFNC treatment, the ROX-HR index in the success group was 6.53 ± 1.32, 6.85 ± 1.44, and 7.57 ± 1.47, respectively. These values were significantly higher than those in the failure group (5.79 ± 1.04, 5.87 ± 1.03, 5.57 ± 0.63) at the corresponding time points ( F = 8.28, 12.61, 58.34, all P < 0.05). At 2, 12, and 24 hours after HFNC treatment, the mROX-HR index in the success group was 6.11 ± 1.30, 6.86 ± 1.13, and 7.79 ± 1.79, respectively. These values were significantly higher than those in the failure group (5.20 ± 1.06, 5.66 ± 1.46, 4.92 ± 0.90) at the corresponding time points ( F = 12.60, 20.87, 78.56, all P < 0.05). The receiver operating characteristic curve analysis revealed that the optimal thresholds were 6.56, 6.02, 6.24, and 5.25 for the ROX index, mROX index, ROX-HR index, and mROX-HR index, respectively. The area under the curve (AUC) values were 0.63, 0.66, 0.68, and 0.72, with sensitivity of 55.4%, 47.7%, 56.9%, and 76.9%, and specificity of 71.4%, 91.4%, 77.1%, and 62.9%, respectively. At 12 hours after treatment, the optimal thresholds were 6.09, 5.53, 6.52, and 5.99, with AUC values of 0.62, 0.70, 0.67, and 0.80, sensitivity of 55.4%, 53.8%, 61.5%, and 80.0%, and specificity of 74.3%, 77.1%, 71.4%, and 74.3%, respectively. At 24 hours after treatment, the optimal thresholds were 6.23, 6.4, 5.99, and 6.22, with AUC values of 0.88, 0.90, 0.91, and 0.93, sensitivity of 81.5%, 80.0%, 87.7%, and 83.1%, and specificity of 91.4%, 94.3%, 80.0%, and 91.4%, respectively.

Conclusion:

The use of the ROX index, mROX index, ROX-HR index, and mROX-HR index can aid in predicting the prognosis of ARDS patients. The predictive value of these indices increases as treatment time progresses. The mROX-HR index offers marked advantages during the initial stages of treatment and could serve as a reliable early predictor.