American Journal of Emergency Medicine 52 (2022) 54–59

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage: www.elsevier.com/locate/ajem

Lactate dehydrogenase to albumin ratio as a prognostic factor in lower

respiratory tract infection patients
Bong-Kyu Lee, MD a, Seung Ryu, MD, PhD a,⁎, Se-Kwang Oh, MD, PhD a, Hong-Joon Ahn, MD, PhD a,
So-Young Jeon, MD a, Won-Joon Jeong, MD, PhD a, Yong-Chul Cho, MD a, Jung-Soo Park, MD, PhD b,
Yeon-Ho You, MD, PhD a, Chang-Shin Kang, MD a
a
Department of Emergency Medicine, Chungnam National University Hospital, Jung-Gu, Daejeon, Republic of Korea
b
Department of Emergency Medicine, School of medicine, Chungnam National University, Jung-Gu, Daejeon, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: To verify the role of lactate dehydrogenase to albumin (LDH/ALB) ratio as an independent prognostic
Received 4 July 2021 factor for mortality due to the lower respiratory tract infection (LRTI) in the emergency department (ED).
Received in revised form 24 October 2021 Methods: We reviewed the electronic medical records of patients who were admitted to the ED for the
Accepted 15 November 2021 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-
Keywords: based ratios (blood urea nitrogen to albumin ratio, C-reactive protein to albumin ratio, and lactate to albu-
Lactate dehydrogenase
min ratio) and severity scales (pneumonia severity index, modified early warning score, CURB-65 scores),
Albumin
which are being used as prognostic factors for in-hospital mortality. Multivariable logistic regression was
Pneumonia
Mortality 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 (AUC) 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.
© 2021 Elsevier Inc. All rights reserved.

1. Introduction ratio), lactate to albumin ratio (Lac/ALB ratio), C-reactive protein

to albumin ratio (CRP/ALB ratio), and procalcitonin to albumin
Lower respiratory tract infection (LRTI) is one of the most common ratio were used to predict the severity and prognosis of patients
infectious diseases worldwide; it includes a range of respiratory infec- with pneumonia and sepsis [11-19].
tions such as pneumonia, bronchitis, pneumonitis, and empyema/ab- The LDH to albumin ratio (LDH/ALB ratio) can be used as a prognos-
scess. Furthermore, it is associated with considerable morbidity and tic factor of pneumonia with improved accuracy. However, no studies
mortality. Because delays in the ICU admission of LRTI patients have have been conducted on whether the LDH/ALB ratio is useful as a prog-
been shown to be associated with increased mortality, it can be helpful nostic factor for mortality in LRTI, although it has been evaluated as a
to predict the mortality of LRTI patients in the emergency department prognostic factor for malignancy [20-22]. Thus, we hypothesized that
(ED). Therefore, it is important to increase the accuracy of prognostic the LDH/ALB ratio would make up for LDH and could be a better prog-
factors that predict the severity of LRTI. nostic factor compared to other albumin-based ratios for LRTI. This
Serum lactate dehydrogenase level (LDH) has been mentioned in study aimed to verify whether the LDH/ALB ratio can be used as an in-
some studies as a prognostic factor of pneumonia and sepsis [1-6]. It dependent biomarker prognostic factor of LRTI in the ED.
tends to rise with infection severity. Moreover, serum albumin was
utilized as a prognostic factor of infection because it tends to
2. Materials and methods
decrease with infection aggravation [7-10]. By using albumin, sev-
eral indexes like blood urea nitrogen to albumin ratio (BUN/ALB
2.1. Study design and setting
⁎ Corresponding author at: Department of Emergency Medicine, Chungnam National
University Hospital, 282 Munhwa-ro, Jung-Gu, Daejeon 35015, Republic of Korea. The study was conducted by reviewing the secondary data extracted
E-mail address: [email protected] (S. Ryu). from electronic medical records (EMR) of patients who visited the

https://doi.org/10.1016/j.ajem.2021.11.028
0735-6757/© 2021 Elsevier Inc. All rights reserved.
B.-K. Lee, S. Ryu, S.-K. Oh et al. American Journal of Emergency Medicine 52 (2022) 54–59

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.

