Evaluation of the AID TB resistance line probe assay for rapid detection of genetic alterations associated with drug resistance in Mycobacterium tuberculosis strains.

The rapid accurate detection of drug resistance mutations in Mycobacterium tuberculosis is essential for optimizing the treatment of tuberculosis and limiting the emergence and spread of drug-resistant strains. The TB Resistance line probe assay from Autoimmun Diagnostika GmbH (AID) (Strassburg, Germany) was designed to detect the most prevalent mutations that confer resistance to isoniazid, rifampin, streptomycin, amikacin, capreomycin, fluoroquinolones, and ethambutol. This assay detected resistance mutations in clinical M. tuberculosis isolates from areas with low and high levels of endemicity (Switzerland, n=104; South Africa, n=52) and in selected Mycobacterium bovis BCG 1721 mutant strains (n=5) with 100% accuracy. Subsequently, the line probe assay was shown to be capable of rapid genetic assessment of drug resistance in MGIT broth cultures, the results of which were in 100% agreement with those of DNA sequencing and phenotypic drug susceptibility testing. Finally, the line probe assay was assessed for direct screening of smear-positive clinical specimens. Screening of 98 clinical specimens demonstrated that the test gave interpretable results for >95% of them. Antibiotic resistance mutations detected in the clinical samples were confirmed by DNA sequencing. We conclude that the AID TB Resistance line probe assay is an accurate tool for the rapid detection of resistance mutations in cultured isolates and in smear-positive clinical specimens.

A Phase II study of the sterilising activities of ofloxacin, gatifloxacin and moxifloxacin in pulmonary tuberculosis.

SETTING: Current treatment for pulmonary tuberculosis (TB) might be shortened by the incorporation of fluoroquinolones (FQs). OBJECTIVES: A Phase II study aimed to assess the sterilising activities of three novel regimens containing FQs before a Phase III trial of a 4-month regimen containing gatifloxacin (GFX). DESIGN: A total of 217 newly diagnosed smear-positive patients were randomly allocated to one of four regimens: isoniazid (INH), pyrazinamide and rifampicin (RMP) with either ethambutol, GFX, moxifloxacin (MFX) or ofloxacin (OFX) for 2 months. At the end of the study, RMP and INH were given for 4 months. The rates of elimination of Mycobacterium tuberculosis were compared in the regimens using non-linear mixed effects modelling of the serial sputum colony counts (SSCC) during the first 8 weeks. RESULTS: After adjustment for covariates, MFX substitution appeared superior during the early phase of a bi-exponential fall in colony counts, but significant and similar acceleration of bacillary elimination during the late phase occurred with both GFX and MFX (P = 0.002). Substitution of OFX had no effect. These findings were supported by estimates of time to conversion, using Cox regression, but there were no significant differences in proportions culture-negative at 8 weeks. CONCLUSIONS: GFX and MFX improve the sterilising activity of regimens and might shorten treatment; their progression into Phase III trials therefore seems warranted.

Monitoring the ingestion of anti-tuberculosis drugs by simple non-invasive methods.

This investigation retrospectively assessed inexpensive non-invasive qualitative methods to monitor the ingestion of anti-tuberculosis drugs isoniazid, rifampicin and rifapentine. Results showed that commercial test strips detected the isoniazid metabolites isonicotinic acid and isonicotinylglycine as efficiently as the isonicotinic acid method in 150 urine samples. The presence of rifamycins in urine samples (n=1085) was detected by microbiological assay techniques and the sensitivity compared to the n-butanol extraction colour test in 91 of these specimens. The proportions detected by the two methods were significantly different and the sensitivity of the n-butanol procedure was only 63.8% (95% CL 51.2-76.4%) as compared to that of the superior microbiological method. Final validation (n=691) showed that qualitative assays measure isoniazid and rifamycin ingestion with an efficiency similar to high-performance liquid chromatography. The qualitative procedures may therefore be valuable in clinical trials and in tuberculosis clinics to confirm drug ingestion.

The influence of human N-acetyltransferase genotype on the early bactericidal activity of isoniazid.

