04 July 2025: Database Analysis
Predictive Nomogram for Acute Kidney Injury Risk with Vancomycin and Piperacillin Tazobactam in Sepsis Treatment
Guohua Liu ABCDEF 1, Ji Li ABCFG 2, Lin Chen DEFG 3, Hao Ding ABCDFG 4*, Sufang Yang ABCDEFG 1
DOI: 10.12659/MSM.949340
Med Sci Monit 2025; 31:e949340
Abstract
BACKGROUND: The combination of vancomycin (VAN) and piperacillin-tazobactam (TZP) is commonly used to treat sepsis, but it is associated with a high risk of acute kidney injury (AKI). This article describes our development of a nomogram to predict the probability of AKI caused by the combination of the VAN and TZP in the treatment of sepsis.
MATERIAL AND METHODS: Patients with sepsis treated with VAN and TZP from the MIMIC-IV database were included. The patients were randomly divided into a training set and a validation set at a 7: 3 ratio. Key variables were identified through the integration of least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression analysis. The performance of the nomogram was evaluated using area under the receiver operating characteristic curves (AUC), calibration curves, and decision curve analysis in the training set, and was further assessed in the validation set.
RESULTS: We included 618 patients, with 469 developing AKI. Six risk factors – body mass index, SOFA score, mechanical ventilation, antihypertensive drugs, serum potassium, and total vancomycin dosage – were identified as predictors of AKI occurrence. The AUC was 0.75 in the training set and 0.74 in the validation set. Calibration curves showed good consistency. Decision curve analysis indicated the nomogram worked well for AKI risk prediction if the threshold in the training set was 3-80% and that in the validation set was 10-95%.
CONCLUSIONS: This study was the first attempt to develop and validate a model that could predict the risk of AKI caused by the combination of VAN and TZP in the treatment of sepsis, providing a reference for clinical decision-making.
Keywords: sepsis, Vancomycin, Piperacillin, Tazobactam Drug Combination, kidney, Predictive Learning Models, Humans, nomograms, Acute Kidney Injury, Male, Female, Middle Aged, Aged, Risk Factors, Anti-Bacterial Agents, ROC Curve
Introduction
Sepsis is a systemic inflammatory response syndrome triggered by infection that can lead to multiorgan dysfunction in patients [1]. Acute kidney injury (AKI) is a common complication in sepsis patients, which who significantly prolongs hospital stays and increases medical costs and mortality rates, thereby imposing substantial physical and economic burdens on families and society [2].
The treatment of sepsis includes early antibiotic administration, control of the source of infection, fluid resuscitation, and organ function support as comprehensive therapeutic measures [3]. Among these, the rational use of antibiotics is crucial for controlling infections and improving patient outcomes. VAN is a glycopeptide antibiotic with strong antibacterial activity against methicillin-resistant
In recent years, several clinical studies have reported that AKI is associated with the combination of the VAN and TZP, highlighting this safety issue [6–9]. Both VAN and TZP alone can cause AKI [10,11], and when used in combination, the risk of AKI can increase to 31.7–35.0% [12,13]. AKI can lead to a greater risk of death. A multicenter observational study of intensive care unit (ICU) patients in 23 countries revealed that the mortality rate among AKI patients was 52%, with an additional 8% of AKI patients dying after discharge, resulting in an overall mortality rate of 60.3% [14]. Moreover, the risk of death increases with increasing severity of AKI [15].
Currently, tools used for the assessment and prediction of AKI mainly include the SOFA score, SAPS II score, RIFLE criteria, AKIN criteria, and KDIGO criteria [16]. The SOFA score, RIFLE criteria, AKIN criteria, and KDIGO criteria are based solely on serum creatinine levels and urine output, without considering other important indicators such as medication use, invasive procedures, and electrolyte levels [17,18]. Therefore, they might not fully reflect subtle changes in renal function. The SAPS II is a general critical illness scoring system and is not specifically designed for AKI. Thus, it may not be as sensitive as specialized AKI scoring criteria in evaluating the severity and prognosis of renal injury [19]. Scoring systems typically only consider a few key indicators while ignoring other clinical data related to AKI, such as patient history and laboratory test results. Their advantage lies in their simplicity and standardization, but their predictive ability is limited. In contrast, predictive models can integrate various types of clinical data, including laboratory test results, medical history, and medication history, thereby having a more comprehensive ability to assess and predict.
