Assessment of a Daily Combined Preparation of Isoniazid, Rifampin, and Pyrazinamide in a Controlled Trial of Three 6-Month Regimens for Smear-positive Pulmonary Tuberculosis 1
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
Introduction In the treatment of pulmonary tuberculosis it is the policy of the Singapore Tuberculosis Service to give every dose of chemotherapy under the full supervision of the nursing staff, almost all treatment being organized on an outpatient basis through 21 neighborhood community health clinics. A previous study in Singapore (1) showed that 6-month regimens of daily treatment with streptomycin, isoniazid, rifampin, and pyrazinamide for 2 months (2SHRZ), for 1 month (1SHRZ), or for 2 months without streptomycin (2HRZ), followed for all patients by three times weekly isoniazid and rifampin (H 3R3 ) for the remainder of the 6 months, were highly effective. All except 1 of 319 patients with drug-susceptible tubercle bacilli pretreatment had a favorable bacteriologic status at the end of chemotherapy, and during 5 years the total subsequent relapse rate was 2.411,10 of 297 patients (95% confidence limits 1.0 to 4.8%). A practical disadvantage of treatment with isoniazid, rifampin, and pyrazinamide when separate drug formulations are used is the number of tablets and capsules that must be assembled by the staff for the patient to swallow. Some patients can experience difficulty in swallowing, nausea, and even vomiting. In the present study, therefore, the same three regimens of chemotherapy were used, but during the initial daily phase of chemotherapy the isoniazid, rifampin, and pyrazinamide were allocated at random to be given either as standard separate formulations or in a combined formulation (Rifater 2; Gruppo Lepetit, Milan, Italy) to compare their acceptability, adverse effects, and therapeutic efficacy. The tablets of the combined formulation each contained 50 mg isoniazid, 120mg rifampin, and 300 mg pyrazinamide. They were formulated as firm, sugar-coated, biconvex tablets with the aim of being easy to
SUMMARY In a study in Singapore 310 patients with sputum smear-positive pulmonary tuberculosis were allocated at random to dally chemotherapy with streptomycin, isoniazid, rlfampln, and pyrazinamide (1) for 2 months (2SHRZ), (2) for 1 month (1SHRZ), or (3) for 2 months without streptomycin (2HRZ). This was followed for all patients by three times weekly Isoniazid and rlfampln to a total duration of 6 months. During the Initial period of dally chemotherapy the patients were also allocated at random to be given their HRZ either as a combined formulation (Rlfater), each tablet containing 50 mg Isoniazid, 120mg rlfampln, and 300 mg pyrazinamide, or as three separate drugs. During the Rlfater versus separate drugs comparison the most common spontaneous complaints were of nausea and vomiting, reported by 8% of 155 patients receiving Rlfater and 7% of 155separate drugs. Other adverse effects were also reported In similar proportions In the two series. Among 271 patients with drug-susceptible strains of tubercle bacilli pretreatment there were no bacteriologIc failures during chemotherapy. During 18 months of SUbsequent follow-up bacteriologic relapse occurred In 3 (7%) of 46 2SHRZ, 2 (5%) of 42 1SHRZ, and 3 (8%) of 40 2HRZ patients allocated to Rlfater and In 0 of 47 2SHRZ, 1 (2%) of 46 1SHRZ, and 1 (2%) of 44 2HRZ patients allocated to separate drugs. There was no evidence of therapeutic benefit from continuing SHRZ administration beyond 1 month or from adding streptomycin to HRZ. The relapse rates were slightly higher In the Rlfater series (p = 0.04). Further follow-up and results from other studies are therefore needed fully to assess the combined preparation. AM REV RESPIR DIS 1991; 143:707-712
swallow, and their drug contents were such that the number of tablets required for each dose ranged from 4 to 6 according to body weight. The bioavailability of the three drugs in the combined formulation had already been shown to be satisfactory in Singapore patients (2). Methods Patients and Treatment Regimens Patients of 15yr of age or more with pulmonary tuberculosis whose sputum was positive for acid-fast bacilli on smear examination and yielded Mycobacterium tuberculosis on culture and who had not received any previous antituberculosis chemotherapy were allocated at random to one of the following three 6-month regimens of chemotherapy. Initial Phase. (1) Streptomycin, isoniazid, rifampin, and pyrazinamide daily for 2 months (2SHRZ regimen); (2) the same as regimen 1 but given for only 1month (1SHRZ regimen); and (3) the same as regimen 1 but without streptomycin (2HRZ regimen). Continuation phase. The continuation phase was the same for all three regimens, namely isoniazid and rifampin three times a week for the remainder of the 6 months. In the initial daily phase the isoniazid, rifampin, and pyrazinamide were allocated at
random to be given either as the combined formulation, Rifater 2 (each tablet containing 50 mg isoniazid, 120 mg rifampin, and 300mg pyrazinamide), or as separate formulations. The dosages weresimilar whether they were given as Rifater or separately (table 1). The streptomycin dosage was 750 mg for all 2SHRZ and ISHRZ patients, regardless of body weight. During the intermittent phase patients weighing 42 kg or less received 6 tablets of isoniazid (600 mg), 43 to 57 kg, 8 tablets (800mg), and 58 kg or more, 10tablets (1,000 mg), and all received 2 capsules (600 mg) of rifampin. All patients received every dose of their chemotherapy under the direct supervision of community health clinic or (rarely) hospital staff.
Assessment of Acceptability, Compliance, and Toxicity Special care was taken to avoid bias in the way data were obtained and recorded on the (Received in originalform August 14, 1990 and in revised form November 6, 1990) I Correspondence and requests for reprints should be addressed to Dr. D. J. Girling, MRC Cardiothoracic Epidemiology Group, Brompton Hospital, Fulham Road, London SW3 6HP, UK or to Dr. S. K. Teo, Tan Tock Seng Hospital, Moulmein Road, Singapore II.
707
708
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
TABLE 1 DOSAGES OF ISONIAZID (H), RIFAMPIN (R), AND PYRAZINAMIDE (Z) IN THE COMBINED AND SEPARATE FORMULATIONS DURING THE INITIAL DAILY PHASE OF CHEMOTHERAPY Separate Formulations Patients by Body Weight
Combined Formulation (Tablets) H (50 mg)
R (120 mg)
Z (300 mg)
H (100 mg) Tablets
R (150 mg) Capsules
R (300 mg) Capsules
Z (500 mg) Tablets
3
3
3
450 11-16
1,500 36-52
42 kg or less 4
No. of tablets or capsules
A
Dose, mg Range, mglkg
(200 5-7
43 to 57 kg No. of tablets or capsules
480 11-17
1,200'1 29-42
300 7-10 3
2
3
300 5-7
600 11-14
1,500 26-35
3
2
4
300 3-5
600 7-10
2,000 22-34
5 A
Dose, mg Range. mglkg
(250 4-6
58 kg or more No. of tablets or capsules
600 11-14
1,500'1 26-35
6 A
Dose, mg Range, mglkg
acceptability to patients of the combined and separate formulations. Each time a patient attended for a supervised dose of treatment, the nurse administering the dose asked: "How are you getting on?" but avoided direct questions about treatment. If the patient made any complaints, she recorded the details on a special treatment card. If a patient did not attend for a dose, the nurse subsequently recorded the reason on the card. The patients were also seen and assessed by a clinician on admission, monthly as a routine, and at other times as necessary to assess, investigate, and treat possible adverse reactions. The clinician recorded on monthly reports the nature and severity of any adverse reactions and the reasons for any modification to chemotherapy.
