Skip to main content

Table 2 Selected list of Predictive Models Currently Available in the Literature

From: Utilizing electronic health records to predict acute kidney injury risk and outcomes: workgroup statements from the 15th ADQI Consensus Conference

Study Population and sample Size Variables in Model Outcome
Aronson A et al, 2007 [91] CABG patients
n = 2381
Age, preoperative CHF, prior MI, preexisting renal disease, intraoperative inotropes, intraoperative intra-aortic balloon pump, bypass time, pre-operative pulse pressure. postoperative creatinine ≥2 mg/dL w/ increase ≥0.7 mg/dL from baseline or dialysis
Basu RK et al. 2014 [29] Pediatric critically ill patients
n = 584 total in 4 cohorts
Vasopressor/inotrope use, invasive mechanical ventilation, percent fluid overload, change in creatinine clearance KDIGO Stage 2/3 AKI
Brown JR et al, 2007 [92] CABG patients
n = 8363
Age, female, diabetes, peripheral vascular disease, CHF, hypertension, prior CABG, preoperative IABP, elevated WBC count eGFR <30 ml/min/1.73 m2
Chawla LS et al, 2013 [37]
Koyner JL et al, 2015 [93]
Critically ill patients
n = 77
Urine output measurement Progression to AKIN stage III
Chong E et al, 2012 [94] Patients w/ eGFR <60 ml/min/1.73 m2 undergoing percutaneous coronary intervention n = 770 Age, baseline eGFR, post-percutaneous coronary intervention, creatinine kinase, contrast volume Contrast-induced nephropathy defined as 25 % or 0.5 mg/dL increase from baseline creatinine within 48 h after PCI
Cruz DN et al. 2014 [95] Critically ill patients n = 506 Age, diabetes, cardiovascular disease, chronic kidney disease, hypertension, obesity, hyperbilirubinemia, cerebrovascular accident, AIDS, cancer, hypotension, high-risk surgery, nephrotoxin exposure, sepsis AKI stage II and III defined by AKIN
Demirjian S et al, 2012 [44] Cardiac surgery patients
n = 25,898
gender, race, weight, pulmonary disease, CHF, diabetes, hypertension, type of surgery, previous cardiac surgery, emergency surgery, eGFR, albumin, bicarbonate, sodium, BUN, hemoglobin, platelet count, bilirubin, BMI, potassium, CPB time, intrasurgical transfusion or vasopressor, intrasurgical UOP AKI requiring dialysis
Forni LG et al. 2013 [39] Patients admitted to an acute medical unit n = 3707 Age, alertness scale, chronic kidney disease, congestive cardiac failure, diabetes, liver disease AKI defined per KDIGO guidelines
Gao et al.
2014 [96]
Coronary angiography intervention n = 3945 Age, hypertension, acute MI, heart failure, use of intra-aortic balloon pump, decreased glomerular filtration rate, contrast volume Increase in serum creatinine level
Grimm JC et al 2015 [97] Lung transplant
n = 10,693
Race, sarcoidosis, diabetes, weight, baseline renal function, Kanofsky performance score, previous ICU stay, ECMO, days on list, double transplant AKI requiring dialysis
Gurm HS et al. 2013 [98] Patients undergoing a percutaneous coronary intervention n = 68,753 age, weight, height, percutaneous coronary intervention status and indication, coronary artery disease presentation, cardiogenic shock, heart failure, ejection fraction, diabetes, CKMB, creatinine, hemoglobin, troponin I and T ≥0.5 mg/dl increase in serum creatinine level from baseline, RRT receipt
Ho J et al, 2012 [99] Cardiac surgery patients n = 350 Bypass time, baseline eGFR, euroSCORE, postoperative serum creatinine AKI by AKIN criteria
Hong SH et al. 2012 [100] Living donor transplant recipients n = 429 age, MELD, hypertension, platelet count, surgical time, packed red blood cell transfusion, lactate, furosemide dose, calcium chloride dose, phosphate level Renal failure was defined according to RIFLE
Kane-Gill et al. 2015 [38] Elderly, critically ill
n = 25,230