The elimination of isoniazid is subject to the influence of the N-acetyltransferase 2 (NAT2) genotype, and individuals may be homozygotic slow, heterozygotic fast, or homozygotic fast acetylators of isoniazid. The early bactericidal activity (EBA) of an antituberculosis agent can be determined by quantitative culture of Mycobacterium tuberculosis in sputum samples obtained from patients with pulmonary tuberculosis during the first days of treatment. In these studies, the EBA of isoniazid during the first 2 days of treatment was determined for 97 patients with sputum smear-positive pulmonary tuberculosis following isoniazid doses of < or =37.5 mg, 75 mg, 150 mg, 300 mg, and 600 mg. The NAT2 genotype was determined in 70 patients, and the association between EBA and genotype was examined in this subgroup. Similarly, the relationship between EBA and isoniazid serum concentration was evaluated in 87 patients. The mean EBA of isoniazid increased with dose, but it levelled off between doses of 150 mg (mean EBA, 0.572) and 300 mg (mean EBA, 0.553). Significant differences were found in the mean EBA of isoniazid between the homozygous slow acetylator group and the heterozygous fast acetylator group and between the homozygous slow acetylator group and the homozygous fast acetylator group, but not between the heterozygous fast acetylator group and the homozygous fast acetylator group. The EBA appeared to reach a maximum at a 2-h isoniazid concentration of 2-3 microg/mL. These data confirm a significant effect of NAT2 genotype on the EBA of isoniazid over a range of doses.

Genome and MIC stability in Mycobacterium tuberculosis and indications for continuation of use of isoniazid in multidrug-resistant tuberculosis.

Mycobacterium tuberculosis strains resistant to two or more of the first line antituberculosis drugs (MDR) are a serious threat to successful tuberculosis control programmes. For this retrospective study, 85 follow-up drug resistant isolates from 23 patients residing in a community with a high incidence of tuberculosis were collected and the level of in-vitro resistance to antibiotics determined quantitatively. PCR-SSCP and sequencing techniques were used to screen for gene mutations associated with resistance in 31 follow-up samples from a smaller group of eight patients. DNA fingerprint analysis was done on sequential isolates to confirm identity. Although treatment had a profound effect on changes in drug resistance patterns, the MIC for a particular agent remained constant in follow-up isolates. DNA fingerprinting and mutational analysis (14 different loci) showed that the genome of MDR strains of M. tuberculosis is relatively stable during the course of therapy. The rpoB gene was the most frequently mutated structural gene involved in drug resistance and a novel C to T mutation upstream of open reading frame (ORF)1 of the inhA operon was detected. No evidence was found of the presence of strain W (New York) in this group of MDR strains. The results stress the importance of confirming individuality of strains for the accurate calculation of frequencies of particular mutations associated with drug resistance, particularly in a high incidence area. Approximately one-half (47.8%) of the patients had isolates resistant to concentrations just above the critical concentration for isoniazid (MICs of 0.2-5 mg/L). Therefore, these patients and their contacts who develop primary drug-resistant tuberculosis may respond to higher dosages of treatment which could have a considerable impact on the cost and the ease of management of resistant tuberculosis.

Primary drug-resistant tuberculosis in children.

SETTING: The Western Cape Province of South Africa, an area with a high tuberculosis (TB) incidence where initial isoniazid (INH) resistance and multidrug resistance (MDR) among adults was 3.9% and 1.1%, respectively, during 1992-1993. OBJECTIVE: To determine the drug resistance incidence among children as compared to adults, to compare the clinical features of drug-resistant and drug-susceptible TB, and the degree of INH resistance in isoniazid-resistant isolates. METHODS: All Mycobacterium tuberculosis cultures obtained from children (0-13 years) at a regional hospital were prospectively collected from August 1994 to April 1998 and susceptibility testing done on each child's specimens. Degree of INH resistance was determined in available resistant isolates. The children's clinical records were reviewed. RESULTS: Susceptibility results were available in 306/338 children with cultures of M. tuberculosis; 21 isolates (6.9%) were INH-resistant, and seven were MDR. Taking into account study limitations, the incidence of INH resistance was 5.6% and MDR 1% in children aged <5 years. Clinical features were similar in children with drug-susceptible and drug-resistant TB. CONCLUSION: The incidence of drug resistance in childhood tuberculosis in Western Cape is low, and probably reflects the level of primary drug resistance amongst organisms currently circulating in the community.

The early bactericidal activity of amikacin in pulmonary tuberculosis.

