a b s t r a c t

Purpose: To verify the role of lactate dehydrogenase to albumin (LDH/ALB) ratio as an independent prognostic factor for mortality due to the lower respiratory tract infection (LRTI) in the emergency department (ED).

Methods: We reviewed the electronic medical records of patients who were admitted to the ED for the management of LRTI between January 2018 and December 2020. Initial vital signs, laboratory data, and patient severity scores in the ED were collected. The LDH/ALB ratio was compared to other albumin- based ratios (blood urea nitrogen to albumin ratio, C-reactive protein to albumin ratio, and lactate to albu- min ratio) and severity scales (pneumonia severity index, modified early warning score, CURB-65 scores), which are being used as prognostic factors for in-hospital mortality. Multivariable logistic regression was performed to identify independent risk factors.

Results: The LDH/ALB ratio was higher in the non-survivor group than in the survivor group (median [interquartile range]: 217.6 [160.3;312.0] vs. 126.4 [100.3;165.1], p < 0.001). In the comparison of the Area Under the Receiver Operating Characteristic Curve for predicting in-hospital mortality, the AUC of the LDH/ALB ratio (0.808, 95% confidence interval: 0.757-0.842, p < 0.001) was wider than other albumin-based ratios and severity scales, except the blood urea nitrogen to albumin ratio. In the multivar- iable logistic regression analysis, the LDH/ALB ratio independently affected in-hospital mortality.

Conclusion: The LDH/ALB ratio may serve as an independent prognostic factor for in-hospital mortality in patients with LRTI.

(C) 2021

1. Introduction

Lower respiratory tract infection (LRTI) is one of the most common infectious diseases worldwide; it includes a range of respiratory infec- tions such as pneumonia, bronchitis, pneumonitis, and empyema/ab- scess. Furthermore, it is associated with considerable morbidity and mortality. Because delays in the ICU admission of LRTI patients have been shown to be associated with increased mortality, it can be helpful to predict the mortality of LRTI patients in the emergency department (ED). Therefore, it is important to increase the accuracy of prognostic factors that predict the severity of LRTI.

Serum lactate dehydrogenase level (LDH) has been mentioned in some studies as a prognostic factor of pneumonia and sepsis [1-6]. It tends to rise with infection severity. Moreover, serum albumin was utilized as a prognostic factor of infection because it tends to decrease with infection aggravation [7-10]. By using albumin, sev- eral indexes like blood urea nitrogen to albumin ratio (BUN/ALB

* Corresponding author at: Department of Emergency Medicine, Chungnam National University Hospital, 282 Munhwa-ro, Jung-Gu, Daejeon 35015, Republic of Korea.

E-mail address: [email protected] (S. Ryu).

ratio), lactate to albumin ratio (Lac/ALB ratio), C-reactive protein to albumin ratio (CRP/ALB ratio), and procalcitonin to albumin ratio were used to predict the severity and prognosis of patients with pneumonia and sepsis [11-19].

The LDH to albumin ratio (LDH/ALB ratio) can be used as a prognos- tic factor of pneumonia with improved accuracy. However, no studies have been conducted on whether the LDH/ALB ratio is useful as a prog- nostic factor for mortality in LRTI, although it has been evaluated as a prognostic factor for malignancy [20-22]. Thus, we hypothesized that the LDH/ALB ratio would make up for LDH and could be a better prog- nostic factor compared to other albumin-based ratios for LRTI. This study aimed to verify whether the LDH/ALB ratio can be used as an in- dependent biomarker prognostic factor of LRTI in the ED.

1. Materials and methods
   1. Study design and setting

The study was conducted by reviewing the secondary data extracted from electronic medical records (EMR) of patients who visited the

https://doi.org/10.1016/j.ajem.2021.11.028

0735-6757/(C) 2021

emergency department at a tertiary academic hospital (a 1365-bed ter- tiary care referral center) between January 2018 and December 2020 and were diagnosed with LRTI. This study was approved by the Institu- tional Review Board of our hospital (No. 2020-10-059). The extracted data included clinical data only and did not include any personally identi- fiable information. Therefore, the need for informed consent was waived.

• 1. Data collection and outcome measures

We analyzed the data of patients aged >18 years and diagnosed with LRTI in the ED. To extract the data of LRTI, we used the Interna- tional Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) code with the following ICD-10 codes: J10 ~ J22, J40 ~ J43, J69, J80, J85, J86 (Supplement Table 1). These ICD-10 codes in- cluded pneumonia, bronchitis, lung abscess, and so on. Patients with a lower respiratory system infection diagnosis at both ED admission and discharge after hospitalization were included. The extracted clinical information included age, sex, and comorbid diseases. Moreover, we calculated the Charlson Comorbidity Index (CCI) for each patient, cate- gorizing the patients’ comorbidities based on the administrative data [23]. The initial vital signs and laboratory data in the ED were also col- lected. The BUN/ALB ratio, CRP/ALB ratio, Lac/ALB ratio, and LDH/ALB ratio were calculated from the initial laboratory data. To verify the pa- tient’s severity, the pneumonia severity index , Modified early warning score and confusion, uremia, elevated respiratory rate, hypotension, and ages >65 years (CURB-65) were also calculated [24-26]. In-hospital mortality was the primary outcome of this study.