2.2. 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 Fig. 1. Flow diagram for the study.
10th Revision (ICD-10) code with the following ICD-10 codes: J10 ~ J22, ED = emergency department; ICD-10 = International Statistical Classification of Diseases
J40 ~ J43, J69, J80, J85, J86 (Supplement Table 1). These ICD-10 codes in- 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.
cluded pneumonia, bronchitis, lung abscess, and so on. Patients with a
lower respiratory system infection diagnosis at both ED admission
ratio was 0.808 (95% confidence interval (CI): 0.757–0.842, p < 0.001).
and discharge after hospitalization were included. The extracted clinical
AUCs of other albumin-based ratios were 0.785 (95% CI: 0.758–0.811,
information included age, sex, and comorbid diseases. Moreover, we
p < 0.001) in the BUN/ALB ratio, 0.762 (95% CI: 0.733–0.781, p < 0.001)
calculated the Charlson comorbidity index (CCI) for each patient, cate-
in the Lac/ALB ratio, 0.729 and (95% CI: 0.698–0.760, p < 0.001) in the
gorizing the patients' comorbidities based on the administrative data
CRP/ALB ratio. The AUC of the LDH/ALB ratio showed better results
[23]. The initial vital signs and laboratory data in the ED were also col-
compared to other albumin-based ratios, except for the BUN/ALB ratio.
lected. The BUN/ALB ratio, CRP/ALB ratio, Lac/ALB ratio, and LDH/ALB
Furthermore, the AUC values of severity indexes like PSI, MEWS and
ratio were calculated from the initial laboratory data. To verify the pa-
CURB-65, were 0.761 (95% CI: 0.734–0.789, p < 0.001), 0.656 (95% CI:
tient's severity, the pneumonia severity index (PSI), modified early
0.624–0.699, p < 0.001), and 0.744 (95% CI: 0.717–0.771, p < 0.001),
warning score (MEWS) and confusion, uremia, elevated respiratory
respectively. The LDH/ALB ratio was a stronger predictor of in-hospital
rate, hypotension, and ages >65 years (CURB-65) were also calculated
mortality compared to severity scales.
[24-26]. In-hospital mortality was the primary outcome of this study.

3.3. Multivariable logistic regression analysis for the in-hospital mortality of

2.3. Statistical analyses
LRTI

Categorical variables were analyzed using the χ2 test or Fisher's

In the univariable analysis, variables affecting in-hospital mortality of
exact test, and continuous variables were analyzed using Student's
LRTI were further analyzed by multivariable logistic regression. Variables
t-test or the Mann-Whitney U test. After univariable analysis, multi-
included vital signs, SpO2 < 90%, white blood cell (WBC) count, platelet,
variable logistic regression was performed to verify the independence
albumin, and albumin-based ratios. Variables affecting LDH and albumin
of the LDH/ALB ratio as a prognostic factor for the in-hospital mortality
were also analyzed; these included sex, age, transfer from long-term care
of LRTI. Variables with p < 0.1 in univariable analysis were used in
facilities, comorbid disease (diabetes, stroke, dementia, and malignancy).
the multivariable logistic regression. Moreover, we performed the
MEWS included vital signs and was developed for grading their severity.
receiver-operating characteristics (ROC) curves analysis and de-
Thus, MEWS was analyzed as a substitute for vital signs such as blood
termined the area under the curve (AUC) for individual measures asso-
pressure, heart rate, respiration rate, body temperature, and Glasgow
ciated with in-hospital mortality. The AUCs of the models were
coma scale. WBC and platelet count were categorized using the
calculated and tested mutually for significance by DeLong or bootstrap
WBC criteria (WBC < 4000 or > 12000 /mm3) and thrombocytopenia
test. All statistical analyses were conducted using statistical software R
(< 150000/mm3). The results are presented in Table 2 and Supplement
version 4.0.5 (The R Foundation for Statistical Computing, Vienna,
Table 2.
Austria). P < 0.05 was considered statistically significant.
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
3. Result 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
3.1. Patient demographics (aOR) of the LDH/ALB ratio for in-hospital mortality was 1.00 (95% CI:
1.00–1.00, p < 0.001).
From January 2018 to December 2020, 4592 patients were diag-
nosed with LRTI according to the ICD-10 code. Of these, 3011 patients 3.4. The OR of the categorized LDH/ALB ratio for the in-hospital mortality
were hospitalized, and 1524 patients were discharged (Fig. 1). The over-
all mortality of the included patients was 8.9% (407/4592). The mortal- The ideal cut-off of the LDH/Albumin ratio for in-hospital mortality
ity of patients admitted to the ICU was 31.6% (74/234). The median using Youden's index was 166.3 (Sensitivity 73.3%, specificity 74.9%)
(IQR) age of the included patients was 71.0 [55.0;80.0] years, and (Table 3). Using the cut-off value, the LDH/Albumin ratio was catego-
2625 patients (57.2%) were men. And the median of LDH/ALB ratio rized into 3 groups: below 150, between 150 and 250, and over 250.
was higher in the non-survivor group (217.6 [160.3;312.0] vs. 126.4 Furthermore, the aOR of the categorized LDH/ALB ratio for in-hospital
[100.3;165.1], p < 0.001) (Table 1, Supplement Fig. 1). The demographic mortality was 1.00 (reference) for LDH/ALB ratios <150, 1.93 (95% CI:
characteristics, comorbid diseases, vital signs, initial laboratory findings, 1.35–2.77, p < 0.001) for LDH/ALB ratios 150–250, and 4.15 (95% CI:
and severity indexes are provided in Table 1. 2.72–6.35, p < 0.001) for LDH/ALB ratios >250 (Table 4).

