Association between glycemic variability and 28-day all-cause mortality in patients with ARDS: A retrospective study based on the MIMIC-IV database

doi:10.21203/rs.3.rs-5032021/v1

Download PDF

Research Article

Association between glycemic variability and 28-day all-cause mortality in patients with ARDS: A retrospective study based on the MIMIC-IV database

https://doi.org/10.21203/rs.3.rs-5032021/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Introduction:

Despite glycemic variability (GV) being a recognized strong independent predictor of mortality in critically ill patients, its association with the prognosis of patients with acute respiratory distress syndrome (ARDS), a common critical condition in the intensive care unit (ICU), remains poorly understood. The objective of the present study was to evaluate the relationship between GV and all-cause mortality in patients with ARDS.

Material and methods

In this retrospective study using the Medical Information Mart for Intensive Care IV database version 2.2 (MIMIC IV v2.2), we evaluated the impact of GV on 28-day all-cause mortality among patients with ARDS admitted to ICU. Patients were included based on the new global definition of ARDS within two days of ICU admission. GV was quantified using the Glycemic Coefficient of Variation (Glu_CV) calculated within the first three days post-admission. Participants were categorized into three tertiles based on their Glu_CV (Low: ≤13.88%, N = 1288; Moderate: 13.88%-22.80%, N = 1291; High: >22.80%, N = 1289). Multivariable Cox proportional hazards models were constructed to analyze the association between Glu_CV and 28-day all-cause mortality in patients with ARDS. Non-linear correlations were explored using restricted cubic splines (RCS). Subgroup analyses were conducted o further investigate the relationship in patient groups defined by age, sex, the severity of ARDS, mean glucose levels with 3 days admission, and history of diabetes.

Results

A total of 3,868 adult ARDS patients were included in this study. Compared to the Low and Moderate groups, the patients in High Glu_CV group exhibited a higher prevalence of diabetes, more use of insulin and glucocorticoids treatment, and higher levels of mean glucose, CCI, SAPS II and SOFA scores. Multivariable Cox proportional hazards analyses revealed a significant association between Glu_CV and 28-day all-cause mortality in ARDS patients [continuous Glu_CV, HR:1.006, 95% CI: 1.002–1.010; tertiles of Glu_CV, High vs Low Glu_CV, HR:1.21, 95%CI: 1.03–1.43]. Three-knots RCS curve showed that the risk of mortality increased linearly with higher Glu_CV. Subgroup analysis revealed that compared with Low Glu_CV, High Glu_CV was associated with a higher risk of 28-day all-cause mortality in subgroups defined by male, mild severity of ARDS, mean glucose < 140 mg/dl within the first 3 days admission and without a diabetes history.

Conclusions

Glu_CV is independently associated with increased 28-day all-cause mortality in ARDS patients and could be used as a valuable tool for mortality risk stratification and guiding the treatment strategies of ARDS in the ICU. Further large prospective clinical studies are needed to evaluate the association between Glu_CV and mortality in ARDS patients.

Acute respiratory distress syndrome

glycemic variability

28-day all-cause mortality

Acute respiratory distress syndrome (ARDS) is defined as acute inflammatory lung injury caused by various diseases, resulting in refractory hypoxemia and ultimately leading to pulmonary dysfunction and high mortality[1]. ARDS accounts for approximately 10% of the intensive care unit (ICU) admissions, and the mortality rate varies depending on the severity of the disease. The mortality of severe ARDS patients up to 40%[2].

