DTI Drug Target Insights 2022; 16: 12-16ISSN 1177-3928 | DOI: 10.33393/dti.2022.2422ORIGINAL RESEARCH ARTICLE Drug Target Insights - ISSN 1177-3928 - www.aboutscience.eu/dti © 2022 The Authors. This article is published by AboutScience and licensed under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). Commercial use is not permitted and is subject to Publisher’s permissions. Full information is available at www.aboutscience.eu The association of ESBL Escherichia coli with mortality in patients with Escherichia coli bacteremia at the emergency department Pariwat Phungoen1, Jessada Sarunyaparit1, Korakot Apiratwarakul1, Lumyai Wonglakorn2, Atibordee Meesing3, Kittisak Sawanyawisuth3 1Department of Emergency Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen - Thailand 2Department of Microbiology, Faculty of Medicine, Khon Kaen University, Khon Kaen - Thailand 3Department of Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen - Thailand ABSTRACT Background: Escherichia coli is a common bloodstream infection pathogen in the emergency department (ED). Patients with extended-spectrum beta-lactamase (ESBL) E. coli have a higher risk of morbidity. However, there is still debate surrounding ESBL E. coli-associated mortality in community, intensive care unit, and tertiary care set- tings. In addition, there have been few studies regarding mortality in ESBL E. coli in ED settings, and results have been contradictory. Methods: This was a retrospective cohort study conducted at the Department of Emergency Medicine, Faculty of Medicine, Khon Kaen University in Thailand aimed at evaluating the possible association between ESBL E. coli bacteremia and mortality in the ED. The inclusion criteria were age 18 years or over, clinical presentation suspi- cious of infection, and positive blood culture for E. coli. Predictors for mortality were analyzed by logistic regres- sion analysis. Results: During the study period, 273 patients presented at the ED with hemoculture positive for E. coli. Of those, 27 (9.89%) died. Five factors remained in the final model, of which plasma glucose levels, serum lactate levels, and ESBL E. coli were significantly associated with 28-day mortality in the ED with adjusted odds ratios of 0.970, 1.258, and 12.885, respectively. Plasma glucose of less than 113 mg/dL yielded a sensitivity of 80.95% and speci- ficity of 64.29%, while serum lactate over 2.4 mmol/L had a sensitivity of 81.48% and specificity of 45.50%. Conclusion: ESBL E. coli, plasma glucose, and serum lactate levels were associated with 28-day mortality in patients with E. coli bacteremia presenting at the ED. Keywords: Extended-spectrum beta-lactamase-producing Escherichia coli, Glucose, Lactate Received: May 11, 2022 Accepted: September 19, 2022 Published online: October 17, 2022 Corresponding author: Kittisak Sawanyawisuth Department of Medicine Khon Kaen University Khon Kaen 40002 - Thailand kittisak@kku.ac.th rate of E. coli BSI to be 9.6%. Male patients aged 70 years or older are at higher risk of 30-day mortality with adjusted inci- dence rate ratios of 1.26 and 10.35 (3). Another study found a mortality rate of 30.6% in patients infected with extended- spectrum beta-lactamase (ESBL) E. coli vs 22.2% in those infected with non-ESBL strains or Klebsiella pneumoniae (4). The prevalence of drug-resistant Gram-negative bacteria is increasing, particularly in in-hospital, intensive care unit (ICU), and tertiary care settings (5-8). A study from a mul- tispecialty hospital in India found that rates of multidrug- resistant Gram-negative bacteria increased from 26.16% in 2012 to 33.33% in 2014 (6). Additionally, urinary tract infec- tion patients with resistant Enterobacteriaceae have been shown to be 1.447 times more likely to have severe sepsis or septic