age, gender, race, eGFR, heart failure, diabetes, hypertension, admission type (medical vs surgical), requirement for vasopressors or mechanical ventilation, sepsis, hypotension, nephrotoxic drugs. AKI by KDIGO criteria
Kim JM et al. 2014 [46] Liver transplant recipients
n = 153
Hepatic encephalopathy, deceased donor liver donations, MELD score,
intraoperative blood loss, and indication for liver transplantation
Patients who needed renal replacement therapy
Kim MY et al, 2011 [101] Isolated off-pump CABG patients
n = 448
High systolic blood pressure, low baseline eGFR, coronary angiography less than 7 days prior to surgery AKI by AKIN criteria
Kim WH et al. 2013 [102] Aortic surgery with cardiopulmonary bypass n = 737 Age, preoperative glomerular filtration, ejection fraction, operation time, intraoperative urine output, intraoperative furosemide use AKI defined by RIFLE
Kristovic D et al. 2015 [26] Cardiac surgery patients
n = 1056
age, atrial fibrillation, CHF classification, previous cardiac surgery, creatinine, endocarditis, weight, gender, COPD, bypass AKI stage by KDIGO
AKI requiring dialysis
Legrand M et al. 2013 [103] Patients with endocarditis/ cardiac surgery with cardiopulmonary bypass n = 202 Age, gender, pre-existing comorbidities, presence of shock, systemic emboli, NYHA classification, hemoglobin, baseline creatinine, need for mechanical ventilation, characteristics of infection/surgery, use of nephrotoxic agents Development or progression of AKI in the 7 days following surgery. AKI defined per AKIN
McMahon GM et al. 2013 [104] Rhabdomyolysis within 3 days of admission n = 2371 Age, female sex, cause of rhabdomyolysis, initial creatinine, creatinine phosphokinase, phosphate, calcium, and bicarbonate Composite endpoint: Renal replacement therapy or mortality
Medha et al. 2013 [41] Trauma patients
n = 4396
hepatic dysfunction, urea, glucose, pulmonary dysfunction, severity of injury serum creatinine level >2.0 mg/dL during the hospital stay
Meersch M et al. 2014 [42] Patients undergoing cardiac surgery with bypass n = 50 Diabetes, severity of illness, ejection fraction, baseline serum creatinine, cross-clamp time, chronic obstructive pulmonary disease AKI defined by RIFLE or AKIN together
Mehran R et al, 2004 [22] Patients undergoing percutaneous coronary interventions n = 8357 Hypotension, intra-aortic balloon pump, congestive heart failure, age >75 years, anemia, diabetes, contrast volume, baseline creatinine or eGFR Increase of ≥25 % or ≥0.5 mg/dL in pre-PCI serum creatinine at 48 h after PCI
Mehta RH et al, 2006 [32] CABG and/or valve surgery patients n = 449,524 Preoperative creatinine, age, race, type of surgery, diabetes, shock, NYHA class, lung disease, recent myocardial infarction, prior cardiovascular surgery AKI requiring dialysis
Ng SY et al. 2014 [105] Cardiac surgery patients n = 28,422 obesity, infective endocarditis, cardiac procedure, preop creatinine, diabetes, urgency status, eGFR, CHF, age, cardiogenic shock, IABP use, bypass time, non-RBC blood product use, gender, reoperation for bleeding, hypercholesterolemia, hypertension, and respiratory disease Increased creatinine > 200 mmol/L (2.26 mg/dL), ≥ 2x increase in creatinine over baseline, a new receipt for RRT
Palomba H et al. 2007 [18] Cardiac surgery patients
n = 603
age, serum creatinine, glucose, heart failure, combined surgeries, cardiopulmonary bypass time, cardiac output, central venous pressure creatinine > 2.0 mg/dl or increase of 50 % above
baseline
Park MH et al. 2015 [106] Living-donor liver transplant
n = 538
weight; diabetes, alcoholic liver disease, albumin <3.5 mg/dL, model for end-stage liver disease score, child-turcotte-pugh- estimated graft to recipient body weight ratio, operation details, calcineurin inhibitor use without mycophenolate AKI as defined by RIFLE