Therefore, the present study was conducted on the risk factors for AKI in patients with sepsis treated with VAN combined with TZP, as well as the construction of a predictive model, could help doctors assess the risk of AKI in patients before treatment. This may allow for the development of more individualized treatment plans and avoided unnecessary renal damage, which has important clinical value and practical significance.
Material and Methods
DATA SOURCE:
The data used in this study were derived from MIMIC-IV 3.0 (Medical Information Mart for Intensive Care), a large, open, and intensive clinical research database. The data included the demographic information of the patients, common laboratory indicators upon admission, comorbidities, medication use, and survival outcomes. One member of our research team, who had undergone training and obtained certification (Certification No. 67219293), was responsible for accessing and extracting data from the database. Our research was conducted in accordance with the Declaration of Helsinki. The MIMIC database adheres to the review board protocol, with all patient personal information deidentified and specific patients marked with random codes. The Ethics Committee of the Affiliated Hospital of Jinggangshan University approved this study to be exempt from informed consent and ethics review requirements.
INCLUSION AND EXCLUSION CRITERIA:
The inclusion criteria were: (1) Patients diagnosed with sepsis, identified using International Classification of Diseases (ICD-9 and ICD-10) codes; (2) Concurrently treated with VAN (intravenous administration) and TZP (8: 1) (intravenous administration), with an overlap of ≥24 hours in the use of both drugs; (3) Complete renal function monitoring data (serum creatinine) available within 48 hours before the start of medication and up to 7 days after discontinuation; (4) Baseline serum creatinine (SCr) ≤1.2 mg/dL (or estimated glomerular filtration rate ≥60 mL/min/1.73 m2) before medication, excluding chronic kidney disease; (5) Age ≥18 years.
Exclusion criteria were: (1) Non-first admission to the ICU and first hospitalization; (2) Received other nephrotoxic drugs (eg, aminoglycosides, contrast agents, NSAIDs) within 48 hours before medication or during the medication period; (3) Presence of other clear causes of AKI (eg, septic shock, cardiogenic shock, urinary tract obstruction, renal vascular disease); (4) Admitted with a diagnosis of AKI or requiring renal replacement therapy (RRT); (5) Total duration of medication use <48 hours or missing key variables (eg, weight, baseline SCr); (6) Discharged or died within 72 hours after the end of medication, making it impossible to fully observe renal toxicity outcomes; (7) Pregnant or breastfeeding patients. A total of 3011 patients were included initially, and 618 patients with sepsis who were treated with a combination of VAN and TZP were ultimately selected for analysis according to the inclusion and exclusion criteria.
DEFINITION OF OUTCOME:
The primary outcome of the study was the occurrence of AKI. The diagnosis and staging of AKI referred to the latest international clinical practice guidelines for AKI, namely the 2024 Kidney Disease: Improving Global Outcomes (KDIGO) criteria [20], which specify that an increase in serum creatinine (SCr) of ≥0.3 mg/dL (≥26.5 μmol/L) within 48 hours, or an increase in SCr of ≥50% from the baseline creatinine level within 7 days, or urine output <0.5 ml/kg/h for ≥6 hours, can be used to diagnose AKI.
DATA EXTRACTION AND MANAGEMENT:
The research team extracted and managed the data using structured query language (SQL) and PostgreSQL 15 tools. The extracted data included patient admission numbers, age, sex, past medical history, medication record, and laboratory indicators. Past medical history included hypertension, diabetes, heart failure, and kidney injury. The laboratory indicators included white blood cells, hemoglobin, platelets, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and serum creatinine. The medications used included antihypertensive drugs, glucocorticoids, nephrotoxic drugs, and immunosuppressants.
