Ethambutol

Background & Objectives: Poor bioavailability of rifampicin (R) in combination with other anti-tuberculosis drugs such as isoniazid (H), pyrazinamide (Z), and etliambutol (E) is a subject of much concern for the last few decades. This could be due to an interaction between R and other drugs. An investigation was therefore undertaken to examine the bioavailability of R in the presence of H, Z and E or a combination of the three drugs.

Methods: The study included eight healthy volunteers, each being investigated on four occasions at weekly intervals once with R alone and with three of the four combinations on the three remaining occasions. A partially balanced incomplete block design was employed and the allocation of R or the drug combinations was random. Plasma concentrations of Rat intervals upto 12 h were determined by microbiological assay using Staphylococcus aitreus as the test organism. The proportion (%) dose of R as R plus desacetyl R (DR) in urine excreted over the periods 0-8 and 8-12 h was also determined. Bioavailability was expressed as an index (BI) of area under time concentration curve (AUC) calculated from the plasma concentrations or proportion of dose of R excreted as R plus DR in urine with the combinations to that with R alone.

Results: The bioavailability indices based on AUC were 0.96 with RE, 0.76 with RH, 1.08 with RZ and 0.65 with REHZ. The indices based on urine estimations (0-8 h) were similar, the values being 0.94, 0.84, 0.94 and 0.75, respectively. A second investigation revealed that the decrease of bioavailability of R with H was not due to the excipients present in H tablets.

Interpretation & conclusion: Isoniazid alone or in combination with E and Z reduces the bioavailability of R. Urinary excretion data offer a simple and non invasive method for the assessment of bioavailability of R.

Key words Bioavailability - rifampicin - treatment - tuberculosis - urinary excretion

Current treatment of tuberculosis involves the use of multiple drug regimens containing rifampicin (R) in addition to ethambutol (E), isoniazid (H) and pyrazinamide (Z). Bioavailability of R is known to be affected by a number of factors such as the manufacturing process1, presence of food in the gastrointestinal tract2, acidity of the gastric juice3 and excipients including those used with companion drugs such as p-ammo salicylic acid4. Ethambutol has been shown to have little or no effect on the gastrointestinal absorption of R5 and concomitant administration of Z resulted in a lower bioavailability of R6. Published results with respect to H are equivocal. While Boman et al7 and Acocella el al8 did not observe any effect, Mouton et al9 demonstrated a decrease in the plasma concentrations of R when administered together with H. In normal adults the peak plasma concentrations after administration of 600 mg R, are in the range of 6-13 µg/ml, while on administration of R along with H and/or Z, the peak concentrations range from 3-6 µg/ml and AUC0^sub 0-8^ values in the range of 30-50 µg/ml.h10-12. Shishoo et al13 demonstrated decreased bioavailability of R in presence of H both by means of in vitro dissolution testing and in vivo bioavailability studies. Grosset et al14 observed an antagonism between R and H in experimental marine tuberculosis and attributed this to a decreased bioavailability of R. To gain a better understanding of the interactions if any. an investigation was done to study the bioavailability of R when administered together with E, H, Z or a combination of the three drugs (EHZ). The conclusions are not only based on the traditional plasma levels (AUC), but also on the excretion of the proportion of the administered drug as R plus its primary metabolite, desacetyl rifampicin (DR) in urine. Since the results of this investigation revealed a significant decrease in the bioavailability of R when administered together with H alone or EHZ, another investigation based on urine excretion of R plus DR was undertaken to examine whether it was H per se or the excipients present in H tablets that was responsible for the decrease.

Material & Methods

The investigations were undertaken in healthy male volunteers (college students and the centre staff) aged 18-56 yr with normal, hepatic and renal function. Rifampicin (Tata Pharma, India), ethambutol (Ciba-Geigy, India), isoniazid (Pfizer, India) and pyrazinamide (Lupin, India) were used in the first investigation. In the second investigation, due to the non-availability of the Tata Pharma Product, rifampicin (Lupin) was used and pure isoniazid powder was a Bayer product. Standard dosages of the drugs15 were used in the investigations: 10 mg/kg body weight for R, 25 mg/kg for E, 12 mg/kg for H and 35 mg/kg for Z.