Rahmanian PB et al, 2011 [107] Cardiac surgery patients
n = 2511
Pulmonary hypertension, preoperative renal dysfunction, bypass time, peripheral vascular disease, recent MI, atrial fibrillation, age, CHF, diabetes AKI requiring dialysis
Rodriguez et al. 2013 [108] Severe Rhabdomyolysis n = 126 Albumin, metabolic acidosis, prothrombin time, peak creatinine phosphokinase RIFLE category
Romano TG et al. 2013 [109] Orthotopic liver transplant patients n = 114 MELD increase ≥ 0.3 mg/dL in serum creatinine
Schneider DF et al, 2012 [110] Critically ill burn patients
n = 309
age, sex, race, % body surface area burned, burn mechanism, intubation, inhalation injury, NROF, fraction of predicted Parkland resuscitation, early transfusion, weight, Charlson Score, drug abuse, smoker, number of preadmission medications, ACEI/ARB, diuretic, NSAIDs, methamphetamine, lowest hematocrit, potassium, sodium, pH, glucose, base deficit, lowest mean arterial pressure, temperature AKI defined using RIFLE classification
Simonini M et al. 2014 [111] Elective cardiac surgery n = 802 Age, gender, ejection fraction, hypertension, diabetes, renal function, reoperation cardiac surgery, surgery type AKIN stage II/III AKI
Slankamenac K et al. 2013 [112] Liver surgery
n = 549
Need for blood transfusion, cirrhosis, oliguria, hepaticojejunostomy, use of colloids, use of diuretics, use of a bolus of catecholamines R of RIFLE
Soto K et al. 2013 [40] Patients admitted from the emergency department n = 616 Age, kidney susceptibility stage, chronic heart failure, hypertension, cardiovascular disease, and diabetes mellitus New onset AKI per RIFLE
Thakar CV et al, 2005 [17] Cardiac surgery patients
n = 31,677
gender, CHF, ejection fraction, preop intra-aortic balloon pump, COPD, diabetes, previous cardiac surgery, emergency surgery, type of surgery, creatinine >1.2 AKI requiring dialysis
Tsai TT et al. 2014 [113] Percutaneous coronary intervention n = 947,012 Age, CKD, prior cardiovascular disease, acute coronary syndrome, cardiac arrest, anemia, CHF, intra-aortic balloon pump prior to procedure, cardiogenic shock AKI defined by AKIN and AKI requiring dialysis
Wang M et al. 2013 [114] Patients with hemorrhagic fever (Hantann virus) n = 112 age, gender, presence of shock, proteinuria, hematuria, platelet count, leukocyte Required dialysis or increased
serum creatinine ≥354 mmol/L
Wang Y et al. 2013 [115] Patients hospitalized with acute heart failure n = 1709 Age, ≥ 3 previous hospital admissions for acute heart failure, systolic blood pressure <90 mmHg, serum sodium <130 mmol/L, heart functional class IV, proteinuria, SCr ≥ 104 mmol/L, intravenous furosemide dose ≥ 80 mg/day increase in serum creatinine (SCr) of
≥26.4 mmol/L or ≥ 50 % within 48 h.
Wijeysundera DN et al, 2007 [19] Cardiac surgery patients
n = 20,131
Preoperative eGFR, diabetes, ejection fraction, previous cardiac surgery, procedures other than isolated CABG or ASD repair, non-elective procedure, preoperative intra-aortic balloon pump AKI requiring dialysis
Wong B et al. 2015 [116] Cardiac surgery patients who developed AKI n = 2316 Age, weight, preoperative creatinine, gender, preoperative intra-aortic balloon pump, ejection fraction, type of surgery, previous cardiac surgery, diabetes, COPD, cardiopulmonary bypass time, clamp time, pump time, number of bypass grafts AKI Stage 1, stage 2, stage 3
Xu X et al, 2010 [117] Liver transplant recipients
n = 102
age, MELD score, preoperative creatinine, BUN, sodium, and potassium, intraoperative UOP, intraoperative hypotension, intraoperative noradrenaline Serum creatinine >1.5 mg/dl with an increase of 50 % above baseline and/or RRT