Published ahead of print on January 6, 2003, doi:10.1164/rccm.200210-1125OC
American Journal of Respiratory and Critical Care Medicine Vol 167. pp. 1348-1354, (2003)
© 2003 American Thoracic Society
Bactericidal and Sterilizing Activities of Antituberculosis Drugs during the First 14 Days
Amina Jindani,
Caroline J. Doré and
Denis A. Mitchison
Clinical Trials Program, International Union Against Tuberculosis and Lung Disease, Paris, France; Medical Research Council Clinical Trials Unit; and Department of Medical Microbiology, St George's Hospital Medical School, London, United Kingdom
Correspondence and requests for reprints should be addressed to Denis A. Mitchison, M.D., Department of Medical Microbiology, St George's Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK. E-mail: dmitchis{at}sghms.ac.uk
 |
ABSTRACT
|
|---|
Colony-forming units of Mycobacterium tuberculosis in sputum were counted at 2-day intervals in 100 patients treated with 22 regimens of isoniazid, rifampin, pyrazinamide, ethambutol, or streptomycin, given alone or in combinations. The exponential fall in colony-forming units was measured by linear regression coefficients of the log counts during the initial 2-day phase of rapid, drug-determined killing and during the subsequent 12 days of much slower sterilizing activity. The regression coefficients during the first 2 days varied significantly according to the drug; the greatest effects in multiple regression analyses were due to isoniazid (p < 0.001) and rifampin (p = 0.027). The rapid kill obtained with isoniazid was unaffected by addition of other drugs, so that a change in activity after adding an unknown drug to isoniazid would not be measurable. In multiple regression analysis of the coefficients during Days 2–14, rifampin and streptomycin had significant effects (p = 0.007 and 0.006, respectively), indicating that both drugs had important sterilizing activity, streptomycin particularly early. Isoniazid and pyrazinamide had no significant effects. In analyses of combined drug regimens only, ethambutol had an effect (p = 0.01) in reverse direction to that of rifampin, suggesting it antagonized the sterilizing activity of other drugs.
Key Words: bactericidal • early bactericidal activity • extended early bactericidal activity • pulmonary tuberculosis • sterilizing
The first study of early bactericidal activity was performed from 1974 onward in Nairobi on a total of 124 patients (1). Counts of viable tubercle bacilli in overnight sputum collections were done pretreatment and at intervals of 2 days over a 14-day period, during which time small groups of patients were treated with 22 different combinations of antituberculosis drugs. The study showed that the fall in colony-forming unit counts per milliliter of sputum per day over the first 2 days differed from drug to drug, and between different dose sizes of the same drug given in monotherapy. On the other hand, all of the drug regimens appeared to have similar bactericidal activities during the succeeding 12 days of treatment. This finding led to the fall in counts during the first 2 days being designated as the standard early bactericidal activity (EBA) of a drug given in monotherapy. Most of the antituberculosis drugs have now been examined by this technique and can be broadly classified for potency in terms of the standard EBA (1–6). Titrations of the standard EBA against dose size can estimate the therapeutic margin between the usual dose size and the dose size that just gives an EBA (6), and the activity of drugs can be compared by standard biological assay methods (3). As a result of these studies, determination of the standard EBA is a rapid, accurate, and inexpensive way of assessing drug activity that is deemed necessary by the Food and Drug Administration for licensing new antituberculosis drugs.
A major drawback of the EBA technique is that it does not measure the sterilizing activity of a drug against bacilli that persist despite effective drug treatment and are responsible for prolonging the treatment period. The possibility of detecting the sterilizing activity of rifampin by prolonging the study period beyond Day 2 (termed extended EBA) has been raised (7, 8). These suggestive results need confirmation. Because, for ethical reasons, it is increasingly difficult to conduct EBA studies with monotherapy for more than 5–7 days, there is also a need to study drugs given in combinations as well. As the original Nairobi data are unique in terms of the numbers of patients and regimens examined and also in the 14-day duration of treatment, we therefore reexamined the data to determine (1) whether the effect of adding another drug to isoniazid, in a combination regimen, could be detected on the basis of EBA values over the first 2 days, (2) whether we could detect the sterilizing activity of rifampin by regression analysis over the remaining 12-day period, and (3) whether we could obtain information about the sterilizing activity of the five main drugs by multiple regression methods and about (4) the benefit, in terms of accuracy, obtained by calculation of regression coefficients from 2 days onward, based on several counts during the period, compared with the EBA estimates based on only two counts, at the beginning and the end of the test period.
|
METHODS
|
|---|
Patients
The methods used in the study have been described fully (1). In brief, African patients with previously untreated, smear-positive pulmonary tuberculosis due to drug-sensitive strains were studied. All gave informed consent. The patients received 14-day regimens of isoniazid (INH [H]), rifampin (RMP [R]), streptomycin (SM [S]), pyrazinamide (PZA [Z]), and ethambutol (EMB [M]) alone, or in various combinations, or no treatment (Nil regimen).
