Validation of the shake test for detecting freeze damage to adsorbed vaccines.

OBJECTIVE To determine the validity of the shake test for detecting freeze damage in aluminium-based, adsorbed, freeze-sensitive vaccines. METHODS A double-blind crossover design was used to compare the performance of the shake test conducted by trained health-care workers (HCWs) with that of phase contrast microscopy as a "gold standard". A total of 475 vials of 8 different types of World Health Organization prequalified freeze-sensitive vaccines from 10 different manufacturers were used. Vaccines were kept at 5 degrees C. Selected numbers of vials from each type were then exposed to -25 degrees C and -2 degrees C for 24-hour periods. FINDINGS There was complete concordance between HCWs and phase-contrast microscopy in identifying freeze-damaged vials and non-frozen samples. Non-frozen samples showed a fine-grain structure under phase contrast microscopy, but freeze-damaged samples showed large conglomerates of massed precipitates with amorphous, crystalline, solid and needle-like structures. Particles in the non-frozen samples measured from 1 microm (vaccines against diphtheria-tetanus-pertussis; Haemophilus influenzae type b; hepatitis B; diphtheria-tetanus-pertussis-hepatitis B) to 20 microm (diphtheria and tetanus vaccines, alone or in combination). By contrast, aggregates in the freeze-damaged samples measured up to 700 microm (diphtheria-tetanus-pertussis) and 350 microm on average. CONCLUSION The shake test had 100% sensitivity, 100% specificity and 100% positive predictive value in this study, which confirms its validity for detecting freeze damage to aluminium-based freeze-sensitive vaccines.