https://orcid.org/0009-0002-5258-1313
Figures
Abstract
Purpose
Acute kidney injury (AKI), a common and severe complication of acute pancreatitis (AP), is significantly linked to patient prognosis. Albumin-corrected anion gap (ACAG) is a modified acid-base balance assessment metric with potential clinical significance in various critical illnesses. However, the role of ACAG in forecasting the risk of AKI in AP patients remains unclear. This study sheds light on the relationship between ACAG levels and AKI risk in the AP population.
Methods
This retrospective study utilized data from the MIMIC-IV database, including 1,552 adult patients diagnosed with AP during their stay in the intensive care unit (ICU). ACAG was calculated using a standard formula, and patients were grouped according to their ACAG levels. Cox proportional hazards and restricted cubic spline (RCS) models were employed to assess the correlation of ACAG levels with AKI risk in AP patients. The incidence of AKI was the primary outcome, and in-hospital mortality was the secondary outcome. Differences in primary and secondary outcomes between ACAG groups were evaluated through Kaplan-Meier (KM) survival analysis. Subgroup analyses were performed for examining the influence of confounding factors.
Results
Higher ACAG levels were significantly related to an elevated risk of AKI. The RCS model demonstrated a nonlinear correlation between higher ACAG levels and increased AKI risk in the AP cohort, and a linear association of ACAG with in-hospital death. KM survival analysis showed that patients exhibiting higher ACAG levels had poorer renal function outcomes and higher ICU mortality. Subgroup analyses further proved this correlation across varied patient characteristics.
Citation: Bian J, Wang X, Chen Y, Lu G, Zhang L, Tu X, et al. (2025) The association between Albumin-Corrected Anion Gap (ACAG) and the risk of acute kidney injury in patients with acute pancreatitis: A retrospective analysis based on the MIMIC-IV database. PLoS One 20(8): e0330458. https://doi.org/10.1371/journal.pone.0330458
Editor: Tanja Grubić Kezele, University of Rijeka Faculty of Medicine: Sveuciliste u Rijeci Medicinski fakultet, CROATIA
Received: April 21, 2025; Accepted: July 31, 2025; Published: August 22, 2025
Copyright: © 2025 Bian et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The raw data were sourced from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database (https://mimic.mit.edu).
Funding: This work was supported by the Fujian Provincial Science and Technology Plan Project (2021Y9083).
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Acute pancreatitis (AP) is a prevalent acute inflammatory abdominal illness with a wide spectrum of severity. It can manifest as mild pancreatitis or progress to severe conditions like systemic inflammatory response syndrome (SIRS) and multiple organ dysfunction syndrome (MODS) [1,2]. Among AP patients, approximately 15–20% have severe acute pancreatitis (SAP), which is a severe condition linked to single or multiple organ failures and local or systemic complications, with a death rate of 36% − 50%. [3–6] Acute kidney injury (AKI), a common and serious complication in AP patients, significantly increases both mortality and healthcare burdens [7,8]. Furthermore, AKI is independently related to higher mortality rates in AP sufferers [9]. Therefore, early identification and risk prediction of AKI in AP patients in the intensive care unit (ICU) hold crucial clinical significance.
Albumin-corrected anion gap (ACAG) is an adjusted anion gap calculated after correcting for serum albumin levels, offering a more accurate assessment method of acid-base balance [10] Traditional anion gap (AG) may lead to misjudgments in acid-base imbalance due to its failure to account for the effects of hypoalbuminemia [11]. In contrast, ACAG, by adjusting for albumin concentration, more accurately reflects the underlying metabolic acidosis status of the body. ACAG is closely correlated with the prognosis of various critical diseases [12–15]. Nevertheless, the link of ACAG to AKI in AP patients is not elucidated.
The MIMIC-IV database is an open-access resource containing extensive clinical data and laboratory indicators from ICU patients, providing a solid data foundation for analyzing the connection of ACAG with AKI in AP patients. This study aims to utilize the MIMIC-IV database to systematically investigate the association between ACAG and AKI in the AP population and further examine the potential clinical value of ACAG in forecasting the risk of AKI. The ultimate goal is to offer a new reference for the early management of critically ill AP sufferers.
2. Methods
2.1. Data source
The raw data were sourced from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database (https://mimic.mit.edu). This database contains detailed clinical data from ICU patients of an academic medical center in the United States between 2008 and 2019. It includes demographic information, laboratory test results, medication records, and prognostic outcomes [16]. To gain access to the database, the author, Jiaxiang Bian, completed the Collaborative Institutional Training Initiative (CITI) course (Record ID: 13501033) and obtained necessary permissions for using the database. The MIMIC-IV project was approved by the Institutional Review Boards of the Massachusetts Institute of Technology (Cambridge, MA, USA) and the Beth Israel Deaconess Medical Center. Patient information has been anonymized, and thus informed consent was not required.
