Quantification of captopril disulphide as a degradation product in captopril tablets using near infrared spectroscopy and chemometrics

Author(s)

Souza, Julia Aparecida L. ^[1] ; Albuquerque, Miracy Muniz ^[1] ; Grangeiro, Jr., Severino ^[1] ; Pimentel, Maria Fernanda ^[2] ; de Santana, Davi Pereira ^[1] ; Simoes, Simone S. ^{[2, 3]}

Affiliation

^[1] Univ Fed Pernambuco, Dept Ciencias Farmaceut, BR-50740521 Recife, PE - Brazil

^[2] Univ Fed Pernambuco, Dept Engn Quim, BR-50740521 Recife, PE - Brazil

^[3] Lab Farmaceut Estado Pernambuco, Recife, PE - Brazil

• Document type: Journal article
• Source: VIBRATIONAL SPECTROSCOPY; v. 62, p. 35-41, SEP 2012.
• Web of Science Citations: 11

Abstract

Captopril disulphide (CD) is the major impurity upon captopril degradation. Besides a reduction of the active principle by degradation, the presence of a high amount of CD in a captopril tablet gives a metallic taste to the tablet, which reduces the therapeutic adhesion. Thus, it is important to quantify this impurity in captopril tablets. This work proposes a new methodology to determine this degradation product through near infrared spectroscopy (NIR) and chemometrics. To conduct this study, tablets of a recently manufactured batch were subjected to accelerated degradation using an environmental chamber to enlarge the range of CD in tablets. Tablets with their batch dates expired were also used. Near infrared (NIR) diffuse reflectance spectra of these tablets were recorded in the region between 14,000 and 3800 cm(-1). The same tablets were also analysed using high performance liquid chromatography (HPLC). Partial least squares (PLS) models were constructed using different pre-processing techniques. The performances of the models were evaluated using an external validation set. In this step, a root mean square error of prediction (RMSEP) of 0.074 mg per tablet was obtained, using Savitzky-Golay 1st derivative spectra (15 points windows and 2nd order polynomial). The value of relative standard error (RSD) under repeat conditions was 12%, which is within the limits of the RSD required to determine traces or impurities (20%). The methodology was validated and can easily be used for routine analysis. (C) 2012 Elsevier B.V. All rights reserved. (AU)

• FAPESP's process: 08/57808-1 - National Institute of Advanced Analytical Science and Technology
• Grantee: Celio Pasquini
• Support Opportunities: Research Projects - Thematic Grants