STATISTICAL ANALYSIS:
Missing values are handled using multiple imputation methods. Patients were grouped based on whether they developed acute kidney injury (AKI), and the statistical significance of differences between continuous variables was assessed. Continuous variables that followed a normal distribution are expressed as mean±standard deviation and compared using the t test; those that did not follow a normal distribution are expressed as median with interquartile range (IQR) and were compared using the Wilcoxon rank-sum test. Categorical variables were expressed as percentages and compared using the chi-square test. All statistical analyses were performed using R software (version 4.3.0) and Zstats v1.0 (www.zstats.net), with two-sided P values <0.05 considered statistically significant. The detailed study methods and procedures are illustrated in Figure 1.
Results
PATIENT CHARACTERISTICS:
A total of 618 eligible patients were included in this study. The entire cohort was randomly assigned to a training set (n=432) or a validation set (n=186). Among them, 469 patients developed AKI, with 325 patients in the training set and 144 patients in the validation set. Statistical analysis revealed no significant differences between the training and validation sets (P>0.05). The baseline characteristics of the patients are listed in Table 1.
IDENTIFICATION OF PREDICTIVE FACTORS:
We performed a multicollinearity test using the Variance Inflation Factor (VIF) for all independent variables, and the results showed that all VIF values were less than 5, indicating no multicollinear variables. Given the large number of variables involved, we employed a two-step approach to screen for clinical characteristics. First, LASSO regression was used for initial screening to identify potential predictive factors, ensuring avoidance of overfitting and enhancing the robustness of the model (Figure 2A, 2B). This process identified 10 candidate predictive factors: HT, cirrhosis, BMI, SOFA score, mechanical ventilation, antihypertensive drugs, WBC, serum potassium, urea nitrogen, and total VAN dosage. Multivariate logistic regression was subsequently performed on these 10 predictive factors (Table 2). The results revealed that BMI, SOFA score, mechanical ventilation, antihypertensive drugs, serum potassium, and total vancomycin dosage were independent predictive factors for AKI caused by the combination of the VAN and TZP in the treatment of sepsis.
PREDICTIVE MODEL OF THE NOMOGRAM:
We constructed a nomogram that integrates 6 important predictive variables (BMI, SOFA score, mechanical ventilation, antihypertensive drugs, serum potassium and total vancomycin dosage) to intuitively assess the risk of developing AKI (Figure 3A). Figure 3B displays an interactive nomogram. For example, for a sample patient with a score of 149, the corresponding probability of infection is 61.6%.
VALIDATION OF THE PREDICTIVE MODEL:
In the training set, the nomogram achieved an AUC of 0.75 (95% CI: 0.70–0.81) for predicting AKI. In the validation set, the AUC remained as high as 0.74 (95% CI: 0.66–0.83), demonstrating the strong discriminatory power of the model (Figure 4).
CALIBRATION OF THE PREDICTIVE MODEL:
The calibration curves demonstrated no significant differences between the predicted and actual probabilities of AKI occurrence in either the training set (χ2=9.65, df=8, P=0.29) or the validation set (χ2=3.21, df=8, P=0.92) (Figure 5).
CLINICAL APPLICATION:
Additionally, the results of decision curve analysis (DCA) indicated that the nomogram could significantly benefit patients who developed AKI. In the training set, these benefits were observed within a risk threshold probability range of 3–80%, and in the validation set they were observed within a range of 10–95% (Figure 6).
Discussion
VAN is a glycopeptide antibiotic that has been widely used in clinical practice for the treatment of methicillin-resistant
Currently, there are no published models for predicting the risk of AKI in sepsis patients treated with a combination of VAN and TZP. Therefore, we conducted a study to develop a predictive model for AKI in patients with sepsis treated with this combination. In this study, we identified BMI, SOFA score, mechanical ventilation, antihypertensive drugs, serum potassium, and total VAN dosage as the main factors associated with AKI. Based on these 6 factors, we developed an AKI predictive model – a nomogram – that demonstrated good discriminatory power, calibration, and clinical utility.