Treatment
Drugs were given in the following daily doses in combinations: INH (300 mg), RMP (10 mg/kg body weight), PZA (2 g), EMB (25 mg/kg), and SM (1 g). For monotherapy only, INH was given either at the standard dose (INH300) or at 600 mg (INH600) or at 150 mg (INH150). Similarly, RMP was given in monotherapy at the standard dose (RMP10) or at 20 mg/kg (RMP20) or at 5 mg/kg (RMP5). SM and PZA were given at standard dose only in monotherapy. Treatment with standard chemotherapy was started immediately after the 14-day study period. All patients in the study eventually attained bacteriological quiescence.
Bacteriology
Sputum was collected overnight on two consecutive nights before treatment was started and then after 2, 4, 6, 8, 10, 12, and 14 days of the allocated regimen. Samples were treated with dithiothreitol to homogenize the sputum. Serial 10-fold dilutions in water were spread on segments of selective Middlebrook 7H10 agar–oleic acid plates. Conventional susceptibility tests were done on cultures before and at the end of treatment; acquired drug resistance arose in only one patient (given INH150), who was then treated successfully with RMP and EMB.
Statistics
The original data sheets and register used in the study were available. Colony counts were entered on an Excel spreadsheet programmed to calculate colony-forming unit count, log colony-forming unit count, various EBA values, and regression coefficients, using the Slope function, of log colony-forming unit count per milliliter of sputum per day for each patient. Appropriate blocks of data were transferred to the Stata statistical package (release 7; Stata, College Station, TX), from which was obtained summary statistics, analyses of variance, and regression analyses. For the multiple regression analyses, the dependent variable (y) was the EBA or b value and the independent variables (x) were INH, RMP, SM, PZA, and EMB. For each regimen, the drug indicator variable was scored 1 if present in the drug combination and 0 if absent. In the case of monotherapy with different dose sizes of INH or RMP, scores of 1 were given to INH600 and INH300 because they have similar EBA0–2 values and a score of 0.5 was given to INH150 (6), whereas scores of 2, 1, and 0.5 were given to RMP20, RMP10 (standard dosage), and RMP5, respectively. The most appropriate estimate of the within-patient standard deviation for assessing linearity of response in Figure 1
was considered to be the square root of the residual mean square of a two-way analysis of variance for each group that had "Between patients" and "Between days" as the main effects. The 95% confidence limits (95% CL) are presented.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1. Colony-forming unit counts of M. tuberculosis in sputum during treatment in Group 1 (isoniazid [H] only), Group 2 (rifampin [R] only), Group 3 (isoniazid and rifampin [HR]), or Group 4 (other drugs). The SDs within patients, suitable for assessing linearity, are 0.492 for Group 1, 0.467 for Group 2, 0.558 for Group 3, and 0.423 for Group 4, each with 117–183 df.
|
|
|
RESULTS
|
|---|
Summary Measures
Missing counts occurred on 31 of the possible 800 occasions, with 22 among the counts at 10, 12, and 14 days, probably because of sputum conversion by treatment in six patients. Various summary measures were calculated for each patient that measured the rate of fall in sputum colony-forming unit counts per milliliter of sputum per day over different treatment periods. These were calculated either as an EBA, based on only two log colony-forming unit counts, at the start and the end of the period, or as b, the linear regression coefficient, based on all the log counts available during the period, calculated separately for each patient. Thus, EBA2–8 = (log count on Day 2 – log count on Day 8)/6, and b2–8 is the regression coefficient based on the four log colony-forming unit counts on Days 2, 4, 6, and 8. These two procedures measure the same daily fall in counts but differ in their accuracy. The Day 0 count used was the mean of the two log colony-forming unit counts obtained pretreatment. To avoid repetition, values of EBA and b are both quoted as numbers without "log10 colony-forming units per milliliter of sputum per day." To ensure easy comparability with EBA0–2, the signs of regression coefficients, whether of b or the partial coefficients (pb) obtained in the multiple regression analyses, were reversed. The use of these summary measures avoids the problem of correlations between repeated observations on the same patients created by conventional group regression methods (9). Summary statistics for EBA0–2 and b2–14, the most important summary measures, are set out in Table 1
classified into four groups according to the presence of INH, RMP, both drugs, or neither in the treatment regimens. Figure 1 shows that there was an initial rapid fall in colony-forming unit counts during the first 2 days, except in Group 4, which contained drugs with low EBA0–2 results. From Day 2 onward there was a slower linear fall in the log counts. A t test, for paired estimates from each patient, showed that each difference between the EBA0–2, measuring the fall during the first 2 days, and b2–14, measuring it over the subsequent 12 days, was highly significant in Groups 1–3.