3.2. The AUC of the LDH/ALB ratio compared to other albumin-based in- 4. Discussion
dexes and severity index
LRTI, including pneumonia, is a frequent cause of sepsis and the
Fig. 2 shows the AUCs of albumin-based ratios and severity scores fourth leading cause of death worldwide [27]. Thus, many studies
for predicting the in-hospital mortality of LRTI. The AUC of the LDH/ALB have been conducted to assess the severity and prognosis of LRTI and

55
B.-K. Lee, S. Ryu, S.-K. Oh et al. American Journal of Emergency Medicine 52 (2022) 54–59

Table 1 and differentiate between these various causes in the busy ED environ-
Basic characteristics of included patients. ment; and the accuracy of the developed indicators was often poor
Total Non-survivors Survivors p when applied to different patient groups. Thus, the fact that our study
included overall causes of LRTI highlights the significance of our results.
N = 4592 N = 407 N = 4185
In our study, the AUC of the LDH/ALB ratio for the in-hospital mortal-
Age 71.0 [55.0;80.0] 78.0 [73.0;84.0] 70.0 [54.0;79.0] <0.001 ity of LRTI was relatively superior to other variables (Fig. 2). Additionally,
Sex (Male) 2625 (57.2%) 269 (66.1%) 2356 (56.3%) <0.001
in the multivariable logistic regression analysis, the LDH/ALB ratio was
Comorbidity
Diabetes 1083 (23.6%) 130 (31.9%) 953 (22.8%) <0.001 an independent prognostic factor for the in-hospital mortality of LRTI
Heart failure 231 (5.0%) 25 (6.1%) 206 (4.9%) 0.339 (Tables 3, 4). These results suggest that the LDH/ALB ratio could be a
Malignancy 709 (15.4%) 104 (25.6%) 605 (14.5%) <0.001 good independent prognostic factor for the in-hospital mortality of LRTI.
Stroke 754 (16.4%) 92 (22.6%) 662 (15.8%) 0.001 In relation to LRTI and sepsis, there have been many attempts to in-
Chronic renal
413 (9.0%) 45 (11.1%) 368 (8.8%) 0.152 crease the accuracy of prognosis with a combination of simple prognos-
disease
Chronic liver tic factors. Among them, several albumin-based ratios were identified as
181 (3.9%) 21 (5.2%) 160 (3.8%) 0.234
disease prognostic factors in several studies [11-19]. Albumin, which is used as
Dementia 877 (19.1%) 132 (32.4%) 745 (17.8%) <0.001 the base for these indicators, is known as a negative acute-phase protein
CCI 4 [2; 6] 5 [4; 6] 4 [1; 5] <0.001 in inflammation reactions. Thus, hypoalbuminemia or decrease of
Transfer 774 (16.9%) 115 (28.3%) 659 (15.7%) <0.001
serum albumin level could be related to infection. However, hypoalbu-
LTCF 483 (10.5%) 100 (24.6%) 383 (9.2%) <0.001
Vital Signs minemia is also related to chronic comorbid disease or lifestyle
SBP, mmHg 126 [110;143] 116 [95;140] 127 [112;143] <0.001 (smoking, alcoholism, and obesity). For this reason, albumin has been
DBP, mmHg 75.0 [66.0;85.0] 68.0 [59.0;79.0] 76.0 [67.0;85.0] <0.001 studied as a single prognostic indicator for infection but also as a combi-
