U.S. Public Health Service Cooperative Trial of Three Rifampin-lsoniazid Regimens in Treatment of Pulmonary Tuberculosis13 MARY W. LONG, DIXIE E. SNIDER, JR., and LAURENCE S. FARER
SUMMARY A total of 822 patients with newly diagnosed pulmonary tuberculosis were assigned randomly to one of 3 daily rifampin-isoniazid (RIF-INH) regimens: 450, 600, or 750 mg of RIF in combination with 300 mg of INH. After an initial 20 weeks of therapy with RIF-INH, patients received 300 mg of INH and 15 mg of ethambutol (EMB) per kg of body weight for either 12 or 18 months after their sputum cultures became negative. The rate of bacteriologic conversion of sputum among the 3 RIF-INH regimens was compared for 552 patients who completed the 20 weeks of RIF-INH therapy. Approximately 60 per cent of these patients also completed their assigned INH-EMB therapy and were examined for relapse for at least one year after therapy was stopped. There was no significant difference in the rate of sputum conversion or rate of relapse between the group of patients who received 600 mg of RIF and those who received 750 mg of RIF. However, the 450-mg RIF regimen was significantly less effective than the other 2 regimens, as manifested by a lower rate of sputum conversion and a higher rate of treatment failures. Further analysis showed that RIF dosages of less than 9 mg per kg of body weight per day may be inadequate for treatment of pulmonary tuberculosis. The acceptability of these regimens was high, and the incidence of adverse reactions requiring discontinuation of RIF-INH therapy was quite low (3.3 per cent). A large proportion of patients (44 per cent) developed increased concentrations of transaminase during therapy with RIF-INH. These abnormalities were usually transient and, in most cases, of no clinical significance. In the relapse analysis, 12 months of chemotherapy after sputum conversion were shown to be as effective as 18 months of therapy after conversion of these RIF-containing regimens.
Introduction
22, 1977 and
d u r a t i o n t h a n the t r a d i t i o n a l 18 to 24 m o n t h s have been shown to be highly effective (1, 2). A l t h o u g h m u c h has b e e n learned a b o u t this drug, substantial questions a b o u t short-term a n d long-term adverse effects a n d the o p t i m a l dosage, r h y t h m , a n d d u r a t i o n of a d m i n i s t r a t i o n rem a i n to be answered. A l t h o u g h r i f a m p i n has been licensed for use in the U n i t e d States since 1971, data available to the C e n t e r for Disease
i From the Tuberculosis Control Division, Bureau of State Services, Center for Disease Control, Atlanta, Ga. 30333. 2 Presented in part at the annual meeting of the American Thoracic Society, Cincinnati, Ohio, May 1974; at the XXIIIrd Conference of the International
Union Against Tuberculosis, Mexico City, September 1975 and at the annual meeting of the American Thoracic Society, San Francisco, May 1977. 3 Requests for reprints should be addressed to Technical Information Services, Bureau of State Services, Center for Disease Control, Atlanta, Ga. 30333.
T h e discovery a n d use of r i f a m p i n (RIF) has b r o u g h t a b o u t a "mini-revolution" in the treatm e n t of tuberculosis. T h e d r u g has proved to be remarkably efficacious, especially w h e n combined with isoniazid ( I N H ) , a n d in recent years R I F - c o n t a i n i n g regimens of m u c h shorter (Received in original form September in revised form February 21,1979)
AMERICAN REVIEW OF RESPIRATORY DISEASE, VOLUME 119, 1979
879
880
LONG, SNIDER, AND FARER
C o n t r o l indicate that in this country, d r u g comb i n a t i o n s that d o n o t c o n t a i n R I F c o n t i n u e to be used m o r e frequently in initial t r e a t m e n t of tuberculosis t h a n R I F - c o n t a i n i n g regimens. Alt h o u g h the high cost of the d r u g may be o n e e x p l a n a t i o n , it may be that the reticence of m a n y clinicians to use r i f a m p i n is d u e to concern a b o u t its p o t e n t i a l toxicity. T h i s U.S. P u b l i c H e a l t h Service therapy trial h a d 2 purposes: (2) to c o m p a r e the efficacy a n d toxicity of 3 different doses of R I F in combination with I N H d u r i n g initial therapy, a n d (2) to c o m p a r e 12 m o n t h s versus 18 m o n t h s of c h e m o t h e r a p y after s p u t u m conversion. T h e patients received R I F - I N H for 20 weeks (approximately 5 m o n t h s ) as initial t r e a t m e n t followed by I N H a n d e t h a m b u t o l (EMB) for varying periods of time, d e p e n d i n g o n the time of sput u m conversion a n d the r a n d o m l y assigned duration of therapy.