In clinical practice, critically ill patients often develop hyperglycemia regardless of whether they have a history of diabetes. A review mentioned that 40% of critically ill patients have hyperglycemia (random blood glucose > 140mg/dl), with 80% of hyperglycemic patients having no history of diabetes before admission[3]. This may be due to insulin resistance or impaired pancreatic β-cell function (stress hyperglycemia) caused by critical illness[4–6], and stress hyperglycemia is strongly associated with adverse clinical outcomes, including mortality, length of hospital stay, infections, and overall complications[7–9]. Stress activation not only leads to the occurrence of hyperglycemia but also increases glycemic variability (GV). In addition to the impact of critical illness, the use of insulin[10], the supplementation of enteral and parenteral nutrition[11], inevitably lead to fluctuations in blood glucose levels in critically ill patients. The fluctuations in blood glucose promote activation of oxidative stress[12], endothelial cell apoptosis associated with oxidative stress[13], monocyte adhesion to the endothelium[14] and endothelial dysfunction [15]. Previous studies have indicated that GV is considered as an independent risk factor for mortality in critically ill patients [16–18]. The pathogenesis of ARDS involved various pathogenic factors triggering innate immune cell–mediated damage of the alveolar endothelial and epithelial barrier, followed by infiltration of immune cells and excessive release of inflammatory mediators exacerbating tissue damage[19]. Association between GV and damage of endothelial cell[20] may elucidate the possible effects of GV on pathophysiological processes of ARDS. However, there were a few studies about the association between GV and prognosis in patients with ARDS.

Therefore, we conducted a retrospective study using the Medical Information Mart for Intensive Care IV database version 2.2 (MIMIC IV v2.2) to evaluate the relationship between GV with all-cause mortality in critically ill patients with ARDS. This may help identify high-risk patients for closer monitoring or timely intervention.

Database

MIMIC-IV database was used for this study. The database was approved by the Institutional Review Board (IRB) of the Massachusetts Institute of Technology (MIT), and it contains comprehensive and high-quality data of well-defined and characterized ICU patients admitted to ICUs at the Beth Israel Deaconess Medical Center between 2008 and 2019. The data included patient demographics and in-hospital mortality, laboratory test results (for example, hematology, chemistry, and microbiology results), discharge summaries and reports of electrocardiogram and imaging studies and billing-related information such as International Classification of Disease, 9th Edition (ICD-9) codes, Diagnosis Related Group (DRG) codes, and Current Procedural Terminology (CPT) codes. Out-of-hospital mortality dates were obtained using the Social Security Administration Death Master File[21]. One author (Yubiao Chen) completed the CITI Data or Specimens Only Research course and obtained permission to the database and performed data extraction and analysis (NO. 40416369). Since this study was conducted using an anonymous public database that satisfies the protocol of the review board, the requirement of ethical consent is not necessary.

Study cohort

ARDS patients was identified by the 2012 Berlin definition[22] and the new global definition[23] respectively. Our investigation focused on the association between Glu_CV and 28-day all-cause mortality in patients with ARDS identified by the new global definition. The method used for data extraction for ARDS patients was established by Fang Qian et al[24]. In brief: (1) ARDS evidence based on ICD-9 code (518.82) and ICD-10 codes (J80)[25]; (2) evidence of bilateral infiltrates suggested by chest imaging; (3) oxygen saturation (Spo₂) to fraction of inspired oxygen (FiO₂) ratio ⩽ 315 (if SpO₂ ⩽ 97%) on supplemental oxygen; or partial pressure of arterial oxygen (PaO₂) to FiO₂ ratio ⩽ 300 mm Hg or SpO₂ to FiO2 ratio ⩽ 315 (if SpO₂ ⩽ 97%) on high-flow nasal oxygen (HFNO) with flow of ⩾30 L/min or noninvasive ventilation (NIV) with at least 5 cm H₂O end-expiratory pressure (PEEP); or PaO₂ to FiO₂ ratio < 300 mmHg with a minimum PEEP of ≥ 5 cmH₂O; (4) absence of heart failure. 4333 patients were identified as ARDS within 2 days after the first admission to ICU identified from MIMIC-IV database. 331 patients were excluded for age less than 18 and staying in ICU less than 24 hours, 134 admissions were excluded for less than 3 times of repeat blood glucose measurements within 3 days after ICU admissions. Finally, we included 3868 ARDS patients in our cohort. In this analysis, the overall GV was evaluated by calculating the coefficient of variation (Glu_CV), which is the ratio of the standard deviation (Glu_CD) to the glycemic average. We grouped the data into 3 group based on Glu_CV: Low group (≤ 13.88%, N = 1288), Moderate group (13.88% – 22.80%, N = 1291) and High group (> 22.80%, N = 1289) (Fig. 1). Details of inclusion and exclusion criteria of ARDS patients identified by the 2012 Berlin definition[22] can be found in supplementary material (Supplementary material. Figure 1).