Assessment of Therapeutic Progress Reports were made on each patient on admission, then monthly to 18months and then every 3 months. Five sputum specimens were examined bacteriologically before treatment (three in London and two in Singapore). Thereafter one specimen was examined monthly in London and one in Singapore at months I to 6 and then in Singapore once every time a progress report was completed to 18 months and two thereafter. If a patient could not produce sputum, pharyngeal secretions were obtained after clearing the back of the throat. Whenever a positive smear or culture result was obtained in the Singapore laboratory after the end of chemotherapy, one specimen was sent to the London laboratory each month for 3 months. All specimens were examined by smear and culture. In addition, in the London laboratory only, positive cultures were identified and tested for susceptibility to streptomycin, isoniazid, and rifampin. The laboratory methods and definitions of resistance have been reported previously (3, 4). A posteroanterior chest radiograph was obtained pretreatment and at 6 and 30 months. The serum alanine aminotransfer-
(300 3-5
720 8-12
1,800'1 20-31
ase (ALT) concentration was measured pretreatment and monthly during the first 8 months.
Results
Population Studied Between October 1983and August 1987, 310 patients were admitted to the study, 155allocated to receivethe combined and 155the separate formulations. Ofthe 310, 3 were excluded from the bacteriologic analyses for reasons present on admission and a further 19because they missed more than 1 wk of treatment during the first month, more than 2 wk during the first two months, or more than 6 wk in all because of adverse reactions (16 patients) or default (3 patients). There remained 288 patients in the bacteriologic analyses: 271 with fully drug-susceptible bacilli pretreatment and 17 (6OJo) with bacilli resistant to isoniazid (4 patients), streptomycin (10patients), or both drugs (3 patients). No patient had bacilli resistant to rifampin pretreatment. The analysis of complaints made by patients, of adverse reactions, and of default was based on all 155 patients allocated to the combined formulation (53 2SHRZ, 50 ISHRZ, and 52 2HRZ) and all 155 to separate drugs (53 2SHRZ, 51 ISHRZ, and 51 2HRZ).
Complaints Made Spontaneously by Patients During the first month of the daily phase of chemotherapy 13 (8%) of the 155 patients on the combined formulation and 15 (10OJo) of the 155 on the separate formulations made spontaneous complaints about their medication. Complaints were first made during the second month by
only 3 (3%) of 101 and 1 (1OJo) of 103 in the two groups, respectively, who were still receiving chemotherapy daily in the 2SHRZ and 2HRZ series. The most common complaints were nausea and vomiting, reported during the first month by 4 (3OJo) and 9 (6OJo), respectively, of patients on the combined formation and by 5 (3070) and 6 (4%) on separate drugs. Less common complaints included too many tablets and capsules, difficulty in swallowing, and tablets and capsules sticking in the throat. Most complaints were made only once, namely by 12 (8OJo) of the patients during treatment with the combined formulation and 17 (11070) the separate drugs. Thus the evidence is that the separate formulations were as acceptable to patients as the combined formulation.
Main Adverse Reactions Reported during the 6 Months The majority of adverse reactions were reported during the first month of chemotherapy, including 25 of the 39 gastrointestinal, 32 of the 44 cutaneous, and 27 of the 34 vestibular reactions. Moreover,the frequency of all types of reaction was similar in the 2SHRZ and ISHRZ regimens. The results for these two regimens have therefore been combined in table 2. Although the proportions of patients with reactions of any type ranged from 25OJo in the HRZ patients receiving separate drugs to 42% in those receiving the combined formulation, the results provide no consistent evidence that one method of formulation was associated with a lower risk of reactions than the other. Indeed, in the SHRZ series 35OJo of the patients on separate drugs had
709
COMBINED FORMULATION OF ISONIAZID, RIFAMPlN, AND PYRAZINAMIDE
TABLE 2 MAIN ADVERSE REACTIONS REPORTED DURING THE 6 MONTHS AND ACTION TAKEN IN THEIR MANAGEMENT· Action Taken in Management Patients with Reaction Reaction Any
Gastrointestinal
Cutaneous
Vestibular
Interruption Nil
7 Days
Total
Drugs
Formulation
Regimen
No.
%
No.
%
No.
%
No.
%
No.
%
S
H
R
Z
Combined
SHRZ HRZ
30 22
29 42
20 18
19 35
2
2 0
3 2
3 4
5 2
5 4
4
0
0 0
1 0
3 2
Separate
SHRZ HRZ
36 13
35 25
27 11
26 22
1 0
1 0
1 2
1 4
7 0
7 0
4
0
2
5
Combined
SHRZ HRZ
13 8
13 15
13 8
13 15
0 0
0 0
0 0
0 0
0 0
0 0
Separate
SHRZ HRZ
14 4
13 8
12
12 6
0 0
0 0
0 1
0 2
2 0
2 0
0
0
3
Combined
SHRZ HRZ
11 8
11 15
6 5
6 10
0 0
0 0
2 2
2 4
3 1
3 2
2
0 0
1 0
3 1
Separate
SHRZ HRZ
18 7
17 14
15 6
14 12
1 0
1 0
0 1
0 2
2 0
2 0
2
0
0
2
Combined
SHRZ HRZ
13 4
13 8
8 4
8
2 0
2 0
1 0
1 0
2 0
2 0
2
0
0
0
8
SHRZ HRZ
14 3
13 6
11
11 6
1 0
1 0
0 0
0 0
2 0
2 0
2
0
0
0
3
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1 4
1
0 0
0 0
0 0
0
0
0
0
0
0
0
0
0
0
0
0
Separate Hepatitis
Combined
HRZ
Separate
SHRZ
Arthralgia
Combined
SHRZ HRZ
Separate
SHRZ
1 4
1
8
8
2 0
2
0
• Based on all 155 patients on the combined formulation (103 SHRZ and 52 HRZ) and all 155 on separate drugs (104 SHRZ and 51 HRZ).
reactions compared with 29070 on the combined formulation. Vestibular reactions were, as expected, more common in patients receiving streptomycin than in those not receiving the drug, occurring in 27 (13070) of 207 compared with 7 (7070) of 103 (P = 0.14), but other reactions occurred with a similar frequency in the SHRZ and HRZ series. The frequency of all types of reaction was similar in the groups of patients given the combined formulation and separate drugs. Hepatitis occurred in only 2 patients: 1 (HRZ) had pyrazinamide withdrawn because of jaundice in Month 2 and the other (SHRZ) rifampin and pyrazinamide because of jaundice in Month 1. Arthralgia occurred in 6 patients during the initial daily phase of chemotherapy. None of the 6 required any modification to their chemotherapy.
Default During the Initial Daily Phase of Chemotherapy During the first month of daily chemotherapy, only 8 of the 310patients, namely 6 receiving the combined formulation (3 with streptomycin) and 2 on separate drugs (both with steptomycin), failed to
attend for treatment on one or more occasions. During the second month only 4 of the 209 patients, namely 2 receiving the combined formulation (1 with streptomycin) and 2 on separate drugs (1 with streptomycin), failed to attend. Moreover, 6 of the 12 patients who failed to attend missed only a single dose and a further 3 only 2 doses. Of the remaining 3, all of whom were receiving the combined formulation, 1 (2HRZ) missed 16doses, 1(2SHRZ) missed 57 doses, and the third (2HRZ) defaulted permanently during the first month. Thus there was a high level of compliance in all 3 regimens whether the 3 orally administered drugs were given combined or separately.