Studies have shown that the above 6 factors are associated with the occurrence of AKI. SOFA score has been an important tool for evaluating the severity of organ dysfunction in critically ill patients [22]. Studies have demonstrated a significant association between the SOFA score and the occurrence of AKI [23,24]. One study included 626 patients with diabetic ketoacidosis and found a significant correlation between the SOFA score and the incidence of AKI. For each 1-point increase in the SOFA score, the risk of AKI increased by 35%, the risk of requiring renal replacement therapy increased by 43.5%, and the risk of in-hospital mortality increased by 43.7% [25]. Mechanical ventilation (MV) has been a common supportive treatment in ICUs, used to improve oxygenation and ventilation function [26]. Studies have shown a significant association between mechanical ventilation and the occurrence of AKI. A systematic review and meta-analysis that included 31 studies demonstrated that mechanical ventilation was significantly correlated with the risk of AKI, and patients who received mechanical ventilation had a 3.16 times higher risk of developing AKI compared to those who did not receive mechanical ventilation [27]. Antihypertensive drugs have been widely used in clinical practice to control hypertension and related diseases. The use of certain antihypertensive drugs has been associated with the occurrence of AKI. A large-scale cohort study showed that patients using angiotensin-converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB) had a 12% higher risk of developing AKI during exposure compared to non-exposure periods [28]. A nationwide prospective cohort study targeting patients with sepsis showed that obesity (BMI ≥30 kg/m2) was significantly associated with a high risk of early sepsis-associated acute kidney injury (SA-AKI). Compared with patients of normal weight, obese patients had a higher incidence of SA-AKI, and for every 10-unit increase in BMI, the risk of AKI further increased [29]. The use of VAN has been significantly associated with the occurrence of AKI, especially when administered at high doses. One study found that the total dose of VAN was significantly correlated with the risk of AKI, particularly in critically ill patients. The study showed that a VAN trough level >20 mg/L was an independent risk factor for AKI [30]. Another study pointed out that dosing regimens aimed at maintaining VAN trough concentrations of 15–20 mg/L (typically requiring 50–60 mg/kg/day) were associated with a higher incidence of nephrotoxicity [31]. Abnormal serum potassium levels (including hyperkalemia and hypokalemia) have been closely related to the occurrence and prognosis of AKI [32]. Hyperkalemia is a common complication of AKI, especially when renal function is impaired and the ability to excrete potassium is reduced, which can easily lead to elevated serum potassium levels [33]. One study showed that patients with higher serum potassium levels at admission (>5.5 mmol/L) had a significantly increased risk of developing AKI, and hyperkalemia was associated with the severity and poor prognosis of AKI [34].
Currently, the tools used to assess AKI mainly include the RIFLE criteria, AKIN criteria, KDIGO criteria, and the SOFA scoring system [16]. These scoring systems primarily rely on serum creatinine levels and urine output to evaluate the severity of renal dysfunction [35]. However, our predictive model integrates the SOFA score, BMI, mechanical ventilation, antihypertensive drugs, serum potassium, and total VAN dosage. It does not rely solely on serum creatinine levels and urine output, but also includes factors such as BMI, medications used, electrolytes, and procedures, providing a more comprehensive assessment of renal damage risk and offering specific risk probabilities. The nomogram predictive model of our study was integrated into the electronic medical record (EMR) system, enabling automated risk assessment and real-time clinical decision support. When the patient’s clinical data were updated, the EMR system automatically called the nomogram model to calculate the risk probability of AKI based on the extracted data. For high-risk patients, the antibiotic dosage was adjusted. For example, the initial dose of VAN was reduced or the duration of combination therapy was shortened. The frequency of renal function monitoring was also increased for high-risk patients, such as daily testing of serum creatinine and urine output, to detect early signs of AKI in a timely manner.
The predictive model was primarily designed for sepsis patients in the ICU who were concurrently treated with VAN and TZP. Sepsis is a common critical illness in ICUs, and VAN and TZP are frequently used antibiotics in clinical practice, with widespread use across different hospitals. The variables included in the model, such as the SOFA score, BMI, serum potassium, and VAN dosage, are clinical data and indicators that can be routinely obtained in the ICUs of most hospitals. Therefore, considering the type of disease, the medications used, and the availability of data, the model has a certain degree of universality and operability across different hospital settings. It can be applied in hospitals with basic ICU facilities and capability for routine laboratory tests. Overall, the model has a generalizability within this specific group of sepsis patients.