Data extraction

Structured Query Language (SQL), PostgreSQL tools (version 9.6) was applied for data extraction and management. The following data were collected on the first day of ICU admission: age, sex, race, body mass index (BMI), comorbidities, mean blood glucose, Charlson Comorbidity Index (CCI), Sequential Organ Failure Assessment (SOFA), Simplified Acute Physiology Score II (SAPSII), Glasgow Coma Scale (GCS) and treatment of insulin and glucocorticoids (prednisone, dexamethasone, hydrocortisone and methylprednisolone). All related comorbidities including cardiovascular disease, coronary artery disease (CAD), chronic pulmonary disease, diabetes, liver disease, malignancy, renal disease, acquired immune deficiency syndrome (AIDS) were identified by ICD-9. Hyperglycemia was defined as mean blood glucose > 140 mg/dl within first day after ICU admission. The following data were collected within 3 days after ICU admission: the use of insulin, and glucocorticoids, mean blood glucose, mean white blood cell counts (WBC), and mean neutrophil-to-lymphocyte ratio (NLR), Endpoints in our study were 7-, 14-, 21- and 28-day all-cause mortality, mortality in the ICU, in-hospital mortality, and length of ICU and hospital stay.

Statistical Analysis

We tested normality distribution by Kolmogorov–Smirnov test. According to the result of the test, continuous variables were summarized as median and interquartile range, and categorical variables are presented as numbers and percentages. Wilcoxon rank-sum test or Chi-square test was performed to compare the differences between groups. Kaplan-Meier curve was used to present the association between Glu_CV and 28-day all-cause mortality, ICU mortality and hospital mortality. Cox proportional-hazards model was used to analyze the independent effects of Glu_CV on 28-day all-cause mortality, ICU mortality and hospital mortality. Additionally, a three-knots restricted cubic spline (RCS) was used to present potential non-linear relationship between Glu_CV and outcomes. The Wald test was used to assess the presence of non-linearity. Subgroup analyses were applied to investigate the association between Glu_CV and outcome, including age (< 60 or ≥ 60 years), sex, the severity of ARDS, mean glucose within 3 days (< 140 or ≥ 140 mg/dL) and with or without history of diabetes. The data were analyzed by R software (version 4.1.0). A two-tailed P < 0.05 was considered as statistically significant. To assess the universality of Glu_CV across different ARDS classifications, we performed same data analysis on both two cohorts. Data analysis of ARDS patients identified by the 2012 Berlin definition were presented in supplementary material.

Population characteristics

As shown in the Fig. 1, 3868 ARDS patients were enrolled in this study. Based on the tertiles of Glu_CV within 3 days after first ICU admission (Low: ≤13.88%, N = 1288; Moderate: 13.88%-22.80%, N = 1291; High: >22.80%, N = 1289), the study participants’ baseline characteristics were analyzed and shown in Table 1. Compared to other two group, High Glu_CV group (> 22.80%) exhibited lower level of BMI (30.00 [25.36, 35.18] vs 29.07 [25.39, 34.30] vs 28.48 [24.77, 33.49], P = 0.003), a higher prevalence of diabetes (21.8% vs 29.3% vs 42.0%, P < 0.001), and more frequent insulin (8.6% vs 21.3% vs 28.6%, P < 0.001) and glucocorticoids treatment (17.1% vs 21.1% vs 31.0%, P < 0.001). An upward trend was observed at higher Glu_CV levels for initial NLR value, mean-glucose (mg/dl) and SAPS II and SOFA scores.

