ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Sept. 2004, p. 36423643 Vol. 48, No. 90066-4804/04/$08.000 DOI: 10.1128/AAC.48.9.36423643.2004

Moxifloxacin Treatment of Tuberculosis

We are pleased to see that the study of Pletz et al. (6)confirms our earlier observations that moxifloxacin is active inpatients with pulmonary tuberculosis with positive sputumsmears (3). However, some of their results differ from ours,and we believe that this is due to the method that they adopted.In our study, we recruited groups of 15 subjects (3) and foundthat moxifloxacin was significantly less active than isoniazidwhen the time to reduce the viable bacillus count by 50%(vt50) was calculated (1), but like Pletz et al., we found nosignificant difference between the two drugs when early bacte-ricidal activity (EBA) was calculated (3, 5, 6). The most im-portant reason for this difference is the number of patients inthe groups recruited by Pletz et al. is small, and thus, theabsence of a statistically significant difference cannot be con-strued as indicating biological equivalence. Second, the EBAmethodology that Pletz et al. use is a variation of the standardmethod (4), which calculates the value for EBA over 5 daysrather than 2 days and has limited power to distinguish be-tween different drugs (5). The EBA methodology produceslarge confidence intervals, and for this reason, studies haveshown that the results for isoniazid are poorly reproduciblebetween centers (7). Third, by grouping the data from differentpatients together, the errors are enlarged. Viable counts ofmycobacteria from sputum vary enormously between patients(often by more than 1 order of magnitude) and from day today. As the EBA measure is half (or a fifth) of the ratiobetween the day 0 and day 2 (or day 5) values expressedlogarithmically, small variations in sputum mycobacterial via-ble counts result in large variations in the EBA values as foundin the study of Pletz et al.

We agree with the authors strategy of measuring the sputummycobacterial viable count over 5 days, but by recording onlythree values and not calculating a regression line, they losemuch of the benefit of the longer period of observation. Asdiscussed in our previous papers (13), by taking measure-ments daily over the 5-day trial period, it is possible to usenonlinear regression, which permits discrepant values to beidentified and removed. Moreover, vt50 has been shown to bea measure which is comparable in different countries (2).

For these reasons, we believe that, although this paper isuseful in that it confirms that moxifloxacin is bactericidal inpatients with pulmonary tuberculosis with positive smears, itsconclusion that this compound is as active as isoniazid is mis-leading. Further studies are required to understand the opti-mal use of new highly active quinolones, such as moxifloxacin,in pulmonary tuberculosis and whether they add to the activityof isoniazid.

REFERENCES

1. Gillespie, S. H., R. D. Gosling, and B. M. Charalambous. 2002. A reiterativemethod for calculating the early bactericidal activity of antituberculosis drugs.Am. J. Respir. Crit. Care Med. 166:3135.

2. Gosling, R. D., L. Heifets, and S. H. Gillespie. 2003. A multicentre compar-ison of a novel surrogate marker for determining the specific potency ofanti-tuberculosis drugs. J. Antimicrob. Chemother. 52:473476.

3. Gosling, R. D., L. O. Uiso, N. E. Sam, E. Bongard, E. G. Kanduma, M.Nyindo, R. W. Morris, and S. H. Gillespie. 2003. The bactericidal activityof moxifloxacin in patients with pulmonary tuberculosis. Am. J. Respir.Crit. Care Med. 168:13421345. (First published 13 August 2003; 10.1164/rccm.200305682OC.)

4. Hafner, R., J. Cohn, D. Wright, N. Dunlap, M. Egorin, M. Enama, K. Muth,C. Peloquin, N. Mor, L. Heifits, and the DATRI 008 Study Group. 1997. Early

bactericidal activity of isoniazid in pulmonary tuberculosis. Optimization ofmethodology. Am. J. Respir. Crit. Care Med. 156:918923.

5. Jindani, A., V. Aber, E. Edwards, and D. Mitchison. 1980. The early bacteri-cidal activity of drugs in patients with pulmonary tuberculosis. Am. Rev.Respir. Dis. 121:939949.

6. Pletz, M. W., A. De Roux, A. Roth, K. H. Neumann, H. Mauch, and H. Lode.2004. Early bactericidal activity of moxifloxacin in treatment of pulmonarytuberculosis: a prospective, randomized study. Antimicrob. Agents Chemo-ther. 48:780782.