Materials and Methods From October 1970 through September 1972, the cooperating investigators (see Appendix) in 18 hospitals and 34 clinics located throughout the United States admitted 822 patients who had consented to participate in this trial. Patients more than 13 years of age whose sputum contained acid-fast organisms were admitted to the study if they had previously received no more than 14 days of antituberculosis chemotherapy. These patients must also have demonstrated roentgenographic evidence of active pulmonary tuberculosis with or without cavitation. Excluded were patients with pre-existing renal, hepatic, hematologic, or ophthalmologic conditions that, in the opinion of the attending physician, introduced undue risks of adverse reactions to the drugs, patients on corticosteroid therapy, and patients who probably could not be observed for 3 years because they were terminally ill, aliens with temporary visas, migrant workers, and so forth. T h e 822 patients admitted to this trial were randomly assigned by hospital and extent of disease to one of the following regimens: (1) R450—450 mg of RIF plus 300 mg of INH given daily; (2) R600— 600 mg of RIF plus 300 mg of INH given daily; (3) R750—750 mg of RIF plus 300 mg of INH given daily. Initially, each patient was prescribed a 20-week regimen of RIF-INH* followed by daily administration
* T h e rifampin, in 150-mg and 300-mg capsules, was donated by CIBA-GEIGY Pharmaceutical Company, Summit, N.J. 07901. T h e isoniazid, in 300-mg tablets, was purchased from Mallinckrodt Chemical Works, St. Louis, Mo. 63160.
of 300 mg of INH and 15 mg of EMBf per kg of body weight for various lengths of time, depending on the randomly assigned duration of chemotherapy after bacteriologic conversion. Most of the patients began their initial therapy in the hospital, where their median length of stay was 14 weeks. After discharge from the hospital, patients were seen at an outpatient clinic at 4-week intervals while on medication. T h e number of days each patient took medications was estimated by counting the capsules and tablets returned to the clinic at each visit. Before treatment and at 4-week intervals throughout the initial 20 weeks of RIF-INH therapy, sputum specimens were obtained for smear, culture, and drug-susceptibility testing; posteroanterior chest roentgenograms, white blood cell counts, and concentrations of serum glutamic oxaloacetic transaminase (SGOT) were obtained; and patients were asked whether they had noted any changes in their condition (in an effort to elicit signs and symptoms of adverse reactions). Total bilirubin determinations were performed on all patients whose SGOT values were 50 Karmen units (KU) or more. Beginning in January 1971, platelet counts were also obtained at 4week intervals. (Blood for platelet counts was not drawn at any fixed interval after drug ingestion.) SGOT, total bilirubin, and mycobacterial drug susceptibility tests were done by the CDC Bureau of Laboratories, using standard procedures (3-5). All other laboratory tests were performed by local treatment facilities. At the start of the second phase of the trial (i.e., after 20 weeks of therapy with RIF-INH), patients from each, of the 3 RIF-INH regimens were assigned randomly to 2 subgroups that received the INHEMB regimen for different durations. Patients in one subgroup received INH-EMB for a length of time sufficient to give 52 weeks (12 months) of daily chemotherapy beyond the time their sputum cultures became negative (sputum conversion). Those in the other subgroup received enough INH-EMB to give 76 weeks (18 months) of chemotherapy after sputum conversion. A schematic diagram of the 12month regimen is shown in figure 1. Each bar represents patients whose sputum converted during the first, second, third, fourth, or fifth month of therapy. T h e unshaded section of each bar represents 20 weeks or 5 months of RIF-INH therapy, and the cross-hatched section of each bar represents INHEMB therapy. T h e part of each bar that falls between the 2 parallel dotted lines represents the 12
f T h e ethambutol, in 400-mg tablets, was purchased from Lederle Laboratories, Pearl River, N.Y. 10965. T h e dosage of 15 mg per kg was rounded to the nearest 200 mg and was based on each patient's body weight at the twentieth week of therapy.