2.2. Study population
This study focused on a cohort comprising 3,850 patients who were diagnosed with AP and had non-continuous ICU admissions (aged ≥18) from MIMIC-IV. The diagnosis of AP was confirmed in adherence to the International Classification of Diseases, 9th edition (ICD-9) and 10th edition (ICD-10) codes. Only the first hospital admission for each patient was considered if they had multiple admissions. To ensure data completeness, patients with missing albumin or AG values, as well as those who stayed fewer than 24 hours in hospital, were removed. At last, 1,552 patients were eligible for the analysis. The specific patient selection is presented in Fig 1.
2.3. Study variables
This study utilized Structured Query Language (SQL) and PostgreSQL (version 14.2) to extract baseline characteristics from MIMIC-IV, including demographic information, laboratory parameters, comorbidities, incidence of AKI, and in-hospital mortality. Demographic data encompassed gender, age, race, marital status, and body mass index (BMI). Vital signs encompassed respiratory rate (RR), heart rate (HR), systolic blood pressure (SBP), and diastolic blood pressure (DBP). Laboratory parameters included red blood cells (RBC), hemoglobin (Hb), white blood cells (WBC), platelets (PLT), pO₂, pCO₂, pH, HCO₃ ⁻ , sodium, potassium, calcium, magnesium, lactate, glucose, triglycerides (TG), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), serum creatinine (SCr), urinary creatinine (UCr), blood urea nitrogen (BUN), AG, and albumin (ALB). Comorbidities and personal medical history determined according to ICD-9 and ICD-10 codes, were hypertension, diabetes mellitus (DM), hyperlipidemia, coronary heart disease (CHD), heart failure (HF), atrial fibrillation (AF), chronic obstructive pulmonary disease (COPD), cirrhosis, chronic pancreatitis (CP), pancreatic cysts and pseudocysts (PC and PPC), chronic kidney disease (CKD), sepsis, and shock. Sequential Organ Failure Assessment (SOFA) score and SIRS score were employed for evaluating the severity of patients after admission. Surgical history included resection of the gallbladder and common bile duct dilation. In addition, to explore whether albumin infusion as an interventional measure might influence outcomes, this variable was incorporated into the analysis.
The exposure variable was ACAG calculated as follows: AG (mmol/l) = (sodium + potassium) − (chloride + bicarbonate) [17]; ACAG = AG + {4.4-[ALB(g/dl)]} × 2.5 [10]. The primary outcome was defined as the occurrence of AKI from ICU admission to hospital discharge. AKI was defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines as an increase in serum creatinine (SCr) to ≥1.5 times the baseline value within the prior 7 days, an absolute increase in SCr of ≥0.3 mg/dL within 48 hours, or a reduction in urine output to <0.5 mL/kg/h for at least 6 hours [18]. The secondary outcome was the in-hospital death rate of AP sufferers.
2.4. Statistical analysis
All statistical analyses were conducted via R 4.4.2. For repeatedly measured laboratory parameters, the first recorded value within 3 days of admission was used. During the data preprocessing stage, timestamps were handled using the as. POSIXctfunction in R. Data entries were subsequently categorized based on the time interval (i.e., AKI onset time minus the parameter measurement time), with the data classified as occurring either before or after the AKI event. To reduce reverse causality bias, data collected after the onset of AKI events were considered invalid. Given the common occurrence of missing laboratory values in MIMIC-IV, the proportions of missing data for each continuous variable were calculated with reference to previous studies [19,20] (S1 Table). For variables with a missing value proportion of less than 20%, missing values were predicted and imputed using a random forest-based multiple imputation method (missForest R). The tbl_summary function was used to compare data before and after multiple imputation (S2 Table). For variables with missing value proportions exceeding 20%, the missing part was treated as a separate category during the analysis.
For continuous variables in normal distribution, means and standard deviations were used, and differences between groups were compared via independent sample t-tests. For continuous variables not in a normal distribution, medians and interquartile ranges (IQR) were used for description, and group differences were compared through non-parametric tests (Mann-Whitney U test). For categorical variables, frequencies and percentages were used, and between-group comparisons were conducted through chi-squared or Fisher’s exact tests.
Kaplan-Meier (KM) survival curves were used to estimate the cumulative probabilities of AKI incidence and in-hospital survival across different ACAG level groups, and the differences between groups were assessed through the log-rank test.
To explore the nonlinear association of ACAG with AKI and in-hospital mortality, restricted cubic splines (RCS) were constructed. A piecewise fitting approach was used to describe the continuous relationship between ACAG and outcome variables, and smoothed curves were plotted to visually present the dose-response relationship. The study employed four default knots located at the 5th, 35th, 65th, and 95th percentiles, as recommended by Harrell et al. [21] to balance model fitting with robustness.
The independent correlations between ACAG and the incidence of AKI and in-hospital death among AP patients were investigated through Cox proportional hazards models. Confounding factors were progressively adjusted for as follows: Model 1 has no adjustment for confounding factors; In Model 2, demographic information like age, gender, race, and marital status were adjusted; In addition to the adjustments in Model 2, further adjustment for clinical variables, including BMI, sodium, potassium, calcium, magnesium, bicarbonate, ALT, AST, SCr, RBC, Hb, WBC, PLT, hypertension, DM, hyperlipidemia, CHD, COPD, CP, CKD, shock and albumin infusion was made in Model 3. To avoid multicollinearity, variance inflation factors (VIFs) were calculated for each variable when constructing multivariable models (Models 2 and 3), and variables with a VIF greater than 5 were excluded.