• 1. Statistical analyses

Categorical variables were analyzed using the ?2 test or Fisher’s exact test, and continuous variables were analyzed using Student’s t-test or the Mann-Whitney U test. After univariable analysis, multi- variable logistic regression was performed to verify the independence of the LDH/ALB ratio as a prognostic factor for the in-hospital mortality of LRTI. Variables with p < 0.1 in univariable analysis were used in the multivariable logistic regression. Moreover, we performed the receiver-operating characteristics (ROC) curves analysis and de- termined the area under the curve (AUC) for individual measures asso- ciated with in-hospital mortality. The AUCs of the models were calculated and tested mutually for significance by DeLong or bootstrap test. All statistical analyses were conducted using statistical software R version 4.0.5 (The R Foundation for Statistical Computing, Vienna, Austria). P < 0.05 was considered statistically significant.

1. Result
   1. Patient demographics

From January 2018 to December 2020, 4592 patients were diag- nosed with LRTI according to the ICD-10 code. Of these, 3011 patients were hospitalized, and 1524 patients were discharged (Fig. 1). The over- all mortality of the included patients was 8.9% (407/4592). The mortal- ity of patients admitted to the ICU was 31.6% (74/234). The median (IQR) age of the included patients was 71.0 [55.0;80.0] years, and 2625 patients (57.2%) were men. And the median of LDH/ALB ratio was higher in the non-survivor group (217.6 [160.3;312.0] vs. 126.4 [100.3;165.1], p < 0.001) (Table 1, Supplement Fig. 1). The demographic characteristics, comorbid diseases, vital signs, initial laboratory findings, and severity indexes are provided in Table 1.

• 1. The AUC of the LDH/ALB ratio compared to other albumin-based in- dexes and severity index

Fig. 2 shows the AUCs of albumin-based ratios and severity scores for predicting the in-hospital mortality of LRTI. The AUC of the LDH/ALB

Fig. 1. Flow diagram for the study.

ED = emergency department; ICD-10 = International Statistical Classification of Diseases and Related Health Problems 10th Revision code. ICD-10 (+): included ICD-10 for the lower respiratory tract infection were J10 ~ J22, J40 ~ J43, J69, J80, J85, J86.

ratio was 0.808 (95% confidence interval (CI): 0.757-0.842, p < 0.001). AUCs of other albumin-based ratios were 0.785 (95% CI: 0.758-0.811, p < 0.001) in the BUN/ALB ratio, 0.762 (95% CI: 0.733-0.781, p < 0.001)

in the Lac/ALB ratio, 0.729 and (95% CI: 0.698-0.760, p < 0.001) in the CRP/ALB ratio. The AUC of the LDH/ALB ratio showed better results compared to other albumin-based ratios, except for the BUN/ALB ratio. Furthermore, the AUC values of severity indexes like PSI, MEWS and CURB-65, were 0.761 (95% CI: 0.734-0.789, p < 0.001), 0.656 (95% CI:

0.624-0.699, p < 0.001), and 0.744 (95% CI: 0.717-0.771, p < 0.001),

respectively. The LDH/ALB ratio was a stronger predictor of in-hospital mortality compared to severity scales.

• 1. Multivariable logistic regression analysis for the in-hospital mortality of LRTI

In the univariable analysis, variables affecting in-hospital mortality of LRTI were further analyzed by multivariable logistic regression. Variables included vital signs, SpO₂ < 90%, white blood cell count, platelet, albumin, and albumin-based ratios. Variables affecting LDH and albumin were also analyzed; these included sex, age, transfer from long-term care facilities, comorbid disease (diabetes, stroke, dementia, and malignancy). MEWS included vital signs and was developed for grading their severity. Thus, MEWS was analyzed as a substitute for vital signs such as blood pressure, heart rate, respiration rate, body temperature, and Glasgow coma scale. WBC and platelet count were categorized using the WBC criteria (WBC < 4000 or > 12000 /mm³) and thrombocytopenia (< 150000/mm³). The results are presented in Table 2 and Supplement Table 2.