USPHS COOPERATIVE TRIAL OF THREE RIFAMPIN-ISONIAZID REGIMENS
months of therapy after conversion. For example, patients whose sputum converted at the first month are represented by the bar at the top of the diagram. After sputum conversion, these patients received RIFINH for 4 more months, followed by 8 months of INH-EMB. Those whose sputum converted at the fifth month (represented by the bottom bar) received 12 months of INH-EMB after sputum conversion. A similar diagram could be drawn of the patients assigned to the 18-month, postconvers ion regimen by extending the cross-hatched section of each bar 6 months. These patients received INH-EMB for 6 more months after sputum conversion than did those assigned to the 12-month postconversion regimen. T h e patients within each RIF dosage regimen were assigned randomly to these duration regimens, but were stratified by age group, radiographic extent of disease, and admitted use of alcohol. All patients were followed for a total of 3 years after therapy was initiated. T h e number of months during which a patient was observed after completion of therapy varied by month of conversion and assigned duration of therapy. Patients whose sputum converted at one month and who were assigned to the 12-month regimen had the longest period of observation after completion of therapy—one year and 48 weeks. Those who converted at the fifth month and were assigned to the 18-month regimen were observed for the shortest period of time while not receiving drugs—one year and 8 weeks. Clinic visits and sputum examinations were made at 4-week intervals while the patients were receiving INH-EMB, and at 12-week intervals after therapy was completed. Chest roentgenograms were taken at 12-week intervals during the INH-EMB phase and at 24-week intervals after drugs were stopped. While taking INH-EMB, patients were asked about signs and symptoms of adverse reactions, but further examinations were performed only if an adverse reaction was suspected. All tests of significance were made using either a
Fig. 1. Diagram of regimen of 12-month duration.
881
Chi-Square Test or Fisher's Exact Test (6). P values of less than 0.05 were considered significant. Multiple-regression analysis was used to assess the effect of various factors on the risk of developing abnormal transaminase values during treatment (7). Relapse rates were estimated using a life table method (8). Results Initial Characteristics of Patients T h e initial characteristics (race, sex, e x t e n t of disease, age, alcohol use, a n d weight) of the 822 patients a d m i t t e d to the study are shown in table 1. T h e s e initial characteristics were similarly distributed a m o n g the p a t i e n t s assigned r a n d o m l y to each of the 3 R I F regimens. T h e n u m b e r of p a t i e n t s assigned to the R600 a n d R750 regimens was nearly equal, 324 a n d 331 patients, respectively. A p p r o x i m a t e l y onehalf as m a n y patients (167) were assigned to the R450 regimen because admission was discontinued w h e n p r e l i m i n a r y analysis indicated a slight lag in conversion r a t e for patients assigned to that regimen. Assessment of the Initial 20-Week Phase Deviations. Of the 822 p a t i e n t s a d m i t t e d to the study, 688 (81.3 per cent) were considered to have completed the initial phase of therapy because they received their assigned regimen of R I F - I N H for at least 80 p e r cent of the time (i.e., for at least 16 weeks d u r i n g the 20-week period, w i t h o u t missing 4 consecutive weeks of therapy). T h e 154 p a t i e n t s (18.7 p e r cent) w h o did not complete their initial therapy deviated from their assigned regimen for the various reasons shown in table 2. T h e most c o m m o n reason {or withdrawal was delinquency, i.e., failure