Stratified analyses were conducted based on key characteristics (e.g., age, gender, race, BMI, hypertension, DM, hyperlipidemia, CHD, COPD, CKD, CP, pancreatic cysts and pseudocysts, HF, AF, liver cirrhosis, and sepsis) to explore whether the effect of ACAG on outcome variables was consistent across subgroups. To elucidate the modifying effects of specific variables on the link between ACAG to outcomes, interaction tests were carried out. P < 0.05 denoted statistical significance.
3. Results
3.1 Population and baseline characteristics
Our study encompassed 1,552 patients, among whom 350 (23%) developed AKI, and 75 (4.8%) experienced in-hospital mortality. Within the entire cohort, a relatively higher proportion of patients were male (53%), 84% of the patients were over 40 years old, and 70% were of Caucasian descent.
S3 Table shows the baseline characteristics of patients stratified into four quartiles according to their ACAG levels (Q1: 8.50–14.69, Q2: 14.69–16.75, Q3: 16.75–19.25, Q4: 19.25–55.00). As ACAG levels increased, the incidence of AKI significantly escalated (9.3% vs. 17% vs. 25% vs. 40%, P < 0.001). Furthermore, the in-hospital death rate was notably higher in the Q4 group, with a significantly greater proportion of patients suffering from diabetes mellitus (DM), hyperlipidemia, HF, AF, COPD, CKD, and sepsis. Regarding clinical and laboratory parameters, the Q4 group exhibited significant increases in length of stay, SOFA score, HR, WBC, glucose (Glu), ALT, AST, TBIL, SCr, urine creatinine (UCr), BUN, and AG, while bicarbonate (HCO₃⁻), calcium (Ca), and ALB levels were significantly lower.
S4 Table summarizes the baseline characteristics of AKI and non-AKI cohorts. Compared with the non-AKI group, the AKI group exhibited a significantly elevated in-hospital mortality rate, and a notably greater proportion of patients with DM, hyperlipidemia, CHD, HF, AF, COPD, liver cirrhosis, pneumonia (PC and PPC), CKD, and sepsis. Clinically, the AKI group showed significantly elevated length of stay, SOFA score, SIRS score, WBC, potassium (K), glucose (Glu), AST, TBIL, SCr, UCr, BUN, and AG, while HCO₃ ⁻ , Ca, and Alb levels were significantly reduced.
3.2 Survival analysis
KM survival curves indicated notable variances in the cumulative incidence of AKI and in-hospital survival rates between the groups stratified by ACAG levels (AKI incidence: Log-rank test, P < 0.0001; in-hospital survival rate: Log-rank test, P = 0.0029). Patients with higher ACAG levels exhibited a higher risk of AKI, along with a significantly lower in-hospital survival rate. The detailed findings are presented in Fig 2.
(A) AKI event occurrence curve; (B) Survival curve for AP patients.
3.3 RCS analysis
RCS analysis revealed a non-linear association between ACAG and AKI in the AP cohort in the unadjusted model (P-overall < 0.001, P-nonlinear = 0.001), partially adjusted model (P-overall < 0.001, P-nonlinear < 0.001), and fully adjusted model (P-overall = 0.003, P-nonlinear = 0.008). Detailed results can be found in Figs 3A–C. Furthermore, a linear correlation of ACAG with inpatient death was observed in AP sufferers in the unadjusted model(P-overall = 0.001, P-nonlinear = 0.072), partially adjusted model (P-overall = 0.001, P-nonlinear = 0.073), and the fully adjusted model (P-overall = 0.002, P-nonlinear = 0.219). Detailed results can be found in Figs 3D–F. As ACAG levels increased, the risk of AKI exhibited an exponential growth trend, with a more pronounced increase beyond a certain threshold (e.g., ACAG > 16.78 mmol/L). In contrast, in-hospital mortality also increased with rising ACAG levels, although the curve was more gradual.
(A-C) AKI incidence; (D-F) in-hospital mortality.
3.4. Cox proportional hazards models
The results of Cox proportional hazards models indicated ACAG as an independent predictor of both AKI incidence and in-hospital mortality. When ACAG was included as a continuous variable in the Cox proportional hazards models, a statistically significant association between ACAG and AKI was demonstrated across unadjusted, partially adjusted, and fully adjusted models. As ACAG levels increased, both AKI risk and in-hospital mortality significantly elevated. Specifically, in the unadjusted model 1, each 1 mmol/L increase in ACAG was linked to a significantly higher risk of AKI (HR = 1.108, 95% CI: 1.090–1.126, P < 0.001), as well as a significantly increased risk of in-hospital death (HR = 1.070, 95% CI: 1.029–1.113, P < 0.001).
