Urea to Albumin Ratio Is an Independent Predictor of In-Hospital Mortality in Patients With Severe Pneumonia: A Retrospective Cohort Study

7 Background: Severe pneumonia (SP) is a major complication 8 of respiratory system disease that is associated with high 9 mortality and morbidity. Our objective was to identify risk 10 factors predictive of SP patients and its mortality in intensive 11 care unit (ICU). 12 Methods: We conducted a single-center retrospective 13 observational study involving 212 patients with SP in ICU from 14 June 1st, 2016 to June 1st, 2020. The receiver operating 15 characteristic (ROC) curve was constructed to assess the 16 predictive significance of urea to albumin ratio (U/A). Kaplan- 17 Meier survival curves were plotted with log-rank tests to 18 compare survival of patients with different value of U/A. 19 Multivariate COX regression models were used to calculate the 20 adjusted hazard ratios (HR). Additionally, interaction analysis 21 showed the association between U/A and in-hospital mortality was influenced by sex. Primary outcome was in-hospital mortality. Results: A total of 212 patients were enrolled in the analysis. In 3 the hospital, 101 (47.6%) patients had died. ROC analysis 4 showed that the current cut-off of 0.2555 had a sensitivity of 5 84.2% for in-hospital mortality (AUC = 0.63, 95%CI: 0.55-0.70, 6 P = 0.001). The multivariate COX analysis showed that the 7 incidence of death was higher with the higher U/A group than 8 the lower group (HR: 2.234, 95%CI: 1.146-4.356, P = 0.018). Besides, this pattern persisted in subgroup analyses considering 10 sex. (HR: 9.380, 95%CI: 2.248-39.138, P = 0.002) 11 Conclusions: A high level of U/A is an independent risk factor 12 for in-hospital mortality in patients with SP.

of high mortality [3]. Therefore, severity evaluation is an 1 essential component of the initial assessment of these patients. 2 However, there is no consensus on the optimal evaluation 3 approach. 4 Risk factors for poor outcomes in patients with CAP include 5 higher blood urea nitrogen and lower albumin [3][4][5][6][7]. B/A levels 6 has also been reported to be associated with a high risk of 30-7 day mortality in ventilator-associated pneumonia (VAP) patients albumin to COVID-19 pneumonia patients to predict the 10 admission to ICU [9]. Moreover, evidence is accumulating that 11 a high blood urea nitrogen/albumin ratio (B/A) is relevant with 12 critical illness [10]. However, there is no study on SP patients. 13 We conducted the SP patients in ICU and U/A to evaluate the 14 in-hospital mortality associated with different levels of U/A. We 15 hypothesized that higher U/A group would be associated with a 16 higher risk of death than the lower group.  Patients who were admitted to participating ICU were screened 4 and, if eligible, were included. We screened the patients 18 5 years of age or older who were admitted to the ICU for SP. 6 Patients were excluded for the reasons: (1) ICU duration<24h;
Statistical analyses were performed with the use of SPSS, 1 version 22.0, and P < 0.05 was considered significant.  Table 1. Of these 212 cases, the median age was 9 73.0 (61.0, 82.8), 0.8% of the patients were male, and 16.0% 10 were hospital-acquired pneumonia (Table 1)  Compare to the U/A ≤ 0.2555 group, patients in the U/A > 20 as well as APACHE II score (P < 0.001, P < 0.001, P < 0.001, 1 respectively).  Table 2. All significant factors identified as predictors of in-6 hospital mortality (P < 0.05 in the COX univariate regression 7 analysis and clinical concern(sex)) were used for the   According to the cutoff value, the 211 SP patients were divided 1 into two groups. The Kaplan-Meier survival curves showed that 2 higher U/A group had a higher in-hospital mortality rate than 3 lower U/A group (Log-rank test chi-square 13.71, P < 0.001).

4
To elucidate the specific relationship between U/A and in-5 hospital mortality, we used different models (

17
Results of interaction analysis between U/A and sex are given in 18   Table 4. There was a significant interaction on in-hospitality Our analysis suggested that the first U/A after admitted to ICU is 6 an independent risk factor for in-hospital mortality in SP patients.

7
Interestingly, this study also demonstrated that U/A was an 8 independent predictor of in-hospital mortality in female 9 subgroups, but not in males. in-hospital mortality and the AUC was 0.63. The risk of death 10 was higher among the patients whose value of U/A was > 11 0.2555 than those whose U/A was ≤ 0.2555 (HR: 2.234, 95%CI: 12 1.146-4.356, P = 0.018). Although the AUC of U/A was not so 13 good, it is easy and quick to use, giving more information to 14 identify the high-risk group.

15
However, the underlying mechanism has remained unclear.

16
Urea is a marker associated with systemic disease. Although 17 urea is not a direct mark of infection, it can be a risk factor 18 because high value leads to high susceptibility to infection.

19
Some previous studies suggested that urea affects the prognosis 20 of critical patients regardless of the creatine level [13,14]. In 21 these prediction model, urea is a significant risk factor for pneumonia. Moreover, urea is an indirect marker of a metabolic 1 systemic pathway [15]. In pneumonia patients, elevations of 2 serum urea are indicators of protein catabolism. Water 3 deficiency appears to be common in pneumonia patients. In the 4 process of dehydration, the concentration of urea increased. In conclusion, our study demonstrated that the U/A is an 16 independent risk factor for in-hospital mortality.

17.
Horwich TB, Kalantar-Zadeh K, MacLellan RW, Fonarow GC. Albumin levels predict 1 survival in patients with systolic heart failure. American heart journal.  Data are mean ± standard or medians(25th-75th percentile)or number and percentage. 1 U/A: urea to albumin ratio; HAP: hospital acquired pneumonia; CHD: coronary heart disease; 2 health evaluation; ALT: alanine aminotransferase; AST: aspartate aminotransferase; WBC: white 1 blood cell; RBC: red blood cell; HCT: hematocrit; LOS: length of stay.   Reference group is U/A ≤ 0.2555 group.    ROC curve for predicting mortality in patients with SP. U/A had a modest power for predicting in-hospital mortality as suggested by AUC of 0.63 (95%CI: 0.55-0.70, P = 0.001), with a sensitivity of 84.2% and a speci city of 37.8% at a cutoff of 0.2555.

Figure 3
Kaplan-Meier survival curve according to U/A level. Compare to the lower group (U/A ≤ 0.2555), patients in the higher group (U/A > 0.2555) showed elevated in-hospital mortality.