The ability of the estimates of EBA or b over various periods of time to discriminate between the different treatment regimens was measured as the value of F (the variance ratio) in a series of one-way analyses of variance (Table 2)
. The EBA0–2 was highly discriminative, with F = 5.44. Over later periods, F was always much lower (less than 2.0). As the treatment period increased, the value of F for the EBAs decreased and remained nonsignificant, whereas F for b increased to 1.96 for the regressions calculated up to Day 14, indicating significant discrimination between regimens with p = 0.017. Thus, b2–14 was the most efficient measure to explore the sterilizing activity of the drugs after the first 2 days, even though it implied more sputum collection from patients and more laboratory work than an EBA study over the same period.
View this table:
[in this window]
[in a new window]
|
TABLE 2. Discrimination, estimated as the variance ratio f, between means in 22 regimens by period of study and type of estimate in one-way analyses of variance
|
|
Multiple Regression Analyses
Multiple regression analyses were performed to determine the effect of the 5 main drugs on the dependent variables EBA0–2 and b2–14 for all 100 patients. For each dependent variable, a further regression analysis was done with the model restricted to the two drugs that had the smallest values of p. Exploratory analyses done with other restricted combinations of drugs did not alter the conclusions. The drug effects in combined treatment were then examined by excluding all regimens of monotherapy and the Nil regimen, to leave 48 patients in the analysis of all drugs and in its restricted set. This exclusion was performed because of the interest in drug effects within combined therapy. The results of the analyses of EBA0–2 (Table 3)
show a large effect of INH with a regression coefficient, pb, of 0.400 (p < 0.0001) and a smaller pb of 0.095 for RMP (p = 0.027). These results were confirmed in the restricted analysis, in which the pb for INH was 0.405 with 95% CL of 0.298 to 0.512 (p < 0.001) and for RMP, the pb was 0.093 with 95% CL of 0.011 to 0.175 (p = 0.027). After exclusion of monotherapy regimens, INH continued to have an overriding bactericidal action. However, the pb for RMP was slightly lower than in the full analysis (0.057) and the effect was nonsignificant, probably because of the exclusion of the large number of patients treated with RMP monotherapy.
View this table:
[in this window]
[in a new window]
|
TABLE 3. Multiple regression analyses of the effects of the five main drugs in 100 patients in all regimens and in 48 patients in combination regimens only (excluding monotherapy and nil groups)
|
|
In the results obtained with b2–14 (Table 3), INH now had no detectable effect either in the complete population or after exclusion of monotherapy. RMP, however, had highly significant effects in both populations (p = 0.007 and 0.009, respectively). In the restricted analysis, the regression coefficient, pb, for RMP was 0.031 with 95% CL of 0.009 to 0.054 (p = 0.007) and similar values were obtained after exclusion for monotherapy. In the initial complete analysis, SM also had a significant effect (p = 0.018), with a pb of 0.042, slightly greater than for RMP (0.033), and 0.044 in the restricted analysis (p = 0.006). However, this effect disappeared when monotherapy was excluded. Of particular interest, after exclusion of monotherapy, the regression due to EMB was significant (p = 0.032) in the complete analyses and highly significant (p = 0.012) in the restricted analysis; the coefficient pb of -0.058 (95% CL of -0.014 to -0.103) was similar in size to the pb of 0.058 for RMP, but was negative rather than positive, suggesting that EMB inhibited the sterilizing activity of other drugs when given in drug combinations.
The same multiple regression analysis was then performed with b2–6, to see what might be achieved with studies lasting only 8 days, the longest that is likely to be allowed today. All of the separate effects were now nonsignificant (including the effect of RMP), except that the SM effect was still significant (p = 0.026) in the analysis of all regimens and in its restricted analysis (p = 0.005), indicating that SM had remarkably early sterilizing activity.
Homogeneity of Combination Regimens Containing Isoniazid
The question of whether drugs added to INH altered the mean EBA0–2 was examined in the regimens of Group 1 in Table 1. There were no significant differences (F[6,76] = 1.7, p = 0.15) between the means of the regimens of INH alone (H150, H300, and H600) and those with the addition of a single drug (HM, HZ, SH, HR). However, heterogeneity of borderline significance (F[9,76] = 2.18, p = 0.033) was found when 3-drug regimens were considered (HRM, SHZ, SHR) together with the 2-drug and monotherapy regimens and again on addition of 4-drug and 5-drug regimens (SHRZ, SHRZM) to make a total of 12 INH-containing regimens (F[11,76] = 1.93, p = 0.048). Heterogeneity seems due mainly to the exceptionally low mean EBA0–2 of 0.320 for the SHR regimen. When this regimen was excluded, the test for all regimens showed no evidence of heterogeneity (F[10,76] = 1.79, p = 0.15).