Heart rate, /min 100 [86.0;113] 105 [89.0;119] 99.0 [85.0;112] <0.001 nation indicator with other prognostic factors such as albumin-based
Respiration
22.0 [20.0;24.0] 26.0 [22.0;30.0] 20.0 [20.0;24.0] <0.001 ratios [7-19]. The prognostic factors combined with albumin were
rate, /min
Body BUN, CRP, lactate, procalcitonin, and others. These factors have worth
37.5 [36.8;38.3] 37.1 [36.5;37.7] 37.5 [36.8;38.4] <0.001
temperature, °C for LRTI or sepsis with different causes. BUN included the end products
SpO2, % 96.0 [95.0;98.0] 94.0 [86.0;97.0] 97.0 [95.0;98.0] <0.001 of nitrogen metabolism and reflects kidney injury among the organ
Glasgow coma damages caused by inflammatory processes and has been studied as a
15.0 [15.0;15.0] 14.0 [10.0;15.0] 15.0 [15.0;15.0] <0.001
scale
prognostic factor for infection [17-19]. CRP is an acute-phase protein
Initial laboratory
data synthesized in the liver in response to inflammation induced by various
WBC, /mm3 9.7 [6.8;13.7] 11.8 [7.8;16.5] 9.5 [6.8;13.4] <0.001 causes. Moreover, it is widely used to evaluate the strength of the sys-
Platelet,
221 [170;285] 204 [134;286] 222 [172;285] <0.001
temic inflammatory response [28]. Serum lactate level may represent
103/mm3 tissue hypoxia associated with signs of organ dysfunction in critically
Glucose, mg/dL 122 [104;158] 139 [112;187] 121 [103;155] <0.001
ill patients and has been studied as a prognostic factor for pneumonia
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 and sepsis [12,14]. Furthermore, procalcitonin is a specified prognostic
15.3 [11.0;23.8] 27.0 [17.0;42.0] 15.0 [11.0;22.0] <0.001 factor for bacterial infection. These prognostic factors showed improved
nitrogen, mg/dL
Lactate accuracy when combined with albumin levels.
dehydrogenase, 426 [354;530] 521 [406;734] 415 [350;510] <0.001 However, an increased level of serum LDH was shown by a different
U/L
mechanism to other factors. LDH is a cytoplasmatic enzyme expressed
C-reactive
6.6 [2.0;13.8] 13.8 [7.9;20.2] 6.10 [1.70;12.8] <0.001 in nearly all types of cells of the body. It is released into the blood
protein, mg/dL
Lactate, mmol/L 1.7 [1.3; 2.4] 2.3 [1.6; 3.9] 1.7 [1.3; 2.3] <0.001 when cells experience injury or death caused by ischemia, excess heat
Severity Scales or cold, starvation, dehydration, bacterial toxins, drugs and chemical
MEWS 3 [2; 5] 5 [3; 7] 3 [2; 5] <0.001 poisonings, and increases in serum LDH level. Moreover, serum LDH
CURB-65 1 [0; 2] 3 [2; 4] 1 [0; 2] <0.001
has been shown to increase during acute and severe lung damage, and
PSI 97 [70;123] 135 [111;163] 93 [67;118] <0.001
Albumin-based elevated serum LDH values have been found in other interstitial lung in-
ratio fections [29]. Thus, LDH was proven to be a prognostic factor for pneu-
BUN/ALB ratio 4.6 [3.1; 7.8] 10.7 [6.7;17.4] 4.3 [3.0; 7.0] <0.001 monia and sepsis [2-4,18,19,23-26,30,31]. However, no studies have