Outcome

The 7-day mortality (10.6% vs 9.7% vs 20.9%, P = 0.007), 14-day mortality (15.8% vs 15.6% vs 29.1%, P < 0.001), 21-day mortality (19.6% vs 20.4% vs 33.0%, P < 0.001), 28-day mortality (21.0% vs 22.9% vs 35.6%, P < 0.001), mortality in ICU (12.7% vs 13.4% vs 28.8%, P < 0.001) and hospital mortality (16.7% vs 17.7% vs 29.6%, P < 0.001) in High Glu_CV group (> 22.80%) were all higher than other two group. Besides, the length of stay in hospital (8.39 [5.21, 14.18] vs 8.93 [5.45, 15.70] vs 8.67 [4.88, 15.26], P = 0.012) and ICU (3.91 [2.14, 7.72] vs 4.07 [2.12, 7.88] vs 4.68 [2.37, 9.24], P < 0.001) (days) showed a tendency of increasing with the Glu_CV tertiles (Table 2).

Analysis of association between Glu_CV with 28-day all-cause mortality

Kaplan-Meier curves in Fig. 2 showed that the highest Glu_CV (> 22.80%) group had worse short-term survival (28-day all-cause mortality, ICU mortality and hospital mortality) compared to those with lower levels of Glu_CV (log-rank P < 0.001). Furthermore, we characterized Glu_CV as continuous variables or categorical variables, and Cox proportional hazards analysis showed a statistically significant association between Glu_CV and survival (28-day all-cause mortality [continuous Glu_CV, HR 1.006, 95%CI: 1.002–1.010; tertiles of Glu_CV, High vs Low, HR 1.21, 95%CI: 1.03–1.43], ICU mortality [continuous Glu_CV, HR 1.009, 95%CI: 1.004–1.014; tertiles of Glu_CV, High vs Low, HR, 1.34, 95%CI: 1.10–1.64] and hospital mortality [continuous Glu_CV, HR 1.006, 95%CI: 1.001–1.011; tertiles of Glu_CV, High vs Low, HR, 1.21, 95%CI: 1.01–1.45]) in model adjusting confounding factors (age, sex, race, the severity of ARDS, CCI, SAPS II, GCS scores and use of insulin and glucocorticoids treatment) (Table 3). Among ARDS patients identified by the 2012 Berlin definition, the highest Glu_CV group also exhibited higher risk of short-term mortality (Supplementary material.eTable 3).

Analysis of non-linear relationship between Glu_CV and 28-day all-cause mortality

RCS curves were performed to detect the nonlinear relationship between the Glu_CV and survival (28-day all-cause mortality, ICU mortality and hospital mortality), we found that the risk of mortality increased linearly with higher Glu_CV in ARDS patients (All of P value for non-linearity > 0.05) (Fig. 3).

Subgroup analyses

We conducted a risk stratification analysis of Glu_CV for the primary endpoint in multiple subgroups, based on age (< 60 or ≥ 60 years), sex, the severity of ARDS, mean glucose within 3 days (< 140 or ≥ 140 mg/kg) and with or without history of diabetes. Compared with Low Glu_CV (> 13.88), High Glu_CV (> 22.80%) was associated with a higher risk of 28-day all-cause mortality, ICU mortality and hospital mortality in subgroups defined by male, mild ARDS, mean glucose within 3 days < 140 mg/dl, without history of diabetes (Table 4).

This study evaluated the association between Glu_CV within 3 days after ICU admission and 28-day all-cause mortality, ICU mortality and hospital mortality in patients with ARDS. Our findings revealed that ARDS patients in high Glu_CV (> 22.80%) group had a higher risk of 28-day mortality compared to other two groups after covariate adjustments. Besides, the short-term mortality (28-day all-cause mortality, ICU mortality and hospital mortality) increased as the Glu_CV levels increased. In addition, subgroup analysis showed that high Glu_CV (> 22.80%) was associated with 28-day mortality compared to low tertile of Glu_CV in ARDS patients without diabetes or with lower levels of mean glucose within 3 days (< 140mg/dl) after ICU admission. These results indicated that Glu_CV is an independent risk factor for 28-day mortality in ARDS patients, especially in those without diabetes or with lower levels of mean glucose within 3 days (< 140mg/dl) after ICU admission.

