ABSTRACT
INTRODUCTION: The COVID-19 pandemic has led to unforeseen and novel manifestations, as illustrated by the management of drug shortages through the development of hospital production of sterile pharmaceutical preparations (P2S). Visual inspection of P2S is a release control whose methods are described in monographs of the European Pharmacopoeia (2.9.20) and the United States Pharmacopeia (1790). However, these non-automated visual methods require training and proficiency testing of personnel. The main objective of this work was to compare the reliability and speed of analysis of two visual methods and an automated method for detecting visible particles by image analysis in P2S. Furthermore, these methods were used to evaluate sources of particulate contamination during pre-production processes (washing, disinfection, depyrogenation) and production (filling, capping). MATERIALS AND METHODS: Three pharmacy technicians examined 41 clear glass vials of type I, 10 and/or 50 mL through manual visual inspection (MVI), semi-automated (SAVI), and automated (AVI) inspection. The vials were distributed as follows: (i) 16 vials of water for injection containing either glass particles (224 µm or 600 µm), stopper fragments, or textile fibres; (ii) five sterile injectable specialties; (iii) 20 vials of water for injection prepared under different pre-production conditions. RESULTS AND DISCUSSION: MVI and SAVI detected 100% of visible particles compared with 28% for AVI, which showed a deficiency in detecting textile fibres. All three methods correctly analysed P2S that did not contain visible particles. The three methods detected particles in vials maintained under International Organization for Standardization (ISO) 9 pre-production conditions. However, detections by (i) MVI and SAVI, and by (ii) AVI of particles contained in vials maintained under ISO 8 pre-production conditions were deemed satisfactory and unsatisfactory, respectively. CONCLUSION: The importance of visual inspection of P2S requires rapid, sensitive, and reliable detection methods. In this context, MVI and SAVI have proven to be more effective than AVI for a more competitive financial, training, and implementation investment.