The investigations were started after getting approval from the institutional ethics committee and informed written consent from all the study subjects.

Plasma R concentrations were determined by the plate diffusion assay of Dickinson et al16 employing a strain of Staphylococcus aureus [Sub Group I -NCTC 10702, a gift from Dr DA Mitchison, British Medical Research Council (BMRC), UK], resistant to streptomycin and other antibiotics. Rifampicin standards ranging from 0.04-1.28 µg/ml were set up in quadruplicate and the concentrations of the drug in plasma (set up in quadruplicate in dilutions of 1 in 5, 1 in 10 and 1 in 20) were obtained from regression line of zone of inhibition on log concentrations of the standards.

Urinary excretion of R plus DR was estimated according to the procedure of Immanuel et al17. In brief, urine samples (3ml or larger according to volume excreted) and standard concentrations of R ranging from 10-50 µg/ml set up in normal urine were mixed with 1.5 ml of a citrate - phosphate buffer, pH 7.0 and extracted with 3 ml of chloroform. The extinction of the chloroform extract was recorded at 475 nm giving an estimate of the concentrations of R plus DR without interference from other metabolites of R.

Conduct of the investigations:

Investigation I-A total of eight volunteers were investigated. Each volunteer was investigated on four occasions, once with R alone (control) and with 3 of the 4 drug combinations, RE, RH, RZ and REHZ on the other three occasions with an interval of at least one week between occasions. A partially balanced incomplete block design was employed and by using this design, information on R alone was available for all the 8 subjects and for 6 each with the four drug combinations. The sequence of administration of R and the combinations was random (Table I).

About eight ml of heparinised blood was collected at 1, 2, 3, 6, 9 and 12 h after administration of drug(s) and total urine excreted over the periods 0-8 and 8-12 h were collected following drug administration on an empty stomach. Uniform breakfast and lunch were provided at the end of 2 and 6 h respectively after drug administration. Plasma was separated and stored along with aliquots of urine at -20°C until analysis. Estimations of plasma R concentrations and those of R plus DR in urine were undertaken within 48 h.

Investigation II-Six volunteers were investigated on two occasions. On the first occasion, 3 volunteers received R(10 mg/kg) plus pure H powder (12 mg/kg), while the other 3 received R plus H tablet in the same dosage. On the second occasion, a week later, the order of administration of H powder or the tablet was reversed, a cross-over design being employed. Urine excreted over the period (0-8 h) was collected and the proportion (%) of the dose of R excreted as R plus DR was calculated.

In both the investigations, all estimations were undertaken after randomising and coding the samples. In investigation-I comparison of the values between the control group (R only) and those who received RE, RH, or REHZ was restricted to the same subjects who received the particular combination and the control.

Pharmacokinetics and statistical analysis: Maximum concentration (C^sub max^) and the time to attain C^sub max^ (T^sub max^) were determined by direct visual inspection of data. Linear trapezoidal rule was used to calculate area under time concentration curve (AUC^sub 0-12^)18. The elimination rate constant (Kel) was calculated from the terminal log-linear decline of concentration. Terminal elimination half-life (t^sub ½^) was calculated as 0.693/Kel.

The bioavailability of R for each of the 4 drug combinations has been expressed as an index (BI) and is the ratio of AUC or the proportion of the dose excreted as R plus DR in urine with the four combinations (RE, RH, RZ and REHZ) to the respective control values obtained after the administration of R alone. Student's t-test (paired and unpaired) was used for testing the significance of the differences between the mean values.

Results

Investigation I: The mean body weight of the 8 volunteers was 66.1 kg±SD (range 56.5-75.4 kg) and the mean dosages of R administered to the RE, RH, RZ and REHZ groups (and the respective controls) were 10.88, 11.43, 11.30 and 10.86 mg/kg, respectively.

The mean serial plasma R concentrations for all the 8 control subjects (R only) and for 6 each with RE, RH, RZ and REHZ are shown in the Fig. The mean peak concentrations and AUC^sub 0-12^ calculated from those concentrations together with the proportion (%) of dose excreted as R plus DR in urine over the periods 0-8 and 0-12 h are presented in Table II.