When ACAG was included as a categorical variable in the model, the results showed that patients in the high ACAG group had a significantly higher AKI risk than the low ACAG cohort. This association remained significant even after all confounding factors were adjusted, validating the strong correlation between high ACAG levels and adverse outcomes. Detailed information is presented in Table 1.
3.5 Subgroup analysis and interaction test
Subgroup analysis revealed that the association between ACAG and AKI risk was consistent across most subgroups (e.g., gender, race, BMI, hypertension, CHD, COPD, CKD, post-cardiac surgery, postoperative pneumonia, HF, AF, liver cirrhosis, and sepsis) (P for interaction > 0.05). However, interaction testing indicated a significant interaction effect between age, diabetes mellitus, hyperlipidemia, and ACAG (P for interaction < 0.05), suggesting that these factors may influence the strength of the effect of ACAG on outcomes. Detailed results are presented in Fig 4.
Notes:HRs are per 1-unit increase in ACAG.
4. Discussion
AKI is considered a major cause of death in AP sufferers in ICU. Plenty of studies reported a wide range of AKI incidence in AP patients, from 7.9% to 69.3%, a disparity that may be attributed to variations in the severity of the conditions of the included patients [7,9,22–26]. In this study, AKI was observed in 23% of AP patients, a result consistent with previous findings. Research has shown that AKI increases the mortality rate in AP patients by approximately fivefold [23]. In our study, the in-hospital death rate in the AKI population was significantly higher than that in the non-AKI cohort(18% vs. 1.1%). Therefore, early detection of high-risk AP patients for AKI and timely intervention are essential in lowering the incidence of AKI and ameliorating outcomes.
AP patients, especially those with SAP, frequently experience disturbances in acid-base homeostasis during disease progression. Experimental studies have previously demonstrated that metabolic acidosis may exacerbate experimental AKI [27]. However, the exact mechanisms underlying the aggravation of AKI by metabolic acidosis remain to be fully elucidated. Bugarski et al. suggested that metabolic acidosis may contribute to AKI by altering NAD and lipid metabolism in proximal renal tubules [28]. As the kidney is a critical organ in maintaining systemic acid-base balance, the development of AKI inevitably disrupts this homeostasis, thereby creating a vicious cycle that further aggravates renal injury [29].
Serum AG, a reliable and easily accessible clinical marker, can be used to determine the type of metabolic acidosis and evaluate the acid-base status [30]. AG correlates with the mortality and length of ICU stay [31–36], and it is widely employed to assess the prognosis of many illnesses like acute myocardial infarction [37], HF [38], influenza [39], acute pesticide intoxication [40], and disseminated intravascular coagulation [41]. However, AG is susceptible to the concentration of ALB, and in cases of hypoalbuminemia, AG may be underestimated [10,42,43]. Specifically, for every 1 g/dL change in serum albumin, a corresponding 2.3 to 2.5 mmol/L change in the serum AG was observed [17]. In our study, the ALB level was much lower in individuals with AKI than in those without AKI, suggesting that the direct use of AG may have limitations under these conditions. The introduction of ACAG can effectively overcome this issue and more accurately reflect the patient’s true metabolic state.
Our findings proved that ACAG was a crucial independent predictor of both AKI and in-hospital mortality. The risk of adverse outcomes significantly increased as ACAG levels rose. Cox proportional hazards models further confirmed this association and demonstrated that even after confounding factors were adjusted, elevated ACAG levels remained an independent risk factor for AKI and in-hospital mortality. Moreover, RCS analysis revealed a nonlinear correlation of ACAG with outcomes, particularly at high ACAG levels, where the risks of AKI and mortality increased at an accelerated rate. This finding suggests that high ACAG may reflect more severe metabolic disturbances or latent pathophysiological abnormalities, with its threshold effect providing an important basis for early clinical intervention.
Notably, ACAG has been proven to be an independent risk factor for mortality, although its threshold varies slightly across different studies. For example, in patients with AP, the critical value of ACAG is 19.03 mmol/L [44], while for severe AKI patients, it is 20 mmol/L [45]; in the critically ill AKI cohort undergoing continuous renal replacement therapy (CRRT), it is 20 mmol/L [46]; for those with acute myocardial infarction [47] and cardiac arrest [48], it is also 20 mmol/L; for trauma patients, it is 20.375 mmol/L [14]; for asthma patients, it is 20.38 mmol/L [49]; and for ICU patients with sepsis, it is 21.25 mmol/L [50]. In this study, ACAG was categorized into four intervals based on median and IQR. The incidence of AKI in AP sufferers significantly increased in the interval of >19.25 mmol/L. Furthermore, the RCS plot demonstrated a marked increase in AKI risk in AP patients once ACAG levels exceeded 16.78 mmol/L. Nonetheless, the findings suggest that clinicians focus more on patients who have higher ACAG values.