Therapeutic Results in Patients with Drug-susceptible Bacilli Pretreatment Condition on admission. The pretreatment distributions of sex, age, weight, and radiographic severity of disease were similar for all three regimens and for the patients receiving the combined formulation and separate drugs (table 3). On independent assessment of a posteroanterior chest radiograph the extent of disease was equivalent to the area of two-
thirds of one lung field or larger in 50%, and 40% of the patients had obvious cavitation. Bacteriologic response during chemotherapy. The proportions of patients with negative cultures (table 4) at months 1 and 2 weresimilar for the patients receiving the combined formulation and separate drugs. At 2 months over 90% of the patients in all six groups had negativecultures, and at 3 months all except one patient. There wereno bacteriologic failures during chemotherapy, no patient having more than one culture positive among the maximum of six available at Months 4, 5, and 6, and only 11 (four combined formulation series, seven separate drugs series) having one positive. Bacteriologic relapseafter chemotherapy. Bacteriologic relapse after chemotherapy was defined as a culture growing 10or more colonies at 2 different months during any 3 consecutive months to 18 months, or 4 months thereafter. Of the 271 patients with a favorable bacteriologic status at the end of chemotherapy, 6 were not assessable 18 months later: 1 had left the country, and 5 had died of causes unrelated to tuberculosis. During
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
710
18 months after the end of chemotherapy (table 5), bacteriologic relapse in patients allocated to the combined formulation occurred in 7C1fo of the patients in the 2SHRZ series, 5C1fo in the ISHRZ, and 8C1fo in the 2HRZ series. The corresponding results in patients allocated to separate drugs were 0, 2, and 2 C1fo, respectively.The proportions werecompared using a logistic model. There was no therapeutic benefit from continuing daily fourdrug (SHRZ) chemotherapy beyond 1 month or from adding streptomycin to three drugs (HRZ) in a 2-month initial daily phase (p > 0.5). The relapse rate was higher in the combined formulation series than when the three component drugs were given separately, but the level of statistical significance was borderline (p = 0.04). Of the 10 relapses 7 started during the first 6 months after the end of chemotherapy and the remaining 3 in Months 7 and 8.
TABLE 3 CONDITION ON ADMISSION OF PATIENTS WITH DRUG-SUSCEPTIBLE BACILLI Patients
Male Age, yr 15-24 25-34 35-44 45-54 ~ 55 Mean weight. kg (SO) Radiographic extent of disease on independent assessment, area* 4 cm2 to 1/6 of 1 lung field 1/6 to 1/3 of 1 lung field 1/3 to 2/3 of 1 lung field 2/3 to the whole of 1 lung field Larger Cavitation present Total patients assessed
No.
%
179
66
63 53 48 64 43 49.5 (9.8)
23 20 18 24 16
12 31 92 73 63 108 271
4 11 34 27 23 40 100
* Based on a single posteroanterior chest radiograph.
TABLE 4 CULTURE NEGATIVITY AT 1 AND 2 MONTHS IN PATIENTS WITH DRUG-SUSCEPTIBLE BACILLI PRETREATMENT*
Month
2
Culture Negative
Regimen
Formulation
Patients Assessed
2SHRZ
Combined Separate
47 47
35 34
74 } 72
1SHRZ
Combined Separate
42 47
32 33
76} 73 70
2HRZ
Combined Separate
41 47
28 29
68 } 62
65
2SHRZ
Combined Separate
46 47
46 45
100 } 96
98
1SHRZ
Combined Separate
42 46
39 42
93 } 91
92
2HRZ
Combined Separate
41 47
39 46
95 } 98
97
No.
%
* Based on the first or only sputum specimen at each month.
73
Therapeutic Results in Patients with Drug-resistant Bacilli Pretreatment Of the 4 patients with isoniazid-resistant bacilli pretreatment, 2 (both ISHRZ) received the combined formulation and 2 (lSHRZ and 2HRZ) separate drugs; of the 10 with streptomycin-resistant bacilli, 8 (l 2SHRZ, 4 ISHRZ, and 3 2HRZ) received the combined formulation and 2 (2SHRZ and ISHRZ) separate drugs, and all 3 of those with double-drug resistance (1 2SHRZ and 2 2HRZ) received separate drugs. Only 1 patient (2SHRZ, double-drug resistance) had an unfavorable bacteriologic status at the end of chemotherapy; this was associated with the emergence of rifampin resistance. The remaining 16 patients were all assessable 18 months after the end of chemotherapy. None of them relapsed during that period. Discussion
TABLE 5 BACTERIOLOGIC RELAPSES DURING 18 MONTHS AFTER THE END OF CHEMOTHERAPY IN PATIENTS WITH DRUG·SUSCEPTIBLE BACILLI PRETREATMENT
Formulation Combined
Separate
Relapses in 18 Months*
Regimen
Patients Assessed
No.
2SHRZ 1SHRZ 2HRZ
46 42 40
3 2 3
2SHRZ 1SHRZ 2HRZ
47 46 44
0 1 1
Months of Relapse
%
D 6
D'
3. 3, 5 3, 7 3,6,7 6 8
* All relapses occurred with strains of tubercle bacilli still susceptible to isoniazid, streptomycin, and rifampin.
The earlier study in Singapore (l) showed that the three regimens compared in the present study, namely streptomycin, isoniazid, rifampin, and pyrazinamide given daily (1) for 2 months, (2) for 1 month, or (3) for 2 months without streptomycin, followed for all patients by isoniazid and rifampin given three times a week to a total duration of 6 months, were highly effective when all three of the orally administered drugs were given in their standard separate formulations and were associated with a low risk of adverse effects. All three regimens are now in use as standard regimens in the Singapore ThberculosisService. The pres-
COMBINED FORMULATION OF ISONIAZID, RIFAMPIN, AND PYRAZINAMIDE
ent study has confirmed the efficacy of these three regimens when separate drug formulations were used. Among 141 patients with drug-susceptible strains of tubercle bacilli pretreatment there were no bacteriologic failures during chemotherapy, and among 137 assessed 18 months after the end of chemotherapy only 2 (1010) experienced bacteriologic relapse (95010 confidence limits 0.2 to 5.2010). The results were less good when the combined formulation of isoniazid, rifampin, and pyrazinamide, namely Rifater 2, was used during the initial daily phase of chemotherapy, although it was given for only the first 1 or 2 months of chemotherapy. All 130patients assessed had a favorable bacteriologic response during chemotherapy, but bacteriologic relapse occurred during the next 18 months in 8 (6.3010) of 128 (95010 confidence limits 2.8 to 12.0; p = 0.04 for the comparison with the patients treated with separate drugs). There is no obvious reason for this difference. The drug dosages in mg/kg were similar in the two groups (see table 1), and the bioavailabilities of all three drugs in the combined formulation weresatisfactory and similar to those of the separate drugs when evaluated before the start of the intake to the study (2). With the comparatively small numbers of patients in the present study, the difference cannot be accepted as firm evidence that the combined formulation is therapeutically inferior. Nevertheless,the preparation used, marketed by Merrell Dow, is widely available for daily use in the treatment of tuberculosis, and it is therefore desirable that other groups both assess it in randomized comparisons with the same drugs given separately in patients with smear-positive pulmonary tuberculosis and report on the results achieved with its use in service programs. In the United States Public Health Service study 21,the use of Rifater compared with separate drugs in a three-drug, 6-month regimen was associated with a more rapid rate of sputum conversion to culture negativity and a very similar, and low, relapse rate (5), but the comparison was not completely concurrent and the Rifater preparation was a different one, each tablet containing 75 mg isoniazid, 150mg rifampin, and 400 mg pyrazinamide (6). The patients in the present study are being followed up to 5 yr from admission. Combined formulations, provided they are shown to be therapeutically effective, have many potential advantages
over separate drugs. With combined formulations prescribing is simpler. Clinicians are less likely to make errors in prescribing, and patients are less likely to be confused about the numbers of tablets and capsules to take in selfadministered chemotherapy. Also, dosages can more easily be adjusted according to the patient's weight, the mg/kg dosages of all the component drugs remaining similar for all patients, whatever their weight; the number of pills to be swallowed is reduced, possiblyencouraging a higher level of compliance, and the component drugs are always taken together at the same time, eliminating the risk of monotherapy. The prescribing of the daily Rifater preparation is particularly simple (7); one tablet is prescribed for every 10 kg of body weight. By this means the dose range in mg/kg varies little, namely for isoniazid from 4.3 to 5.5, for rifampin from 10.3 to 13.3, and for pyrazinamide from 25.7 to 33.3. In the present study special care was taken to avoid bias in the way the data on the acceptability of the preparations studied were obtained because patients, nurses, and medical staff may be biased in their attitudes to "new" drug preparations. In the event, patients made few complaints about the size or number of tablets and capsules or about difficulty in swallowing them, and only 8010 receiving the combined formulation and 7010 on separate drugs complained of nausea (3010 and 5010, respectively, with vomiting), suggesting that the combined formulation and the separate drugs were equally acceptable. When isoniazid, rifampin, and pyrazinamide are given intermittently, larger dosages of isoniazid and pyrazinamide are necessary than in daily regimens, and hence larger numbers of tablets and capsules in each dose. In a study of intermittent chemotherapy in Hong Kong (8), the combined formulation, which in this case was designed for intermittent use and was not the same as that used in the present study, appeared to be more acceptable to patients than the three drugs given separately, but the evidence was slender: 5010 of the patients receiving separate drugs compared with 1010 receiving the combined formulation complained about the size and number of their tablets and capsules or about difficulty in swallowing them, and 45010 compared with 32010 brought their own drink to the clinic to help them swallow their tablets and capsules, a not uncommon practice in Hong Kong.