Peak concentrations were attained before the third hour in all the subjects and the concentrations fell exponentially thereafter with mean half-lives rangingfrom 4.14 in the RZ group to 5.14 in the REHZ group, none of these differences being significant. Administration of E or Z did not appear to have any significant effect on the plasma concentrations or the urinary excretion of R. However, the C^sub max^ and the AUC values were significantly lower in the RH (24%) and REHZ (39%) groups than in the respective controls who received R only (P

None of the differences in the mean values of C^sub max^, AUC^sub 0-12^ or the proportion of dose excreted as R plus DR in urine was significant between the RE and the RZ groups or between the RE and the RH groups. Only the mean AUC^sub 0-12^ value of R was significantly higher in the RZ group than in the RH group (P=0.02). The mean values of all the variables in the RE and the RZ groups were, however, significantly higher than those in the REHZ group (P

The BI (based on AUC and per cent dose excreted in urine) of RE and RZ groups were significantly higher than those obtained for the RH and REHZ groups (P

Investigation II: The mean proportions of the dose of R excreted as R plus DR in urine excreted over the period 0-8 h were 9.3 per cent with R plus pure H powder and 10.6 per cent with R plus H tablets; the difference was not significant.

Discussion

Good bioavailability leading to adequate plasma and tissue concentrations of R (and other drugs) is, an absolute prerequisite for the success of treatment of tuberculosis. It has been postulated that peak plasma R concentrations should be of the order of 10-15 µg/ml with dosages of 10-12 mg/kg for good therapeutic response19. Together with Z, R is one of the key sterilizing drugs and its anti-TB potency is markedly dose dependent20. A United States Public Health System (USPHS) trial has shown that with lower dosages such as 9 mg/kg leading to lower plasma concentrations of R, the speed at which sputum conversion occurs is also reduced21.

Evidence presented in this report suggests that H alone or in combination with E and Z reduces the bioavailability of R. Since the plasma elimination half-lives were similar with the different combinations employed, this could result only from a decreased gastrointestinal absorption of R caused by H. Impaired bioavailability of R in the presence of H as seen in the present study could be due to degradation of R in the presence of H as reported by Jindal et al22 by in vitro disssolution testing. Similar results have been reported by Shishoo et al23 who showed more degradation of R in combination with H (18-21%). They have further observed that after oral administration of Rand H together, an appreciable amount of R is degraded in the acidic conditions of the stomach to 3-formyl rifampicin (3-FR), which is not absorbed and is inactive. 3-formyl rifampicin undergoes a reversible reaction to form a hydrazone with H. As a result, decomposition of R to 3-FR is pushed forward and an overall increase in the degradation of R is observed in the presence of H13. This observation is in agreement with that of Singh and co-workers24, who demonstrated degradation of R in the presence of H using a specific HPLC method showing a 3-FR hydrazone peak in the IIPEC analysis which was confirmed with a synthetic 3-FR hydrazone.

The results of the second investigation show that it is H per se and not the excipient present in H tablet, that is responsible for the decrease in the absorption of R. These findings are in agreement with those reported earlier9,14. Further, our observation that Z does not affect the bioavailability of R contradicts the findings of Jain et al6. Whether the decreased bioavailability of R exerts any therapeutic penalty in the presence of three other powerful anti-tuberculosis drugs during the initial intensive phase of treatment is a matter of conjecture. Combining H with R during the continuation phase is unlikely to have any deleterious effect as there would have been a substantial reduction in the bacterial load by then.

We also examined the bioavailability of E, H and Z in a total of 17 healthy individuals (including the 8 volunteers in the first investigation of this study). Results (not reported) showed that concomitant administration of the other drugs either alone or in combination did not in any way, affect the bioavailability of E, H or Z. This finding is in agreement with that reported by other workers25-27.