7. Sirgel, F. A., P. R. Donald, J. Odhiambo, W. Githui, K. C. Umapathy, C. N.Paramasivan, C. M. Tam, K. M. Kam, C. W. Lam, K. M. Sole, and D. A.Mitchison. 2000. A multicentre study of the early bactericidal activity ofanti-tuberculosis drugs. J. Antimicrob. Chemother. 45:859870.

Roland Gosling*Stephen GillespieDepartment of Medical MicrobiologyRoyal Free and University CollegePond StreetLondon NW3 2PFUnited Kingdom

*Phone: 44 2077940500Fax: 44 2077940433E-mail: rolygos@aol.com

Authors Reply

Both the study of Gosling et al. (1) and our study (4) usedisoniazid as a comparator to moxifloxacin, but different anal-yses were performed in the two studies. Our study was basedon the classical parameter of EBA over 5 days. When theEBAs of moxifloxacin and isoniazid were compared, there wasno statistically significant difference. Our conclusion was thatthe activity of moxifloxacin is comparablewe did not writeequalto that of isoniazid. This was not based on the absenceof a statistically significant difference but rather on the follow-ing data. (i) The reduction in the number of CFU per milliliterover 5 days was more pronounced in the moxifloxacin groupthan in the isoniazid group (from 14 17.2 to 0.7 0.9CFU/ml for moxifloxacin versus 11.5 8.7 to 1.2 0.15CFU/ml for isoniazid). (ii) The EBA of moxifloxacin wasgreater than the EBA of isoniazid (0.273 for moxifloxacin and0.209 for isoniazid). (iii) Calculating 95% confidence intervalsrevealed that moxifloxacin had at least 61.6% of the antimy-cobacterial activity of isoniazid despite the deviation of thedata.

Our conclusion is also supported by results of animal studieswhich showed that the bactericidal activity of moxifloxacin iscomparable to that of isoniazid (2, 3). Furthermore, Nuerm-berger et al. showed in a mouse model that the replacement ofisoniazid by moxifloxacin in the standard regimen increasedthe activity dramatically, resulting in earlier culture negativity(E. Nuermberger, T. Yoshimatsu, S. Tyagi, W. Bishai, and J.Grosset, Abstr. 43rd Intersci. Conf. Antimicrob. Agents Che-mother., abstr. 1035, 2003).

In conclusion, we agree with Gosling et al. that the discrep-ancy in our studies regarding the comparison of moxifloxacinto isoniazid is caused by the different parameters (EBA versusvt50) used, but on the basis of our results and the results ofseveral animal studies, we maintain that moxifloxacin and iso-niazid have comparable activities. However, larger studies us-ing clinical endpoints, such as time to culture negativity, arerequired to identify the drug with the best clinical activity.

3642

on April 29, 2015 by guest

http://aac.asm.org/

Dow

nloaded from

REFERENCES

1. Gosling, R. D., L. O. Uiso, N. E. Sam, E. Bongard, E. G. Kanduma, M.Nyindo, R. W. Morris, and S. H. Gillespie. 2003. The bactericidal activity ofmoxifloxacin in patients with pulmonary tuberculosis. Am. J. Respir. Crit.Care Med. 168:13421345. (First published 13 August 2003; 10.1164/rccm.200305682OC.)

2. Ji, B., N. Lounis, C. Maslo, C. Truffot-Pernot, P. Bonnafous, and J. Grosset.1998. In vitro and in vivo activities of moxifloxacin and clinafloxacin againstMycobacterium tuberculosis. Antimicrob. Agents Chemother. 42:20662069.

3. Miyazaki, E., M. Miyazaki, J. M. Chen, R. E. Chaisson, and W. R. Bishai.1999. Moxifloxacin (BAY128039), a new 8-methoxyquinolone, is active in amouse model of tuberculosis. Antimicrob. Agents Chemother. 43:8589.

4. Pletz, M. W., A. De Roux, A. Roth, K. H. Neumann, H. Mauch, and H. Lode.2004. Early bactericidal activity of moxifloxacin in treatment of pulmonarytuberculosis: a prospective, randomized study. Antimicrob. Agents Che-mother. 48:780782.

Mathias W. R. Pletz*Department of International HealthRollins School of Public HealthEmory UniversityAtlanta, Georgia

Hartmut LodeDepartment of Chest and Infectious DiseasesChest Hospital HeckeshornBerlin, Germany

*Phone: (404) 727-3984Fax: (404) 712-8419E-mail: mpletz@sph.emory.edu

VOL. 48, 2004 LETTERS TO THE EDITOR 3643

on April 29, 2015 by guest

http://aac.asm.org/

Dow

nloaded from