711
Despite there being little objective, unbiased evidence that patients find combined preparations more acceptable than separate drugs, their use is recommended by the Committee on Treatment of the International Union Against Thberculosis and Lung Disease (lUATLD) (9) on the grounds of the advantages discussed previously. It is important, however,that only those preparations are used that have been shown to be adequately absorbed from the gastrointestinal tract. Tests of the human bioavailability of three brands of rifampin and isoniazid combined and four of isoniazid, rifampin, and pyrazinamide combined, all intended for daily use and commercially available in developing countries, have shown that one two-drug and all four of the three-drug formulations produced unacceptably low plasma rifampin concentrations (10). Both the American Thoracic Society (ATS) (11) and the IUATLD (9) recommend a 6-month regimen of isoniazid, rifampin, and pyrazinamide for 2 months (without a fourth drug), followed by isoniazid and rifampin for 4 months, in the treatment of patients with active tuberculosis who are known or likely to be infected with fully drug-susceptible strains. As far as we are aware, the present and the previous study in Singapore (1) are the only studies that have shown this regimen to be effective in a randomized controlled comparison with a regimen in which four drugs were given for the initial 2-month intensive phase, and in these Singapore studies the continuation phase was given three times a week, not daily. It is therefore desirable that results obtained with the use of the three-drug recommended regimen, both in controlled trials and in service programs, are published in the medical literature. Both the American Thoracic Society and the IUATLD recommend adding a fourth drug to the initial phase if initial drug resistance is suspected, and the IUATLD if the regimen is given intermittently from the start. There is now evidence from a study in Hong Kong that the recommended regimen may be marginally inferior therapeutically if it is given intermittently from the start without a fourth drug (streptomycin in the Hong Kong study) during the initial intensive phase (12). In view of the comparatively small numbers of patients in the previous study in Singapore (l) and in the present study, it is still possible that the fourdrug combination may prove to be marginally superior to the recommended
712
three-drug combination, even in daily regimens, if comparative studies could be conducted in sufficiently large numbers of patients. Such studies are highly desirable because even a small deficiency in a widely recommended and widely used regimen would be important. Acknowledgment In Singapore the following physicians participated in the study: from 'Ian Tock Seng Hospital, Dr. Ng Kwok Choy, Dr. Poh Soo Chuan, Dr. F. J. Jayaratnam, Dr. Lee SiewKhow, Dr. Teo Seng Kee, Dr. lan Tiong Har, and Mrs. K. L. Lo (bacteriologist); from the Department of Thberculosis Control, Dr. S. Devi and Dr. Thn Kai Kiat. In London the bacteriologic laboratory was the Medical Research Council Unit for Laboratory Studies of Thberculosis(subsequently the Department of Bacteriology, Royal Postgraduate Medical School, London; Professor D. A. Mitchison, Mr. V. R. Aber, and Mr. B. W. Allen). Dr. I. H. Kerr was the radiologic assessor. The investigation was coordinated in Singapore by Dr. Teo Seng Kee, and the independent assessor was Dr. Chew Chin Hin. In London the coordinator was Dr. D. J. Girling of the Medical Research Council Thberculosis and Chest Diseases Unit (now the MRC Cardiothoracic Epidemiology Group). The report was prepared on behalf of the cooperating physicians by Dr. D. J. Girling, Mr. A. J. Nunn, Professor Wallace Fox, and Professor
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
D. A. Mitchison. The participants are gratetul to Dr. Kwa Soon Bee, Permanent Secretary (Health)/Director of Medical Services, Singapore, for his encouragement and support and the Medical Clinical Research Committee, Ministry of Health, for a research grant. The study would not have been possible but for the dedicated work of Nurses Foo Gwek Eng and Chang Seow Kiak. Gruppo Lepetit of Milan provided the Rifater and Bracco Industria Chimica of Milan the pyrazinamide as gifts.
References 1. Singapore Tuberculosis Service/British Medical Research Council. Five-yearfollow-up of a clinical trial of three 6-month regimens of chemotherapy given intermittently in the continuation phase in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 1988; 137:1147-50. 2. Ellard GA, Ellard DR, Allen BW, et al. The bioavailability of isoniazid, rifampin, and pyrazinamide in two commercially available combined formulations designed for use in the short-course treatment of tuberculosis. Am Rev Respir Dis 1986; 133:1076-80. 3. Singapore Thberculosis Service/British Medical Research Council. Clinical trial of six-month and four-month regimens of chemotherapy in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 1979; 119:579-85. 4. Singapore Thberculosis Service/British Medical ResearchCouncil. Clinicaltrial of three 6-month regimens of chemotherapy given intermittently in the continuation phase in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 1985; 132: 374-8.
5. Combs DL, O'Brien RJ, Geiter LJ. USPHS tuberculosis short-course chemotherapy trial 21:effectiveness, toxicity, and acceptability: the report of final results. Ann Intern Moo 1990;112:397-406. 6. Geiter LJ, O'Brien RJ, Combs DL, Snider DE. United States Public Health Service Thberculosis therapy trial 21: preliminary results of an evaluation of a combined tablet of isoniazid, rifampin and pyrazinamide. Thbercle 1987;68(Suppl: 41-6). 7. Acocella G, Angel JH. Short-course chemotherapy of pulmonary tuberculosis: a new approach to drug dosage in the intial intensive phase. Am Rev Respir Dis 1986; 134:1283-6. 8. Hong Kong Chest Service/British Medical Research Council. Acceptability, compliance, and adverse reactions when isoniazid, rifampin, and pyrazinamide are given as a combined formulation or separately during three-times-weekly antituberculosis chemotherapy. Am Rev Respir Dis 1989; 140:1618-22. 9. International Union Against Thberculosis and Lung Disease. Antituberculosis regimens of chemotherapy: recommendations from the Committee on Treatment. Bull IUATLD 1988;63(2):60-4. 10. Acocella G. Quality control of antituberculosis drugs: human bioavailability studies. Bull IUATLD 1989; 64(1):38-40. 11. American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children: joint statement of the American Thoracic Society and the Centers for Disease Control. Am Rev Respir Dis 1986; 134:355-63. 12. Hong Kong Chest Service/British Medical Research Council. Controlled trial of 2, 4, and 6 months of pyrazinamide in 6-month 3-times-weekly regimens for smear-positive pulmonary tuberculosis, including an assessment of a combined preparation of isoniazid, rifampin, and pyrazinamide. Am Rev Respir Dis.