shock at presentation than those with nonresistant strains (9). Data regarding the association of ESBL E. coli and mortality in community, ICU, and tertiary care settings have been inconclusive. Two studies conducted in community set- tings, for example, found differences in mortality between Introduction Bloodstream infection (BSI) with Gram-negative bacteria is common in the emergency department (ED), accounting for 39.4% of ED patients with suspected infection (1). A study from China found that Escherichia coli was the most common Gram-negative BSI in 3,199 patients and accounted for 34.3% of cases (2). One population-based study found the mortality https://doi.org/10.33393/dti.2022.2422 https://creativecommons.org/licenses/by-nc/4.0/legalcode mailto:kittisak@kku.ac.th Phungoen et al Drug Target Insights 2022; 16: 13 © 2022 The Authors. Published by AboutScience - www.aboutscience.eu patients with ESBL and non-ESBL bacteremia (10,11), whereas a study from a tertiary care setting found compara- ble rates (9.7% vs 9.2%), as did a study in a teaching hospital in China (12,13). However, another study in a teaching hos- pital in Japan found higher rate of mortality in patients with ESBL strains (14), as did a study in an ICU (37.5% vs 15.6%; p = 0.04) (15). This study thus aimed to evaluate if ESBL E. coli bacteremia was associated with mortality in an ED setting. Methods This was a retrospective cohort study conducted at the Department of Emergency Medicine, Faculty of Medicine, Khon Kaen University in Thailand as part of an ED infection project. The inclusion criteria were age 18 years or over, clini- cal presentation suspicious of infection, and positive blood culture for E. coli. Patients who received prophylactic antibi- otics, presented with cardiac arrest or symptoms related to trauma, were referred from other hospitals, or had missing clinical data were excluded. The study period was between 2016 and 2018. Eligible patients were selected from the hospital database. We reviewed participants’ clinical data at the time of presen- tation as well as mortality data over the following 28 days. Clinical data included baseline characteristics, laboratory results, and treatment. Baseline characteristics reviewed were age, sex, comorbid diseases, Charlson Comorbidity Index, physical signs, and quick Sepsis Related Organ Failure Assessment (qSOFA) score. Laboratory results included complete blood count, chemistry, arterial blood gas, serum lactate levels, and blood culture results (for ESBL E. coli posi- tivity). The primary outcome was 28-day mortality. Statistical analyses Eligible patients were categorized into two groups by mortality. Descriptive statistics were used to calculate differ- ences between the two groups. Predictors for mortality were analyzed using logistic regression analysis. Univariate logistic analysis was used to calculate the unadjusted odds ratio with 95% confidence interval and p value for each factor. Factors with a p value less than 0.05 by univariate logistic regression analysis or those that were clinically significant were subse- quently subjected to stepwise, multivariate logistic regression analysis. The final model was tested for goodness of fit using the Hosmer-Lemeshow method. Results were reported as unadjusted/adjusted odds ratios with their 95% confidence intervals. A numerical predictor for mortality as an appropri- ate diagnostic cutoff point was computed with its sensitivity and specificity. All statistical analyses were performed using STATA version 10.1 (College Station, Texas, USA). Results During the study period, 273 patients presented at the ED with hemoculture positive for E. coli. Of those, 27 (9.89%) died. In terms of baseline characteristics and physical signs, there were 12 factors that differed