In this study, hyperglycemia (> 140 mg/dL) at ICU admission was presented in 54.0% of ARDS patients, and 56.0% of those did not have a history of diabetes. This phenomenon is similar to that observed in critically ill patients. Farnoosh Farrokhi et al.[3] summarized that 40% of critically ill patients have hyperglycemia, and 80% of these patients were without history of diabetes. It suggests that stress-induced hyperglycemia is also common among patients with ARDS.

This study found that higher Glu_CV was an independent risk factor for short-term mortality in ARDS patients. This is consistent with the findings of two previous study. Chiara Lazzeri et al. [26]analyzed clinical data from 83 patients with COVID-19-related ARDS, and the results indicated that non-survivors of COVID-19-related ARDS exhibited significantly higher GV within the first 48 hours of ICU admission. Bojan Hartmann et al.[27] collected all fasting blood glucose measurement data from COVID-19-related ARDS patients during their ICU stay to calculate Glu_CV. They found that high Glu_CV during ICU admission was significantly associated with increased 30-day all-cause mortality in Covid-19 ARDS patients. However, these two studies were conducted in COVID-19-related ARDS patients. This may be different from the mechanism in this study. For patients with COVID-19, acute elevation of blood glucose promoted the binding of SARS-COV2 to ACE2, which facilitated viral invasion of cells and may lead to a more severe disease[28]. This study is a retrospective study based on the MIMIC-IV database, which contains clinical data of critically ill patients admitted to the ICU between 2009 and 2018. Therefore, this study could have a more accurate assessment of the relationship between GV and ARDS mortality for avoiding COVID-19-related potential confounding factors. The fluctuation in blood glucose levels is associated with endothelial cell apoptosis[20], endothelial dysfunction[15], oxidative stress[13], and inflammation-related processes[29], which exacerbate damage to the alveolar-capillary barrier in ARDS patients. This may be the reason why higher Glu_CV increased the risk of 28-day mortality in ARDS patients in this study. Furthermore, Bojan Hartmann et al.[27] calculated Glu_CV using all measurements of blood glucose during ICU stay. However, over an extended period, levels of blood glucose may be influenced by more clinical interventions, such as the use of insulin and supplementation of enteral and parenteral nutrition, resulting in less accurate predictive outcomes. In this study, Glu_CV was calculated using blood glucose measurements within 3 days after ICU admission. Therefore, it could be obtained simply and be more reproducible and accurate to predict the prognosis of ARDS patients in clinical practice. Besides, in this study we identified patients with ARDS by the new global definition and the 2012 Berlin definition and performed same data analyzed in these two cohorts. And the both of two results suggested that higher level of Glu_CV associated with worse 28-day survival. It indicated that Glu_CV may have great universality and stability for prognostic assessment in patients with ARDS.

In subgroup analysis, we found that Glu_CV was not associated with 28-day mortality in ARDS patients with average blood glucose levels within 3 days after ICU admission more than 140mg/dl or those with history of diabetes. Conversely, compared to low Glu_CV, high Glu_CV (> 22.80%) was positively associated with 28-day mortality in ARDS patients with average blood glucose levels within 3 days after ICU admission less than 140mg/dl or those without history of diabetes. This is consistent with previous studies. In a retrospective study, James Stephen Krinsley[30] investigated the impact of diabetes or its absence on Glu_CV as a risk factor for mortality and found that in critically ill patients without history of diabetes, high Glu_CV (> 50%) was associated with an increased risk of mortality [OR (95% CI), 2.12 (1.23–3.65)], while in the population with history of diabetes, high Glu_CV (> 50%) was not independently associated with mortality. It suggested that patients with diabetes may develop relative tolerance to cellular and microvascular complications associated with moderate to high levels of blood glucose. In contrast, the same degree of hyperglycemia may cause greater stress response in critically ill patients without history of diabetes. Additionally, GV reflects fluctuations in blood glucose, and high glycemic variability may lead to severe hyperglycemia or hypoglycemia. It is believed that hypoglycemia is more harmful than hyperglycemia[31], which may explain the differential effect of Glu_CV on mortality between patients with average blood glucose levels within 3 days after ICU admission below 140mg/dl and those above 140mg/dl. In summary, Glu_CV may be a better predictor of prognosis in ARDS patients with blood glucose levels below 140mg/dl or without history of diabetes.