The crude odds ratio (OR) was 1.01 (95% CI: 1.01-1.01, p < 0.001) for the LDH/ALB ratio, 1.13 (95% CI: 1.11-1.14, p < 0.001) for the BUN/ALB

ratio, 1.26 (95% CI: 1.23-1.30, p < 0.001) for the CRP/ALB ratio and 2.87 (95% CI: 2.50-3.32, p < 0.01) for the Lac/ALB ratio. The adjusted OR (aOR) of the LDH/ALB ratio for in-hospital mortality was 1.00 (95% CI: 1.00-1.00, p < 0.001).

• 1. The OR of the categorized LDH/ALB ratio for the in-hospital mortality

The ideal cut-off of the LDH/Albumin ratio for in-hospital mortality using Youden’s index was 166.3 (Sensitivity 73.3%, specificity 74.9%) (Table 3). Using the cut-off value, the LDH/Albumin ratio was catego- rized into 3 groups: below 150, between 150 and 250, and over 250. Furthermore, the aOR of the categorized LDH/ALB ratio for in-hospital mortality was 1.00 (reference) for LDH/ALB ratios <150, 1.93 (95% CI:

1.35-2.77, p < 0.001) for LDH/ALB ratios 150-250, and 4.15 (95% CI:

2.72-6.35, p < 0.001) for LDH/ALB ratios >250 (Table 4).

1. Discussion

LRTI, including pneumonia, is a frequent cause of sepsis and the fourth leading cause of death worldwide [27]. Thus, many studies have been conducted to assess the severity and prognosis of LRTI and

Table 1

Basic characteristics of included patients.

Total Non-survivors Survivors p

	N = 4592		N = 407		N = 4185
Age Sex (Male)	71.0 [55.0;80.0] 2625 (57.2%)		78.0 [73.0;84.0] 269 (66.1%)		70.0 [54.0;79.0] 2356 (56.3%)	<0.001 <0.001
Comorbidity Diabetes 1083 (23.6%)		130 (31.9%)		953 (22.8%)		<0.001
Heart failure 231 (5.0%)		25 (6.1%)		206 (4.9%)		0.339
Malignancy 709 (15.4%)		104 (25.6%)		605 (14.5%)		<0.001
Stroke 754 (16.4%)		92 (22.6%)		662 (15.8%)		0.001

Chronic renal disease Chronic liver disease Dementia

413 (9.0%)	45 (11.1%)	368 (8.8%)	0.152
181 (3.9%)	21 (5.2%)	160 (3.8%)	0.234
877 (19.1%)	132 (32.4%)	745 (17.8%)	<0.001

CCI	4 [2; 6]	5 [4; 6]	4 [1; 5]	<0.001
Transfer	774 (16.9%)	115 (28.3%)	659 (15.7%)	<0.001
LTCF	483 (10.5%)	100 (24.6%)	383 (9.2%)	<0.001
Vital Signs
SBP, mmHg	126 [110;143]	116 [95;140]	127 [112;143]	<0.001
DBP, mmHg	75.0 [66.0;85.0]	68.0 [59.0;79.0]	76.0 [67.0;85.0]	<0.001
Heart rate, /min	100 [86.0;113]	105 [89.0;119]	99.0 [85.0;112]	<0.001
Respiration rate, /min	22.0 [20.0;24.0]	26.0 [22.0;30.0]	20.0 [20.0;24.0]	<0.001
Body temperature, ?C	37.5 [36.8;38.3]	37.1 [36.5;37.7]	37.5 [36.8;38.4]	<0.001
SpO2, %	96.0 [95.0;98.0]	94.0 [86.0;97.0]	97.0 [95.0;98.0]	<0.001
Glasgow coma scale	15.0 [15.0;15.0]	14.0 [10.0;15.0]	15.0 [15.0;15.0]	<0.001
Initial laboratory
data WBC, /mm3	9.7 [6.8;13.7]	11.8 [7.8;16.5]	9.5 [6.8;13.4]	<0.001
Platelet, 103/mm3	221 [170;285]	204 [134;286]	222 [172;285]	<0.001
Glucose, mg/dL	122 [104;158]	139 [112;187]	121 [103;155]	<0.001
Albumin, g/dL	3.4 [2.8; 3.9]	2.5 [2.1; 2.9]	3.5 [2.9; 3.9]	<0.001
Blood urea nitrogen, mg/dL	15.3 [11.0;23.8]	27.0 [17.0;42.0]	15.0 [11.0;22.0]	<0.001
Lactate
dehydrogenase,	426 [354;530]	521 [406;734]	415 [350;510]	<0.001
U/L
C-reactive protein, mg/dL	6.6 [2.0;13.8]	13.8 [7.9;20.2]	6.10 [1.70;12.8]	<0.001
Lactate, mmol/L	1.7 [1.3; 2.4]	2.3 [1.6; 3.9]	1.7 [1.3; 2.3]	<0.001
Severity Scales
MEWS	3 [2; 5]	5 [3; 7]	3 [2; 5]	<0.001
CURB-65	1 [0; 2]	3 [2; 4]	1 [0; 2]	<0.001
PSI	97 [70;123]	135 [111;163]	93 [67;118]	<0.001
Albumin-based
ratio
BUN/ALB ratio	4.6 [3.1; 7.8]	10.7 [6.7;17.4]	4.3 [3.0; 7.0]	<0.001
CRP/ALB ratio	2.0 [0.5; 4.6]	5.4 [2.9; 9.1]	1.8 [0.5; 4.2]	<0.001
Lac/ALB ratio	0.5 [0.4; 0.8]	1.0 [0.6; 1.6]	0.5 [0.4; 0.7]	<0.001
LDH/ALB ratio	132.7 [104.0;181.5]	217.6 [160.3;312.0]	126.4 [100.3;165.1]	<0.001
ICU admission	234 (5.1%)	74 (18.2%)	160 (3.8%)	<0.001
ED-LOS, hours		8.2 [5.1;18.4]	5.20 [3.3; 9.9]	<0.001