significantly between those who survived and those who died (Tab. I). For example, nonsurvivors had a significantly higher Charlson Comorbidity Index (5 vs 4), respiratory rate (28 vs 24 breaths/min), and qSOFA score (2 vs 1), but oxygen saturation at presentation was lower (96% vs 97%; p 0.040). qSOFA scores were sig- nificantly higher in those who died than those who survived (2 vs 1; p < 0.001). With regard to laboratory tests and treatment, seven factors differed significantly between groups (Tab. II). For example, the nonsurvival group had significantly lower lev- els of serum bicarbonate (17 vs 21 mEq/L) and plasma glu- cose (94 vs 131 mg/dL), higher serum lactate levels (4.5 vs 2.6 mmol/L), and a greater percentage of patients with ESBL TABLE I - Baseline characteristics of patients with Escherichia coli bacteremia presenting at the emergency department categorized by mortality at 28 days Factors Survivors n = 246 Nonsurvivors n = 27 p- Value Age, years 66 (18-100) 73 (19-93) 0.161 Male sex 125 (50.81) 10 (37.04) 0.224 Comorbid diseases Liver disease 50 (20.33) 10 (37.04) 0.053 Diabetes 57 (23.85) 2 (7.41) 0.053 CKD (moderate-severe) 24 (9.76) 4 (14.81) 0.499 Solid organ tumor 74 (30.08) 15 (55.56) 0.010 Palliative care 4 (1.63) 4 (14.81) 0.004 Leukemia 2 (0.81) 1 (3.70) 0.269 Lymphoma 2 (0.81) 2 (7.41) 0.050 Hypertension 90 (36.59) 7 (25.93) 0.299 HIV infection 2 (0.81) 0 0.999 Cholangiocarcinoma 34 (13.82) 8 (29.63) 0.045 Charlson Comorbidity Index 4 (0-12) 5 (1-12) <0.001 Temperature, °C 38.6 (35.9-41.5) 38.2 (35.6-41.0) 0.184 Pulse rate, beats/min 96 (58-190) 96 (52-148) 0.898 Respiratory rate, breaths/ min 24 (18-50) 28 (18-40) 0.008 SBP, mm Hg 126 (64-218) 112 (80-167) 0.003 DBP, mm Hg 70 (33-112) 67 (37-95) 0.071 MAP, mm Hg 91 (48-138) 80 (56-119) 0.009 Oxygen saturation, % 97 (60-100) 96 (65-100) 0.040 GCS 15 (4-15) 15 (7-15) <0.001 Sepsis score qSOFA 1 (0-3) 2 (1-3) <0.001 Data are presented as median (range) or number (percentage). CKD = chronic kidney disease; DBP = diastolic blood pressure; GCS = Glasgow coma scale; MAP = mean arterial pressure; qSOFA = quick Sepsis Related Organ Failure Assessment; SBP = systolic blood pressure. Data presented as number (percentage) unless indicated otherwise. ESBL E. coli ED14 © 2022 The Authors. Drug Target Insights - ISSN 1177-3928 - www.aboutscience.eu/dti E. coli (14.81% vs 2.03%) than the survival group. In addition, patients in the nonsurvival group underwent significantly more aggressive treatment (such as vasopressor treatment) and had a higher rate of ICU admission. However, duration of hospital stay in the nonsurvival group was shorter (8 vs 11 days; p 0.013). Five factors remained in the final model for predicting death (Tab. III). Plasma glucose, serum lactate levels, and ESBL E. coli were significantly associated with mortality, with adjusted odds ratios of 0.970, 1.258, and 12.885, respectively. The final model had a Hosmer-Lemeshow Chi square of 6.73 (p = 0.565). Plasma glucose of 113 mg/dL or lower yielded a sensitivity of 80.95% and specificity of 64.29%, while serum lactate level of over 2.4 mmol/L had a sensitivity of 81.48% and specificity of 45.50%. Discussion The prevalence of ESBL E. coli bacteremia at the ED in this study was 3.29%, which is lower than previously reported in community settings (6.7%-9.5%) (10,11,16). In addition to the difference in setting, these results may indicate differ- ing rates among countries, as higher rates have been found in developed countries (South Korea and Spain). A previous report found that frequent visits to the ED increased the risk of ESBL bacteremia by a factor of 9.98, including in those patients who had undergone previous antibiotic treatment. In Thailand, the rate of previous antibiotic use may be lower than in some