Search

A quantitative analysis method based on near infrared (NIR) spectroscopy was developed for determining the azithromycin content and its degradation product impurity (decladinosylazithromycin) in azithromycin for injection. Two quantitative models were obtained by evaluating 18 laboratory samples and 44 commercial samples formulated as hydrochlorate, phosphate, lactate and citrate salts. The samples had an azithromcyin content ranging from 74.2% to 91.0% (w/w) and decladinosylazithromycin ranging from 0.1% to 7.7% (w/w). The root mean square error of cross-validation for the azithromycin model and decladinosylazithromycin model were 1.54% and 0.49%, respectively. The estimated limit of detection and limit of quantification for the decladinosylazithromycin calibration were about 1.2% and 4.1% (w/w), respectively. In addition, the quantitative models were validated in terms of specificity, linearity, accuracy, repeatability and intermediate precision, according to the International Conference on Harmonisation guidelines, by evaluating the characteristics of the NIR spectra. The results showed that the models were satisfactory. Our results demonstrated that a universal quantitative model based on NIR diffuse reflectance spectroscopy could be used as a quick and reliable method for quality control of low-dose impurities in pharmaceutical drugs.

A quantitative analysis method based on near infrared (NIR) spectroscopy was developed for determining the azithromycin content and its degradation product impurity (decladinosylazithromycin) in azithromycin for injection. Two quantitative models were obtained by evaluating 18 laboratory samples and 44 commercial samples formulated as hydrochlorate, phosphate, lactate and citrate salts. The samples had an azithromcyin content ranging from 74.2% to 91.0% (w/w) and decladinosylazithromycin ranging from 0.1% to 7.7% (w/w). The root mean square error of cross-validation for the azithromycin model and decladinosylazithromycin model were 1.54% and 0.49%, respectively. The estimated limit of detection and limit of quantification for the decladinosylazithromycin calibration were about 1.2% and 4.1% (w/w), respectively. In addition, the quantitative models were validated in terms of specificity, linearity, accuracy, repeatability and intermediate precision, according to the International Conference on Harmonisation guidelines, by evaluating the characteristics of the NIR spectra. The results showed that the models were satisfactory. Our results demonstrated that a universal quantitative model based on NIR diffuse reflectance spectroscopy could be used as a quick and reliable method for quality control of low-dose impurities in pharmaceutical drugs.