Detailed Effects of Isoniazid and Rifampin
An examination of the means in Table 1 and the slopes in Figure 1 indicate that the initial fall in counts over the first 2 days (EBA0–2) is greater in Groups 1 and 2 (INH with or without RMP) than in Group 3 (RMP but no INH). However, the regression from 2 days onward (b2–14) is greater in Groups 2 and 3 (RMP with or without INH), having an overall mean of 0.140, than in Group 1 (INH but no RMP) with a mean of 0.089. There is a suggestion from Figure 1 that acceleration in killing in the two Groups 2 and 3 (RMP) compared with Group 1 (INH only) starts only after Day 6.
Detailed Effects of Streptomycin and Ethambutol
To examine in greater detail the effects of SM and EMB after 2 days, which appeared in the multiple regression analysis, basic regimens were chosen for which b2–14 results were available with or without the addition of SM or EMB (Table 4)
. Thus, a basic regimen of INH300 could be compared with SM added as the SH regimen and with EMB added as the HM regimen in the second column of means headed by INH. Again, the basic regimen of R10 could be compared with the SR and RM regimens in the third column. The addition of SM increased the b2–14 values on average from 0.107 to 0.147. This increase is evident in all except the contrast of the INH300 and SH regimens. The addition of EMB to form combined regimens had an opposite effect, decreasing b2–14 values, on average, from 0.134 to 0.103. However, because the effect was significant only when monotherapy regimens were excluded, the greatest weight should be given to the contrasts between HR and HRM and between SHRZ and SHRZM.
Effects of Pyrazinamide
The EBA0–2 of PZA in monotherapy was 0.054 (Table 1), the lowest of any of the drugs, although not much lower than with RMP at the marginally low dose of 5 mg/kg (0.062) or with SM (0.071). During the succeeding 12 days, there was slow sterilizing activity with a b2–14 of 0.114, a low value, but not lower than the mean of 0.089 for all regimens that contain INH but not RMP (Group 1 in Table 1). The 95% CL values of 0.085 to 0.143 clearly establishes that PZA has sterilizing activity, as they do not include 0.0. It is remarkable that there is no evidence from the multiple regression analysis (Table 3) that PZA either increases or decreases sterilizing activity in the five regimens in which it is incorporated (SZ, HZ, SHZ, SHRZ, SHRZM).
Comparison of Results of EBA0–2 Values with Previous Estimates
Table 5
sets out values of EBA0–2 obtained in the current and previous studies performed by the Stellenbosch University group in South Africa, where the great majority of EBA studies have been done (2, 10). The first two rows of Table 5 detail results with untreated patients, obtained as the two pretreatment estimates in the current study and in separate patients who received no drug in Nil groups in previous studies. The next two rows of Table 5 compare results with INH, using the pooled data with all three dose levels of INH in the current study with previous results, using a 300-mg dose throughout, in several studies that were found to be homogeneous by analysis of variance. The two sets of results for Nil and INH300 are the "low" and "high" controls recommended in any EBA study. They show remarkable agreement between the current results and the Stellenbosch results, both in their means and their measures of variation (SD and 95% CL). It can be concluded that the precision of the current study was as good as that obtained by the Stellenbosch group. An additional estimate of the EBA0–2 for PZA indicates that the mean of 0.054 in the current study may be a small overestimate. The overall mean in both studies is 0.022. The lower 95% CL values for both data sets are negative, indicating that PZA might have no EBA over the first 2 days.
|
DISCUSSION
|
|---|
The first point to make is that the means of EBA0–2 and their estimates of error are similar in this study and in later studies at Stellenbosch University (10, 11), in which there has been much greater care in selecting patients suitable for EBA studies, the period of sputum collection has been 16 hours rather than overnight, and more accurate colony-forming unit counts have been done. It can be concluded that the accuracy of the present study is as good as in later studies and that great attention to these details has not improved results, and may well be unnecessary.