other countries. Despite the inconsistency in the ESBL E. coli mortality rate in other settings, this study found that patients with ESBL E. coli had a 13 times higher risk of mortality than those with non-ESBL strains. Other fac- tors associated with mortality in patients with E. coli infection may be personal characteristics and inappropriate antibiotic use. A report from Korea found that presenting with septic shock or malignancy increased mortality risk by 26.6 and 11.9 times, respectively, while another study found that mortal- ity rates were comparable in patients with ESBL and non- ESBL E. coli if antibiotics were administered appropriately (p = 0.23) (11,15). Hypoglycemia has been shown to be related with higher mortality in sepsis patients and critically ill patients (17-19). Although the causal relationship between hypoglycemia and mortality is not well understood, several mechanisms have been proposed including the inhibition of the physiologi- cal responses of hormones such as insulin and epinephrine, TABLE II - Laboratory results and treatment of patients with Escherichia coli bacteremia presenting at the emergency depart- ment categorized by mortality at 28 days Factors Survivors n = 246 Nonsurvivors n = 27 p- Value Hb, g/dL 11.0 (4.6-16.0) 9.6 (4.8-13.8) 0.002 WBC, ×103/mm3 33.8 (13.0-51.7) 26.6 (14.9-41.9) 0.010 Platelet, ×106 179 (4-584) 138 (13-451) 0.098 BUN, mg/dL 17.9 (3.7-144.8) 27.3 (6.7-153.2) 0.009 Creatinine, mg/dL 1.1 (0.4-10.4) 1.5 (0.5-11.1) 0.118 Bicarbonate, mEq/L 21 (7-30) 17 (7-27) <0.001 Total bilirubin, mg/dL 1.4 (0.2-33.8) 2.1 (0.3-33.8) 0.251 Glucose, mg/dL 131 (53-548) 94 (35-172) <0.001 PaO 2 , mmHg 76 (23-512) 89 (33-253) 0.702 pH 7.44 (7.16-7.58) 7.40 (7.11-7.56) 0.194 Lactate level, mmol/L 2.6 (0.5-18.3) 4.5 (1.3-17.9) 0.003 ESBL E. coli 5 (2.03) 4 (14.81) 0.007 Treatment Mechanical ventilator 21 (8.54) 5 (18.52) 0.155 ICU admission 83 (3.74) 17 (62.96) 0.005 Vasopressor* 62 (25.20) 21 (77.78) <0.001 LOS 11 (2-56) 8 (1-54) 0.013 Data are presented as median (range) or number (percentage). BUN = blood urea nitrogen; ESBL E. coli = extended-spectrum beta-lactamase- producing Escherichia coli; Hb = hemoglobin; ICU = intensive care unit; LOS = length of stay; PaO 2 = partial pressure of oxygen; pH = power of hydro- gen; WBC = white blood cell. *indicates that the patient received norepinephrine, adrenaline, or dopamine. TABLE III - Factors associated with a 28-day mortality in patients with Escherichia coli bacteremia presenting at the emergency department Factors Unadjusted odds ratio (95% confidence interval) Adjusted odds ratio (95% confidence interval) Oxygen saturation 0.933 (0.883, 0.985) 0.934 (0.860, 1.015) Hemoglobin 0.732 (0.605, 0.885) 0.856 (0.662, 1.106) Plasma glucose 0.971 (0.955, 0.987) 0.970 (0.954, 0.987) Serum lactate level 1.185 (1.067, 1.317) 1.258 (1.090, 1.451) ESBL E. coli* 8.382 (2.103, 33.407) 12.885 (1.082, 153.338) Age 1.008 (0.981, 1.036) Not retained Sex 0.569 (0.251, 1.293) Not retained Liver disease 2.306 (0.995, 5.344) Not retained Diabetes 0.255 (0.058, 1.111) Not retained Cholangiocarcinoma 2.625 (1.065, 6.469) Not retained Solid organ tumor 2.905 (1.297, 6.508) Not retained qSOFA 3.761 (1.974, 7.164) Not retained DBP 0.974 (0.947, 1.001) Not retained GCS 0.770 (0.630, 0.941) Not retained Hemoglobin 0.732 (0.605, 0.885) Not retained WBC 1.014 (0.995, 1.032) Not retained Serum bicarbonate 0.836 (0.762, 0.918) Not retained Factors in the model included age, sex, liver disease, diabetes, cholangiocar- cinoma, solid organ tumor, qSOFA, DBP, GCS, WBC, and serum bicarbonate. DBP = diastolic blood pressure; ESBL = extended-spectrum beta-lactamase; GCS = Glasgow coma scale; qSOFA = quick Sepsis Related Organ Failure Assessment; WBC = white blood cell. Phungoen et al Drug