About 20 per cent of the administered dose of R is excreted as R plus DR over a 24 h period (TRC, unpublished findings). Of this, more than 70 per cent (i.e. about 14% of the dose) is excreted during the first 8 h period. Determination of plasma R concentrations is a complex process and the variation associated with these determinations is larger than that with the estimations of R plus DR in urine. Thus, on the basis of the results obtained in the eight volunteers following administration of R alone, the coefficient of variation associated with the AUC values (based on the determination of plasma concentrations) was 24.9 per cent as against coefficient of variations of 16.1 and 18.7 per cent with the estimation of R plus DR in urine collected over the periods 0-8 and 0-12 h respectively. Thus, the urine method reported in this paper, provides bioavailability indices similar to those based on plasma concentrations, which is in agreement with that reported by others28-30. The urine method, therefore, offers a simple and reliable non invasive procedure for investigations requiring an assessment of the bioavailability of R in double and triple drug formulations containing the drug.

Acknowledgment

The authors acknowledge Smt. S. Vijayalakshmi for technical assistance, and the volunteers for their cooperation.

References

1. Cavenaghi R. Rifampicin raw material characteristics and their effect of bioavailability. Bull lnt Union Tuberc Lung Dis 1989; 64 : 36-7.

2. Siegler Dl, Bryant M, Burley DM, Citron KM, Standen SM. Effect of meals on nfampicin absorption. Lancet 1974; ii : 197-8.

3. Velo GP, Vetturi B, Ricerche. Sull assorbimenlo orale e sulla eliminazione urinaria della rifampicine. Gaz Int di Med Chir 1968; 73: 2799-804.

4. Boman G, Lundgren P, Stjernstrom G. Mechanism of the inhibitory effect of PAS granules on the absorption of rifampicin: Adsorption of rifampicin by an excipient, bentonite. Eur J Clin Pharmacol 1975; 8 : 293-9.

5. Verbist L. Pharmacological study of rifampicin after repeated high dosage during intermittent combined therapy. I. Variation of the rifampicin serum levels (947 determinations). Respiration 1971; 28 (Suppl) : 7-16.

6. Jain A, Melha VL, Kiilshrestha S. Effect of pyrazinamide on rifampicin kinetics in patients with tuberculosis. Tuberc Lung Dis 1993: 77:87-90.

7. Boman G, Borga O, Hanngren A, Malmborg AS, Sjoqvist F. Pharmacokinetic interactions between the tuberculostatics rifampicin. para-aminosalicylic acid and isoniazid. Acta Pharmacol Toxicol (Copenh) 1970;28: 15.

8. Acocella G. Bonollo L, Garimoldi M, Mainardi M, Tenconi LT, Nicolis FB. Kinetics of rifampicin and isoniazid administered alone and in combination to normal subjects and patients with liver disease. Gut 1972; 13 : 47-53.

9. Mouton RR Mattie H, Swart K, Kreukniet J, de Wael J. Blood levels of rifampicin, desacetyl rifampicin and isoniazid during combined therapy. J Antimicrob Chemother 1979; 5 : 447-54.

10. Doshi BS. Bhate AD, Chauhan BL, Parkar TA, Kulkarni RD. Pharmacokinetic interaction of oral rifampicin and isoniazid in normal subjects. Indian Drugs. 1986; 23: 672-6.

11. Zwolska Z. Niemirowska-Mikulska H, Augustynowicz-Kopec E, Walkiewicz R, Stambrowska H, Safianowska A, et al. Bioavailability of rifampicin, isoniazid and pyrazinamide from fixed dose combination capsules. Int J Tuberc Lung Dis 1998; 2 : 824-30.

12. Padgaonkar KA, Rcvankar SN, Bhatt AD, Vaz JA, Desai ND, D'sa S, et al. Comparative bioequivalence study of rifampicin and isoniazid combinations in healthy volunteers. Int J Tuberc Lung Dis 1999; 3: 627-31.

13. Shishoo CJ, Shah SA, Rathod IS, Savale SS. Impaired bioavailability of rifampicin from fixed dose combination (FDC) formulations with isoniazid. Indian J Pharm Sd 2001; 63 : 443-9.

14. Grosset J. TrulTot-Pernot C, Eacroix C, Ji B. Antagonism between isoniazid and the combination of pyrazinamide-rifampirn against tuberculosis infection in mice. Antimicrob Agents Chemother 1992; 36 : 548-51.