SETTING: Stellenbosch University, a tertiary care hospital in Cape Town, South Africa. OBJECTIVE: To determine the early bactericidal activity (EBA) of amikacin in dosages of 5 mg/kg, 10 mg/kg and 15 mg/kg body weight in comparison to that of isoniazid 6 mg/kg body weight or no drug. DESIGN: An open, randomised trial. PATIENTS: Patients with previously untreated, sputum smear-positive pulmonary tuberculosis. INTERVENTION: Patients received amikacin 5 mg/kg (12 patients), 10 mg/kg (13 patients) or 15 mg/kg (15 patients), isoniazid 6 mg/kg (9 patients) or no drug (10 patients). RESULTS: The rate of decrease in log viable colony forming units of Mycobacterium tuberculosis per ml of sputum per day during the first 2 days of treatment with amikacin 5 mg/kg, 10 mg/kg and 15 mg/kg was 0.041 (SD 0.100), 0.045 (SD 0.144) and 0.052 (SD 0.096), respectively, 0.515 (SD 0.173) in the patients receiving isoniazid 6 mg/kg, and 0.041 (SD 0.113) in those receiving no drug. The EBA found in patients receiving amikacin did not differ significantly from that of the no drug group. However, as the EBA in the no drug group was the highest ever encountered at Stellenbosch University, the mean in patients receiving drug was tested against 0 and found to differ significantly (P = 0.03), suggesting minimal activity. Mean amikacin serum concentrations 1 hour after intramuscular drug administration were 13.5 microg/ml, 26.7 microg/ml and 39.2 microg/ml in the patients receiving 5 mg, 10 mg and 15 mg per kg body weight, respectively. CONCLUSION: Despite serum concentrations well in excess of the minimal inhibitory concentration of 2-4 microg/ml, the EBA of amikacin in patients with smear-positive pulmonary tuberculosis was only just detectable.

The early bactericidal activity of streptomycin.

SETTING: Patients with sputum smear-positive, newly diagnosed pulmonary tuberculosis studied at Tygerburg Hospital, Cape Town, for their early response to streptomycin (SM). OBJECTIVE: To determine the standard early bactericidal activity (EBA), namely the fall in viable counts of tubercle bacilli in 16-hour sputum collections during the first 2 days of treatment with SM. DESIGN: Patients were randomised to logarithmically spaced daily doses of 7.5, 15 or 30 mg/kg SM. A comparison by standard biological assay methods was then made with previous estimations of the EBA of paromomycin in doses of 7.5 and 15 mg/kg. RESULTS: An EBA of 0.133 obtained with 30 mg/kg SM differed significantly from zero (P = 0.0009), while the EBAs of 0.043 with 15 mg/kg and -0.025 with 7.5 mg/kg did not so differ. A linear regression equation of EBA = -0.2587 + 0.2627 log10 dose was obtained with significant slope (P = 0.007). Paromomycin was estimated to be 1.745 more potent than SM with wide 95% confidence limits (0.6-28.6), indicating that it cannot be considered more potent than SM. CONCLUSIONS: The low EBAs show that SM has low, dose-related, bactericidal activity in cavities, consistent with results from clinical trials. If streptomycin-resistant bacilli are present, paromomycin is probably the aminoglycoside of choice.

Early bactericidal activity of antituberculosis agents.

The early bactericidal activity (EBA) of an antituberculosis agent is arbitrarily defined as the fall in log(10) colony forming units (cfu) of Mycobacterium tuberculosis per ml sputum per day during the first 2 days of treatment. Determining the EBA is an important preliminary step in the clinical evaluation of an antituberculosis agent. We review the results of eight published studies of the EBA of different antituberculosis agents, the impact of these results on our understanding of the actions of the respective agents, the clinical characteristics and sputum findings of patients included in these studies, and explore sources of variation in the EBA results. Patients in these studies had a mean age of 31-36 years, a mean weight of 50-57 kg, 67% were male and 56% had lung involvement covering an area of more than one lung, and 90% had multicavitary disease. None of these findings were related to EBA in any study. The mean log(10) cfu per ml sputum in the first specimen was 6.474. This was related to radiological extent of disease and cavity size in one study (p < 0.001) and, in the case of isoniazid to EBA with a rise in EBA of 0.094 (95% CL 0.029-0.158) for each tenfold rise in cfu counts/ml sputum. The overall variation in EBA in these studies was 0.0303, that due to laboratory processing of specimens was 0.0011, and due to patient characteristics and sputum sampling 0.0212. The EBA is a reproducible investigation that has contributed significantly to our knowledge of the actions and characteristics of both established and new antituberculosis agents. The greatest source of variation in EBA results appears to be that due to interpatient variation in disease characteristics and sputum sampling.