Target Insights 2022; 16: 15 © 2022 The Authors. Published by AboutScience - www.aboutscience.eu increased inflammatory response, and cellular damage from glucose administration (19). Previous studies have also found low plasma glucose to be associated with mortality in patients with sepsis (11,20). This study found that glucose of 113 mg/dL or lower yielded a sensitivity of 80.95% compared to a previous study, in which plasma glucose of 40-69 mg/dL resulted in an adjusted odds ratio of 3.43 (95% confidence interval of 1.51, 7.82) for mortality (19,20). These results may imply that patients with E. coli bacteremia and hypoglycemia may have as high of a risk of mortality as other patients with sepsis. The different plasma glucose cutoff points in the two studies may be due to differences in study population. This study enrolled only patients with E. coli bacteremia at the ED, while the previous study included patients with sepsis, which may have been caused by various pathogens. The plasma glucose cutoff point in this study may be more specific to patients with E. coli bacteremia at the ED. As previously reported, serum lactate is an indicator for mortality in patients with infection at the ED (21-23). A previ- ous study found that serum lactate greater than 4 mmol/L was associated with higher mortality than at 2 mmol/L (40.7% vs 2.7%) (24). In this study, we found that serum lactate over 2.4 mmol/L yielded sufficient sensitivity to predict fatality in patients with E. coli bacteremia at the ED. Another study found a serum lactate cutoff point of 5.80 mmol/L in patients with necrotizing fasciitis (25). This indicates that E. coli bacte- remia may be severe and that the serum lactate cutoff point may vary depending on the causative agents. Although oxygen saturation and hemoglobin were sig- nificantly associated with mortality by univariate logistic regression analysis (Tab. III), they were no longer significant in the final model. These results may indicate that neither factor was a strong predictor compared with the other three. Additionally, there might have been some related confounding factors. Other factors included in the model that had p values of less than 0.05 by univariate analysis were not retained in the final model for the same reasons. Some comorbid diseases, such as diabetes, were found to be significant predictors for mortality in a previous observa- tional study (26). However, comorbid diseases were not sig- nificant in this study, as previously mentioned. Additionally, the model used in this study differed from that in the pre- vious study. In this study, we included clinical factors such as ESBL E. coli in the model, while the previous study did not include ESBL E. coli and included treatment-related factors such as peak inspiratory pressure and positive end- expiratory pressure. There were some limitations to this study. First, the ED at which it was conducted was a single site at a university hospital. Further prospective studies in other settings may be required to confirm the results. In addition, this was an exploratory study without validation. The final predictive model included more factors than event outcomes. There were five factors in the model with only 27 nonsurvivors, resulting in a ratio of more than 1:10. Moreover, the total number of patients in the final model was 130. These limi- tations could have caused the model to be unbalanced or biased. However, the final model had a high goodness of fit. Another limitation was that some factors were not studied such as previous antibiotic use, previous history of resistant pathogens, or special conditions (27-35). Finally, mortality was defined as 28-day mortality. ESBL E. coli, plasma glucose, and serum lactate levels were associated with 28-day mortality in