BackgroundAcute Kidney Injury (AKI) occurs in at least 5 % of hospitalized patients and can result in 40–70 % morbidity and mortality. Even following recovery, many subjects may experience progressive deterioration of renal function. The heterogeneous etiology and pathophysiology of AKI complicates its diagnosis and medical management and can add to poor patient outcomes and incur substantial hospital costs. AKI is predictable and may be avoidable if early risk factors are identified and utilized in the clinical setting. Timely detection of undiagnosed AKI in hospitalized patients can also lead to better disease management.MethodsData from 25,521 hospital stays in one calendar year of patients 60 years and older was collected from a large health care system. Four machine learning models (logistic regression, support vector machines, decision trees and naïve Bayes) along with their ensemble were tested for AKI prediction and detection tasks. Patient demographics, laboratory tests, medications and comorbid conditions were used as the predictor variables. The models were compared using the area under ROC curve (AUC) evaluation metric.ResultsLogistic regression performed the best for AKI detection (AUC 0.743) and was a close second to the ensemble for AKI prediction (AUC ensemble: 0.664, AUC logistic regression: 0.660). History of prior AKI, use of combination drugs such as ACE inhibitors, NSAIDS and diuretics, and presence of comorbid conditions such as respiratory failure were found significant for both AKI detection and risk prediction.ConclusionsThe machine learning models performed fairly well on both predicting AKI and detecting undiagnosed AKI. To the best of our knowledge, this is the first study examining the difference between prediction and detection of AKI. The distinction has clinical relevance, and can help providers either identify at risk subjects and implement preventative strategies or manage their treatment depending on whether AKI is predicted or detected.

Generation of excessive reactive oxygen species (ROS) leads to mitochondrial dysfunction, apoptosis and necrosis in renal ischemia-reperfusion (IR) injury. Previously we showed that lentiviral vector mediated over-expression of superoxide dismutase-1 (SOD1) in proximal tubular epithelial cells (LLC-PK 1 ) reduced cytotoxicity in an in vitro model of IR injury. Here, we examined the effects of SOD1 over-expression on mitochondrial signaling following ATP depletion-recovery (ATP-DR). To examine the role of mitochondrial ROS, a subset of cells were treated with the mitochondrial antioxidant, MitoTEMPO. ATP-DR-mediated increases in mitochondrial calcium ([Ca 2+ ] m ), loss of mitochondrial membrane potential (ΔΨ m ) and increase in mitochondrial permeability transition pore (MPTP) were attenuated by SOD1 and MitoTEMPO (P < 0.01). SOD1 prevented ATP-DR induced mitochondrial Bax translocation, although the release of pro-apoptotic proteins from mitochondria was not prevented by SOD1 alone and required the presence of both SOD1 and MitoTEMPO. SOD1 suppressed the increase in c-jun phosphorylation suggesting that JNK signaling regulates Bax translocation to mitochondria via ROS. ATP-DRmediated changes in MPTP and mitochondrial signaling increased necrosis and apoptosis, both of which were partially attenuated by SOD1 and MitoTEMPO. These studies show that SOD1 and MitoTEMPO preserve mitochondrial integrity and attenuate ATP-DR mediated necrosis and apoptosis.

Cardioprotection conferred by hyperoxia+hyperbaria is directly dependent on oxygen availability and mediated by NOS.

, a cell-permeant SOD mimetic, reduces oxidative stress and apoptosis following renal ischemia-reperfusion. Am J Physiol Renal Physiol 296: F266 -F276, 2009. First published December 17, 2008 doi:10.1152/ajprenal.90533.2008.-Oxidative stress and apoptosis are important factors in the etiology of renal ischemia-reperfusion (I/R) injury. The present study tested the hypothesis that the cell-permeant SOD mimetic manganese(III) tetrakis(1-methyl-4-pyridyl)porphyrin (MnTMPyP) protects the kidney from I/R-mediated oxidative stress and apoptosis in vivo. Male Sprague-Dawley rats (175-220 g) underwent renal I/R by bilateral clamping of the renal arteries for 45 min followed by reperfusion for 24 h. To examine the role of reactive oxygen species (ROS) in renal I/R injury, a subset of animals were treated with either saline vehicle (I/R Veh) or MnTMPyP (I/R Mn) (5 mg/kg ip) 30 min before and 6 h after surgery. MnTMPyP significantly attenuated the I/R-mediated increase in serum creatinine levels and decreased tubular epithelial cell damage following I/R. MnTMPyP also decreased TNF-␣ levels, gp 91phox , and lipid peroxidation after I/R. Furthermore, MnTMPyP inhibited the I/R-mediated increase in apoptosis and caspase-3 activation. Interestingly, although MnTMPyP did not increase expression of the antiapoptotic protein Bcl-2, it decreased the expression of the proapoptotic genes Bax and FasL. These results suggest that MnTMPyP is effective in reducing apoptosis associated with renal I/R injury and that multiple signaling mechanisms are involved in ROS-mediated cell death following renal I/R injury. acute ischemia-reperfusion; manganese(III) tetrakis(1-methyl-4-pyridyl)porphyrin; reactive oxygen species

20-HETE, a metabolite of arachidonic acid, has been implicated as a mediator of free radical formation and tissue death following ischemia-reperfusion (IR) injury in the brain and heart. The present study examined the role of this pathway in a simulated IR renal injury model in vitro. Modified self-inactivating lentiviral vectors were generated to stably overexpress murine Cyp4a12 following transduction into LLC-PK(1) cells (LLC-Cyp4a12). We compared the survival of control and transduced LLC-PK(1) cells following 4 h of ATP depletion and 2 h of recovery in serum-free medium. ATP depletion-recovery of LLC-Cyp4a12 cells resulted in a significantly higher LDH release (P < 0.05) compared with LLC-enhanced green fluorescent protein (EGFP) cells. Treatment with the SOD mimetic MnTMPyP (100 microM) resulted in decreased cytotoxicity in LLC-Cyp4a12 cells. The selective 20-HETE inhibitor HET-0016 (10 microM) also inhibited cytotoxicity significantly (P < 0.05) in LLC-Cyp4a12 cells. Dihydroethidium fluorescence showed that superoxide levels were increased to the same degree in LLC-EGFP and LLC-Cyp4a12 cells after ATP depletion-recovery compared with control cells and that this increase was inhibited by MnTMPyP. There was a significant increase (P < 0.05) of caspase-3 cleavage, an effector protease of the apoptotic pathway, in the LLC-Cyp4a12 vs. LLC-EGFP cells (P < 0.05). This was abolished in the presence of HET-0016 (P < 0.05) or MnTMPyP (P < 0.01). These results demonstrate that 20-HETE overexpression can significantly exacerbate the cellular damage that is associated with renal IR injury and that the programmed cell death is mediated by activation of caspase-3 and is partially dependent on enhanced CYP4A generation of free radicals.