to keep clinic a p p o i n t m e n t s (43 p a t i e n t s or 5.2 p e r cent). Twenty-seven patients (3.3 p e r cent) deviated because of adverse d r u g reactions, 23 (2.8 per cent) because of d e a t h u n r e l a t e d to tuberculosis, 18 (2.2 p e r cent) because of administrative errors, 9 (1.1 p e r cent) because of c o n c u r r e n t diseases that necessitated a change in chemotherapy, 8 (1.0 p e r cent) because of t r e a t m e n t failure, 7 (0.9 p e r cent) because n o n t u b e r c u l o u s mycobacterial disease was diagnosed, 6 (0.7 p e r cent) because of irregular ingestion of medication, a n d 5 patients (0.6 p e r cent) refused t r e a t m e n t . Eight a d d i t i o n a l p a t i e n t s were w i t h d r a w n from the R450 regimen w h e n a p r e l i m i n a r y analysis indicated a lag in conversion rates for the R450 regimen g r o u p . For several reasons, a larger pro-
882
LONG, SNIDER, AND FARER
3,500 cells per mm 3 . Eight asymptomatic patients demonstrated abnormal SGOT values ranging from 60 to 310 KU after 26 to 111 days of therapy. Two patients developed abnormal SGOT values and clinical jaundice without gastrointestinal symptoms after 17 and 140 days of therapy, respectively. Eight patients developed gastrointestinal symptoms and peak abnormal SGOT values of 73 to 700 KU during the first 3 weeks of therapy (average onset, 12 days). Two of these 8 patients did not develop jaundice, but six of them became jaundiced with peak total bilirubin values of 3.1 to 28.0 mg per 100 ml. One of the 27 patients with adverse reactions
portion of patients assigned to the R450 regimen were withdrawn as compared to the other 2 regimens, but none of the differences between regimens was significant. Treatment was changed or interrupted for 27 patients (3.3 per cent) because of "adverse reactions" (table 3). Five such reactions were attributed to INH, seven to RIF, and 15 to both drugs. There was no significant difference in the incidence of adverse reactions among the 3 RIFINH regimens. Two asymptomatic patients developed leukopenia, one after 58 and the other after 93 days of therapy. In neither case did the white blood cell count decrease to less than
TABLE 1 I N I T I A L CHARACTERISTICS OF PATIENTS R A N D O M L Y ASSIGNED TO R I F A M P I N REGIMENS Rifampin Isoniaz id Regimen Code Total
Patients Race and sex White male Other races, male White female Other races, female Extent of disease Minimal MA without cavity MA with cavity FA without cavity FA with cavity Age, years 65 Alcohol use None Infrequent Moderate Heavy Excessive Unknown Weight, kg < 50 50-59 60-67 68-76 > 11
R450
R600
R750
(no.)
(%)
(no.)
(%)
(no.)
(no.)
(%)
822
100.0
167
100.0
324
100.0
331
100.0
294
35.8 42.8
53
131
39.5 40.4
113 146
34.1 44.1
61
7.4
75 15
31.7 44.9
128
352
9.0
23
7.1
23
6.9
115
14.0
24
14.4
42
13.0
49
14.8
13
1.6
0
0.0
8
2.5
5
1.5
56 204
6.8
17 42
10.2 25.1
16 82
4.9
24.8
25.3
23 80
24.2
26 523
3.2
3 105
1.8
15
4.5
62.9
8 210
2.5
63.6
64.8
208
62.8
3.7
5 78 141
19 191 330
2.3
2 37 66
1.2
12 76 123
266
23.2 40.1 32.4
60
22.2 39.5 35.9
16
1.9
2
1.2
135
16.4 20.1 30.3 21.9 10.9
31 56 34
18.6 20.4 33.5 20.4
12
7.2
0.4
0
0.0
37 1
37
110
22.1 32.1 27.5 13.4
48 19
22.2 35.3 28.7 11.4
40
4.9
4
2.4
165 249 180
90 3 182 264 226
34
59
(%)
6.9
1.5
108
23.5 38.0 33.3
98
23.6 42.6 29.6
5
1.5
9
2.7
54
16.7 21.0 30.9 19.8 11.4
50 63
41
15.1 19.0 28.1 24.8 12.4
0.3
2
0.6
22.8 27.8 29.0 14.5
71
90 94 47
115 84
44
21.5 34.7 25.4 13.3
19
5.9
17
5.1
68 100
64
74
93 82
Definitions of abbreviations: R450 = daily administration of 450 mg of rifampin plus 300 mg of isoniazid; R600 = daily administration of 600 mg of rifampin plus 300 mg of isoniazid; R750 = daily administration of 750 mg of rifampin plus 300 mg of isoniazid; MA = moderately advanced; FA = far advanced.