patients with E. coli bacteremia presenting at the ED. Acknowledgments The authors would like to acknowledge Dr. Dylan Southard for editing the manuscript via the Khon Kaen University Publication Clinic (Thailand). Disclosures Conflict of interest: The authors declare that they have no conflicts of interest. Financial support: This study was funded by a grant from the Khon Kaen University, Faculty of Medicine, Thailand (Grant Number: MN63307). References 1. Chase M, Klasco RS, Joyce NR, Donnino MW, Wolfe RE, Shapiro NI. Predictors of bacteremia in emergency department patients with suspected infection. Am J Emerg Med. 2012;30(9): 1691-1697. CrossRef PubMed 2. Zhu Q, Zhu M, Li C, et al. Epidemiology and microbiology of Gram-negative bloodstream infections in a tertiary-care hospi- tal in Beijing, China: a 9-year retrospective study. Expert Rev Anti Infect Ther. 2021;19(6):769-776. CrossRef PubMed 3. MacKinnon MC, McEwen SA, Pearl DL, et al. Mortality in Escherichia coli bloodstream infections: a multinational pop- ulation-based cohort study. BMC Infect Dis. 2021;21(1):606. CrossRef PubMed 4. Sianipar O, Asmara W, Dwiprahasto I, Mulyono B. Mortality risk of bloodstream infection caused by either Escherichia coli or Klebsiella pneumoniae producing extended-spectrum β-lactamase: a prospective cohort study. BMC Res Notes. 2019; 12(1):719. CrossRef PubMed 5. Hawkey PM. Multidrug-resistant Gram-negative bacteria: a product of globalization. J Hosp Infect. 2015;89(4):241-247. CrossRef PubMed 6. Nagvekar V, Sawant S, Amey S. Prevalence of multidrug- resistant Gram-negative bacteria cases at admission in a multi- speciality hospital. J Glob Antimicrob Resist. 2020;22:457-461. CrossRef PubMed 7. Morris S, Cerceo E. Trends, epidemiology, and management of multi-drug resistant Gram-negative bacterial infections in the hospitalized setting. Antibiotics (Basel). 2020;9(4):E196. CrossRef PubMed 8. Siwakoti S, Subedi A, Sharma A, Baral R, Bhattarai NR, Khanal B. Incidence and outcomes of multidrug-resistant gram-negative bacteria infections in intensive care unit from Nepal – a pro- spective cohort study. Antimicrob Resist Infect Control. 2018; 7(1):114. CrossRef PubMed 9. Lee YC, Hsiao CY, Hung MC, et al. Bacteremic urinary tract infection caused by multidrug-resistant Enterobacteriaceae are associated with severe sepsis at admission: implication for empirical therapy. Medicine (Baltimore). 2016;95(20):e3694. CrossRef PubMed 10. Kang CI, Song JH, Chung DR, et al; Korean Network for Study of Infectious Diseases (KONSID). Risk factors and treatment out- comes of community-onset bacteraemia caused by extended- spectrum beta-lactamase-producing Escherichia coli. Int J Antimicrob Agents. 2010;36(3):284-287. CrossRef PubMed https://doi.org/10.1016/j.ajem.2012.01.018 https://www.ncbi.nlm.nih.gov/pubmed/22626814 https://doi.org/10.1080/14787210.2021.1848544 https://www.ncbi.nlm.nih.gov/pubmed/33187451 https://doi.org/10.1186/s12879-021-06326-x https://www.ncbi.nlm.nih.gov/pubmed/34172003 https://doi.org/10.1186/s13104-019-4751-9 https://www.ncbi.nlm.nih.gov/pubmed/31675991 https://doi.org/10.1016/j.jhin.2015.01.008 https://www.ncbi.nlm.nih.gov/pubmed/25737092 https://doi.org/10.1016/j.jgar.2020.02.030 https://www.ncbi.nlm.nih.gov/pubmed/32165287 https://doi.org/10.3390/antibiotics9040196 https://www.ncbi.nlm.nih.gov/pubmed/32326058 https://doi.org/10.1186/s13756-018-0404-3 https://www.ncbi.nlm.nih.gov/pubmed/30275945 https://doi.org/10.1097/MD.0000000000003694 https://www.ncbi.nlm.nih.gov/pubmed/27196480 https://doi.org/10.1016/j.ijantimicag.2010.05.009 https://www.ncbi.nlm.nih.gov/pubmed/20580534 ESBL E. coli ED16 © 2022 The Authors. Drug Target Insights - ISSN 1177-3928 - www.aboutscience.eu/dti 11. Park SH, Choi SM, Lee DG, et al. Emergence of extended- spectrum β-lactamase-producing Escherichia coli as a cause of community-onset bacteremia in South Korea: risk factors and clinical outcomes. Microb Drug Resist. 