Categorical variables are expressed by number (%) and continuous variables are expressed by median [interquartile range (IQR)] or mean +- standard deviation (SD). CCI=Charlson comorbidity index; LTCF = long-term care facility; SBP = systolic blood pressure; DBP= Diastolic Blood Pressure; SpO₂ = oxygen saturation measured by pulse oximetry; WBC = white blood cell; MEWS = modified early warning score; CURB-65 = confusion, urea nitrogen, respiration rate, blood pressure, age equal or over 65 yrs.; PSI = pneumonia severity index; BUN/ALB ratio = blood urea nitrogen to albumin ratio; CRP/ALB ratio = C-reactive protein to albumin ratio; Lac/ALB ratio = lactate to albumin ratio; LDH/ALB ratio = lactate dehydrogenase to albumin ratio; ICU = intensive care unit; ED-LOS = length of stay in the emergency department.

provide adequate management and disposition according to its severity. Furthermore, several studies have been performed to find and evaluate new independent prognostic factors and to evaluate a combination of independent prognostic factors to improve the accuracy and speed of prognosis assessment. However, LRTI has similar respiratory symptoms regardless of its causes (such as community-acquired, nosocomial, aspi- ration, bacteria, virus, fungus). For this reason, it is difficult to identify

and differentiate between these various causes in the busy ED environ- ment; and the accuracy of the developed indicators was often poor when applied to different patient groups. Thus, the fact that our study included overall causes of LRTI highlights the significance of our results. In our study, the AUC of the LDH/ALB ratio for the in-hospital mortal- ity of LRTI was relatively superior to other variables (Fig. 2). Additionally, in the multivariable logistic regression analysis, the LDH/ALB ratio was an independent prognostic factor for the in-hospital mortality of LRTI (Tables 3, 4). These results suggest that the LDH/ALB ratio could be a good independent prognostic factor for the in-hospital mortality of LRTI. In relation to LRTI and sepsis, there have been many attempts to in- crease the accuracy of prognosis with a combination of simple prognos- tic factors. Among them, several albumin-based ratios were identified as prognostic factors in several studies [11-19]. Albumin, which is used as the base for these indicators, is known as a negative acute-phase protein in inflammation reactions. Thus, hypoalbuminemia or decrease of serum Albumin level could be related to infection. However, hypoalbu- minemia is also related to chronic comorbid disease or lifestyle (smoking, alcoholism, and obesity). For this reason, albumin has been studied as a single prognostic indicator for infection but also as a combi- nation indicator with other prognostic factors such as albumin-based ratios [7-19]. The prognostic factors combined with albumin were BUN, CRP, lactate, procalcitonin, and others. These factors have worth for LRTI or sepsis with different causes. BUN included the end products of nitrogen metabolism and reflects kidney injury among the organ damages caused by Inflammatory processes and has been studied as a prognostic factor for infection [17-19]. CRP is an acute-phase Protein Synthesized in the liver in response to inflammation induced by various causes. Moreover, it is widely used to evaluate the strength of the sys- temic inflammatory response [28]. Serum lactate level may represent tissue hypoxia associated with signs of organ dysfunction in critically ill patients and has been studied as a prognostic factor for pneumonia and sepsis [12,14]. Furthermore, procalcitonin is a specified prognostic factor for bacterial infection. These prognostic factors showed improved

accuracy when combined with albumin levels.

However, an increased level of serum LDH was shown by a different mechanism to other factors. LDH is a cytoplasmatic enzyme expressed in nearly all types of cells of the body. It is released into the blood when cells experience injury or death caused by ischemia, excess heat or cold, starvation, dehydration, bacterial toxins, drugs and chemical poisonings, and increases in serum LDH level. Moreover, serum LDH has been shown to increase during acute and severe lung damage, and elevated serum LDH values have been found in other interstitial lung in- fections [29]. Thus, LDH was proven to be a prognostic factor for pneu- monia and sepsis [2-4,18,19,23-26,30,31]. However, no studies have been conducted on whether the LDH/ALB ratio is useful as a prognostic factor for mortality in LRTI. Thus, we hypothesized that the LDH/ALB ratio would make up for LDH and could be a better prognostic factor compared to other albumin-based ratios for LRTI. Notably, the result of our study supported this hypothesis.