It is clear that there were two distinct phases in the early bactericidal activity of antituberculosis drugs. In Phase 1, lasting about 2 days, exponential killing was more rapid than in Phase 2. Furthermore, the predominant drug in Phase 1 was INH, with a killing rate that was the greatest among all that have been measured, because it is given in a dose about 20 times greater than the minimal effective dose (6). INH activity was unaffected by other drugs, even RMP, given together with it. Thus, there is little prospect that an EBA study on a new drug X, which compares monotherapy with X to a combination including INH (HX) will be fruitful. In Phase 2, the killing rate was slower and the dominant drug was RMP, whose rate of kill was unaffected by INH. It is probable that rapidly growing bacilli in cavitary lesions are killed, or at least prevented from growing, during the first 2 days of treatment. The speed of this process is determined by drug action, with priority given to the drug with greatest bactericidal activity against log-phase organisms. After 2 days, the remaining, persisting bacilli are killed in a process of slow sterilization, in which speed is determined by bacterial factors rather than drug action. Whenever a bacillus starts to metabolize, it is killed by any drug that is active; priority between drugs is then determined by the speed with which bactericidal activity starts (12). The bactericidal and sterilizing properties of drugs are separate attributes and need separate measurement. The overall response, in its two phases, could be measured by an exponential equation with two rate constants and therefore five parameters such that colony-forming unit count at time
 |
where C1 and k1 are parameters for the initial phase of killing, C2 and k2 are parameters for the later sterilizing phase, and S is the asymptotic value for incomplete killing (13). However, it requires each patient to be studied repeatedly over long periods to obtain estimates of all five parameters for each patient by iterative fitting. It is clear that a procedure (14) that uses a model with a single rate constant over a period of at least 5 days is incorrect, because this attempts to fit a single straight line over the entire period. The procedure is inadvisable for the reasons given in correspondence (15, 16). The calculation of two straight lines is a pragmatic solution to the problem of estimating both parts of the equation. However, there must be a time, at or soon after 2 days, when both rates are having an effect so that there should be a short, smooth transition from the early rapid kill to early sterilizing activity. Bacterial metabolism then diminishes during the remainder of the long treatment period. It is generally agreed that shortening the treatment period is the main way in which the control of tuberculosis can be improved. This requires the development of new drugs with high sterilizing activity (17). Efficient methods for detecting sterilizing activity are therefore central to lightening the burden of tuberculosis worldwide. Sterilizing activity may, at least in its initial stages, be measurable by the rate of kill (b) after 2 days. However, even with a total study duration of 6–8 days, significant differences in b could not be found between the regimens (with the exception of the early effect of SM) and were found only when the study period was increased to 10 or 14 days, with p = 0.033 and 0.017, respectively (Table 2). Because such a prolongation would not be allowed on the basis of current ethical considerations, estimation of sterilizing activity of a new drug will probably have to be performed by a serial sputum colony-forming unit count study in which patients are treated with combined regimens and the small differences in sterilizing activity between them are measured by regression analysis of serial sputum counts starting at Day 2 and continuing for a period of at least 1 month (8). Nevertheless, the results in the current study with b2–14 estimates provide information indicating that the drug with the main effect in determining the rate of kill is RMP. This finding agrees with results of an earlier clinical trial (18), and a serial sputum colony-forming unit count study (8), proving the importance of rifampin in shortening treatment.
The evidence that SM also acts as a sterilizing drug is less strong because it is apparent only when all regimens are analyzed by multiple regression. However, the effect was detectable uniquely early in the b2–6 estimates, indicating a particularly early onset of sterilizing activity while there was still appreciable bacillary metabolism. These findings are in keeping with evidence from a clinical trial that the proportion of negative sputum cultures was significantly greater (p = 0.03) at 1 month in a SHRZ regimen than in the corresponding HRZ regimen, although the difference was not maintained at 2 months (19). This apparent decrease in sterilizing activity was never found in studies on the addition of RMP or PZA, where the effect, as measured by sputum culture, was always greater at 2 months than at 1 month (20). Attempts to show that SM decreases relapse rates have, however, failed, indicating that despite early activity during the first 14 days, SM cannot compete with RMP in sterilizing activity throughout the long treatment period (21).
Finally, the presence of EMB in regimens with combinations of two or more drugs appears to have slowed the rate of kill by the other drugs. This effect, as judged by the values of the regression coefficients, is as great as the effect of RMP, but in the opposite direction. The effect does not appear at all in the full analysis with all regimens, perhaps because the b2–14 for EMB alone is higher at 0.162 than the overall means of 0.066, 0.067, 0.084, and 0.080 in each of the treatment groups (Table 1). Thus EMB seems to have an inhibitory effect on the sterilizing action of other drugs whereas it has fairly high sterilizing activity on its own. The strong antagonism shown in vitro by EMB to the bactericidal activity of RMP and the combination of RMP and INH (22) may not be relevant, because no such antagonism was shown in the EBA0–2 findings. Clinical studies have shown no suggestion of a decreased proportion of negative sputum cultures at 2 months when EMB is added to the initial phase of a regimen in place of SM (23, 24), although there is a slight suggestion (not statistically significant) that relapse rates might be slightly higher (2 of 137 patients receiving SHRZ compared with 13 of 283 when EMB was added [23]). We also must account for a much higher relapse rate in the 8-month 2HRZE/6EH regimen (12.4%) than in the 6-month 2HRZE/4RH regimen (2.7%) described in the presentation of preliminary findings in the first study undertaken by the International Union against Tuberculosis and Lung Disease (A. Jindani, personal communication). That this difference is unlikely to be due solely to an absence of RMP in the 8-month regimen is shown by (1) the finding of 0 and 3% relapse rates in the two earlier studies of 2SHRZ/TH (25) and 2SHRZ/H (26) and (2) an absence of any sterilizing activity due to thiacetazone (H in the above-described regimen) (18).