15. Balasubramanian R. Sivasubramanian S, Vijayan VK, Ramachandran R, Jawahar MS, Paramasivan CN et al. Five year results of a 3-months and two 5-month regimens for the treatment of sputum-positive pulmonary tuberculosis in south India Tubercle 1990:71 : 253-8.

16. Dickinson JM, Aber VR, Allen BW, Ellard GA, Mitchison DA. Assay of rifampicin in serum. J Clin Pathol 1994; 27 : 457-62.

17. Immanuel C, Jayasankar K, Narayana AS,Sarma GR. Self-induction of rifampicin metabolism in man. Indian J Med Res 1985; 82 : 381-7.

18. Gibaldi M, Perrier D. Introduction to Pharmacokinetics. In: Pharmacokinetics. New York: Morcel Dekker; 1982 p. 1-13.

19. Acocella G. Human bioavailability studies. Bull Int Union Tuberc Lung Dis 1989; 64: 38-40.

20. Mitchison DA. Mechanisms of the action of drugs in the short course chemotherapy. Bull Int Union Tuberc 1985; 60 : 36-40.

21. Long MW, Snider DR Jr., Farer LS. US Public Health Service cooperative trial of three rifampicin-isoniazid regimens in treatment of pulmonary tuberculosis. Am Rev Respir Dis 1979; 119 : 879-94.

22. Jindal KC, Chaudhary RS, Singla AK, Gangwal SS, Khanna S. Dissolution test method for rifampicin 1/n isoniazid fixed dose formulations. J Pharm Biomed Anal 1994; 12 : 493-7.

23. Shishoo CJ, Shah SA, Rathod IS, Savale SS, Kotecha JS, Shah PB. Stability of rifampicin in dissolution medium in presence of isoniazid. Int J Pharmacol, 1999; 190 : 109-23.

24. Singh S, Mariappan TT, Sharda N, Kumar S, Chakraborti AK. The reason for an increase in decomposition of rifampicin in the presence of isoniazid under acid conditions. Pharm Pharmacol Commun 2000; 6:405-10.

25. Ellard GA, Ellard DR, Allen BW, Girling DJ, Nunn AJ, Teo SK, et al. The bioavailability of isoniazid, rifampirn and pyrazinamide in two commercially available combined formulations designed for the short-course treatment of tuberculosis. Am Rev Respir Dis 1986; 133: 1076-80.

26. Advenier C, Gobert C, Houin G, Bidet D, Richelet S, Tillement JP. Pharmacokinetic studies of rifampicin in the elderly. Ther Drug Monit 1983; 5 : 61-5.

27. Gelber R, Jacobsen P, Levy L. A study of the availability of six commercial formulations of isoniazid. Clin Pharmacol Ther 1969; 10:841-8.

28. Brechbuhler S, Fluehler H, Riess W, Theobald W. The renal elimination of rifampicin as a function of the oral dose. A convenient way to assess relative bioavailability. Arzneimi itelforschung 1978; 28 : 480-3.

29. Panchagnula R, Kaur KJ, Singh I, Kaul CL. The WHO simplified study protocol in practice: investigation of combined formulations supplied by the WHO. Int J Tuberc Lung Dis 1999; 3 (11 Suppl 3) : S336-42.

30. Pillai G, Ellard GA, Smith PJ, Fourie PB. The potential use of urinary excretion data for assessing the relative bioavailability of rifampicin in fixed dose combination anti-tuberculosis formulations. Int J Tuberc Lung Dis 2001; J : 691-5.

Chandra Immanuel, Prema Gurumurthy, Geetha Ramachandran, P. Venkatesan, V. Chandrasekaran & R. Prabhakar

Tuberculosis Research Centre (ICMR), Chennai, India

Received February 27, 2003

Reprint requests : Dr Geetha Ramachandran, Research Officer, Department of Biochemistry, Tuberculosis Research Centre Mayor V.R. Ramanathan Road, Chetput, Chennai 600031, India

e-mail : icmrtrc@vsnl.com, trcicmr@md3.vsnl.net.in (biochemistry)

Copyright Indian Council of Medical Research Sep 2003
Provided by ProQuest Information and Learning Company. All rights Reserved