The influence of dose and N-acetyltransferase-2 (NAT2) genotype and phenotype on the pharmacokinetics and pharmacodynamics of isoniazid.

OBJECTIVE: This study evaluated the pharmacokinetics of isoniazid (INH) associated with optimal early bactericidal activity (EBA), defined as 90% of the maximum EBA (EBA(90)) and the influence of N-acetyltransferase-2 (NAT2) subtype on the ability of pulmonary tuberculosis (PTB) patients to reach the identified pharmacokinetic values after INH doses ranging from 0.2 to 10-12 mg/kg body weight. METHODS: INH serum concentrations and NAT2 subtype were determined during four studies of PTB patients in three of whom the EBA of INH was determined. The relationship of EBA to area under the curve (AUC) (AUC(0-infinity)) and 2-h serum concentrations was examined by exponential regression and fitted curves estimated the AUC(0-infinity) and 2-h serum concentrations at which EBA(90) was reached. RESULTS: EBA(90) was reached at an AUC(0-infinity) of 10.52 microg/ml per hour and 2-h serum concentrations of 2.19 microg/ml. An AUC(0-infinity) of 10.52 microg/ml per hour was reached by all 66 patients receiving a 10-12 mg/kg INH dose and all 21 receiving 6 mg/kg, except 1 of 10 (10%) homozygous fast (FF) acetylators; however, at 5 mg/kg, 4 of 12 (33%) FF and 26 of 27 (96%) heterozygous fast (FS), but all 21 homozygous slow (SS) acetylators did so; and 1 of 3 (33%) FF, 2 of 6 (33%) FS, but all 4 SS acetylators at dose 3 mg/kg. An INH 2-h serum concentration of 2.19 microg/ml was reached by all 66 patients receiving 10-12 mg/kg and all 21 receiving 6 mg/kg, except for 2 (20%) FF acetylators at a dose of 5 mg/kg; however, only 3 (25%) of 12 FF acetylators, but 26 (96%) of 27 FS acetylators, and all 21 SS acetylators reached this concentration; and at a dose of 3 mg/kg, 1 (33%) of 3 FF acetylators, 2 (33%) of 6 FF, but all 4 SS acetylators. CONCLUSIONS: At a 6 mg/kg dose, all except a minority of FF NAT2 acetylators, achieve an INH AUC(0-infinity) and 2-h INH serum concentrations associated with EBA(90), as did all 4 SS acetylators receiving 3 mg/kg. Any dose reduction below 6 mg/kg body weight will tend to disadvantage a significant proportion of faster acetylators, but, conversely, SS acetylators require only a 3 mg/kg dose to achieve a satisfactory exposure to INH.

Characteristics and classification of incC group plasmid-dependent phage C-1.

Incompatibility group C plasmid-specific bacteriophage C-1 has a buoyant density of 1.43 g/cm3, a sedimentation coefficient of 80-82S, and a molecular weight of 4 x 10(6). It contains approximately 33% nucleic acid, which has been identified as linear single-stranded RNA of molecular weight (1.3 +/- 0.1) x 10(6). These characteristics, in conjunction with particle morphology and resistance to chloroform and diethyl ether, suggest that phage C-1 belongs to the Leviviridae group of phages.

Genetic recombination in fused spheroplasts of Providence alcalifaciens.

Spheroplasts of Providence alcalifaciens strain P29 auxotrophs were prepared by combined treatment with glycine and lysozyme-EDTA. About 15% of spheroplasts had areas of cytoplasmic membrane exposed where cell wall was absent. The spheroplasts of different auxotrophs were mixed pairwise and fusion was attempted with polyethylene glycol or nascent calcium phosphate. After spheroplasts had regenerated to bacterial forms selection was made for recombinants. Recombinants arose at frequencies of 3.8 X 10(-6) to 1.7 X 10(-7) per spheroplast initially present, by both methods of fusion. The frequency was strongly dependent on the number of chromosomal loci used in selection. The possible order of five loci was determined and this corresponded to that on the closely related Proteus mirabilis chromosome. Control experiments excluded possibilities of auxotrophic reversion, conjugation, transformation, transfection or transduction as explanations of the results. Analysis of prototrophic clones yielded stable prototrophs or mixtures of stable prototrophs and stable recombinants. Parental types were not encountered. Unselected markers segregated among recombinants. It was concluded that the formation of recombinant bacteria was due to spheroplast fusion and that only stable products of the very temporary heteroploid state were haploid recombinants. The low frequency of recombination was ascribed to the limited number of spheroplasts with areas of exposed cytoplasmic membrane.