This study is the first to report the association between Glu_CV and 28-day mortality in non-COVID-related ARDS patients identified respectively by two different definitions. The calculation of Glu_CV requires at least three blood glucose measurements. Repeated blood glucose measurements are often performed in critically ill patients during ICU stay. Therefore, Glu_CV is an available and simple indicator. This study explored the association between Glu_CV within 3 days after ICU admission and 28-day mortality, which allowed clinician for obtaining Glu_CV in a short period of time. This study provides additional information for ARDS prognosis prediction. However, this study has several limitations. Firstly, due to database limitations, we were unable to attain the fasting time of the patients. Thus, each blood glucose record in this study should be considered as random blood glucose. Secondly, MIMIC-IV database lacks data on biomarkers of inflammation, endothelial damage, and oxidative stress. Therefore, we explained the reason for association between high Glu_CV and increased risk of 28-day all-cause mortality in ARDS patients based on previous literature, and it is still hypothetical and requires further investigation through appropriately designed studies. Finally, this study is a single-center, retrospective cohort study. Our findings need to be validated in external populations.

Our study provided evidence that Glu_CV is an independent risk factor for 28-day all-cause mortality in ARDS patients, especially in patients without history of diabetes or mean glucose within 3 days after ICU admission lower than 140 mg/dl. Therefore, Glu_CV could be used as a valid indicator for grading and treating patients with ARDS in the ICU. However, this study is a single-center retrospective study and has inherent limitations. Further large appropriately designed studies are needed to confirm the relationship of Glu_CV and clinical outcome of ARDS patients.

ARDS

Acute respiratory distress syndrome

Glycemic variability

Glu_CV

Glycemic coefficient of variation

ICU

Intensive care unit,ICU

AFIB

Atrial fibrillation

CAD

Coronary artery disease

CHF

Congestive heart failure

COPD

Chronic obstructive pulmonary disease

NLR

Neutrophil-to-Lymphocyte Ratio

WBC

White blood cell counts

SOFA

Sequential Organ Failure Assessment

SAPSII

Simplified Acute Physiology Score II

PaO2

Partial pressure of oxygen

FiO2

Fraction inspired oxygen concentration.

Data availability

Available upon request. Corresponding to Drs. Yongbo Huang ([email protected]) or Yimin Li ([email protected])

Ethics approval and consent to participate

The MIMIC-III databases have received ethical approval from the Institutional Review Boards at Beth Israel Deaconess Medical Center and Massachusetts Institute of Technology. As the databases do not contain protected health information, a waiver of informed consent was included in the approval from the Institutional Review Boards at Beth Israel Deaconess Medical Center and Massachusetts Institute of Technology. Therefore, this manuscript does not involve a research protocol requiring approval by the relevant institutional review board or ethics committee. All methods in this study were carried out in accordance with relevant guidelines and regulations (declarations of Helsinki).

Consent for publication

Not applicable.

Competing interests

There are no competing interests.

Author contributions

YBC, LFL, BYL and XMD conceived the study idea participated in the study conception and design, statistical analysis, data interpretation, drafting and revisions of the manuscript. YBC, YXZ, YZ, KW, HSJ, XYG, TZ and WYY participated in the data acquisition and revisions of data analysis and interpretation. YBH and YML participated in study conception and design, data interpretation and revisions of the manuscript, also supervised the whole study. All authors provided their final approval for manuscript submission.

Funding

This work was supported by the National Natural Science Foundation of China [grant numbers 82070084, 82270085]; National Key R&D Program of China [grant number 2023YFC3041700]; the Science and Technology Program of Guangzhou [grant number 202201020444]; the Emergency Key Program of Guangzhou Laboratory [grant number SRPG23-001]; Guangdong Marine Economy Development Special Project [grant number GDNRC[2022]35].

Acknowledgments

We appreciate the excellent work of MIMIC team (Massachusetts Institute of Technology Laboratory for Computational Physiology) to collect bedside data continuously and make the database available for every intensive care researcher.