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
Introduction In the treatment of pulmonary tuberculosis it is the policy of the Singapore Tuberculosis Service to give every dose of chemotherapy under the full supervision of the nursing staff, almost all treatment being organized on an outpatient basis through 21 neighborhood community health clinics. A previous study in Singapore (1) showed that 6-month regimens of daily treatment with streptomycin, isoniazid, rifampin, and pyrazinamide for 2 months (2SHRZ), for 1 month (1SHRZ), or for 2 months without streptomycin (2HRZ), followed for all patients by three times weekly isoniazid and rifampin (H 3R3 ) for the remainder of the 6 months, were highly effective. All except 1 of 319 patients with drug-susceptible tubercle bacilli pretreatment had a favorable bacteriologic status at the end of chemotherapy, and during 5 years the total subsequent relapse rate was 2.411,10 of 297 patients (95% confidence limits 1.0 to 4.8%). A practical disadvantage of treatment with isoniazid, rifampin, and pyrazinamide when separate drug formulations are used is the number of tablets and capsules that must be assembled by the staff for the patient to swallow. Some patients can experience difficulty in swallowing, nausea, and even vomiting. In the present study, therefore, the same three regimens of chemotherapy were used, but during the initial daily phase of chemotherapy the isoniazid, rifampin, and pyrazinamide were allocated at random to be given either as standard separate formulations or in a combined formulation (Rifater 2; Gruppo Lepetit, Milan, Italy) to compare their acceptability, adverse effects, and therapeutic efficacy. The tablets of the combined formulation each contained 50 mg isoniazid, 120mg rifampin, and 300 mg pyrazinamide. They were formulated as firm, sugar-coated, biconvex tablets with the aim of being easy to
SUMMARY In a study in Singapore 310 patients with sputum smear-positive pulmonary tuberculosis were allocated at random to dally chemotherapy with streptomycin, isoniazid, rlfampln, and pyrazinamide (1) for 2 months (2SHRZ), (2) for 1 month (1SHRZ), or (3) for 2 months without streptomycin (2HRZ). This was followed for all patients by three times weekly Isoniazid and rlfampln to a total duration of 6 months. During the Initial period of dally chemotherapy the patients were also allocated at random to be given their HRZ either as a combined formulation (Rlfater), each tablet containing 50 mg Isoniazid, 120mg rlfampln, and 300 mg pyrazinamide, or as three separate drugs. During the Rlfater versus separate drugs comparison the most common spontaneous complaints were of nausea and vomiting, reported by 8% of 155 patients receiving Rlfater and 7% of 155separate drugs. Other adverse effects were also reported In similar proportions In the two series. Among 271 patients with drug-susceptible strains of tubercle bacilli pretreatment there were no bacteriologIc failures during chemotherapy. During 18 months of SUbsequent follow-up bacteriologic relapse occurred In 3 (7%) of 46 2SHRZ, 2 (5%) of 42 1SHRZ, and 3 (8%) of 40 2HRZ patients allocated to Rlfater and In 0 of 47 2SHRZ, 1 (2%) of 46 1SHRZ, and 1 (2%) of 44 2HRZ patients allocated to separate drugs. There was no evidence of therapeutic benefit from continuing SHRZ administration beyond 1 month or from adding streptomycin to HRZ. The relapse rates were slightly higher In the Rlfater series (p = 0.04). Further follow-up and results from other studies are therefore needed fully to assess the combined preparation. AM REV RESPIR DIS 1991; 143:707-712
swallow, and their drug contents were such that the number of tablets required for each dose ranged from 4 to 6 according to body weight. The bioavailability of the three drugs in the combined formulation had already been shown to be satisfactory in Singapore patients (2). Methods Patients and Treatment Regimens Patients of 15yr of age or more with pulmonary tuberculosis whose sputum was positive for acid-fast bacilli on smear examination and yielded Mycobacterium tuberculosis on culture and who had not received any previous antituberculosis chemotherapy were allocated at random to one of the following three 6-month regimens of chemotherapy. Initial Phase. (1) Streptomycin, isoniazid, rifampin, and pyrazinamide daily for 2 months (2SHRZ regimen); (2) the same as regimen 1 but given for only 1month (1SHRZ regimen); and (3) the same as regimen 1 but without streptomycin (2HRZ regimen). Continuation phase. The continuation phase was the same for all three regimens, namely isoniazid and rifampin three times a week for the remainder of the 6 months. In the initial daily phase the isoniazid, rifampin, and pyrazinamide were allocated at
random to be given either as the combined formulation, Rifater 2 (each tablet containing 50 mg isoniazid, 120 mg rifampin, and 300mg pyrazinamide), or as separate formulations. The dosages weresimilar whether they were given as Rifater or separately (table 1). The streptomycin dosage was 750 mg for all 2SHRZ and ISHRZ patients, regardless of body weight. During the intermittent phase patients weighing 42 kg or less received 6 tablets of isoniazid (600 mg), 43 to 57 kg, 8 tablets (800mg), and 58 kg or more, 10tablets (1,000 mg), and all received 2 capsules (600 mg) of rifampin. All patients received every dose of their chemotherapy under the direct supervision of community health clinic or (rarely) hospital staff.
Assessment of Acceptability, Compliance, and Toxicity Special care was taken to avoid bias in the way data were obtained and recorded on the (Received in originalform August 14, 1990 and in revised form November 6, 1990) I Correspondence and requests for reprints should be addressed to Dr. D. J. Girling, MRC Cardiothoracic Epidemiology Group, Brompton Hospital, Fulham Road, London SW3 6HP, UK or to Dr. S. K. Teo, Tan Tock Seng Hospital, Moulmein Road, Singapore II.
707
708
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
TABLE 1 DOSAGES OF ISONIAZID (H), RIFAMPIN (R), AND PYRAZINAMIDE (Z) IN THE COMBINED AND SEPARATE FORMULATIONS DURING THE INITIAL DAILY PHASE OF CHEMOTHERAPY Separate Formulations Patients by Body Weight
Combined Formulation (Tablets) H (50 mg)
R (120 mg)
Z (300 mg)
H (100 mg) Tablets
R (150 mg) Capsules
R (300 mg) Capsules
Z (500 mg) Tablets
3
3
3
450 11-16
1,500 36-52
42 kg or less 4
No. of tablets or capsules
A
Dose, mg Range, mglkg
(200 5-7
43 to 57 kg No. of tablets or capsules
480 11-17
1,200'1 29-42
300 7-10 3
2
3
300 5-7
600 11-14
1,500 26-35
3
2
4
300 3-5
600 7-10
2,000 22-34
5 A
Dose, mg Range. mglkg
(250 4-6
58 kg or more No. of tablets or capsules
600 11-14
1,500'1 26-35
6 A
Dose, mg Range, mglkg
acceptability to patients of the combined and separate formulations. Each time a patient attended for a supervised dose of treatment, the nurse administering the dose asked: "How are you getting on?" but avoided direct questions about treatment. If the patient made any complaints, she recorded the details on a special treatment card. If a patient did not attend for a dose, the nurse subsequently recorded the reason on the card. The patients were also seen and assessed by a clinician on admission, monthly as a routine, and at other times as necessary to assess, investigate, and treat possible adverse reactions. The clinician recorded on monthly reports the nature and severity of any adverse reactions and the reasons for any modification to chemotherapy.