883
USPHS COOPERATIVE TRIAL OF THREE RIFAMPIN-ISONIAZID REGIMENS
TABLE 2 DEVIATIONS FROM ASSIGNED RIFAMPIN-ISONIAZID THERAPY Rifamp in-lsonizaid Regimen TotalI
R750
R600
R450
(no.)
(%)
(no.)
(%)
(no.)
(%)
(no.)
(%)
822
100.0 18.7
167
59
100.0 17.8
43 27
5.2 3.3
9 4
5.4
55 17
100.0 17.0
331
40
100.0 24.0
324
154
5.2
2.4
10
3.1
17 13
3.9
23
2.8
7
4.2
11
3.4
5
1.5
18 9
2.2
5
3.0
6 4
1.9
7
2.1
1.2
8 8
1.0
4
2.4
1
0.3
5 3
0.9
1.0
8
4.8
7
0.9 0.7
2
1.2
2
0.6
3
0.9
0.9
3
0.9
0.6
1
0.6
3 1
0.3
3
0.9
81.3
127
76.0
269
83.0
272
82.2
Patients admitted Deviations Delinquent Adverse reaction Deaths unrelated to tuberculosis Administrative error Concurrent illness Treatment failure Protocol modification Nontuberculous mycobacterial disease Irregular pill taker Refused treatment " C o m p l e t e d " rifampin-isoniazid therapy
6 5 668
1.1
5.1
1.5
For definitions of abbreviations, see table 1.
developed arthritis and anemia after 46 days of therapy, and another patient had a febrile reaction after 5 days of therapy. Five patients developed rashes after 30 to 101 days of therapy, which ranged from a minimal skin reaction with pruritis to an erythema multiforme-type reaction. Of the 8 patients who deviated from their assigned regimen because of "treatment failure"
(table 3), five were judged to be clinically worse, although their sputum cultures remained negative. Three additional patients died of tuberculosis; two of them received the R750 regimen for only 2 or 3 days, and the other received the R450 regimen for only 10 days. Hematologic monitoring. During RIF-INH therapy, 6 patients had a white blood cell count of less than 3,000 cells per mm 3 : one received
TABLE 3 PATIENTS WITH ADVERSE REACTIONS OR TREATMENT FAILURES DURING RIFAMPIN-ISONIAZID THERAPY Rifa mpin-lsoniazi id Regimen
Adverse reactions Asymptomatic leukopenia Abnormal SGOT* concentration w i t h o u t symptoms Abnormal SGOT cbncentration with jaundice G l * symptoms w i t h o u t jaundice* * Gl symptoms w i t h jaundice Arthritis and anemia Fever Rash Treatment failures Clinically worse Death f r o m tuberculosis
Total
R450
R600
R750
27 2
4
10
13 2
8
1
5
2
2 2
1
6
1 1
1
3
3 1
1 1
1
3
5
2
8
4
1
5
3 1
1
3
Serum glutamic oxaloacetic transaminase. * Gastrointestinal. * *Abnormal SGOT concentration in addition to Gl symptoms. For definitions of other abbreviations, see table 1.
3 1 2
884
LONG, SNIDER, AND FARER
R450, 3 received R600, and 2 received R750. None of these patients had 2 or more counts of less than 3,000 cells per mm^, and RIF-INH therapy was continued in all 6 patients without subsequent adverse effects. During RIF-INH therapy, 84 patients (14.0 per cent) had one platelet count of less than 150,000 cells per mm*, and 12 (2.0 per cent) had 2 or more counts of less than 150,000 cells per mm 3 on consecutive monthly measurements. The incidence of these persistently low counts was 1.9, 1.2, and 2.7 per cent in the R450, R600, and R750 regimens, respectively. Eight patients (three on R600 and five on R750) had a single platelet count of less than 100,000 cells per mm 3 during therapy, and one R750 patient had 2 consecutive counts of this magnitude. Although there was a trend toward a dose-related effect of RIF on platelet counts, the differences among the regimens did not achieve significance. None of the patients with low white blood cell or platelet counts had clinical signs or symptoms attributable to the abnormal laboratory finding. SGOT monitoring. Of the 822 patients admitted to the study, 804 had serum for SGOT determination shipped to the Center for Disease Control for at least one of the 5 specified intervals during the 20 weeks of RIF-INH therapy. In table 4, these patients are listed by regimen according to the highest SGOT value obtained. Before the initiation of RIF-INH therapy, 233 of 751 patients (31 per cent) had increased SGOT values (> 40 KU). During therapy, 56 per cent of the 804 patients had SGOT values that never exceeded 40 KU (level 0).