As mentioned above, the AUCs of PSI and CURB-65 were lower than those of the LDH/ALB ratio in our study. PSI and CURB-65 were more complex scoring systems for pneumonia than LDH/ALB ratio. Thus, it was expected that the PSI and CURB-65 would be better pre- dictors of prognosis. However, in our study, the AUC of LDH/ALB ratio for the in-hospital mortality was higher than those of PSI and CURB-

65. This could be explained in several ways. First, it seems that the LDH/ALB ratio could better reflect the severity of the overall cause of LRTI. Second, CURB-65 and PSI were dependent on the age and dis- ease severity of the study group. Thus, their accuracy was lower in the older age and severe sepsis groups [32,33]. Furthermore, our study did not include outpatients. Thus, our study group could have a rela- tively slight increase in severity and age compared to studies that in- cluded outpatients. Third, the diagnosis criteria of this study included the overall causes of LRTI; however, severity scores such as CURB-65 were developed for the prognostication of community-acquired pneumonia. Thus, it might not be able to reflect the severity of

Variables	AUC (95%CI)	p1	p2		Variables	AUC (95%CI)	p1	p2
LDH/ALB ratio	0.808	<0.001			LDH/ALB ratio	0.808	<0.001

(0.757-0.842)

BUN/ALB ratio 0.785

(0.758-0.811)

CRP/ALB ratio 0.729

(0.698-0.760)

Lac/ALB ratio 0.762

(0.757-0.842)

<0.001 0.157 CURB-65 0.744

(0.717-0.771)

<0.001 <0.001 MEWS 0.656

(0.624-0.689)

<0.001 0.010 PSI 0.761

<0.001 <0.001

<0.001 <0.001

<0.001 0.011

(0.733-0.781) (0.734-0.789)

Fig. 2. Comparison of AUCs for predicting in-hospital mortality.

The AUCs of the models were calculated and tested mutually for significance by DeLong or bootstrap tests (p₁ = p-value for the AUC of each variable; p₂ = p-value for the equality compared to LDH/ALB ratio).

AUC = area under the curve; CI = confidence interval; LDH/ALB ratio = lactate dehydrogenase to albumin ratio; BUN/ALB ratio = blood urea nitrogen to albumin ratio; CRP/ALB ratio = C-reactive protein to albumin ratio; Lac/ALB ratio = lactate to albumin ratio; CURB-65 = confusion, urea nitrogen, respiration rate, blood pressure, age equal or over 65 yrs.; MEWS = modified early warning score; PSI = pneumonia severity index.

Table 2

Multivariable logistic regression analysis for the in-hospital mortality.

Crude OR (95% CI) p Adjusted OR (95% CI) p

Age category

< 50 yrs Reference Reference

50-59 yrs 3.28 (1.57-7.19) 0.002 1.73 (0.53-6.25) 0.376

60-69 yrs 4.83 (2.55-9.94) <0.001 2.57 (0.95-8.33) 0.084

70-79 yrs 11.46 (6.47-22.58) <0.001 4.41 (1.74-13.68) 0.004

over 80 yrs 14.61 (8.29-28.68) <0.001 4.74 (1.86-14.77) 0.003

Sex: Male 1.51 (1.22-1.88) <0.001 1.00 (0.73-1.36) 0.987

LTCF 3.23 (2.51-4.13) <0.001 1.21 (0.85-1.72) 0.289

Diabetes 1.59 (1.27-1.98) <0.001 0.99 (0.72-1.35) 0.958

Malignancy 2.03 (1.59-2.57) <0.001 1.29 (0.92-1.80) 0.131

Stroke 1.55 (1.21-1.98) <0.001 0.85 (0.58-1.23) 0.398

Dementia 2.22 (1.77-2.76) <0.001 0.90 (0.63-1.27) 0.540

MEWS 1.37 (1.31-1.43) <0.001 1.11 (1.04-1.19) 0.001

SpO₂ < 90% 7.33 (5.72-9.36) <0.001 2.49 (1.80-3.45) <0.001

WBC criteria 2.24 (1.82-2.76) <0.001 1.50 (1.12-1.99) 0.006

Thrombocytopenia 2.26 (1.79-2.85) <0.001 1.54 (1.11-2.14) 0.010

Albumin 0.14 (0.12-0.17) <0.001 0.36 (0.26-0.49) <0.001

other causes of LRTI. Last, this study was a retrospective study. All data were extracted from electronic medical records, and all severity scores were calculated from these data. Therefore, each score could be different from real scores and might have a bias. And these could be our study limitations.