Fan E, Brodie D, Slutsky AS: Acute Respiratory Distress Syndrome: Advances in Diagnosis and Treatment.JAMA 2018, 319:698–710.
Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, Gattinoni L, van Haren F, Larsson A, McAuley DF, et al: Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries.JAMA 2016, 315:788–800.
Farrokhi F, Smiley D, Umpierrez GE: Glycemic control in non-diabetic critically ill patients.Best Pract Res Clin Endocrinol Metab 2011, 25:813–824.
Smith FG, Sheehy AM, Vincent JL, Coursin DB: Critical illness-induced dysglycaemia: diabetes and beyond.Crit Care 2010, 14:327.
Dungan KM, Braithwaite SS, Preiser JC: Stress hyperglycaemia.Lancet 2009, 373:1798–1807.
Marik PE, Bellomo R: Stress hyperglycemia: an essential survival response!Crit Care 2013, 17:305.
Badawi O, Waite MD, Fuhrman SA, Zuckerman IH: Association between intensive care unit-acquired dysglycemia and in-hospital mortality.Crit Care Med 2012, 40:3180–3188.
Bruno A, Levine SR, Frankel MR, Brott TG, Lin Y, Tilley BC, Lyden PD, Broderick JP, Kwiatkowski TG, Fineberg SE, Group Nr-PSS: Admission glucose level and clinical outcomes in the NINDS rt-PA Stroke Trial.Neurology 2002, 59:669–674.
Capes SE, Hunt D, Malmberg K, Gerstein HC: Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview.Lancet 2000, 355:773–778.
Arabi YM, Dabbagh OC, Tamim HM, Al-Shimemeri AA, Memish ZA, Haddad SH, Syed SJ, Giridhar HR, Rishu AH, Al-Daker MO, et al: Intensive versus conventional insulin therapy: a randomized controlled trial in medical and surgical critically ill patients.Crit Care Med 2008, 36:3190–3197.
Doola R, Greer RM, Hurford R, Flatley C, Forbes JM, Todd AS, Joyce CJ, Sturgess DJ: Glycaemic variability and its association with enteral and parenteral nutrition in critically ill ventilated patients.Clin Nutr 2019, 38:1707–1712.
Monnier L, Mas E, Ginet C, Michel F, Villon L, Cristol JP, Colette C: Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes.JAMA 2006, 295:1681–1687.
Quagliaro L, Piconi L, Assaloni R, Martinelli L, Motz E, Ceriello A: Intermittent high glucose enhances apoptosis related to oxidative stress in human umbilical vein endothelial cells: the role of protein kinase C and NAD(P)H-oxidase activation.Diabetes 2003, 52:2795–2804.
Azuma K, Kawamori R, Toyofuku Y, Kitahara Y, Sato F, Shimizu T, Miura K, Mine T, Tanaka Y, Mitsumata M, Watada H: Repetitive fluctuations in blood glucose enhance monocyte adhesion to the endothelium of rat thoracic aorta.Arterioscler Thromb Vasc Biol 2006, 26:2275–2280.
Ceriello A, Esposito K, Piconi L, Ihnat MA, Thorpe JE, Testa R, Boemi M, Giugliano D: Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients.Diabetes 2008, 57:1349–1354.
Al-Dorzi HM, Tamim HM, Arabi YM: Glycaemic fluctuation predicts mortality in critically ill patients.Anaesth Intensive Care 2010, 38:695–702.
Eslami S, Taherzadeh Z, Schultz MJ, Abu-Hanna A: Glucose variability measures and their effect on mortality: a systematic review.Intensive Care Med 2011, 37:583–593.
Waeschle RM, Moerer O, Hilgers R, Herrmann P, Neumann P, Quintel M: The impact of the severity of sepsis on the risk of hypoglycaemia and glycaemic variability.Crit Care 2008, 12:R129.
Thompson BT, Chambers RC, Liu KD: Acute Respiratory Distress Syndrome.N Engl J Med 2017, 377:562–572.
Risso A, Mercuri F, Quagliaro L, Damante G, Ceriello A: Intermittent high glucose enhances apoptosis in human umbilical vein endothelial cells in culture.Am J Physiol Endocrinol Metab 2001, 281:E924-930.
Johnson AEW, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, Pollard TJ, Hao S, Moody B, Gow B, et al: MIMIC-IV, a freely accessible electronic health record dataset.Sci Data 2023, 10:1.
Force ADT, Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, Fan E, Camporota L, Slutsky AS: Acute respiratory distress syndrome: the Berlin Definition.JAMA 2012, 307:2526–2533.
Matthay MA, Arabi Y, Arroliga AC, Bernard G, Bersten AD, Brochard LJ, Calfee CS, Combes A, Daniel BM, Ferguson ND, et al: A New Global Definition of Acute Respiratory Distress Syndrome.Am J Respir Crit Care Med 2024, 209:37–47.
Qian F, van den Boom W, See KC: The new global definition of acute respiratory distress syndrome: insights from the MIMIC-IV database.Intensive Care Med 2024, 50:608–609.
Wang Z, Zhang L, Huang T, Yang R, Cheng H, Wang H, Yin H, Lyu J: Developing an explainable machine learning model to predict the mechanical ventilation duration of patients with ARDS in intensive care units.Heart Lung 2023, 58:74–81.
Lazzeri C, Bonizzoli M, Batacchi S, Di Valvasone S, Chiostri M, Peris A: The prognostic role of hyperglycemia and glucose variability in covid-related acute respiratory distress Syndrome.Diabetes Res Clin Pract 2021, 175:108789.
Hartmann B, Verket M, Balfanz P, Hartmann NU, Jacobsen M, Brandts J, Dreher M, Kossack N, Hackl D, Marx N, Muller-Wieland D: Glycaemic variability is associated with all-cause mortality in COVID-19 patients with ARDS, a retrospective subcohort study.Sci Rep 2022, 12:9862.
Brufsky A: Hyperglycemia, hydroxychloroquine, and the COVID-19 pandemic.J Med Virol 2020, 92:770–775.
Skrha J, Soupal J, Skrha J, Jr., Prazny M: Glucose variability, HbA1c and microvascular complications.Rev Endocr Metab Disord 2016, 17:103–110.
Krinsley JS: Glycemic variability and mortality in critically ill patients: the impact of diabetes.J Diabetes Sci Technol 2009, 3:1292–1301.
Gunst J, De Bruyn A, Van den Berghe G: Glucose control in the ICU.Curr Opin Anaesthesiol 2019, 32:156–162.