Thirty-four per cent demonstrated SGOT values between 41 and 99 KU (level I); 8 per cent had SGOT values between 100 and 249 KU (level II); and the remaining 2 per cent (16 patients) had SGOT values equal to or greater than 250 KU (level III). These data were analyzed by stepwise multiple-regression analysis, using the highest value of SGOT observed at 4-week intervals during treatment as the dependent variable. The set of independent variables included RIF regimen, dosage of RIF per body weight (mg per kg), pretreatment weight, age, sex, race, admitted use of alcohol, extent of disease, presence of concurrent diagnoses at initiation of therapy, SGOT value before treatment, and presence of signs or symptoms of hepatic dysfunction during treatment. Of these independent variables, only race, use of alcohol, extent of disease, SGOT value before treatment, and presence of signs or symptoms were significantly related to the highest SGOT value. Nonwhite patients had higher SGOT values during treatment than did whites. As expected, the SGOT value before treatment, admitted use of alcohol, and presence of signs and symptoms of hepatic dysfunction were correlated positively with the highest SGOT value. Somewhat surprisingly, the extent of disease was inversely related to the highest SGOT value; i.e., patients with more extensive disease before treatment tended to have lower SGOT values during therapy. Although the preceding relationships were significant, only approximately 7.5 per cent of the variation in the dependent variable (highest SGOT value during treatment) was explained by this par-
TAB LE 4 NUMBER A N D PERCENTAGE OF PATIENTS WHOSE HIGHEST SGOT V A L U E OCCURRED A T SPECIFIED LEVELS D U R I N G 20 WEEKS OF R I F A M P I N - I S O N I A Z I D T H E R A P Y BY R E G I M E N Rifampin!Isoniazid Regi men Level of Highest SGOT Value Patients with SGOT data* 0: S G O T t < 4 0 K U * * 1: SGOT 4 1 - 9 9 KU II: SGOT 1 0 0 - 2 4 9 KU III: SGOT > 2 5 0 KU
Total
(no.) 804 451 270 67 16
R450
R750
R600
(%)
(no.)
(%)
(no.)
(%)
(no.)
(%)
100.0 56.1 33.6 8.3 2.0
162 89 60 7 6
100.0 54.9 37.0 4.3 3.7
318 175 111 29 3
100.0 55.0 34.9 9.1 0.9
324 187 99 31 7
100.0 57.7 30.6 9.6 2.2
Eighteen of 822 patients were excluded because no SGOT determinations were available at 4, 8, 12, 16, or 20 weeks of therapy. * Serum glutamic oxaloacetic transaminase. Karmen Units, 32° C. For definitions of other abbreviations, see table 1.