This study had several additional limitations. First, this is a single- center study, leading to limitations in the generalizability of the results. Second, the study’s retrospective nature might have distorted the results due to selection bias or missing data. Third, the procalcitonin to albumin ratio could not be compared because of insufficient procalcitonin data. Fourth, although we checked the in-hospital mortality, we did not con- sider long-term and other prognoses. This limitation was also caused by the retrospective nature of our study. Fifth, we could not confirm the usefulness of the LDH/ALB ratio for each cause because classification according to the cause of infection was not performed. Thus, the LDH/ ALB ratio could be useless in some subgroups of LRTI. However, since identifying the exact causes of LRTI infection in the ED is usually impos- sible, this limitation should not be a problem in actual clinical practice. Fi- nally, the LDH/ALB ratio seEMS useful as an initial prognostic indicator because it showed a good AUC value for the overall LRTI. Thus, despite

BUN/ALB ratio CRP/ALB ratio Lac/ALB ratio	1.13 (1.11-1.14) 1.26 (1.23-1.30) 2.87 (2.50-3.32)	<0.001 <0.001 <0.001	1.03 (1.01-1.05) 1.03 (0.99-1.07) 1.14 (0.95-1.38)	0.003 0.194 0.154	this limitation, the LDH/ALB ratio could be an independent prognostic factor for the mortality of LRTI in the ED. A Prospective analysis is needed
LDH/ALB ratio	1.01 (1.01-1.01)	<0.001	1.00 (1.00-1.00)	<0.001	in the future to overcome the overall limitations of this study.

Adjusted ORs were the result of the multivariable logistic regression analysis by using the stepwise selection method.

OR = odds ratio; CI = confidence interval; LTCF = long-term care facility; MEWS = mod- ified early warning score; SpO₂ = oxygen saturation measured by pulse oximetry; WBC criteria = white blood cell criteria (WBC < 4000 or > 12,000/mm³); Thrombocytopenia

= platelet <150,000/mm³; BUN/ALB ratio = blood urea nitrogen to albumin ratio; CRP/ ALB ratio = C-reactive protein to albumin ratio; Lac/ALB ratio = lactate to albumin ratio; LDH/ALB ratio = lactate dehydrogenase to albumin ratio.

1. Conclusion

The LDH/ALB ratio showed better results compared with other albumin-based ratios and other severity scales. Therefore LDH/ALB ratio could be used as an independent prognostic factor for the in- hospital mortality in LRTI patients.

Table 3

Comparison in each cut-off values of LDH/ALB ratio for the in-hospital mortality of lower respiratory tract infection.

Cut-off Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI)

100	0.969	(0.949-0.986)	0.227	(0.208-0.245)	0.157	(0.153-0.161)	0.981	(0.967-0.991)
110	0.955	(0.932-0.979)	0.326	(0.306-0.345)	0.174	(0.168-0.179)	0.980	(0.969-0.990)
150	0.801	(0.753-0.849)	0.660	(0.639-0.681)	0.259	(0.243-0.276)	0.957	(0.948-0.967)
166.3 ?	0.733	(0.682-0.781)	0.749	(0.732-0.769)	0.302	(0.282-0.325)	0.950	(0.941-0.959)
200	0.562	(0.503-0.616)	0.845	(0.829-0.862)	0.350	(0.320-0.383)	0.929	(0.920-0.938)
250	0.404	(0.356-0.462)	0.923	(0.910-0.934)	0.436	(0.384-0.488)	0.913	(0.906-0.920)
300	0.257	(0.205-0.308)	0.961	(0.952-0.970)	0.497	(0.423-0.571)	0.897	(0.891-0.904)

LDH/ALB ratio = lactate dehydrogenase to albumin ratio; CI = confidence interval; PPV = positive predictive value; NPV = negative predictive value.

* Youden index.

Table 4

The adjusted odds ratio of the categorized LDH/ALB ratio for the in-hospital mortality. LDH/ALB ratio Adjusted OR (95% CI) p

<150	Reference
150-250	1.93	(1.35-2.77)	<0.001
>250	4.15	(2.72-6.35)	<0.001

The odds ratio was adjusted with age, sex, transferred from a long-term care facility, dia- betes, malignancy, stroke, dementia, modified early warning scores, pulse oximetry satu- ration < 90%, white blood cell criteria, platelet criteria, albumin, blood urea nitrogen to albumin ratio, C-reactive protein to albumin ratio, lactate to albumin ratio.

LDH/ALB ratio = lactate dehydrogenase to albumin ratio; CI = confidence interval.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influ- ence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2021.11.028.