Assessment of Therapeutic Progress Reports were made on each patient on admission, then monthly to 18months and then every 3 months. Five sputum specimens were examined bacteriologically before treatment (three in London and two in Singapore). Thereafter one specimen was examined monthly in London and one in Singapore at months I to 6 and then in Singapore once every time a progress report was completed to 18 months and two thereafter. If a patient could not produce sputum, pharyngeal secretions were obtained after clearing the back of the throat. Whenever a positive smear or culture result was obtained in the Singapore laboratory after the end of chemotherapy, one specimen was sent to the London laboratory each month for 3 months. All specimens were examined by smear and culture. In addition, in the London laboratory only, positive cultures were identified and tested for susceptibility to streptomycin, isoniazid, and rifampin. The laboratory methods and definitions of resistance have been reported previously (3, 4). A posteroanterior chest radiograph was obtained pretreatment and at 6 and 30 months. The serum alanine aminotransfer-
(300 3-5
720 8-12
1,800'1 20-31
ase (ALT) concentration was measured pretreatment and monthly during the first 8 months.
Results
Population Studied Between October 1983and August 1987, 310 patients were admitted to the study, 155allocated to receivethe combined and 155the separate formulations. Ofthe 310, 3 were excluded from the bacteriologic analyses for reasons present on admission and a further 19because they missed more than 1 wk of treatment during the first month, more than 2 wk during the first two months, or more than 6 wk in all because of adverse reactions (16 patients) or default (3 patients). There remained 288 patients in the bacteriologic analyses: 271 with fully drug-susceptible bacilli pretreatment and 17 (6OJo) with bacilli resistant to isoniazid (4 patients), streptomycin (10patients), or both drugs (3 patients). No patient had bacilli resistant to rifampin pretreatment. The analysis of complaints made by patients, of adverse reactions, and of default was based on all 155 patients allocated to the combined formulation (53 2SHRZ, 50 ISHRZ, and 52 2HRZ) and all 155 to separate drugs (53 2SHRZ, 51 ISHRZ, and 51 2HRZ).
Complaints Made Spontaneously by Patients During the first month of the daily phase of chemotherapy 13 (8%) of the 155 patients on the combined formulation and 15 (10OJo) of the 155 on the separate formulations made spontaneous complaints about their medication. Complaints were first made during the second month by
only 3 (3%) of 101 and 1 (1OJo) of 103 in the two groups, respectively, who were still receiving chemotherapy daily in the 2SHRZ and 2HRZ series. The most common complaints were nausea and vomiting, reported during the first month by 4 (3OJo) and 9 (6OJo), respectively, of patients on the combined formation and by 5 (3070) and 6 (4%) on separate drugs. Less common complaints included too many tablets and capsules, difficulty in swallowing, and tablets and capsules sticking in the throat. Most complaints were made only once, namely by 12 (8OJo) of the patients during treatment with the combined formulation and 17 (11070) the separate drugs. Thus the evidence is that the separate formulations were as acceptable to patients as the combined formulation.
Main Adverse Reactions Reported during the 6 Months The majority of adverse reactions were reported during the first month of chemotherapy, including 25 of the 39 gastrointestinal, 32 of the 44 cutaneous, and 27 of the 34 vestibular reactions. Moreover,the frequency of all types of reaction was similar in the 2SHRZ and ISHRZ regimens. The results for these two regimens have therefore been combined in table 2. Although the proportions of patients with reactions of any type ranged from 25OJo in the HRZ patients receiving separate drugs to 42% in those receiving the combined formulation, the results provide no consistent evidence that one method of formulation was associated with a lower risk of reactions than the other. Indeed, in the SHRZ series 35OJo of the patients on separate drugs had
709
COMBINED FORMULATION OF ISONIAZID, RIFAMPlN, AND PYRAZINAMIDE
TABLE 2 MAIN ADVERSE REACTIONS REPORTED DURING THE 6 MONTHS AND ACTION TAKEN IN THEIR MANAGEMENT· Action Taken in Management Patients with Reaction Reaction Any
Gastrointestinal
Cutaneous
Vestibular
Interruption Nil
7 Days
Total
Drugs
Formulation
Regimen
No.
%
No.
%
No.
%
No.
%
No.
%
S
H
R
Z
Combined
SHRZ HRZ
30 22
29 42
20 18
19 35
2
2 0
3 2
3 4
5 2
5 4
4
0
0 0
1 0
3 2
Separate
SHRZ HRZ
36 13
35 25
27 11
26 22
1 0
1 0
1 2
1 4
7 0
7 0
4
0
2
5
Combined
SHRZ HRZ
13 8
13 15
13 8
13 15
0 0
0 0
0 0
0 0
0 0
0 0
Separate
SHRZ HRZ
14 4
13 8
12
12 6
0 0
0 0
0 1
0 2
2 0
2 0
0
0
3
Combined
SHRZ HRZ
11 8
11 15
6 5
6 10
0 0
0 0
2 2
2 4
3 1
3 2
2
0 0
1 0
3 1
Separate
SHRZ HRZ
18 7
17 14
15 6
14 12
1 0
1 0
0 1
0 2
2 0
2 0
2
0
0
2
Combined
SHRZ HRZ
13 4
13 8
8 4
8
2 0
2 0
1 0
1 0
2 0
2 0
2
0
0
0
8
SHRZ HRZ
14 3
13 6
11
11 6
1 0
1 0
0 0
0 0
2 0
2 0
2
0
0
0
3
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1 4
1
0 0
0 0
0 0
0
0
0
0
0
0
0
0
0
0
0
0
Separate Hepatitis
Combined
HRZ
Separate
SHRZ
Arthralgia
Combined
SHRZ HRZ
Separate
SHRZ
1 4
1
8
8
2 0
2
0
• Based on all 155 patients on the combined formulation (103 SHRZ and 52 HRZ) and all 155 on separate drugs (104 SHRZ and 51 HRZ).
reactions compared with 29070 on the combined formulation. Vestibular reactions were, as expected, more common in patients receiving streptomycin than in those not receiving the drug, occurring in 27 (13070) of 207 compared with 7 (7070) of 103 (P = 0.14), but other reactions occurred with a similar frequency in the SHRZ and HRZ series. The frequency of all types of reaction was similar in the groups of patients given the combined formulation and separate drugs. Hepatitis occurred in only 2 patients: 1 (HRZ) had pyrazinamide withdrawn because of jaundice in Month 2 and the other (SHRZ) rifampin and pyrazinamide because of jaundice in Month 1. Arthralgia occurred in 6 patients during the initial daily phase of chemotherapy. None of the 6 required any modification to their chemotherapy.
Default During the Initial Daily Phase of Chemotherapy During the first month of daily chemotherapy, only 8 of the 310patients, namely 6 receiving the combined formulation (3 with streptomycin) and 2 on separate drugs (both with steptomycin), failed to
attend for treatment on one or more occasions. During the second month only 4 of the 209 patients, namely 2 receiving the combined formulation (1 with streptomycin) and 2 on separate drugs (1 with streptomycin), failed to attend. Moreover, 6 of the 12 patients who failed to attend missed only a single dose and a further 3 only 2 doses. Of the remaining 3, all of whom were receiving the combined formulation, 1 (2HRZ) missed 16doses, 1(2SHRZ) missed 57 doses, and the third (2HRZ) defaulted permanently during the first month. Thus there was a high level of compliance in all 3 regimens whether the 3 orally administered drugs were given combined or separately.