885
USPHS COOPERATIVE TRIAL OF THREE RIFAMPIN-ISONIAZID REGIMENS
ticular set of independent variables. Thus, other unknown factors must have contributed most of the observed variation. Furthermore, in this multivariate analysis, significant interrelationships among the "independent variables" cannot be excluded. Of the 353 patients who demonstrated increased SGOT during RIF-INH therapy, 121 (34.3 per cent) developed abnormal SGOT values during the first 4 weeks, 187 (53.0 per cent) during the first 8 weeks, and 254 (72.0 per cent) had abnormal SGOT values during the first 12 weeks of therapy. Supplemental hepatic dysfunction data. One hundred three patients were reported to have "suspected hepatic disease" because they had routine SGOT values of 100 KU or more, total bilirubin values of 2.5 mg per 100 ml or more, or jaundice reported during therapy. To evaluate these patients further, supplemental data were obtained from their hospital and clinic records. Only 32 of these 103 patients (31.1 per cent) developed overt signs or symptoms of an adverse reaction. Twenty-one became clinically jaundiced, and 18 developed gastrointestinal symptoms. Two of the 32 patients had a pruritic rash. Nearly all developed signs or symptoms during the first month of therapy. Slightly more than one half (56 or 54.3 per cent) of the 103 selected patients had other factors identified that may have caused or contributed to the hepatic dysfunction. Forty of them abused alcohol, 14 were addicted to heroin, and two were taking hepatotoxic medications in addition to RIF-INH. Sputum conversion during the initial phase. Of the 668 patients who completed initial therapy, 116 were excluded from this analysis of efficacy for the following reasons: (1) the pretreatment sputum cultures from 39 patients contained Mycobacterium tuberculosis organisms that were resistant to EMB, INH, and/or RIF; # 33 patients had negative cultures; nine had contaminated cultures; and 26 had pretreatment cultures containing nontuberculous mycobacteria. The remaining 9 patients were excluded because they had previously received antituberculosis drugs for more than 14 days. Thus, the
analysis of efficacy, as measured by the rate of bacteriologic conversion, was based on 552 patients. Approximately two thirds of the patients admitted to each regimen were included in this efficacy analysis; their initial characteristics had a distribution similar to that shown in table 1. Shown in figure 2 is the per cent of cultures that contained M. tuberculosis organisms at 4week intervals among the patients in the 3 RIF regimens. At 12 weeks, 12 per cent of those on the R450 regimen, 7 per cent of those on the R600 regimen, and 6 per cent of those on the R750 regimen had positive cultures. By 20 weeks, 7.7 per cent of those on the R450 regimen but only 0.5 per cent on the R600 and 0.9 per cent on the R750 regimen had positive cultures. There was no significant difference in the percentages between the R600 and R750 regimens at any 4-week interval. However, at 16 and 20 weeks, the percentages of positive cultures for the R450 regimen were significantly greater than were those for the R600 and R750 regimens (P < 0.01). The per cent of cultures positive for M. tuberculosis among patients who received RIF, based on dosage, in mg per kg, is shown in figure 3. At 12 weeks, 9.5 per cent of those who received less than 9.0 mg per kg of body weight, 7.6 per cent of those who received 9.0 to 12.9 mg per kg, and 5.4 per cent of those who received 13.0 mg or more per kg had positive cultures. By 20 weeks, 4.7 per cent of those who received less than 9.0 mg per kg, 1.4 per cent of those who re-
8
* Of the 39 patients who had resistant organisms in the pretreatment culture, 20 showed resistance to INH alone, seven to RIF alone, one to EMB alone, three to RIF and INH, six to INH and EMB, and two to all 3 drugs.
12
16
20
WEEKS
Fig. 2. Per cent of cultures containing Mycobacterium tuberculosis organisms among patients completing rifampin-isoniazid (RIF-INH) therapy, by week and regimen.
886
LONG, SNIDER, AND FARER
RIFAMPIN mg/kg
signed to the R450 regimen, but only 19 per cent of those assigned to R600 and 2 per cent of those assigned to R750 received less than 9.0 mg per kg per day.
SUMMARY A total of 822 patients with newly diagnosed pulmonary tuberculosis were assigned randomly to one of 3 daily rifampin-isoniazid (RIF-INH) regimens: 450, 600, or 750 mg of RIF in combination with 300 mg of INH. After an initial 20 weeks of therapy with RIF-INH, patients received 300 mg of INH and 15 mg of ethambutol (EMB) per kg of body weight for either 12 or 18 months after their sputum cultures became negative. The rate of bacteriologic conversion of sputum among the 3 RIF-INH regimens was compared for 552 patients who completed the 20 weeks of RIF-INH therapy. Approximately 60 per cent of these patients also completed their assigned INH-EMB therapy and were examined for relapse for at least one year after therapy was stopped. There was no significant difference in the rate of sputum conversion or rate of relapse between the group of patients who received 600 mg of RIF and those who received 750 mg of RIF. However, the 450-mg RIF regimen was significantly less effective than the other 2 regimens, as manifested by a lower rate of sputum conversion and a higher rate of treatment failures. Further analysis showed that RIF dosages of less than 9 mg per kg of body weight per day may be inadequate for treatment of pulmonary tuberculosis. The acceptability of these regimens was high, and the incidence of adverse reactions requiring discontinuation of RIF-INH therapy was quite low (3.3 per cent). A large proportion of patients (44 per cent) developed increased concentrations of transaminase during therapy with RIF-INH. These abnormalities were usually transient and, in most cases, of no clinical significance. In the relapse analysis, 12 months of chemotherapy after sputum conversion were shown to be as effective as 18 months of therapy after conversion of these RIF-containing regimens.