In the subgroup analysis, this study revealed that although the overall association between ACAG and outcomes remained consistent across most subgroups, significant interactions between ACAG and factors such as age, diabetes mellitus (DM), hyperlipidemia, and CP were observed. This suggests that these variables might modulate the predictive effectiveness of ACAG for AKI risk in AP patients. Previous research has shown that age influences the composition of blood acid-base balance, and with increasing age, the severity of metabolic acidosis may be exacerbated due to declining renal function [51]. Among patients with diabetes, chronic hyperglycemia, insulin resistance, impaired lactate and ketone metabolism, and the effects of various antidiabetic medications may all influence metabolic acidosis and the AG [52–58]. Moreover, diabetes itself can lead to varying degrees of renal impairment [59,60]. In patients with hyperlipidemia, elevated circulating lipid levels may interfere with ion measurements in laboratory assays, potentially leading to abnormal AG results [61–63]. Existing pancreatic dysfunction in CP patients may worsen the acid-base imbalance [64]. Moreover, although racial differences did not exhibit significant interaction effects in subgroup analyses, prior studies have reported a higher prevalence of metabolic acidosis among CKD patients who are Asian, Black, or Hispanic compared to non-Hispanic Whites [65]. Variations in acid-base metabolism across racial groups may be attributable to differences in dietary habits and genetic predispositions [66]. These findings underscore the importance of precise assessment and individualized management for specific high-risk patient groups. Future studies could focus on patient populations with differing comorbidities to further clarify the impact of ACAG on AKI occurrence and prognosis in patients with AP.
Our study also has some limitations. First, due to its retrospective nature, causal relationships cannot be definitively established, and only associations between variables can be explored. Additionally, several unmeasured confounding factors, such as fluid resuscitation practices and medication usage, may have influenced the results. Second, the exclusion of cases with uncomputable ACAG values may have introduced residual selection bias. Moreover, certain laboratory indicators were missing, and although multiple imputation methods were employed to address this, potential biases remain difficult to fully eliminate. Additionally, the study sample was predominantly drawn from a single-center public database with a majority of White patients, which may lead to selection bias and limit the applicability of our conclusions to other populations, such as those in Africa or Asia.
Our results unveil that ACAG, as a readily accessible laboratory marker, can effectively forecast AKI risk and in-hospital mortality in the AP cohort. Future multicenter prospective cohort studies are warranted to validate the predictive value of ACAG across diverse populations and explore whether interventions aimed at reducing ACAG levels (e.g., albumin infusion, management of acidosis) may improve clinical outcomes. Furthermore, existing medical databases can be further leveraged for in-depth data mining, in conjunction with machine learning and other advanced analytical techniques [67,68] to further optimize prediction models of ACAG combined with other clinical variables, thereby enhancing the accuracy and efficacy of risk assessment.
5. Conclusion
This study confirms that ACAG is a significant prognostic indicator for the AP population, and elevated levels significantly raise the risk of AKI and in-hospital mortality. Further research may pave the way for the integration of ACAG into standardized clinical risk assessment tools and provide a more scientific basis for personalized treatment in AP patients.
Supporting information
S1 Table. Proportion of missing values for each variable.
https://doi.org/10.1371/journal.pone.0330458.s001
(DOCX)
S2 Table. Sensitivity analysis (comparison of data before and after missing value interpolation).
https://doi.org/10.1371/journal.pone.0330458.s002
(DOCX)
S3 Table. The baseline characteristics of patients (According to ACAG levels).
https://doi.org/10.1371/journal.pone.0330458.s003
(DOCX)
S4 Table. The baseline characteristics of patients (According to AKI and non-AKI cohorts).
https://doi.org/10.1371/journal.pone.0330458.s004
(DOCX)
References
- 1. Mederos MA, Reber HA, Girgis MD. Acute Pancreatitis: A Review. JAMA. 2021;325(4):382–90. pmid:33496779
- 2. Garg PK, Singh VP. Organ Failure Due to Systemic Injury in Acute Pancreatitis. Gastroenterology. 2019;156(7):2008–23. pmid:30768987