CRP/ALB ratio 2.0 [0.5; 4.6] 5.4 [2.9; 9.1] 1.8 [0.5; 4.2] <0.001 been conducted on whether the LDH/ALB ratio is useful as a prognostic
Lac/ALB ratio 0.5 [0.4; 0.8] 1.0 [0.6; 1.6] 0.5 [0.4; 0.7] <0.001
factor for mortality in LRTI. Thus, we hypothesized that the LDH/ALB
132.7 217.6 126.4
LDH/ALB ratio
[104.0;181.5] [160.3;312.0] [100.3;165.1]
<0.001 ratio would make up for LDH and could be a better prognostic factor
ICU admission 234 (5.1%) 74 (18.2%) 160 (3.8%) <0.001 compared to other albumin-based ratios for LRTI. Notably, the result of
ED-LOS, hours 8.2 [5.1;18.4] 5.20 [3.3; 9.9] <0.001 our study supported this hypothesis.
Categorical variables are expressed by number (%) and continuous variables are expressed As mentioned above, the AUCs of PSI and CURB-65 were lower
by median [interquartile range (IQR)] or mean ± standard deviation (SD). CCI=Charlson than those of the LDH/ALB ratio in our study. PSI and CURB-65 were
comorbidity index; LTCF = long-term care facility; SBP = systolic blood pressure; DBP= more complex scoring systems for pneumonia than LDH/ALB ratio.
Diastolic Blood Pressure; SpO2 = oxygen saturation measured by pulse oximetry; WBC = Thus, it was expected that the PSI and CURB-65 would be better pre-
white blood cell; MEWS = modified early warning score; CURB-65 = confusion, urea
dictors of prognosis. However, in our study, the AUC of LDH/ALB ratio
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 = for the in-hospital mortality was higher than those of PSI and CURB-
C-reactive protein to albumin ratio; Lac/ALB ratio = lactate to albumin ratio; LDH/ALB 65. This could be explained in several ways. First, it seems that the
ratio = lactate dehydrogenase to albumin ratio; ICU = intensive care unit; ED-LOS = LDH/ALB ratio could better reflect the severity of the overall cause
length of stay in the emergency department. 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
provide adequate management and disposition according to its severity. older age and severe sepsis groups [32,33]. Furthermore, our study
Furthermore, several studies have been performed to find and evaluate did not include outpatients. Thus, our study group could have a rela-
new independent prognostic factors and to evaluate a combination of tively slight increase in severity and age compared to studies that in-
independent prognostic factors to improve the accuracy and speed of cluded outpatients. Third, the diagnosis criteria of this study included
prognosis assessment. However, LRTI has similar respiratory symptoms the overall causes of LRTI; however, severity scores such as CURB-65
regardless of its causes (such as community-acquired, nosocomial, aspi- were developed for the prognostication of community-acquired
ration, bacteria, virus, fungus). For this reason, it is difficult to identify pneumonia. Thus, it might not be able to reflect the severity of