References

1. Fernandez P, Torres A, Miro JM, Vieigas C, Mallolas J, Zamora L, et al. Prognostic fac- tors influencing the outcome in pneumocystis carinii pneumonia in patients with AIDS. Thorax. 1995;50:668-71. https://doi.org/10.1136/thx.50.6.668.
2. Takano Y, Sakamoto O, Suga M, Muranaka H, Ando M. Prognostic factors of nosoco- mial pneumonia in general wards: a prospective multivariate analysis in Japan. Respir Med. 2002;96:18-23. https://doi.org/10.1053/rmed.2001.1201.
3. Zein JG, Lee GL, Tawk M, Dabaja M, Kinasewitz GT. Prognostic significance of ele- vated serum Lactate dehydrogenase in patients with severe Sepsis. CHEST. 2004;126:873S. https://doi.org/10.1378/chest.126.4_MeetingAbstracts.873S.
4. Lu J, Wei Z, Jiang H, Cheng L, Chen Q, Chen M, et al. Lactate dehydrogenase is asso- ciated with 28-day mortality in patients with sepsis: a retrospective observational study. J Surg Res. 2018;228:314-21. https://doi.org/10.1016/j.jss.2018.03.035.
5. Miyashita N, Horita N, Higa F, Aoki Y, Kikuchi T, Seki M, et al. Validation of a diagnos- tic score model for the prediction of legionella pneumophila pneumonia. J Infect Chemother. 2019;25:407-12. https://doi.org/10.1016/j.jiac.2019.03.009.
6. Pan F, Yang L, Li Y, Liang B, Li L, Ye T, et al. Factors associated with death outcome in patients with severe coronavirus disease-19 (COVID-19): a case-control study. Int J Med Sci. 2020;17:1281-92. https://doi.org/10.7150/ijms.46614.
7. Yin M, Si L, Qin W, Li C, Zhang J, Yang H, et al. Predictive value of Serum albumin level for the prognosis of severe Sepsis without exogenous human albumin administra- tion: a prospective cohort study. J Intensive Care Med. 2018;33:687-94. https:// doi.org/10.1177/0885066616685300.
8. Arnau-Barres I, Guerri-Fernandez R, Luque S, Sorli L, Vazquez O, Miralles R. Serum albumin is a strong predictor of sepsis outcome in elderly patients. Eur J Clin Microbiol Infect Dis. 2019;38:743-6. https://doi.org/10.1007/s10096-019-03478-2.
9. Godinez-Vidal AR, Correa-Montoya A, Enriquez-Santos D, Perez-Escobedo SU, Lopez-Romero SC, Gracida-Mancilla NI. Is albumin a predictor of severity and mor- tality in patients with abdominal sepsis? Cir Cir. 2019;87:485-9. https://doi.org/10. 24875/CIRU.180003903.
10. Takegawa R, Kabata D, Shimizu K, Hisano S, Ogura H, Shintani A, et al. Serum albu- min as a risk factor for death in patients with prolonged sepsis: an observational study. J Crit Care. 2019;51:139-44. https://doi.org/10.1016/j.jcrc.2019.02.004.
11. Hwang YJ, Chung SP, Park YS, Chung HS, Lee HS, Park JW, et al. Newly designed delta neutrophil index-to-serum albumin ratio prognosis of early mortality in severe sep- sis. Am J Emerg Med. 2015;33:1577-82. https://doi.org/10.1016/j.ajem.2015.06.012.
12. Wang B, Chen G, Cao Y, Xue J, Li J, Wu Y. Correlation of lactate/albumin ratio level to organ failure and mortality in severe sepsis and septic shock. J Crit Care. 2015;30: 271-5. https://doi.org/10.1016/j.jcrc.2014.10.030.
13. Luo X, Yang X, Li J, Zou G, Lin Y, Qing G, et al. The procalcitonin/albumin ratio as an early diagnostic predictor in discriminating urosepsis from patients with febrile uri- nary tract infection. Medicine (Baltimore). 2018;97:e11078. https://doi.org/10. 1097/MD.0000000000011078.
14. Shin J, Hwang SY, Jo IJ, Kim WY, Ryoo SM, Kang GH, et al. Prognostic value of the lac- tate/albumin ratio for predicting 28-day mortality in critically ILL Sepsis patients. Shock. 2018;50:545-50. https://doi.org/10.1097/SHK.0000000000001128.
15. Deng S, Gao J, Zhao Z, Tian M, Li Y, Gong Y. Albumin/Procalcitonin ratio is a sensitive early marker of nosocomial blood stream infection in patients with intra-cerebral hemorrhage. Surg Infect (Larchmt). 2019;20:643-9. https://doi.org/10.1089/sur. 2018.260.
16. Gong Y, Li D, Cheng B, Ying B, Wang B. Increased neutrophil percentage-to-albumin ratio is associated with all-cause mortality in patients with severe sepsis or septic shock. Epidemiol Infect. 2020;148:e87. https://doi.org/10.1017/S0950268820000771.