- 3. Banks PA, Bollen TL, Dervenis C, Gooszen HG, Johnson CD, Sarr MG, et al. Classification of acute pancreatitis--2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013;62(1):102–11. pmid:23100216
- 4. Wilmer A. ICU management of severe acute pancreatitis. Eur J Intern Med. 2004;15(5):274–80. pmid:15450983
- 5. Bollen TL, van Santvoort HC, Besselink MG, van Leeuwen MS, Horvath KD, Freeny PC, et al. The Atlanta Classification of acute pancreatitis revisited. Br J Surg. 2008;95(1):6–21. pmid:17985333
- 6. Swaroop VS, Chari ST, Clain JE. Severe acute pancreatitis. JAMA. 2004;291(23):2865–8. pmid:15199038
- 7. Nassar TI, Qunibi WY. AKI Associated with Acute Pancreatitis. Clin J Am Soc Nephrol. 2019;14(7):1106–15. pmid:31118209
- 8. Wajda J, Dumnicka P, Maraj M, Ceranowicz P, Kuźniewski M, Kuśnierz-Cabala B. Potential Prognostic Markers of Acute Kidney Injury in the Early Phase of Acute Pancreatitis. Int J Mol Sci. 2019;20(15):3714. pmid:31366007
- 9. Devani K, Charilaou P, Radadiya D, Brahmbhatt B, Young M, Reddy C. Acute pancreatitis: Trends in outcomes and the role of acute kidney injury in mortality- A propensity-matched analysis. Pancreatology. 2018;18(8):870–7. pmid:30337224
- 10. Figge J, Jabor A, Kazda A, Fencl V. Anion gap and hypoalbuminemia. Crit Care Med. 1998;26(11):1807–10. pmid:9824071
- 11. Durward A, Mayer A, Skellett S, Taylor D, Hanna S, Tibby SM, et al. Hypoalbuminaemia in critically ill children: incidence, prognosis, and influence on the anion gap. Arch Dis Child. 2003;88(5):419–22. pmid:12716714
- 12. Wang Y, Tao Y, Yuan M, Yu P, Zhang K, Ying H, et al. Relationship between the albumin-corrected anion gap and short-term prognosis among patients with cardiogenic shock: a retrospective analysis of the MIMIC-IV and eICU databases. BMJ Open. 2024;14(10):e081597. pmid:39357986
- 13. Chen Y, You M-Y, Chu L. Association of albumin-corrected anion gap with severe consciousness disorders and outcomes in ischemic stroke: a retrospective MIMIC analysis. Sci Rep. 2024;14(1):26006. pmid:39472602
- 14. Yin F, Qiao Z, Wu X, Shi Q, Jin R, Xu Y. Association between albumin-corrected anion gap and in-hospital mortality of intensive care patients with trauma: A retrospective study based on MIMIC-Ⅲ and Ⅳ databases. PLoS One. 2024;19(3):e0300012. pmid:38452113
- 15. Lu Z, Yao Y, Xu Y, Zhang X, Wang J. Albumin corrected anion gap for predicting in-hospital death among patients with acute myocardial infarction: A retrospective cohort study. Clinics (Sao Paulo). 2024;79:100455. pmid:39079461
- 16. Johnson AEW, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10(1):1. pmid:36596836
- 17. Kraut JA, Madias NE. Serum anion gap: its uses and limitations in clinical medicine. Clin J Am Soc Nephrol. 2007;2(1):162–74. pmid:17699401
- 18. Kellum JA, Lameire N, Aspelin P, Barsoum RS, Burdmann EA, Goldstein SL, et al. Improving global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guideline for acute kidney injury. Kidney International Supplements. 2012;2(1): 1–138.
- 19. Yang Z, Gong H, Kan F, Ji N. Association between the triglyceride glucose (TyG) index and the risk of acute kidney injury in critically ill patients with heart failure: analysis of the MIMIC-IV database. Cardiovasc Diabetol. 2023;22(1):232. pmid:37653418
- 20. Chen T, Qian Y, Deng X. Triglyceride glucose index is a significant predictor of severe disturbance of consciousness and all-cause mortality in critical cerebrovascular disease patients. Cardiovasc Diabetol. 2023;22(1):156. pmid:37386419
- 21.
Harrell F. Regression modeling strategies: With applications to linear models, logistic and ordinal regression, and survival analysis. 2015.
- 22. Kes P, Vucicević Z, Ratković-Gusić I, Fotivec A. Acute renal failure complicating severe acute pancreatitis. Ren Fail. 1996;18(4):621–8. pmid:8875688
- 23. Li H, Qian Z, Liu Z, Liu X, Han X, Kang H. Risk factors and outcome of acute renal failure in patients with severe acute pancreatitis. J Crit Care. 2010;25(2):225–9. pmid:19781906
- 24. Lin H-Y, Lai J-I, Lai Y-C, Lin P-C, Chang S-C, Tang G-J. Acute renal failure in severe pancreatitis: A population-based study. Ups J Med Sci. 2011;116(2):155–9. pmid:21250932
- 25. Zhou J, Li Y, Tang Y, Liu F, Yu S, Zhang L, et al. Effect of acute kidney injury on mortality and hospital stay in patient with severe acute pancreatitis. Nephrology (Carlton). 2015;20(7):485–91. pmid:25726708
- 26. Tran DD, Oe PL, de Fijter CWH, van der Meulen J, Cuesta MA. Acute renal failure in patients with acute pancreatitis: prevalence, risk factors, and outcome. Nephrology Dialysis Transplantation. 1993;8(10):1079–84.
