DocumentCode :
2002088
Title :
Identifying diseases assciated with a high risk for acute kidney injury using a hospital information system database
Author :
Otomo, K. ; Ishibashi, Takayuki ; Kataoka, Haruno ; Hatakeyama, Yutaka ; Okuhara, Yoshiyasu
Author_Institution :
Center for Innovative & Translational Med., Kochi Univ., Nankoku, Japan
fYear :
2012
fDate :
20-24 Nov. 2012
Firstpage :
560
Lastpage :
563
Abstract :
Acute kidney injury (AKI) occurs when a patient cannot maintain fluid homeostasis because of acute deterioration in renal function. Several studies have indicated that patients with AKI have higher risks for developing chronic kidney disease and mortality compared with patients without AKI. Although AKI is a serious disorder, few studies have identified the diseases that cause AKI. The purpose of this study was to identify the diseases associated with AKI. METHODS The data of 68,504 hospitalized patients at Kochi Medical School Hospital from 1981 to 2010 were analyzed. All laboratory test results were automatically processed and saved in a hospital information system database. Episodes of AKI were identified by the serum creatinine level as defined by Acute Kidney Injury Network (AKIN) criteria. All diseases that were diagnosed within 30 days prior to the development of AKI were collected from the hospital information system database. Compared with patients who were not affected with AKI, the odds ratios were calculated, and a multivariate analysis was conducted. RESULTS AND DISCUSSION The highest odds ratio was AKI (odds ratio 46.44, 95% confidence interval 36.88-58.49) This means that our method has enough usability and validity to identify the diseases associated with a high risk for AKI. In addition, cardiovascular diseases (e.g., shock syndrome), respiratory diseases (e.g., pneumonia), and infection diseases (e.g., septicemia) showed high odds ratio for AKI as expected (odds ratio 4.44, 95% confidence interval 4.07-4.84; odds ratio 2.38, 95% confidence interval 2.23-2.56; odds ratio 4.54, 95% confidence interval 4.17-4.93 respectively). Furthermore, hyperuricemia also had a significant odds ratio (odds ratio 1.79, 95% confidence interval 1.62-1.98). Few studies have shown a relationship between AKI and hyperuricemia. This study identified not only diseases expected to be risks for AKI, but also diseases that have never been regarded as risks for AKI.
Keywords :
cardiovascular system; data mining; diseases; kidney; medical disorders; medical information systems; AKI patients; AKIN criteria; Kochi Medical School Hospital; acute kidney injury network criteria; cardiovascular diseases; chronic kidney disease; chronic kidney mortality; disease identification; fluid homeostasis; hospital information system database; hospitalized patient data; hyperuricemia; infection diseases; multivariate analysis; pneumonia; renal function deterioration; respiratory diseases; septicemia; serum creatinine level; shock syndrome; acute kidney injury; hospital information system database; medical data mining; risk disease;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Soft Computing and Intelligent Systems (SCIS) and 13th International Symposium on Advanced Intelligent Systems (ISIS), 2012 Joint 6th International Conference on
Conference_Location :
Kobe
Print_ISBN :
978-1-4673-2742-8
Type :
conf
DOI :
10.1109/SCIS-ISIS.2012.6505062
Filename :
6505062
Link To Document :
بازگشت