Tables 1 to 4 are available in the Supplementary Files section.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
09 Sep, 2024
Editor assigned by journal
05 Sep, 2024
Submission checks completed at journal
05 Sep, 2024
First submitted to journal
04 Sep, 2024

You are reading this latest preprint version

Association between glycemic variability and 28-day all-cause mortality in patients with ARDS: A retrospective study based on the MIMIC-IV database

Status:

Version 1

Abstract

Introduction:

Material and methods

Results

Conclusions

Figures

Introduction

Material and methods

Database

Study cohort

Data extraction

Statistical Analysis

Results

Population characteristics

Outcome

Analysis of association between Glu_CV with 28-day all-cause mortality

Analysis of non-linear relationship between Glu_CV and 28-day all-cause mortality

Subgroup analyses

Discussion

Conclusions

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1

Association between glycemic variability and 28-day all-cause mortality in patients with ARDS: A retrospective study based on the MIMIC-IV database

Status:

Version 1

Abstract

Introduction:

Material and methods

Results

Conclusions

Figures

Introduction

Material and methods

Database

Study cohort

Data extraction

Statistical Analysis

Results

Population characteristics

Outcome

Analysis of association between GluCV with 28-day all-cause mortality

Analysis of non-linear relationship between GluCV and 28-day all-cause mortality

Subgroup analyses

Discussion

Conclusions

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1

Analysis of association between Glu_CV with 28-day all-cause mortality

Analysis of non-linear relationship between Glu_CV and 28-day all-cause mortality