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.

Background:Trastuzumab targets the human epidermal growth factor receptor 2 oncogene and in combination with first-line therapy results in significantly improved survival outcomes and has thus become standard of care in both adjuvant and metastatic settings. While it is estimated that 1% to 4% of patients treated with trastuzumab will develop heart failure and ∼10% will experience a reduction in left ventricular ejection fraction (LVEF), the patient risk factors associated with trastuzumab-induced cardiotoxicity (TIC) are unclear. This meta-analysis aims to consolidate previously published data to identify the risk factors most likely leading to TIC.Methods:A search of the MEDLINE literature database using the keywords trastuzumab/Herceptin, risk factors, outcomes, cardiac, cardiotoxicity, cardiomyopathy, LVEF, and chemotherapy was performed. Only prospective/retrospective human studies were included, with additional studies excluded if they reported baseline LVEF > 68%, a cohort of <50 patients, or results that were not stratified based on cardiotoxic events. Pooled odds ratio (OR) and 95% confidence interval (CI) for each potential risk factor were calculated, with heterogeneity of data and samples explored using random-effects modeling.Results:Data were collected from 17 articles, capturing 6527 patients. Hypertension (OR 1.61, 95% CI 1.14–2.26; P < 0.01), diabetes (OR 1.62; 95% CI 1.10–2.38; P < 0.02), previous anthracycline use (OR 2.14; 95% CI 1.17–3.92; P < 0.02), and older age (P = 0.013) were all shown to be associated with TIC.Conclusion:Cardiac performance should be closely monitored in women treated with trastuzumab. Recognizing potential risk factors along with careful attention to symptoms/LVEF measurements could minimize the occurrence of TIC in this population.

Hantavirus Cardiopulmonary Syndrome is a severe disease caused by exposure to New World hantaviruses. Early diagnosis is difficult due to the lack of specific initial symptoms. Anti-hantavirus antibodies are usually negative until late in the febrile prodrome or the beginning of cardiopulmonary phase while Andes hantavirus (ANDV) RNA genome can be detected before symptoms onset. We analyzed the effectiveness of RTqPCR as a diagnostic tool detecting ANDV-Sout genome in peripheral blood cells from 78 confirmed hantavirus patients and 166 negative controls. Our results indicate that RTqPCR had a low detection limit (~10 copies), with a specificity of 100% and a sensitivity of 94.9%. This suggests the potential for establishing RT-qPCR as the assay of choice for early diagnosis, promoting early effective care of patients and improve other important aspects of ANDV infection management, such as compliance of biosafety recommendations for health personnel in order to avoid nosocomial transmission.

Follow this and additional works at: https://aurora.org/jpcrr Part of the Cardiology Commons, Chemicals and Drugs Commons, and the Oncology Commons Journal of Patient-Centered Research and Reviews ( JPCRR) is a peerreviewed scientific journal whose mission is to communicate clinical and bench research findings, with the goal of improving the quality of human health, the care of the individual patient, and the care of populations.