Therapeutic Results in Patients with Drug-susceptible Bacilli Pretreatment Condition on admission. The pretreatment distributions of sex, age, weight, and radiographic severity of disease were similar for all three regimens and for the patients receiving the combined formulation and separate drugs (table 3). On independent assessment of a posteroanterior chest radiograph the extent of disease was equivalent to the area of two-
thirds of one lung field or larger in 50%, and 40% of the patients had obvious cavitation. Bacteriologic response during chemotherapy. The proportions of patients with negative cultures (table 4) at months 1 and 2 weresimilar for the patients receiving the combined formulation and separate drugs. At 2 months over 90% of the patients in all six groups had negativecultures, and at 3 months all except one patient. There wereno bacteriologic failures during chemotherapy, no patient having more than one culture positive among the maximum of six available at Months 4, 5, and 6, and only 11 (four combined formulation series, seven separate drugs series) having one positive. Bacteriologic relapseafter chemotherapy. Bacteriologic relapse after chemotherapy was defined as a culture growing 10or more colonies at 2 different months during any 3 consecutive months to 18 months, or 4 months thereafter. Of the 271 patients with a favorable bacteriologic status at the end of chemotherapy, 6 were not assessable 18 months later: 1 had left the country, and 5 had died of causes unrelated to tuberculosis. During
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
710
18 months after the end of chemotherapy (table 5), bacteriologic relapse in patients allocated to the combined formulation occurred in 7C1fo of the patients in the 2SHRZ series, 5C1fo in the ISHRZ, and 8C1fo in the 2HRZ series. The corresponding results in patients allocated to separate drugs were 0, 2, and 2 C1fo, respectively.The proportions werecompared using a logistic model. There was no therapeutic benefit from continuing daily fourdrug (SHRZ) chemotherapy beyond 1 month or from adding streptomycin to three drugs (HRZ) in a 2-month initial daily phase (p > 0.5). The relapse rate was higher in the combined formulation series than when the three component drugs were given separately, but the level of statistical significance was borderline (p = 0.04). Of the 10 relapses 7 started during the first 6 months after the end of chemotherapy and the remaining 3 in Months 7 and 8.
TABLE 3 CONDITION ON ADMISSION OF PATIENTS WITH DRUG-SUSCEPTIBLE BACILLI Patients
Male Age, yr 15-24 25-34 35-44 45-54 ~ 55 Mean weight. kg (SO) Radiographic extent of disease on independent assessment, area* 4 cm2 to 1/6 of 1 lung field 1/6 to 1/3 of 1 lung field 1/3 to 2/3 of 1 lung field 2/3 to the whole of 1 lung field Larger Cavitation present Total patients assessed
No.
%
179
66
63 53 48 64 43 49.5 (9.8)
23 20 18 24 16
12 31 92 73 63 108 271
4 11 34 27 23 40 100
* Based on a single posteroanterior chest radiograph.
TABLE 4 CULTURE NEGATIVITY AT 1 AND 2 MONTHS IN PATIENTS WITH DRUG-SUSCEPTIBLE BACILLI PRETREATMENT*
Month
2
Culture Negative
Regimen
Formulation
Patients Assessed
2SHRZ
Combined Separate
47 47
35 34
74 } 72
1SHRZ
Combined Separate
42 47
32 33
76} 73 70
2HRZ
Combined Separate
41 47
28 29
68 } 62
65
2SHRZ
Combined Separate
46 47
46 45
100 } 96
98
1SHRZ
Combined Separate
42 46
39 42
93 } 91
92
2HRZ
Combined Separate
41 47
39 46
95 } 98
97
No.
%
* Based on the first or only sputum specimen at each month.
73
Therapeutic Results in Patients with Drug-resistant Bacilli Pretreatment Of the 4 patients with isoniazid-resistant bacilli pretreatment, 2 (both ISHRZ) received the combined formulation and 2 (lSHRZ and 2HRZ) separate drugs; of the 10 with streptomycin-resistant bacilli, 8 (l 2SHRZ, 4 ISHRZ, and 3 2HRZ) received the combined formulation and 2 (2SHRZ and ISHRZ) separate drugs, and all 3 of those with double-drug resistance (1 2SHRZ and 2 2HRZ) received separate drugs. Only 1 patient (2SHRZ, double-drug resistance) had an unfavorable bacteriologic status at the end of chemotherapy; this was associated with the emergence of rifampin resistance. The remaining 16 patients were all assessable 18 months after the end of chemotherapy. None of them relapsed during that period. Discussion
TABLE 5 BACTERIOLOGIC RELAPSES DURING 18 MONTHS AFTER THE END OF CHEMOTHERAPY IN PATIENTS WITH DRUG·SUSCEPTIBLE BACILLI PRETREATMENT
Formulation Combined
Separate
Relapses in 18 Months*
Regimen
Patients Assessed
No.
2SHRZ 1SHRZ 2HRZ
46 42 40
3 2 3
2SHRZ 1SHRZ 2HRZ
47 46 44
0 1 1
Months of Relapse
%
D 6
D'
3. 3, 5 3, 7 3,6,7 6 8
* All relapses occurred with strains of tubercle bacilli still susceptible to isoniazid, streptomycin, and rifampin.
The earlier study in Singapore (l) showed that the three regimens compared in the present study, namely streptomycin, isoniazid, rifampin, and pyrazinamide given daily (1) for 2 months, (2) for 1 month, or (3) for 2 months without streptomycin, followed for all patients by isoniazid and rifampin given three times a week to a total duration of 6 months, were highly effective when all three of the orally administered drugs were given in their standard separate formulations and were associated with a low risk of adverse effects. All three regimens are now in use as standard regimens in the Singapore ThberculosisService. The pres-
COMBINED FORMULATION OF ISONIAZID, RIFAMPIN, AND PYRAZINAMIDE
ent study has confirmed the efficacy of these three regimens when separate drug formulations were used. Among 141 patients with drug-susceptible strains of tubercle bacilli pretreatment there were no bacteriologic failures during chemotherapy, and among 137 assessed 18 months after the end of chemotherapy only 2 (1010) experienced bacteriologic relapse (95010 confidence limits 0.2 to 5.2010). The results were less good when the combined formulation of isoniazid, rifampin, and pyrazinamide, namely Rifater 2, was used during the initial daily phase of chemotherapy, although it was given for only the first 1 or 2 months of chemotherapy. All 130patients assessed had a favorable bacteriologic response during chemotherapy, but bacteriologic relapse occurred during the next 18 months in 8 (6.3010) of 128 (95010 confidence limits 2.8 to 12.0; p = 0.04 for the comparison with the patients treated with separate drugs). There is no obvious reason for this difference. The drug dosages in mg/kg were similar in the two groups (see table 1), and the bioavailabilities of all three drugs in the combined formulation weresatisfactory and similar to those of the separate drugs when evaluated before the start of the intake to the study (2). With the comparatively small numbers of patients in the present study, the difference cannot be accepted as firm evidence that the combined formulation is therapeutically inferior. Nevertheless,the preparation used, marketed by Merrell Dow, is widely available for daily use in the treatment of tuberculosis, and it is therefore desirable that other groups both assess it in randomized comparisons with the same drugs given separately in patients with smear-positive pulmonary tuberculosis and report on the results achieved with its use in service programs. In the United States Public Health Service study 21,the use of Rifater compared with separate drugs in a three-drug, 6-month regimen was associated with a more rapid rate of sputum conversion to culture negativity and a very similar, and low, relapse rate (5), but the comparison was not completely concurrent and the Rifater preparation was a different one, each tablet containing 75 mg isoniazid, 150mg rifampin, and 400 mg pyrazinamide (6). The patients in the present study are being followed up to 5 yr from admission. Combined formulations, provided they are shown to be therapeutically effective, have many potential advantages
over separate drugs. With combined formulations prescribing is simpler. Clinicians are less likely to make errors in prescribing, and patients are less likely to be confused about the numbers of tablets and capsules to take in selfadministered chemotherapy. Also, dosages can more easily be adjusted according to the patient's weight, the mg/kg dosages of all the component drugs remaining similar for all patients, whatever their weight; the number of pills to be swallowed is reduced, possiblyencouraging a higher level of compliance, and the component drugs are always taken together at the same time, eliminating the risk of monotherapy. The prescribing of the daily Rifater preparation is particularly simple (7); one tablet is prescribed for every 10 kg of body weight. By this means the dose range in mg/kg varies little, namely for isoniazid from 4.3 to 5.5, for rifampin from 10.3 to 13.3, and for pyrazinamide from 25.7 to 33.3. In the present study special care was taken to avoid bias in the way the data on the acceptability of the preparations studied were obtained because patients, nurses, and medical staff may be biased in their attitudes to "new" drug preparations. In the event, patients made few complaints about the size or number of tablets and capsules or about difficulty in swallowing them, and only 8010 receiving the combined formulation and 7010 on separate drugs complained of nausea (3010 and 5010, respectively, with vomiting), suggesting that the combined formulation and the separate drugs were equally acceptable. When isoniazid, rifampin, and pyrazinamide are given intermittently, larger dosages of isoniazid and pyrazinamide are necessary than in daily regimens, and hence larger numbers of tablets and capsules in each dose. In a study of intermittent chemotherapy in Hong Kong (8), the combined formulation, which in this case was designed for intermittent use and was not the same as that used in the present study, appeared to be more acceptable to patients than the three drugs given separately, but the evidence was slender: 5010 of the patients receiving separate drugs compared with 1010 receiving the combined formulation complained about the size and number of their tablets and capsules or about difficulty in swallowing them, and 45010 compared with 32010 brought their own drink to the clinic to help them swallow their tablets and capsules, a not uncommon practice in Hong Kong.