Nevertheless, our study had certain limitations. Firstly, this was a retrospective study, which inevitably leads to selection bias resulting in the exclusion of certain variables, such as the detection of TIMP-2 and IGFBP7 in urine [36] and the Systemic Immune-Inflammation Index [37]. Second, since the data were derived solely from the MIMIC-IV database, this could affect the generalizability of the predictive model to other populations. Therefore, it is necessary to conduct large-scale, multicenter studies for external validation. Finally, due to the limited types of variables in the public database, some variables of interest, such as VAN serum concentrations, detailed fluid management data, and specific sources of infection, were not included in the study. Failure to consider VAN trough concentrations can lead to underestimation or overestimation of the risk of AKI in certain patient groups [38]. For patients with well-controlled trough concentrations, the model might overestimate their risk; whereas for those with excessively high trough concentrations, the model might underestimate their risk [39]. The model might also fail to distinguish patients at high risk due to fluid overload, thereby affecting its discriminative ability and calibration. Sepsis has a variety of infection sources, and different sources may have different impacts on renal function [40]. For example, urinary tract infections can directly cause kidney damage, while pulmonary infections can indirectly affect kidney function through systemic inflammatory responses, which could lead to inaccurate predictions of AKI risk by the model.
Conclusions
To the best of our knowledge, this is the first study to develop and validate a nomogram for predicting the risk of AKI associated with the combination of VAN and TZP in the treatment of sepsis. Our nomogram incorporates the SOFA score, BMI, mechanical ventilation, antihypertensive drugs, serum potassium, and total VAN dosage. The nomogram was internally validated as a useful tool for risk assessment. The predictive model was developed and integrated into the EMR system to enable physicians to assess the risk of AKI in patients before treatment, thereby facilitating the development of more individualized treatment plans and avoiding unnecessary renal injury.
Figures
Figure 1. Study flowchart. 
Figure 2. (A) The optimal parameter (λ) selection in the LASSO model employed 5-fold cross-validation using a minimum criteria approach. The optimal values of λ are represented by dotted vertical lines. Among these values, λ=0.0063, corresponding to an algorithm of λ equal to −3.2, was selected as the optimal choice. (B) LASSO coefficient profiles of 40 clinical features. The plot was created using a logarithmic scale for the lambda values. A vertical line was added to indicate the lambda value selected through 5-fold cross-validation. This optimal lambda value led to the identification of 7 features with non-zero coefficients. The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). 
Figure 3. (A) Nomogram with BMI, SOFA, Ventilation, Antihypertensive, Potassium, and Vantotalval. (B) The dynamic nomogram reveals the risk of AKI in sepsis patients treated with PTZ and VAN. This patient has a BMI of 30 kg/m2, a SOFA score of 2, is not on antihypertensive medication, has a serum potassium level of 4.0 mEq/L, has received a total vancomycin dose of 1 g, and is on mechanical ventilation. The probability of AKI occurrence for this case is 0.616. SOFA – sequential organ failure assessment score; Vantotalval – total VAN dosage; Ventilation – mechanical ventilation; BMI – body mass index. The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). 
Figure 4. The area under the receiver operating characteristic curve (AUC) for the discrimination of the model. (A) The training set, 0.75 (95% CI: 0.70–0.81). (B) The validation set, 0.74 (95% CI: 0.66–0.83). The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). 
Figure 5. Calibration curves for the predicting probability of AKI in the training set (A, B) in the validation set, all P value >0.05 in the Hosmer-Lemeshow test suggested an agreement between the predicted probabilities and observed outcomes. The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). 