The failure to show any effect of PZA in increasing the b2–14 estimates is not surprising despite the clear evidence from clinical trials that its sterilizing effect is nearly as great as that of RMP (18) and, unlike RMP, occurs only during the first 2 months of treatment (26, 27). PZA is a prodrug converted to pyrazinoic acid by bacillary amidase. The uptake of the toxic protonated form of pyrazinoic acid is by passive diffusion, whereas its excretion from the bacilli is due to an inefficient active pump, so that it slowly accumulates and is eventually lethal (28). For this reason, PZA is particularly active against old, resting bacilli whose excretory pump has slowed down even further. Unlike other antituberculosis drugs, PZA is unique in having virtually no bactericidal action against rapidly growing bacilli but has increasing activity as the growth rate slows down. Because the lethal process is slow, PZA has no detectable bactericidal activity during the first 2 days, and is then lethal at much the same rate as other drugs. However, it continues to kill after other drugs have little bactericidal effect; this increased sterilizing action is probably evident only after about 14 days and so does not influence the multiple regression analyses (8).
In summary, we have shown that the EBA0–2 and the fall in counts after 2 days estimate different properties of drugs. In the first 2 days, there was a rapid bactericidal action in which INH had most effect and was not influenced by adding other drugs in combination. After 2 days, regression analysis has uncovered trends not detectable by calculation of EBAs. It has demonstrated the important action of RMP, the absence of INH activity, and the early onset of the action of SM in sterilization. There also appeared to be possible bactericidal antagonism due to EMB. Nevertheless, the regression analyses do not agree well with the results of clinical trials. There is disagreement between the apparent early sterilizing activity of SM and its failure to have a long-term sterilizing role, as exemplified by ultimate relapse rates, possibly because its sterilizing activity depends on the existence of some bacterial metabolism. The role played by EMB needs further exploration in clinical studies. Furthermore, without an extension of the total treatment period to more than 5–7 days, it is unlikely that even the sterilizing activity of RMP could have been demonstrated. Even with an extension, the considerable sterilizing activity of PZA could not be shown in EBA studies.
Received in original form October 2, 2002;
accepted in final form January 2, 2003
|
REFERENCES
|
|---|
- Jindani A, Aber VR, Edwards EA, Mitchison DA. The early bactericidal activity of drugs in patients with pulmonary tuberculosis. Am Rev Respir Dis 1980;121:939–949.[Medline]
- Mitchison DA, Sturm WA. The measurement of early bactericidal activity. In: Malin A, McAdam KPWJ, editors. Baillière's clinical infectious diseases: Mycobacterial diseases. Part II. London: Baillière Tindall; 1997. p. 185–206.
- Donald PR, Sirgel FA, Venter A, Smit E, Parkin DP, van de Wal BW, Doré CJ, Mitchison DA. The early bactericidal activity of streptomycin. Int J Tuberc Lung Dis 2002;6:693–698.[Medline]
- Donald PR, Sirgel FA, Kanyok TP, Danziger LH, Venter A, Botha FJ, Parkin DP, Seifart HI, van de Wal BW, Maritz JS, et al. The early bactericidal activity of paromomycin (aminosidine) in patients with smear-positive pulmonary tuberculosis. Antimicrob Agents Chemother 2000;44:3285–3287.[Abstract/Free Full Text]
- Donald PR, Sirgel FA, Venter A, Smith E, Parkin DP, van de Wal BW, Mitchison DA. The early bactericidal activity of amikacin in pulmonary tuberculosis. Int J Tuberc Lung Dis 2001;5:533–538.[Medline]
- Donald PR, Sirgel FA, Botha FJ, Seifart HI, Parkin DP, Vandenplas ML, Van de Wal BW, Maritz JS, Mitchison DA. The early bactericidal activity of isoniazid related to its dose size in pulmonary tuberculosis. Am J Respir Crit Care Med 1997;156:895–900.[Abstract/Free Full Text]
- Sirgel FA, Donald PR, Odhiambo J, Githui W, Umapathy KC, Paramasivan CN, Tam CM, Lam KM, Lam CW, Sole KM, Mitchison DA, EBA Collaborative Study Group. A multicentre study of the early bactericidal activity of anti-tuberculosis drugs. J Antimicrob Chemother 2000;45:859–870.[Abstract/Free Full Text]
- Brindle R, Odhiambo J, Mitchison D. Serial counts of Mycobacterium tuberculosis in sputum as surrogate markers of the sterilising activity of rifampicin and pyrazinamide in treating pulmonary tuberculosis. BMC Pulm Med 2001;1:2.[CrossRef][Medline]
- Matthews JNS, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990;300:230–235.
