Comparison of a Neural Network Approach with Five Traditional Methods for Predicting Creatinine Clearance in Patients with Human Immunodeficiency Virus Infection

• Published in: Pharmacotherapy 1999-06, Vol.19 (6), p.734-740
• Main Authors: Herman, Ronald A., Noormohamed, Saleem, Hirankarn, Sarapee, Shelton, Mark J., Huang, Eric, Morse, Gene D., Hewitt, Ross G., Stapleton, Jack T.
• Format: Article
• Language: English
• Subjects: Adult; AIDS/HIV; Biological and medical sciences; Creatinine - urine; Female; General pharmacology; HIV Infections - urine; Human immunodeficiency virus 1; Human viral diseases; Humans; Infectious diseases; Iowa; Male; Medical sciences; Middle Aged; Neural Networks (Computer); Pharmacokinetics. Pharmacogenetics. Drug-receptor interactions; Pharmacology. Drug treatments; Predictive Value of Tests; Prospective Studies; Viral diseases; Viral diseases of the lymphoid tissue and the blood. Aids
• ISSN: 0277-0008; 1875-9114
• DOI: 10.1592/phco.19.9.734.31545

Study Objective. To compare the results of an artificial neural network approach with those of five published creatinine clearance (Clcr) prediction equations and with the measured (true) Clcr in patients infected with the human immunodeficiency virus (HIV). Design. Six‐month prospective study. Settings. Two university medical centers. Patients. Sixty‐five HIV‐infected patients: 18 relatively healthy outpatients and 47 inpatients. Interventions. All subjects had urine collected for 24 hours to determine Clcr. Measurements and Main Results. The 16 input variables were age, ideal body weight, actual body weight, body surface area, height, and the following blood chemistries: sodium, potassium, aspartate aminotransferase, alanine aminotransferase, red blood cell count, platelet count, white blood cell count, glucose, serum creatinine, blood urea nitrogen, and albumin. The only output variable was Clcr. A training set of 55 subjects was used to develop the relationship between input variables and the output variable. The trained neural network was then used to predict Clcr of a validation set of 10 subjects. Mean differences between predicted Clcr and actual Clcr (bias) were 4.1, 28.7, 29.4, 26.0, 31.8, and 55.8 ml/min/1.73 m2 for the artificial neural network, Cockcroft and Gault, Jelliffe 1, Jelliffe 2, Mawer et al, and Hull et al methods, respectively. Conclusion. The accuracy of predicting Clcr in subjects with HIV infection by the artificial neural network is superior to that of the five equations that are currently used in clinical settings.