Figure 6. Decision curve analysis (DCA) of the nomogram for predicting AKI. The green line represents the scenario in which no patients develop AKI, whereas the red line represents the scenario in which all patients develop AKI. The blue line corresponds to the nomogram model. The analysis was performed for both training set (A) and validation set (B). The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). References
1. Gauer R, Forbes D, Boyer N, Sepsis: Diagnosis and management: Am Fam Physician, 2020; 101(7); 409-18
2. White KC, Serpa-Neto A, Hurford R, Sepsis-associated acute kidney injury in the intensive care unit: Incidence, patient characteristics, timing, trajectory, treatment, and associated outcomes. A multicenter, observational study: Intensive Care Med, 2023; 49(9); 1079-89
3. Kellum JA, Wen X, de Caestecker MP, Hukriede NA, Sepsis-associated acute kidney injury: A problem deserving of new solutions: Nephron, 2019; 143(3); 174-78
4. Tabah A, Laupland KB: Curr Opin Crit Care, 2022; 28(5); 495-504
5. Chanderraj R, Admon AJ, He Y, Mortality of patients with sepsis administered piperacillin-tazobactam vs cefepime: JAMA Intern Med, 2024; 184(7); 769-77
6. Miano TA, Hennessy S, Yang W, Association of vancomycin plus piperacillin-tazobactam with early changes in creatinine versus cystatin C in critically ill adults: A prospective cohort study: Intensive Care Med, 2022; 48(9); 1144-55
7. Luther MK, Timbrook TT, Caffrey AR, Vancomycin plus piperacillin-tazobactam and acute kidney injury in adults: A systematic review and meta-analysis: Crit Care Med, 2018; 46(1); 12-20
8. Holsen MR, Meaney CJ, Hassinger AB, Fusco NM, Increased risk of acute kidney injury in critically ill children treated with vancomycin and piperacillin/tazobactam: Pediatr Crit Care Med, 2017; 18(12); e585-e91
9. Covert KL, Knoetze D, Cole M, Lewis P, Vancomycin plus piperacillin/tazobactam and acute kidney injury risk: A review of the literature: J Clin Pharm Ther, 2020; 45(6); 1253-63
10. Kunming P, Can C, Zhangzhang C, Vancomycin associated acute kidney injury: A longitudinal study in China: Front Pharmacol, 2021; 12; 632107
11. Tang Girdwood S, Hasson D, Caldwell JT, Relationship between piperacillin concentrations, clinical factors and piperacillin/tazobactam-associated acute kidney injury: J Antimicrob Chemother, 2023; 78(2); 478-87
12. Blair K, Covington EW, Incidence and risk factors of acute kidney injury in patients receiving concomitant vancomycin and continuous-infusion piperacillin/tazobactam: A retrospective cohort study: Ann Pharmacother, 2020; 54(11); 1096-101
13. Robertson AD, Li C, Hammond DA, Dickey TA, Incidence of acute kidney injury among patients receiving the combination of vancomycin with piperacillin-tazobactam or meropenem: Pharmacotherapy, 2018; 38(12); 1184-93
14. Uchino S, Kellum JA, Bellomo R, Acute renal failure in critically ill patients: A multinational, multicenter study: JAMA, 2005; 294(7); 813-18
15. Hoste EA, Clermont G, Kersten A, RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: A cohort analysis: Crit Care, 2006; 10(3); R73
16. Gong Y, Ding F, Zhang F, Gu Y, Investigate predictive capacity of in-hospital mortality of four severity score systems on critically ill patients with acute kidney injury: J Investig Med, 2019; 67(8); 1103-109
17. Kellum JA, Lameire NKDIGO AKI Guideline Work Group, Diagnosis, evaluation, and management of acute kidney injury: A KDIGO summary (Part 1): Crit Care, 2013; 17(1); 204
18. Lameire N, Van Biesen W, Vanholder R, Acute renal failure: Lancet, 2005; 365(9457); 417-30
19. Jansen G, Entz S, Holland FO, A comparison of simplified acute physiology score II and sepsis-related organ failure assessment score for prediction of mortality after intensive care unit cardiac arrest: Minerva Anestesiol, 2024; 90(5); 359-68
20. Madero M, Levin A, Ahmed SB, Evaluation and management of chronic kidney disease: Aynopsis of the kidney disease: Improving global outcomes 2024 clinical practice guideline: Ann Intern Med, 2025 [Online ahead of print]