- Sirgel F, Venter A, Mitchison D. Sources of variation in studies of the early bactericidal activity of antituberculosis drugs. J Antimicrob Chemother 2001;47:177–182.[Abstract/Free Full Text]
- Botha FJH, Sirgel FA, Parkin DP, van de Wal BW, Donald PR, Mitchison DA. Early bactericidal activity of ethambutol, pyrazinamide and the fixed combination of isoniazid, rifampicin and pyrazinamide (Rifater) in patients with pulmonary tuberculosis. S Afr Med J 1996;86:155–158.[Medline]
- Dickinson JM, Mitchison DA. Experimental models to explain the high sterilizing activity of rifampin in the chemotherapy of tuberculosis. Am Rev Respir Dis 1981;123:367–371.[Medline]
- Snedecor GW, Cochran WG. Statistical methods, 8th ed. Ames, IA: Iowa State University Press; 1989. p. 400–404.
- Gillespie SH, Gosling RD, Charalambous BM. A reiterative method for calculating the early bactericidal activity of antituberculosis drugs. Am J Respir Crit Care Med 2002;166:31–35.[Abstract/Free Full Text]
- Mitchison DA, Gillespie SH, Charalambous BM, Gosling RD. A reiterative method for calculating bactericidal activity. Am J Respir Crit Care Med 2003;163:663–664.
- Doré CJ, Nunn AJ, Gillespie SH, Gosling RD, Charalambous BM. Bactericidal activity of antituberculosis drugs. Am J Respir Crit Care Med 2003;167:663.
- Global Alliance for Tuberculosis Drug Development. Tuberculosis: scientific blueprint for TB drug development. Tuberculosis 2001;81:1–52.
- East African/British Medical Research Council. Controlled clinical trial of four short-course (6-month) regimens of chemotherapy for treatment of pulmonary tuberculosis. Lancet 1973;1:1331–1339.[CrossRef][Medline]
- Hong Kong Chest Service/British Medical Research Council. Controlled trial of 2, 4 and 6 months of pyrazinamide in 6-month, 3x weekly regimens for smear-positive pulmonary tuberculosis, including an assessment of a combined preparation of isoniazid, rifampin and pyrazinamide. Am Rev Respir Dis 1991;143:700–706.[Medline]
- Mitchison DA. Modern methods for assessing the drugs used in the chemotherapy of mycobacterial disease. J Appl Bacteriol Symp Suppl 1996;81(Suppl.):72S–80S.
- Fox W, Ellard GA, Mitchison DA. Studies on the treatment of tuberculosis undertaken by the British Medical Research Council Tuberculosis Units, 1946–1984, with relevant subsequent publications. Int J Tuberc Lung Dis 1999;3:S231–S279.[Medline]
- Dickinson JM, Aber VR, Mitchison DA. Bactericidal activity of streptomycin, isoniazid, rifampin, ethambutol, and pyrazinamide alone and in combination against Mycobacterium tuberculosis. Am Rev Respir Dis 1977;116:627–635.[Medline]
- Hong Kong Chest Service/British Medical Research Council. Controlled trial of 4 three-times weekly regimens and a daily regimen all given for 6 months for pulmonary tuberculosis. Second report: the results up to 24 months. Tubercle 1982;63:89–98.[Medline]
- British Thoracic Association. A controlled trial of six months chemotherapy in pulmonary tuberculosis. Br J Dis Chest 1981;75:141–153.[CrossRef][Medline]
- East African/British Medical Research Council. Controlled clinical trial of four short-course regimens of chemotherapy for two durations in the treatment of pulmonary tuberculosis: second report. Tubercle 1980;61:59–69.[CrossRef][Medline]
- East and Central African/British Medical Research Councils. Controlled clinical trial of 4 short-course regimens of chemotherapy (three 6-month and one 8-month) for pulmonary tuberculosis (fifth collaborative study): final report. Tubercle 1986;67:5–15.[Medline]
- East African/British Medical Research Councils. Controlled clinical trial of five short-course (4-month) chemotherapy regimens in pulmonary tuberculosis: second report of the 4th study. Am Rev Respir Dis 1981;123:165–170.[Medline]
- Zhang Y, Mitchison DA. The curious characteristics of pyrazinamide: a review. Int J Tuberc Lung Dis 2003;7:6–21.[Medline]
This article has been cited by other articles:
|
|
|
|
 
S. R. Ritchie, A. C. Harrison, R. H. Vaughan, L. Calder, and A. J. Morris
New recommendations for duration of respiratory isolation based on time to detect Mycobacterium tuberculosis in liquid culture
Eur. Respir. J.,
September 1, 2007;
30(3):
501 - 507.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. Gumbo, A. Louie, W. Liu, D. Brown, P. G. Ambrose, S. M. Bhavnani, and G. L. Drusano
Isoniazid Bactericidal Activity and Resistance Emergence: Integrating Pharmacodynamics and Pharmacogenomics To Predict Efficacy in Different Ethnic Populations
Antimicrob. Agents Chemother.,
July 1, 2007;
51(7):
2329 - 2336.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. R. Davies, R. Brindle, S. H. Khoo, and L. J. Aarons
Use of nonlinear mixed-effects analysis for improved precision of early pharmacodynamic measures in tuberculosis treatment.