- 27. Magalhães PA da F, de Brito TS, Freire RS, da Silva MTB, dos Santos AA, Vale ML, et al. Metabolic acidosis aggravates experimental acute kidney injury. Life Sci. 2016;146:58–65. pmid:26773857
- 28. Bugarski M, Ghazi S, Polesel M, Martins JR, Hall AM. Changes in NAD and Lipid Metabolism Drive Acidosis-Induced Acute Kidney Injury. J Am Soc Nephrol. 2021;32(2):342–56. pmid:33478973
- 29. Doyle JF, Forni LG. Acute kidney injury: short-term and long-term effects. Crit Care. 2016;20(1):188. pmid:27373891
- 30. Achanti A, Szerlip HM. Acid-Base Disorders in the Critically Ill Patient. Clin J Am Soc Nephrol. 2023;18(1):102–12. pmid:35998977
- 31. Yang DE, Jo S, Lee DH, An WS, Jeong MJ, Son M. Dynamics of serum anion gaps with in-hospital mortality: Analysis of the multi-open databases. PLoS One. 2024;19(4):e0302206. pmid:38625899
- 32. Lipnick MS, Braun AB, Cheung JT-W, Gibbons FK, Christopher KB. The difference between critical care initiation anion gap and prehospital admission anion gap is predictive of mortality in critical illness. Crit Care Med. 2013;41(1):49–59. pmid:23190721
- 33. Li R, Jin X, Ren J, Deng G, Li J, Gao Y, et al. Relationship of Admission Serum Anion Gap and Prognosis of Critically Ill Patients: A Large Multicenter Cohort Study. Dis Markers. 2022;2022:5926049. pmid:36569219
- 34. Kim MJ, Kim YH, Sol IS, Kim SY, Kim JD, Kim HY, et al. Serum anion gap at admission as a predictor of mortality in the pediatric intensive care unit. Sci Rep. 2017;7(1):1456. pmid:28469150
- 35. Sun X, Lu J, Weng W, Yan Q. Association between anion gap and all-cause mortality of critically ill surgical patients: a retrospective cohort study. BMC Surg. 2023;23(1):226. pmid:37559030
- 36. Glasmacher SA, Stones W. Anion gap as a prognostic tool for risk stratification in critically ill patients - a systematic review and meta-analysis. BMC Anesthesiol. 2016;16(1):68. pmid:27577038
- 37. Lu J, Zhong L, Yuan M, Min J, Xu Y. Association between serum anion gap and all-cause mortality in patients with acute myocardial infarction: A retrospective study based on MIMIC-IV database. Heliyon. 2023;9(7):e17397. pmid:37539277
- 38. Xu H, Xia J, Wang A, Zong L, An X, Sun X. Serum anion gap is associated with mortality in intensive care unit patients with diastolic heart failure. Sci Rep. 2023;13(1):16670. pmid:37794229
- 39. Huang Y, Ao T, Zhen P, Hu M. Association between the anion gap and mortality in critically ill patients with influenza: A cohort study. Heliyon. 2024;10(15):e35199. pmid:39170390
- 40. Lee S-H, Park S, Lee J-W, Hwang I-W, Moon H-J, Kim K-H, et al. The Anion Gap is a Predictive Clinical Marker for Death in Patients with Acute Pesticide Intoxication. J Korean Med Sci. 2016;31(7):1150–9. pmid:27366016
- 41. Hu B, Cao J, Hu Y, Qin Z, Wang J. The Association Between Serum Anion Gap and All-Cause Mortality in Disseminated Intravascular Coagulation Patients: A Retrospective Analysis. Int J Gen Med. 2021;14:4535–44. pmid:34429638
- 42. Bruegger D, Jacob M, Scheingraber S, Conzen P, Becker BF, Finsterer U, et al. Changes in acid-base balance following bolus infusion of 20% albumin solution in humans. Intensive Care Med. 2005;31(8):1123–7. pmid:15999255
- 43. Pratumvinit B, Lam L, Kongruttanachok N, Hemthong C, Kost GJ, Kamkang P, et al. Anion gap reference intervals show instrument dependence and weak correlation with albumin levels. Clin Chim Acta. 2020;500:172–9. pmid:31669932
- 44. Li P, Shi L, Yan X, Wang L, Wan D, Zhang Z, et al. Albumin Corrected Anion Gap and the Risk of in-Hospital Mortality in Patients with Acute Pancreatitis: A Retrospective Cohort Study. J Inflamm Res. 2023;16:2415–22. pmid:37313307
- 45. Gao P, Min J, Zhong L, Shao M. Association between albumin corrected anion gap and all-cause mortality in critically ill patients with acute kidney injury: a retrospective study based on MIMIC-IV database. Ren Fail. 2023;45(2):2282708. pmid:37975171
- 46. Zhong L, Xie B, Ji X-W, Yang X-H. The association between albumin corrected anion gap and ICU mortality in acute kidney injury patients requiring continuous renal replacement therapy. Intern Emerg Med. 2022;17(8):2315–22. pmid:36112320
- 47. Sheng H, Lu J, Zhong L, Hu B, Sun X, Dong H. The correlation between albumin-corrected anion gap and prognosis in patients with acute myocardial infarction. ESC Heart Fail. 2024;11(2):826–36. pmid:38164072