56
B.-K. Lee, S. Ryu, S.-K. Oh et al. American Journal of Emergency Medicine 52 (2022) 54–59

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) (0.757-0.842)
BUN/ALB ratio 0.785 <0.001 0.157 CURB-65 0.744 <0.001 <0.001
(0.758-0.811) (0.717-0.771)
CRP/ALB ratio 0.729 <0.001 <0.001 MEWS 0.656 <0.001 <0.001
(0.698-0.760) (0.624-0.689)
Lac/ALB ratio 0.762 <0.001 0.010 PSI 0.761 <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.

other causes of LRTI. Last, this study was a retrospective study. All
data were extracted from electronic medical records, and all severity
Table 2 scores were calculated from these data. Therefore, each score could
Multivariable logistic regression analysis for the in-hospital mortality. be different from real scores and might have a bias. And these could
Crude OR (95% CI) p Adjusted OR (95% CI) p be our study limitations.
Age category
This study had several additional limitations. First, this is a single-
< 50 yrs Reference Reference center study, leading to limitations in the generalizability of the results.
50–59 yrs 3.28 (1.57–7.19) 0.002 1.73 (0.53–6.25) 0.376 Second, the study's retrospective nature might have distorted the results
60–69 yrs 4.83 (2.55–9.94) <0.001 2.57 (0.95–8.33) 0.084 due to selection bias or missing data. Third, the procalcitonin to albumin
70–79 yrs 11.46 (6.47–22.58) <0.001 4.41 (1.74–13.68) 0.004 ratio could not be compared because of insufficient procalcitonin data.
over 80 yrs 14.61 (8.29–28.68) <0.001 4.74 (1.86–14.77) 0.003
Fourth, although we checked the in-hospital mortality, we did not con-
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 sider long-term and other prognoses. This limitation was also caused
Diabetes 1.59 (1.27–1.98) <0.001 0.99 (0.72–1.35) 0.958 by the retrospective nature of our study. Fifth, we could not confirm
Malignancy 2.03 (1.59–2.57) <0.001 1.29 (0.92–1.80) 0.131 the usefulness of the LDH/ALB ratio for each cause because classification
Stroke 1.55 (1.21–1.98) <0.001 0.85 (0.58–1.23) 0.398 according to the cause of infection was not performed. Thus, the LDH/
Dementia 2.22 (1.77–2.76) <0.001 0.90 (0.63–1.27) 0.540
ALB ratio could be useless in some subgroups of LRTI. However, since
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 identifying the exact causes of LRTI infection in the ED is usually impos-
WBC criteria 2.24 (1.82–2.76) <0.001 1.50 (1.12–1.99) 0.006 sible, this limitation should not be a problem in actual clinical practice. Fi-
Thrombocytopenia 2.26 (1.79–2.85) <0.001 1.54 (1.11–2.14) 0.010 nally, the LDH/ALB ratio seems useful as an initial prognostic indicator
Albumin 0.14 (0.12–0.17) <0.001 0.36 (0.26–0.49) <0.001 because it showed a good AUC value for the overall LRTI. Thus, despite
BUN/ALB ratio 1.13 (1.11–1.14) <0.001 1.03 (1.01–1.05) 0.003
this limitation, the LDH/ALB ratio could be an independent prognostic
CRP/ALB ratio 1.26 (1.23–1.30) <0.001 1.03 (0.99–1.07) 0.194
Lac/ALB ratio 2.87 (2.50–3.32) <0.001 1.14 (0.95–1.38) 0.154 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. 5. Conclusion
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 The LDH/ALB ratio showed better results compared with other
= platelet <150,000/mm3; BUN/ALB ratio = blood urea nitrogen to albumin ratio; CRP/ albumin-based ratios and other severity scales. Therefore LDH/ALB
ALB ratio = C-reactive protein to albumin ratio; Lac/ALB ratio = lactate to albumin ratio could be used as an independent prognostic factor for the in-
ratio; LDH/ALB ratio = lactate dehydrogenase to albumin ratio. hospital mortality in LRTI patients.

57
B.-K. Lee, S. Ryu, S.-K. Oh et al. American Journal of Emergency Medicine 52 (2022) 54–59

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.