Properties of a filamentous phage which adsorbs to pili coded by plasmids of the IncI complex.

A filamentous phage, PR64FS, which adsorbs to tips of I pili was isolated. PR64FS is shorter than the I pilus-adsorbing phage If1 and differs from it in plaque morphology. Phages PR64FS and Ifl are serologically related and, like the latter, PR64FS adsorbs to pili coded by all Inc groups of the I plasmid complex.

Phage C-1: an IncC group; plasmid-specific phage.

A phage was isolated from sewage and shown to form plaques on Salmonella typhimurium strains carrying C plasmids. It failed to multiply on strains lacking plasmids of this group. It also plated on Proteus mirabilis and Serratia marcescens strains carrying various IncC plasmids but failed to form plaques on Escherichia coli strains harbouring most of these plasmids, although in all cases, phage multiplication on the strains was demonstrated. No phage increase occurred in any of the strains which lacked an IncC plasmid or contained plasmids of other incompatibility groups. The phage is small, hexagonal in outline, contains RNA, is resistant to chloroform and adsorbs to the shafts of pili coded for by IncC plasmids.

Rifampicin resistance in Mycobacterium tuberculosis--rapid detection and implications in chemotherapy.

OBJECTIVES: Tuberculosis treatment and susceptibility testing are cumbersome, especially in the case of multidrug-resistant (MDR) Mycobacterium tuberculosis. It is known that mutations in the rpoB gene of M. tuberculosis lead to resistance to rifampicin (RMP). In this study, an attempt was made to apply molecular techniques for rapid detection of antibiotic resistance in clinical isolates of M. tuberculosis. DESIGN, SETTINGS AND SUBJECTS: RMP-resistant clinical isolates of M. tuberculosis from South Africa (N = 120) with unique resistant patterns were selected for calculation of resistance frequencies, and 74 MDR isolates of M. tuberculosis from different geographical origins were used for microbiological and molecular analysis. The polymerase chain reaction (PCR) technique was applied for amplification of a previously described region around a cluster of mutations in the rpoB gene, and single-stranded conformational polymorphism (SSCP) analysis was optimised to screen for mutations in the amplified region. RESULTS: The results showed that an optimised PCR-SSCP procedure could detect a cluster of mutations in the rpoB gene (for RMP resistance) in 95% of RMP-resistant isolates. This procedure could therefore be used in the prediction of RMP resistance. Evidence was obtained that these mutations can be screened for directly from BACTEC cultures or even directly from Ziehl-Neelsen-positive sputum samples. Statistical analysis also showed that this locus can be used to predict the presence of an MDR isolate, which may have important implications in decisions concerning chemotherapy. CONCLUSION: It is currently not feasible to test all tuberculosis cases, but application of the PCR-SSCP technology in the prediction of multidrug resistance in M. tuberculosis isolates may be important in patients, especially where frequencies are high for drug-resistant isolates This methodology could reduce the time required for sensitivity testing from approximately 6-12 weeks to a few days.

In vitro investigation of the antimicrobial activities of novel tetramethylpiperidine-substituted phenazines against Mycobacterium tuberculosis.

The intra- and extracellular activities of 5 novel tetramethylpiperidine (TMP)-substituted phenazines against Mycobacterium tuberculosis H37Rv (ATCC 27294) were determined and compared with those of clofazimine and rifampicin. Two of these agents, together with clofazimine, were also tested for their activities against drug-resistant strains of M. tuberculosis. Three of the TMP-substituted phenazine compounds were significantly more active than clofazimine against M. tuberculosis, including multidrug-resistant clinical strains of this microbial pathogen, demonstrating a lack of cross-resistance between the riminophenazines and standard anti-tuberculous drugs. Using M. tuberculosis-infected monocyte-derived macrophages, all of the TMP-substituted phenazines were found to possess intracellular activity which was superior to that of both clofazimine and rifampicin. In this model of intracellular bioactivity, the experimental compounds inhibited bacterial growth at concentrations which were approximately 10-fold lower than the corresponding minimal inhibitory concentration values obtained using conventional in vitro sensitivity testing procedures. These results demonstrate that the novel TMP phenazines are active against multidrug-resistant M. tuberculosis strains, and particularly effective intracellularly.