17. Ugajin M, Yamaki K, Iwamura N, Yagi T, Asano T. Blood urea nitrogen to serum albu- min ratio independently predicts mortality and severity of community-acquired pneumonia. Int J Gen Med. 2012;5:583-9. https://doi.org/10.2147/IJGM.S33628.
18. Feng DY, Zhou YQ, Zou XL, Zhou M, Yang HL, Chen XX, et al. Elevated blood urea nitrogen-to-serum albumin ratio as a factor that negatively affects the mortality of patients with Hospital-acquired pneumonia. Can J Infect Dis Med Microbiol. 2019; 2019:1547405. https://doi.org/10.1155/2019/1547405.
19. Ryu S, Oh SK, Cho SU, You Y, Park JS, Min JH, et al. Utility of the blood urea nitrogen to serum albumin ratio as a prognostic factor of mortality in Aspiration pneumonia patients. Am J Emerg Med. 2020. https://doi.org/10.1016/j.ajem.2020.02.045.
20. Feng JF, Wang L, Yang X, Jiang YH. Prognostic value of lactate dehydrogenase to al- bumin ratio (LAR) in patients with resectable esophageal Squamous cell carcinoma. Cancer Manag Res. 2019;11:7243. https://doi.org/10.2147/CMAR.S208320.
21. Aday U, Tatli F, Akpulat FV, Inan M, Kafadar MT, Bilge H, et al. Prognostic significance of pretreatment serum lactate dehydrogenase-to-albumin ratio in gastric cancer. Contemp Oncol (Pozn). 2020;24:145. https://doi.org/10.5114/wo.2020.100219.
22. Aday U, Boyuk A, Akkoc H. The prognostic significance of serum lactate dehydrogenase-to-albumin ratio in Colorectal cancer. Ann Surg Treat Res. 2020; 99:161. https://doi.org/10.4174/astr.2020.99.3.161.
23. Charlson ME, Pompei P, Ales KL, MacKenzie CRJJocd.. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-83. https://doi.org/10.1016/0021-9681(87)90171-8.
24. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336:243-50. https://doi.org/10.1056/NEJM199701233360402.
25. Subbe CP, Kruger M, Rutherford P, Gemmel LJQ. Validation of a modified early warn- ing score in medical admissions. QJM. 2001;94:521-6. https://doi.org/10.1093/ qjmed/94.10.521.
26. Lim W, Van der Eerden M, Laing R, Boersma W, Karalus N, Town G, et al. Defining community acquired pneumonia severity on presentation to hospital: an interna- tional derivation and validation study. Thorax. 2003;58:377-82. https://doi.org/10. 1136/thorax.58.5.377.
27. World Health Organization. The top ten causes of death; 2020, December 9 Re- trieved from https://www.who.int/news-room/fact-sheets/detail/the-top-10- causes-of-death [asccessed 10 July 2021].
28. Ansar W, Ghosh S. Inflammation and inflammatory diseases, markers, and media- tors: Role of CRP in some inflammatory diseases. Biology of C Reactive Protein in Health and Disease: Springer. 2016:67-107. https://doi.org/10.1007/978-81-322- 2680-2_4.
29. Luna CM, Perin MMJAoTM.. Can Lactate dehydrogenase be used as a marker of severity of pneumonia in patients with renal transplant? Ann Transl Med. 2020;8. https://doi.org/10.21037/atm-2020-34.
30. Wu M-y, Yao L, Wang Y, Zhu X-y, Wang X-f, Tang P-j, et al. Clinical evaluation of po- tential usefulness of serum lactate dehydrogenase (LDH) in 2019 novel coronavirus (COVID-19) pneumonia. Respir Res 2020;21:1-6. doi:https://doi.org/10.1186/s12 931-020-01427-8.
31. Ewig S, Bauer T, Hasper E, Pizzulli L, Kubini R, Luderitz B. Prognostic analysis and pre- dictive rule for outcome of hospital-treated community-acquired pneumonia. Eur Respir J. 1995;8:392-7. https://doi.org/10.1183/09031936.95.08030392.
32. Parsonage M, Nathwani D, Davey P. Barlow GJCM, infection. Evaluation of the perfor- mance of CURB-65 with increasing age. Clin Microbiol Infect. 2009;15:858-64. https://doi.org/10.1111/j.1469-0691.2009.02908.x.
33. Richards G, Levy H, Laterre P-F, Feldman C, Woodward B, Bates BM, et al. CURB-65, PSI, and APACHE II to assess mortality risk in patients with severe sepsis and com- munity acquired pneumonia in PROWESS. J Intensive Care Med. 2011;26:34-40. https://doi.org/10.1177/0885066610383949.