Lawrence Garrod had a life-long interest in the way that antibiotics worked and was one of the first to examine experimentally various aspects of their antibacterial action and the behaviour of the bacteria against which they were aimed within the context of the type of lesion and the immune response in the patient. As far as I am aware, he only once turned his vigorous and pertinent attention to the treatment of tuberculosis and that was when streptomycin was introduced in the early 1950s. It is, I hope, appropriate that I should attempt to fill this gap by reviewing aspects of the chemotherapy of tuberculosis which are currently controversial or for which new experimental evidence has accumulated recently.

The first phase in the development of the chemotherapy of tuberculosis started in the late 1940s with the introduction of several potent antibacterial drugs, streptomycin first and then p-aminosalicylic acid, isoniazid, ethionamide and pyrazinamide, as well as several minor drugs. At this stage, the main emphasis was on the use of double-drug combinations, supplemented later by an initial triple-drug intensive phase, to prevent the emergence of drug-resistant tubercle bacilli. With the demonstration in the late 1930s that chemotherapy given under adverse domiciliary conditions was just as effective as when given under excellent conditions in sanatoria (Tuberculosis Chemotherapy Centre, Madras, 1959), our attention turned to methods for improving compliance during periods of therapy that lasted for at least one year and often for as long as two years. The first of these methods was the use of intermittent regimens which facilitated fully supervised drug administration, as well as reducing drug costs and toxicity. The second method was the introduction in the early 1970s of short-course regimens which allowed the treatment period to be reduced to six months (East African/British Medical Research Councils, 1974). Short-course regimens, using the potent sterilizing drugs rifampicin and pyrazinamide, are now universal in technically advanced countries and are being introduced gradually in many developing countries. Theory now changed from its earlier concentration on drug-resistance to consideration of the sterilizing activity of drugs, since this determines the duration of treatment. This is still an aspect that is often misunderstood.