[13] Luo X, Yang X, Li J, Zou G, Lin Y, Qing G, et al. The procalcitonin/albumin ratio as an
Table 4 early diagnostic predictor in discriminating urosepsis from patients with febrile uri-
The adjusted odds ratio of the categorized LDH/ALB ratio for the in-hospital mortality. nary tract infection. Medicine (Baltimore). 2018;97:e11078. https://doi.org/10.
1097/MD.0000000000011078.
LDH/ALB ratio Adjusted OR (95% CI) p
[14] Shin J, Hwang SY, Jo IJ, Kim WY, Ryoo SM, Kang GH, et al. Prognostic value of the lac-
<150 Reference tate/albumin ratio for predicting 28-day mortality in critically ILL Sepsis patients.
150–250 1.93 (1.35–2.77) <0.001 Shock. 2018;50:545–50. https://doi.org/10.1097/SHK.0000000000001128.
>250 4.15 (2.72–6.35) <0.001 [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
The odds ratio was adjusted with age, sex, transferred from a long-term care facility, dia- hemorrhage. Surg Infect (Larchmt). 2019;20:643–9. https://doi.org/10.1089/sur.
betes, malignancy, stroke, dementia, modified early warning scores, pulse oximetry satu- 2018.260.
ration < 90%, white blood cell criteria, platelet criteria, albumin, blood urea nitrogen to [16] Gong Y, Li D, Cheng B, Ying B, Wang B. Increased neutrophil percentage-to-albumin
albumin ratio, C-reactive protein to albumin ratio, lactate to albumin ratio. ratio is associated with all-cause mortality in patients with severe sepsis or septic
LDH/ALB ratio = lactate dehydrogenase to albumin ratio; CI = confidence interval. 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
Declaration of Competing Interest 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
The authors declare that they have no known competing financial
patients with hospital-acquired pneumonia. Can J Infect Dis Med Microbiol. 2019;
interests or personal relationships that could have appeared to influ- 2019:1547405. https://doi.org/10.1155/2019/1547405.
ence the work reported in this paper. [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.
Appendix A. Supplementary data [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.
Supplementary data to this article can be found online at https://doi. Cancer Manag Res. 2019;11:7243. https://doi.org/10.2147/CMAR.S208320.
[21] Aday U, Tatlı F, Akpulat FV, İnan M, Kafadar MT, Bilge H, et al. Prognostic significance
org/10.1016/j.ajem.2021.11.028. 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.
References [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;
[1] Fernandez P, Torres A, Miro JM, Vieigas C, Mallolas J, Zamora L, et al. Prognostic fac- 99:161. https://doi.org/10.4174/astr.2020.99.3.161.
tors influencing the outcome in pneumocystis carinii pneumonia in patients with [23] Charlson ME, Pompei P, Ales KL, MacKenzie CRJJocd.. A new method of classifying
AIDS. Thorax. 1995;50:668–71. https://doi.org/10.1136/thx.50.6.668. prognostic comorbidity in longitudinal studies: development and validation. J
[2] Takano Y, Sakamoto O, Suga M, Muranaka H, Ando M. Prognostic factors of nosoco- Chronic Dis. 1987;40:373–83. https://doi.org/10.1016/0021-9681(87)90171-8.
mial pneumonia in general wards: a prospective multivariate analysis in Japan. [24] Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction
Respir Med. 2002;96:18–23. https://doi.org/10.1053/rmed.2001.1201. rule to identify low-risk patients with community-acquired pneumonia. N Engl J
[3] Zein JG, Lee GL, Tawk M, Dabaja M, Kinasewitz GT. Prognostic significance of ele- Med. 1997;336:243–50. https://doi.org/10.1056/NEJM199701233360402.
vated serum lactate dehydrogenase (LDH) in patients with severe Sepsis. CHEST. [25] Subbe CP, Kruger M, Rutherford P, Gemmel LJQ. Validation of a modified early warn-
2004;126:873S. https://doi.org/10.1378/chest.126.4_MeetingAbstracts.873S. ing score in medical admissions. QJM. 2001;94:521–6. https://doi.org/10.1093/
qjmed/94.10.521.
[4] Lu J, Wei Z, Jiang H, Cheng L, Chen Q, Chen M, et al. Lactate dehydrogenase is asso-
[26] Lim W, Van der Eerden M, Laing R, Boersma W, Karalus N, Town G, et al. Defining