Phages C-2 and J: IncC and IncJ plasmid-dependent phages, respectively.

Phages C-2 and J were isolated from sewage. Phage C-2 was filamentous and formed plaques on Salmonella typhimurium strains carrying various C plasmids. It also plated on Proteus mirabilis and Serratia marcescens strains carrying particular C plasmids, but failed to form plaques on lines of Escherichia coli K12 strains harbouring most of these plasmids, although in all cases, phage multiplication on the strains was demonstrated. No phage increase occurred in any strain which lacked a C plasmid or contained plasmids of other incompatibility groups. The phage was sensitive to chloroform and, unlike other filamentous bacterial viruses, adsorbed to shafts of conjugative pili. It had a disc-like structure at the end which attached to the pilus. Phage C-2 had a buoyant density of 1 . 30 g cm-3 and a single-stranded circular DNA genome of 3 . 0 MDal. Phage J had an hexagonal head with an inter-apical distance of 40 nm and a short noncontractile tail. It was resistant to chloroform and diethyl ether. The phage formed plaques or propagated on E. coli strains harbouring some IncC plasmids and all IncJ and IncD plasmids tested. The phage did not form plaques but propagated on P. mirabilis and Ser. marcescens strains carrying these plasmids. It did not plate or propagate on S. typhimurium strains harbouring the plasmids. The plaques were very hazy and variable in size. The phage attached sparsely, at a site which appeared to be located at the base of the tail, to sides of conjugative pili.

The early bactericidal activity of ciprofloxacin in patients with pulmonary tuberculosis.

The early bactericidal activity (EBA) of ciprofloxacin (CIP) was measured in 80 patients with previously untreated, smear-positive pulmonary tuberculosis by counting viable bacilli in sputum collections during the first 2 d of treatment. Groups of about 10 patients were treated daily with graded doses of CIP or with 300 mg isoniazid or with no drug. The mean EBA, defined as the fall in log CFU/ ml sputum/d, increased from -0.011 in the no drug group to 0.046, 0.092, 0.121, and 0.205 in the groups receiving 250, 500, 1,000, or 1,500 mg CIP, respectively, a highly significant trend. These results demonstrate the antimycobacterial activity of CIP in high dosage, though the mean EBAs of 0.55 and 0.66 in two groups receiving isoniazid were much higher.

The early bactericidal activity of a low-clearance liposomal amikacin in pulmonary tuberculosis.

The early bactericidal activity (EBA) of a liposomal preparation of amikacin (MiKasome) with a long plasma half-life of 120-200 h was examined in seven patients with newly diagnosed, smear-positive pulmonary tuberculosis. Liposomal amikacin was given in slow iv infusions of 30 mg total amikacin/kg body weight on three successive days. Cfu counts were set up on 16 h sputum collections preceding the first dose and following each dose and were used for calculating the EBA. Despite the high concentrations of total amikacin, >1000 mg/L, obtainable in plasma, no evidence of EBA was obtained. In view of the considerable activity of liposomal amikacin in experimental murine tuberculosis, this finding indicates that liberation of amikacin from the long-life liposomes occurs only in macrophages that are not usually present in the vicinity of the large extracellular clumps of bacilli in the cavity caseum.

The early bactericidal activity of rifabutin in patients with pulmonary tuberculosis measured by sputum viable counts: a new method of drug assessment.

The activity of rifabutin and rifampicin against rapidly growing, extra-cellular Mycobacterium tuberculosis in cavity walls was measured by counting colony-forming units (cfu) in the sputum of 74 patients with newly diagnosed, severe pulmonary tuberculosis during the first 2 days of daily chemotherapy. The fall in counts, (log10 cfu/mL sputum/day), was termed the early bactericidal activity (EBA). The EBA, a highly reproducible measure within groups of 10-13 patients, was -0.015 for a low EBA reference group (who received no chemotherapy) and 0.495 for a high EBA reference group (who received 300 mg isoniazid daily). The EBAs in patients receiving 300 and 600 mg rifabutin were 0.014 and 0.075, and for those taking 150, 300 and 600 mg rifampicin 0.021, 0.150 and 0.204, respectively. Weight-for-weight, the ratio rifabutin to rifampicin producing the same EBA was estimated to be 2.73 (95% confidence limits 1.96-3.78). Determination of the EBA is a rapid and economical method of comparing the potency in human lesions of drugs of the same type before embarking on a conventional clinical trial.