- 48. Hu B, Zhong L, Yuan M, Min J, Ye L, Lu J, et al. Elevated albumin corrected anion gap is associated with poor in-hospital prognosis in patients with cardiac arrest: A retrospective study based on MIMIC-IV database. Front Cardiovasc Med. 2023;10:1099003. pmid:37034339
- 49. Wang S, Teng H, Han H, Xu Y. The relationship between albumin corrected anion gap levels and mortality in patients with asthma in the ICU. Sci Rep. 2023;13(1):16903. pmid:37803051
- 50. Hu T, Zhang Z, Jiang Y. Albumin corrected anion gap for predicting in-hospital mortality among intensive care patients with sepsis: A retrospective propensity score matching analysis. Clin Chim Acta. 2021;521:272–7. pmid:34303712
- 51. Frassetto L, Sebastian A. Age and systemic acid-base equilibrium: analysis of published data. J Gerontol A Biol Sci Med Sci. 1996;51(1):B91-9. pmid:8548506
- 52. Fenves AZ, Emmett M. Approach to Patients With High Anion Gap Metabolic Acidosis: Core Curriculum 2021. Am J Kidney Dis. 2021;78(4):590–600. pmid:34400023
- 53. Kraut JA, Xing SX. Approach to the evaluation of a patient with an increased serum osmolal gap and high-anion-gap metabolic acidosis. Am J Kidney Dis. 2011;58(3):480–4. pmid:21794966
- 54. Bo J, Li W, Chen Z, Wadden DG, Randell E, Zhou H, et al. D-lactate: a novel contributor to metabolic acidosis and high anion gap in diabetic ketoacidosis. Clin Chem. 2013;59(9):1406–7. pmid:23836843
- 55. Calimag APP, Chlebek S, Lerma EV, Chaiban JT. Diabetic ketoacidosis. Dis Mon. 2023;69(3):101418. pmid:35577617
- 56. Steinhorn B, Wiener-Kronish J. Dose-dependent relationship between SGLT2 inhibitor hold time and risk for postoperative anion gap acidosis: a single-centre retrospective analysis. Br J Anaesth. 2023;131(4):682–6. pmid:37541949
- 57. Connelly PJ, Lonergan M, Soto-Pedre E, Donnelly L, Zhou K, Pearson ER. Acute kidney injury, plasma lactate concentrations and lactic acidosis in metformin users: A GoDarts study. Diabetes Obes Metab. 2017;19(11):1579–86. pmid:28432751
- 58. Donnan K, Segar L. SGLT2 inhibitors and metformin: Dual antihyperglycemic therapy and the risk of metabolic acidosis in type 2 diabetes. Eur J Pharmacol. 2019;846:23–9. pmid:30639796
- 59. Gui Y, Palanza Z, Fu H, Zhou D. Acute kidney injury in diabetes mellitus: Epidemiology, diagnostic, and therapeutic concepts. FASEB J. 2023;37(4):e22884. pmid:36943403
- 60. Rhee CM, Kalantar-Zadeh K. Diabetes mellitus: Complex interplay between metformin, AKI and lactic acidosis. Nat Rev Nephrol. 2017;13(9):521–2. pmid:28736433
- 61. Graber ML, Quigg RJ, Stempsey WE, Weis S. Spurious hyperchloremia and decreased anion gap in hyperlipidemia. Ann Intern Med. 1983;98(5 Pt 1):607–9. pmid:6846974
- 62. Akkawi AR, Fawcett M, Challa A, Nguyen T, Jackson J, Muraga R. Hypertriglyceridemia-Induced Pseudo-High Anion Gap Metabolic Acidosis Manifesting as Ischemic Stroke. Mayo Clin Proc. 2023;98(9):1425–6. pmid:37661152
- 63. Bonatto NCM, Alves LSS, Silva LE, Milhorine CA, de Barros LD, Santos JA, et al. Does postprandial lipemia interfere with blood gas analysis and assessment of acid-base status in dogs?. Res Vet Sci. 2023;154:52–8. pmid:36459719
- 64. Kleeff J, Whitcomb DC, Shimosegawa T, Esposito I, Lerch MM, Gress T, et al. Chronic pancreatitis. Nat Rev Dis Primers. 2017;3:17060. pmid:28880010
- 65. Tangri N, Mathur V, Reaven N, Funk SE, Wesson DE. Relationship between Metabolic Acidosis and Chronic Kidney Disease Progression across Racial and Ethnic Groups: An Observational, Retrospective Cohort Study. Am J Nephrol. 2022;53(8–9):603–13. pmid:36349769
- 66. Abdel Goad EH, Bereczky ZB. Metabolic risk factors in patients with renal stones in KwaZulu Natal: an inter-racial study (Asian and Whites). BJU Int. 2004;93(1):120–3. pmid:14678382
- 67. Che W-Q, Li Y-J, Tsang C-K, Wang Y-J, Chen Z, Wang X-Y, et al. How to use the Surveillance, Epidemiology, and End Results (SEER) data: research design and methodology. Mil Med Res. 2023;10(1):50. pmid:37899480
- 68. Xu Y, Zheng X, Li Y, Ye X, Cheng H, Wang H, et al. Exploring patient medication adherence and data mining methods in clinical big data: A contemporary review. J Evid Based Med. 2023;16(3):342–75. pmid:37718729