Antimicrob. Agents Chemother.,
September 1, 2006;
50(9):
3154 - 3156.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. R. Davies, S. H. Khoo, and L. J. Aarons
Optimal sampling strategies for early pharmacodynamic measures in tuberculosis
J. Antimicrob. Chemother.,
September 1, 2006;
58(3):
594 - 600.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. F. Warner and V. Mizrahi
Tuberculosis Chemotherapy: the Influence of Bacillary Stress and Damage Response Pathways on Drug Efficacy
Clin. Microbiol. Rev.,
July 1, 2006;
19(3):
558 - 570.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Ji, S. Lefrancois, J. Robert, A. Chauffour, C. Truffot, and V. Jarlier
In Vitro and In Vivo Activities of Rifampin, Streptomycin, Amikacin, Moxifloxacin, R207910, Linezolid, and PA-824 against Mycobacterium ulcerans.
Antimicrob. Agents Chemother.,
June 1, 2006;
50(6):
1921 - 1926.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. H. Gillespie, R. D. Gosling, L. Uiso, N. E. Sam, E. G. Kanduma, and T. D. McHugh
Early bactericidal activity of a moxifloxacin and isoniazid combination in smear-positive pulmonary tuberculosis
J. Antimicrob. Chemother.,
December 1, 2005;
56(6):
1169 - 1171.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: Controlling Tuberculosis in the United States
Am. J. Respir. Crit. Care Med.,
November 1, 2005;
172(9):
1169 - 1227.
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. A. Sirgel, P. B. Fourie, P. R. Donald, N. Padayatchi, R. Rustomjee, J. Levin, G. Roscigno, J. Norman, H. McIlleron, D. A. Mitchison, et al.
The Early Bactericidal Activities of Rifampin and Rifapentine in Pulmonary Tuberculosis
Am. J. Respir. Crit. Care Med.,
July 1, 2005;
172(1):
128 - 135.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. A. Mitchison
The Diagnosis and Therapy of Tuberculosis During the Past 100 Years
Am. J. Respir. Crit. Care Med.,
April 1, 2005;
171(7):
699 - 706.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. N. Paramasivan, S. Sulochana, G. Kubendiran, P. Venkatesan, and D. A. Mitchison
Bactericidal Action of Gatifloxacin, Rifampin, and Isoniazid on Logarithmic- and Stationary-Phase Cultures of Mycobacterium tuberculosis
Antimicrob. Agents Chemother.,
February 1, 2005;
49(2):
627 - 631.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. Janulionis, C. Sofer, H.-Y. Song, and R. S. Wallis
Lack of Activity of Orally Administered Clofazimine against Intracellular Mycobacterium tuberculosis in Whole-Blood Culture
Antimicrob. Agents Chemother.,
August 1, 2004;
48(8):
3133 - 3135.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. A. Mitchison
Fluoroquinolones in the Treatment of Tuberculosis: A Study in Mice
Am. J. Respir. Crit. Care Med.,
February 1, 2004;
169(3):
334 - 335.
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. J. Tobin
Tuberculosis, Lung Infections, Interstitial Lung Disease, Social Issues and Journalology in AJRCCM 2003
Am. J. Respir. Crit. Care Med.,
January 15, 2004;
169(2):
288 - 300.
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. D. Gosling, L. O. Uiso, N. E. Sam, E. Bongard, E. G. Kanduma, M. Nyindo, R. W. Morris, and S. H. Gillespie
The Bactericidal Activity of Moxifloxacin in Patients with Pulmonary Tuberculosis
Am. J. Respir. Crit. Care Med.,
December 1, 2003;
168(11):
1342 - 1345.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. A. Mitchison
Role of Isoniazid in Once-Weekly Rifapentine Treatment of Pulmonary Tuberculosis
Am. J. Respir. Crit. Care Med.,
May 15, 2003;
167(10):
1298 - 1299.
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. J. Burman
The Hunt for the Elusive Surrogate Marker of Sterilizing Activity in Tuberculosis Treatment
Am. J. Respir. Crit. Care Med.,
May 15, 2003;
167(10):
1299 - 1301.
[Full Text]
[PDF]
|
|
|
Copyright © 2003 American Thoracic Society
|
TOP
ABSTRACT
METHODS