21. Bamgbola O, Review of vancomycin-induced renal toxicity: An update: Ther Adv Endocrinol Metab, 2016; 7(3); 136-47
22. Chang CH, Fan PC, Chang MY, Acute kidney injury enhances outcome prediction ability of sequential organ failure assessment score in critically ill patients: PLoS One, 2014; 9(10); e109649
23. Lee CC, Chen CW, Yen HK, Comparison of two modern survival prediction tools, SORG-MLA and METSSS, in patients with symptomatic long-bone metastases who underwent 24. Local treatment with surgery followed by radiotherapy and with radiotherapy alone: Clin Orthop Relat Res, 2024; 482(12); 2193-208
25. Wang Y, Zhong L, Min J, Albumin corrected anion gap and clinical outcomes in elderly patients with acute kidney injury caused or accompanied by sepsis: A MIMIC-IV retrospective study: Eur J Med Res, 2025; 30(1); 11
26. van den Akker JP, Egal M, Groeneveld AB, Invasive mechanical ventilation as a risk factor for acute kidney injury in the critically ill: A systematic review and meta-analysis: Crit Care, 2013; 17(3); R98
27. Frydman S, Freund O, Zornitzki L, Relation of mechanical ventilation to acute kidney injury in myocardial infarction patients: Cardiorenal Med, 2023; 13(1); 263-70
28. Mansfield KE, Nitsch D, Smeeth L, Prescription of renin-angiotensin system blockers and risk of acute kidney injury: A population-based cohort study: BMJ Open, 2016; 6(12); e012690
29. Ahn YH, Yoon SM, Lee J, Early sepsis-associated acute kidney injury and obesity: JAMA Netw Open, 2024; 7(2); e2354923
30. Chang J, Pais GM, Engel PL, Impact of vancomycin loading doses and dose escalation on glomerular function and kidney injury biomarkers in a translational rat model: Antimicrob Agents Chemother, 2023; 67(2); e0127622
31. Chang J, Pais GM, Marianski S, Iohexol-measured glomerular filtration rate and urinary biomarker changes between vancomycin and vancomycin plus piperacillin-tazobactam in a translational rat model: Antimicrob Agents Chemother, 2023; 67(8); e0030423
32. Marahrens B, Patschan S, Patschan D, Stratification of acute kidney injury risk, disease severity, and outcomes by electrolyte disturbances: J Clin Med Res, 2023; 15(2); 59-67
33. Guo Y, Qiu Y, Xue T, Association between admission baseline blood potassium levels and all-cause mortality in patients with acute kidney injury combined with sepsis: A retrospective cohort study: PLoS One, 2024; 19(11); e0309764
34. Gao XP, Zheng CF, Liao MQ, Admission serum sodium and potassium levels predict survival among critically ill patients with acute kidney injury: A cohort study: BMC Nephrol, 2019; 20(1); 311
35. Luo X, Jiang L, Du B, A comparison of different diagnostic criteria of acute kidney injury in critically ill patients: Crit Care, 2014; 18(4); R144
36. Wang K, Xie S, Xiao K, Biomarkers of sepsis-induced acute kidney injury: Biomed Res Int, 2018; 2018; 6937947
37. Sun Jie, Qi Ying, Wang Wenxin, Systemic immune-inflammation index (SII) as a predictor of short-term mortality risk in sepsis-associated acute kidney injury: A retrospective cohort study: Med Sci Monit, 2024; 30; e943414
38. Yue S, Li S, Huang X, Construction and validation of a risk prediction model for acute kidney injury in patients suffering from septic shock: Dis Markers, 2022; 2022; 9367873
39. Deng F, Peng M, Li J, Nomogram to predict the risk of septic acute kidney injury in the first 24 h of admission: An analysis of intensive care unit data: Ren Fail, 2020; 42(1); 428-36
40. Ma Z, Liu W, Deng F, An early warning model to predict acute kidney injury in sepsis patients with prior hypertension: Heliyon, 2024; 10(2); e24227
Figures