2011;17(4):537-544. CrossRef 12. Namikawa H, Yamada K, Fujimoto H, et al. Clinical characteristics of bacteremia caused by extended-spectrum beta-lactamase- producing Escherichia coli at a tertiary hospital. Intern Med. 2017;56(14):1807-1815. CrossRef PubMed 13. Xiao T, Wu Z, Shi Q, et al. A retrospective analysis of risk fac- tors and outcomes in patients with extended-spectrum beta- lactamase-producing Escherichia coli bloodstream infections. J Glob Antimicrob Resist. 2019;17:147-156. CrossRef PubMed 14. Hattori H, Maeda M, Nagatomo Y, et al. Epidemiology and risk factors for mortality in bloodstream infections: a single-center retrospective study in Japan. Am J Infect Control. 2018;46(12): e75-e79. CrossRef PubMed 15. Haruki Y, Hagiya H, Haruki M, Sugiyama T. Clinical characteristics and outcome of critically ill patients with bacteremia caused by extended-spectrum β-lactamase-producing and non-producing Escherichia coli. J Infect Chemother. 2018;24(11):944-947. CrossRef PubMed 16. Rodríguez-Baño J, Picón E, Gijón P, et al; Spanish Network for Research in Infectious Diseases (REIPI). Community-onset bac- teremia due to extended-spectrum beta-lactamase-producing Escherichia coli: risk factors and prognosis. Clin Infect Dis. 2010;50(1):40-48. CrossRef PubMed 17. Yao RQ, Ren C, Wu GS, Zhu YB, Xia ZF, Yao YM. Is intensive glucose control bad for critically ill patients? A systematic review and meta-analysis. Int J Biol Sci. 2020;16(9):1658-1675. CrossRef PubMed 18. Wang J, Zhu CK, Yu JQ, Tan R, Yang PL. Hypoglycemia and mor- tality in sepsis patients: a systematic review and meta-analysis. Heart Lung. 2021;50(6):933-940. CrossRef PubMed 19. Park S, Kim DG, Suh GY, et al. Mild hypoglycemia is indepen- dently associated with increased risk of mortality in patients with sepsis: a 3-year retrospective observational study. Crit Care. 2012;16(5):R189. CrossRef PubMed 20. Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in the medical ICU. N Engl J Med. 2006;354(5):449-461. CrossRef PubMed 21. Shapiro NI, Howell MD, Talmor D, et al. Serum lactate as a predic- tor of mortality in emergency department patients with infec- tion. Ann Emerg Med. 2005;45(5):524-528. CrossRef PubMed 22. Vorwerk C, Loryman B, Coats TJ, et al. Prediction of mortality in adult emergency department patients with sepsis. Emerg Med J. 2009;26(4):254-258. CrossRef PubMed 23. Puskarich MA, Illich BM, Jones AE. Prognosis of emergency department patients with suspected infection and intermedi- ate lactate levels: a systematic review. J Crit Care. 2014;29(3): 334-339. CrossRef PubMed 24. Bou Chebl R, El Khuri C, Shami A, et al. Serum lactate is an inde- pendent predictor of hospital mortality in critically ill patients in the emergency department: a retrospective study. Scand J Trauma Resusc Emerg Med. 2017;25(1):69. CrossRef PubMed 25. Chang CP, Fann WC, Wu SR, Lin CN, Hsiao CT. Lactate on emer- gency department arrival as a predictor of in-hospital mortality in necrotizing fasciitis: a retrospective study. J Orthop Surg Res. 2019;14(1):73. CrossRef PubMed 26. Rezoagli E, McNicholas BA, Madotto F, Pham T, Bellani G, Laffey JG; LUNG SAFE Investigators, the ESICM Trials Group. Presence of comorbidities alters management and worsens outcome of patients with acute respiratory distress syndrome: insights from the LUNG SAFE study. Ann Intensive Care. 2022;12(1):42. CrossRef PubMed 27. Charoentanyarak S, Sawunyavisuth B, Deepai S, Sawanyawisuth K. A point-of-care serum lactate level and mortality in adult sepsis patients: a community hospital setting. J Prim Care Community Health. 