A review of the determination of the early bactericidal activity of various antituberculosis agents

  • P. R. Donald Pediatrie en Kindergesondheid, Universiteit van Stellenbosch
  • F. A. Sirgel Die Nasionale Tuberkulose-navorsingsprogram
  • A. Venter Die Nasionale Tuberkulose-navorsingsprogram
  • P. B. Fourie Die Nasionale Tuberkulose-navorsingsprogram
  • D. P. Parkin Parkin Farmakologie, Universiteit van Stellenbosch en Tygerberg Hospitaal
  • H. I. Seifart Farmakologie, Universiteit van Stellenbosch en Tygerberg Hospitaal
  • B. W. van de Wal Interne Geneeskunde, Universiteit van Stellenbosch en Tygerberg Hospitaal
  • J. S. Maritz Sentrum vir Epidemiologiese Navorsing van die Suid-Afrikaanse Mediese Navorsingsraad


The early bactericidal activity (EBA) of an antituberculosis agent is the daily decline in log10 colony forming units of M tuberculosis per ml of sputum during the first two days of treatment with the agent. It reflects the capacity of an agent to kill the actively metabolising organisms in tuberculosis lung cavities. It offers a relatively cheap means to evaluate the antituberculosis activity of an agent in a small group of patients within a matter of months. This article summarizes the authors’ experience in seven published EBA studies and identifies sources of variation in the procedure. The patients who participated in these studies had a mean age of 33 years, a mean weight of 50 kg and there was extensive or massive involvement of the lungs in 55% of patients. The highest EBA values (0,50-0,66) were found in groups of patients receiving isoniazid and the lowest values (0,05 and 0,09 respectively), in patients receiving  the aminoglycosides amikacin and paromomycin in a dose of 15 mg/kg body weight. The variation in EBA in 248 patients was 0,0312 and the variation ascribable to the process of sputum production and collection was 0,0233. This implies that the different aspects of sputum production and collection involved in obtaining a representative sputum sample are responsible for most of the variation in EBA results. The selection of patients for inclusion in EBA studies and their ability to co-operate in producing a representative sputum specimen are of critical importance in the successful completion of EBA studies.
Original Research