Figure 1. Study flowchart.Figure 2. (A) The optimal parameter (λ) selection in the LASSO model employed 5-fold cross-validation using a minimum criteria approach. The optimal values of λ are represented by dotted vertical lines. Among these values, λ=0.0063, corresponding to an algorithm of λ equal to −3.2, was selected as the optimal choice. (B) LASSO coefficient profiles of 40 clinical features. The plot was created using a logarithmic scale for the lambda values. A vertical line was added to indicate the lambda value selected through 5-fold cross-validation. This optimal lambda value led to the identification of 7 features with non-zero coefficients. The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net).Figure 3. (A) Nomogram with BMI, SOFA, Ventilation, Antihypertensive, Potassium, and Vantotalval. (B) The dynamic nomogram reveals the risk of AKI in sepsis patients treated with PTZ and VAN. This patient has a BMI of 30 kg/m2, a SOFA score of 2, is not on antihypertensive medication, has a serum potassium level of 4.0 mEq/L, has received a total vancomycin dose of 1 g, and is on mechanical ventilation. The probability of AKI occurrence for this case is 0.616. SOFA – sequential organ failure assessment score; Vantotalval – total VAN dosage; Ventilation – mechanical ventilation; BMI – body mass index. The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net).Figure 4. The area under the receiver operating characteristic curve (AUC) for the discrimination of the model. (A) The training set, 0.75 (95% CI: 0.70–0.81). (B) The validation set, 0.74 (95% CI: 0.66–0.83). The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net).Figure 5. Calibration curves for the predicting probability of AKI in the training set (A, B) in the validation set, all P value >0.05 in the Hosmer-Lemeshow test suggested an agreement between the predicted probabilities and observed outcomes. The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net).Figure 6. Decision curve analysis (DCA) of the nomogram for predicting AKI. The green line represents the scenario in which no patients develop AKI, whereas the red line represents the scenario in which all patients develop AKI. The blue line corresponds to the nomogram model. The analysis was performed for both training set (A) and validation set (B). The software used for creation of the figure: Produced by Professor Zheng’s team from Zhejiang Chinese Medical University: R version 4.3.3 (2024-02-29), along with Zstats 1.0 (www.zstats.net). Tables
Table 1. Baseline characteristics of participants in the training and validation sets.
Table 2. Multivariate regression model based on LASSO regression analysis results.Table 1. Baseline characteristics of participants in the training and validation sets.Table 2. Multivariate regression model based on LASSO regression analysis results. In Press
Clinical Research
Institutional and Regional Variations in Access to Clinical Trials and Next-Generation Sequencing in Turkis...Med Sci Monit In Press; DOI: 10.12659/MSM.951027
Clinical Research
Low-Intensity Blood Flow-Restricted Multi-Joint Exercise Improves Muscle Function in Patients With Patellof...Med Sci Monit In Press; DOI: 10.12659/MSM.950516
Review article
Musculoskeletal Ultrasound and MRI in the Evaluation of Chemotherapy-Induced Peripheral Neuropathy: A ReviewMed Sci Monit In Press; DOI: 10.12659/MSM.951283
Clinical Research
Sensory Processing, Dissociation, and Affective Symptoms in Misophonia: A Cross-Sectional Study of 35 AdultsMed Sci Monit In Press; DOI: 10.12659/MSM.950938
Most Viewed Current Articles
17 Jan 2024 : Review article 10,187,196
Vaccination Guidelines for Pregnant Women: Addressing COVID-19 and the Omicron VariantDOI :10.12659/MSM.942799
Med Sci Monit 2024; 30:e942799
13 Nov 2021 : Clinical Research 3,708,487
Acceptance of COVID-19 Vaccination and Its Associated Factors Among Cancer Patients Attending the Oncology ...DOI :10.12659/MSM.932788
Med Sci Monit 2021; 27:e932788
14 Dec 2022 : Clinical Research 2,341,643
Prevalence and Variability of Allergen-Specific Immunoglobulin E in Patients with Elevated Tryptase LevelsDOI :10.12659/MSM.937990
Med Sci Monit 2022; 28:e937990
16 May 2023 : Clinical Research 706,524
Electrophysiological Testing for an Auditory Processing Disorder and Reading Performance in 54 School Stude...DOI :10.12659/MSM.940387
Med Sci Monit 2023; 29:e940387