2021;12:21501327211000233. CrossRef PubMed 28. Jeerasuwannakul B, Sawunyavisuth B, Khamsai S, Sawanyawisuth K. Prevalence and risk factors of proteinuria in patients with type 2 diabetes mellitus. Asia-Pac J Sci Technol. 2021 Jul 1; 26(04):APST-26-04-02. Online (Accessed May 2022). 29. Soontornrungsun B, Khamsai S, Sawunyavisuth B, et al. Obstructive sleep apnea in patients with diabetes less than 40 years of age. Diabetes Metab Syndr. 2020;14(6):1859-1863. CrossRef PubMed 30. Tongdee S, Sawunyavisuth B, Sukeepaisarnjaroen W, Boonsawat W, Khamsai S, Sawanyawisuth K. Clinical factors predictive of appropriate treatment in COPD: a community hospital setting. Drug Target Insights. 2021;15:21-25. CrossRef PubMed 31. Namwaing P, Chaisuksant S, Sawadpanich R, et al. Oxygen satu- ration is associated with recurrent primary spontaneous pneu- mothorax treated by intercostal chest drainage. Asian J Surg. 2022;45(1):431-434. CrossRef PubMed 32. Namwaing P, Chaisuksant S, Sawadpanich R, et al. Factors asso- ciated with duration of intercostal chest drainage in patients with primary spontaneous pneumothorax and the role of pulmonary rehabilitation. Open Access Emerg Med. 2021;13: 569-573. CrossRef PubMed 33. Boonwang T, Namwaing P, Srisaphonphusitti L, et al. Esports may improve cognitive skills in soccer players: a systematic review. Asia-Pac J Sci Technol. 2022;27(3). Online (Accessed May 2022). 34. Ruangritkul P, Tiamkao S, Chainirun N, et al. The efficacy and safety profile of generic intravenous levetiracetam in a real- world setting. Curr Ther Res Clin Exp. 2021;95:100648. CrossRef PubMed 35. Sanlung T, Sawanyawisuth K, Silaruks S, et al. Clinical char- acteristics and complications of obstructive sleep apnea in Srinagarind Hospital. J Med Assoc Thai. 2020;103(1):36-39. Online (Accessed May 2022). https://doi.org/10.1089/mdr.2011.0072 https://doi.org/10.2169/internalmedicine.56.7702 https://www.ncbi.nlm.nih.gov/pubmed/28717075 https://doi.org/10.1016/j.jgar.2018.12.014 https://www.ncbi.nlm.nih.gov/pubmed/30634054 https://doi.org/10.1016/j.ajic.2018.06.019 https://www.ncbi.nlm.nih.gov/pubmed/30172607 https://doi.org/10.1016/j.jiac.2018.04.016 https://www.ncbi.nlm.nih.gov/pubmed/29803763 https://doi.org/10.1086/649537 https://www.ncbi.nlm.nih.gov/pubmed/19995215 https://doi.org/10.7150/ijbs.43447 https://www.ncbi.nlm.nih.gov/pubmed/32226310 https://doi.org/10.1016/j.hrtlng.2021.07.017 https://www.ncbi.nlm.nih.gov/pubmed/34433111 https://doi.org/10.1186/cc11674 https://www.ncbi.nlm.nih.gov/pubmed/23062226 https://doi.org/10.1056/NEJMoa052521 https://www.ncbi.nlm.nih.gov/pubmed/16452557 https://doi.org/10.1016/j.annemergmed.2004.12.006 https://www.ncbi.nlm.nih.gov/pubmed/15855951 https://doi.org/10.1136/emj.2007.053298 https://www.ncbi.nlm.nih.gov/pubmed/19307384 https://doi.org/10.1016/j.jcrc.2013.12.017 https://www.ncbi.nlm.nih.gov/pubmed/24559577 https://doi.org/10.1186/s13049-017-0415-8 https://www.ncbi.nlm.nih.gov/pubmed/28705203 https://doi.org/10.1186/s13018-019-1108-y https://www.ncbi.nlm.nih.gov/pubmed/30841912 https://doi.org/10.1186/s13613-022-01015-7 https://www.ncbi.nlm.nih.gov/pubmed/35596885 https://doi.org/10.1177/21501327211000233 https://www.ncbi.nlm.nih.gov/pubmed/33733925 https://so01.tci-thaijo.org/index.php/APST/article/view/248718 https://doi.org/10.1016/j.dsx.2020.09.008 https://www.ncbi.nlm.nih.gov/pubmed/32992217 https://doi.org/10.33393/dti.2021.2291 https://www.ncbi.nlm.nih.gov/pubmed/34803374 https://doi.org/10.1016/j.asjsur.2021.07.021 https://www.ncbi.nlm.nih.gov/pubmed/34312054 https://doi.org/10.2147/OAEM.S336637 https://www.ncbi.nlm.nih.gov/pubmed/34938130 https://so01.tci-thaijo.org/index.php/APST/article/view/257467 https://doi.org/10.1016/j.curtheres.2021.100648 https://www.ncbi.nlm.nih.gov/pubmed/34840633 http://www.jmatonline.com/index.php/jmat/article/view/10743