711
Despite there being little objective, unbiased evidence that patients find combined preparations more acceptable than separate drugs, their use is recommended by the Committee on Treatment of the International Union Against Thberculosis and Lung Disease (lUATLD) (9) on the grounds of the advantages discussed previously. It is important, however,that only those preparations are used that have been shown to be adequately absorbed from the gastrointestinal tract. Tests of the human bioavailability of three brands of rifampin and isoniazid combined and four of isoniazid, rifampin, and pyrazinamide combined, all intended for daily use and commercially available in developing countries, have shown that one two-drug and all four of the three-drug formulations produced unacceptably low plasma rifampin concentrations (10). Both the American Thoracic Society (ATS) (11) and the IUATLD (9) recommend a 6-month regimen of isoniazid, rifampin, and pyrazinamide for 2 months (without a fourth drug), followed by isoniazid and rifampin for 4 months, in the treatment of patients with active tuberculosis who are known or likely to be infected with fully drug-susceptible strains. As far as we are aware, the present and the previous study in Singapore (1) are the only studies that have shown this regimen to be effective in a randomized controlled comparison with a regimen in which four drugs were given for the initial 2-month intensive phase, and in these Singapore studies the continuation phase was given three times a week, not daily. It is therefore desirable that results obtained with the use of the three-drug recommended regimen, both in controlled trials and in service programs, are published in the medical literature. Both the American Thoracic Society and the IUATLD recommend adding a fourth drug to the initial phase if initial drug resistance is suspected, and the IUATLD if the regimen is given intermittently from the start. There is now evidence from a study in Hong Kong that the recommended regimen may be marginally inferior therapeutically if it is given intermittently from the start without a fourth drug (streptomycin in the Hong Kong study) during the initial intensive phase (12). In view of the comparatively small numbers of patients in the previous study in Singapore (l) and in the present study, it is still possible that the fourdrug combination may prove to be marginally superior to the recommended
712
three-drug combination, even in daily regimens, if comparative studies could be conducted in sufficiently large numbers of patients. Such studies are highly desirable because even a small deficiency in a widely recommended and widely used regimen would be important. Acknowledgment In Singapore the following physicians participated in the study: from 'Ian Tock Seng Hospital, Dr. Ng Kwok Choy, Dr. Poh Soo Chuan, Dr. F. J. Jayaratnam, Dr. Lee SiewKhow, Dr. Teo Seng Kee, Dr. lan Tiong Har, and Mrs. K. L. Lo (bacteriologist); from the Department of Thberculosis Control, Dr. S. Devi and Dr. Thn Kai Kiat. In London the bacteriologic laboratory was the Medical Research Council Unit for Laboratory Studies of Thberculosis(subsequently the Department of Bacteriology, Royal Postgraduate Medical School, London; Professor D. A. Mitchison, Mr. V. R. Aber, and Mr. B. W. Allen). Dr. I. H. Kerr was the radiologic assessor. The investigation was coordinated in Singapore by Dr. Teo Seng Kee, and the independent assessor was Dr. Chew Chin Hin. In London the coordinator was Dr. D. J. Girling of the Medical Research Council Thberculosis and Chest Diseases Unit (now the MRC Cardiothoracic Epidemiology Group). The report was prepared on behalf of the cooperating physicians by Dr. D. J. Girling, Mr. A. J. Nunn, Professor Wallace Fox, and Professor
SINGAPORE TUBERCULOSIS SERVICE/BRITISH MEDICAL RESEARCH COUNCIL
D. A. Mitchison. The participants are gratetul to Dr. Kwa Soon Bee, Permanent Secretary (Health)/Director of Medical Services, Singapore, for his encouragement and support and the Medical Clinical Research Committee, Ministry of Health, for a research grant. The study would not have been possible but for the dedicated work of Nurses Foo Gwek Eng and Chang Seow Kiak. Gruppo Lepetit of Milan provided the Rifater and Bracco Industria Chimica of Milan the pyrazinamide as gifts.
References 1. Singapore Tuberculosis Service/British Medical Research Council. Five-yearfollow-up of a clinical trial of three 6-month regimens of chemotherapy given intermittently in the continuation phase in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 1988; 137:1147-50. 2. Ellard GA, Ellard DR, Allen BW, et al. The bioavailability of isoniazid, rifampin, and pyrazinamide in two commercially available combined formulations designed for use in the short-course treatment of tuberculosis. Am Rev Respir Dis 1986; 133:1076-80. 3. Singapore Thberculosis Service/British Medical Research Council. Clinical trial of six-month and four-month regimens of chemotherapy in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 1979; 119:579-85. 4. Singapore Thberculosis Service/British Medical ResearchCouncil. Clinicaltrial of three 6-month regimens of chemotherapy given intermittently in the continuation phase in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 1985; 132: 374-8.
5. Combs DL, O'Brien RJ, Geiter LJ. USPHS tuberculosis short-course chemotherapy trial 21:effectiveness, toxicity, and acceptability: the report of final results. Ann Intern Moo 1990;112:397-406. 6. Geiter LJ, O'Brien RJ, Combs DL, Snider DE. United States Public Health Service Thberculosis therapy trial 21: preliminary results of an evaluation of a combined tablet of isoniazid, rifampin and pyrazinamide. Thbercle 1987;68(Suppl: 41-6). 7. Acocella G, Angel JH. Short-course chemotherapy of pulmonary tuberculosis: a new approach to drug dosage in the intial intensive phase. Am Rev Respir Dis 1986; 134:1283-6. 8. Hong Kong Chest Service/British Medical Research Council. Acceptability, compliance, and adverse reactions when isoniazid, rifampin, and pyrazinamide are given as a combined formulation or separately during three-times-weekly antituberculosis chemotherapy. Am Rev Respir Dis 1989; 140:1618-22. 9. International Union Against Thberculosis and Lung Disease. Antituberculosis regimens of chemotherapy: recommendations from the Committee on Treatment. Bull IUATLD 1988;63(2):60-4. 10. Acocella G. Quality control of antituberculosis drugs: human bioavailability studies. Bull IUATLD 1989; 64(1):38-40. 11. American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children: joint statement of the American Thoracic Society and the Centers for Disease Control. Am Rev Respir Dis 1986; 134:355-63. 12. Hong Kong Chest Service/British Medical Research Council. Controlled trial of 2, 4, and 6 months of pyrazinamide in 6-month 3-times-weekly regimens for smear-positive pulmonary tuberculosis, including an assessment of a combined preparation of isoniazid, rifampin, and pyrazinamide. Am Rev Respir Dis.