ciated with 28-day mortality in patients with sepsis: a retrospective observational
community acquired pneumonia severity on presentation to hospital: an interna-
study. J Surg Res. 2018;228:314–21. https://doi.org/10.1016/j.jss.2018.03.035.
tional derivation and validation study. Thorax. 2003;58:377–82. https://doi.org/10.
[5] Miyashita N, Horita N, Higa F, Aoki Y, Kikuchi T, Seki M, et al. Validation of a diagnos-
1136/thorax.58.5.377.
tic score model for the prediction of legionella pneumophila pneumonia. J Infect
[27] World Health Organization. The top ten causes of death; 2020, December 9 Re-
Chemother. 2019;25:407–12. https://doi.org/10.1016/j.jiac.2019.03.009.
trieved from https://www.who.int/news-room/fact-sheets/detail/the-top-10-
[6] Pan F, Yang L, Li Y, Liang B, Li L, Ye T, et al. Factors associated with death outcome in
causes-of-death [asccessed 10 July 2021].
patients with severe coronavirus disease-19 (COVID-19): a case-control study. Int J
[28] Ansar W, Ghosh S. Inflammation and inflammatory diseases, markers, and media-
Med Sci. 2020;17:1281–92. https://doi.org/10.7150/ijms.46614.
tors: Role of CRP in some inflammatory diseases. Biology of C Reactive Protein in
[7] Yin M, Si L, Qin W, Li C, Zhang J, Yang H, et al. Predictive value of serum albumin level Health and Disease: Springer. 2016:67–107. https://doi.org/10.1007/978-81-322-
for the prognosis of severe Sepsis without exogenous human albumin administra- 2680-2_4.
tion: a prospective cohort study. J Intensive Care Med. 2018;33:687–94. https:// [29] Luna CM, Perín MMJAoTM.. Can lactate dehydrogenase (LDH) be used as a marker of
doi.org/10.1177/0885066616685300. severity of pneumonia in patients with renal transplant? Ann Transl Med. 2020;8.
[8] Arnau-Barres I, Guerri-Fernandez R, Luque S, Sorli L, Vazquez O, Miralles R. Serum https://doi.org/10.21037/atm-2020-34.
albumin is a strong predictor of sepsis outcome in elderly patients. Eur J Clin [30] Wu M-y, Yao L, Wang Y, Zhu X-y, Wang X-f, Tang P-j, et al. Clinical evaluation of po-
Microbiol Infect Dis. 2019;38:743–6. https://doi.org/10.1007/s10096-019-03478-2. tential usefulness of serum lactate dehydrogenase (LDH) in 2019 novel coronavirus
[9] Godinez-Vidal AR, Correa-Montoya A, Enriquez-Santos D, Perez-Escobedo SU, (COVID-19) pneumonia. Respir Res 2020;21:1–6. doi:https://doi.org/10.1186/s12
Lopez-Romero SC, Gracida-Mancilla NI. Is albumin a predictor of severity and mor- 931-020-01427-8.
tality in patients with abdominal sepsis? Cir Cir. 2019;87:485–9. https://doi.org/10. [31] Ewig S, Bauer T, Hasper E, Pizzulli L, Kubini R, Luderitz B. Prognostic analysis and pre-
24875/CIRU.180003903. dictive rule for outcome of hospital-treated community-acquired pneumonia. Eur
[10] Takegawa R, Kabata D, Shimizu K, Hisano S, Ogura H, Shintani A, et al. Serum albu- Respir J. 1995;8:392–7. https://doi.org/10.1183/09031936.95.08030392.
min as a risk factor for death in patients with prolonged sepsis: an observational [32] Parsonage M, Nathwani D, Davey P. Barlow GJCM, infection. Evaluation of the perfor-
study. J Crit Care. 2019;51:139–44. https://doi.org/10.1016/j.jcrc.2019.02.004. mance of CURB-65 with increasing age. Clin Microbiol Infect. 2009;15:858–64.
[11] Hwang YJ, Chung SP, Park YS, Chung HS, Lee HS, Park JW, et al. Newly designed delta https://doi.org/10.1111/j.1469-0691.2009.02908.x.
neutrophil index-to-serum albumin ratio prognosis of early mortality in severe sep- [33] Richards G, Levy H, Laterre P-F, Feldman C, Woodward B, Bates BM, et al. CURB-65,
sis. Am J Emerg Med. 2015;33:1577–82. https://doi.org/10.1016/j.ajem.2015.06.012. PSI, and APACHE II to assess mortality risk in patients with severe sepsis and com-
[12] Wang B, Chen G, Cao Y, Xue J, Li J, Wu Y. Correlation of lactate/albumin ratio level to munity acquired pneumonia in PROWESS. J Intensive Care Med. 2011;26:34–40.
organ failure and mortality in severe sepsis and septic shock. J Crit Care. 2015;30: https://doi.org/10.1177/0885066610383949.
271–5. https://doi.org/10.1016/j.jcrc.2014.10.030.

58