Introduction
22, 1977 and
d u r a t i o n t h a n the t r a d i t i o n a l 18 to 24 m o n t h s have been shown to be highly effective (1, 2). A l t h o u g h m u c h has b e e n learned a b o u t this drug, substantial questions a b o u t short-term a n d long-term adverse effects a n d the o p t i m a l dosage, r h y t h m , a n d d u r a t i o n of a d m i n i s t r a t i o n rem a i n to be answered. A l t h o u g h r i f a m p i n has been licensed for use in the U n i t e d States since 1971, data available to the C e n t e r for Disease
i From the Tuberculosis Control Division, Bureau of State Services, Center for Disease Control, Atlanta, Ga. 30333. 2 Presented in part at the annual meeting of the American Thoracic Society, Cincinnati, Ohio, May 1974; at the XXIIIrd Conference of the International
Union Against Tuberculosis, Mexico City, September 1975 and at the annual meeting of the American Thoracic Society, San Francisco, May 1977. 3 Requests for reprints should be addressed to Technical Information Services, Bureau of State Services, Center for Disease Control, Atlanta, Ga. 30333.
T h e discovery a n d use of r i f a m p i n (RIF) has b r o u g h t a b o u t a "mini-revolution" in the treatm e n t of tuberculosis. T h e d r u g has proved to be remarkably efficacious, especially w h e n combined with isoniazid ( I N H ) , a n d in recent years R I F - c o n t a i n i n g regimens of m u c h shorter (Received in original form September in revised form February 21,1979)
AMERICAN REVIEW OF RESPIRATORY DISEASE, VOLUME 119, 1979
879
880
LONG, SNIDER, AND FARER
C o n t r o l indicate that in this country, d r u g comb i n a t i o n s that d o n o t c o n t a i n R I F c o n t i n u e to be used m o r e frequently in initial t r e a t m e n t of tuberculosis t h a n R I F - c o n t a i n i n g regimens. Alt h o u g h the high cost of the d r u g may be o n e e x p l a n a t i o n , it may be that the reticence of m a n y clinicians to use r i f a m p i n is d u e to concern a b o u t its p o t e n t i a l toxicity. T h i s U.S. P u b l i c H e a l t h Service therapy trial h a d 2 purposes: (2) to c o m p a r e the efficacy a n d toxicity of 3 different doses of R I F in combination with I N H d u r i n g initial therapy, a n d (2) to c o m p a r e 12 m o n t h s versus 18 m o n t h s of c h e m o t h e r a p y after s p u t u m conversion. T h e patients received R I F - I N H for 20 weeks (approximately 5 m o n t h s ) as initial t r e a t m e n t followed by I N H a n d e t h a m b u t o l (EMB) for varying periods of time, d e p e n d i n g o n the time of sput u m conversion a n d the r a n d o m l y assigned duration of therapy.
Materials and Methods From October 1970 through September 1972, the cooperating investigators (see Appendix) in 18 hospitals and 34 clinics located throughout the United States admitted 822 patients who had consented to participate in this trial. Patients more than 13 years of age whose sputum contained acid-fast organisms were admitted to the study if they had previously received no more than 14 days of antituberculosis chemotherapy. These patients must also have demonstrated roentgenographic evidence of active pulmonary tuberculosis with or without cavitation. Excluded were patients with pre-existing renal, hepatic, hematologic, or ophthalmologic conditions that, in the opinion of the attending physician, introduced undue risks of adverse reactions to the drugs, patients on corticosteroid therapy, and patients who probably could not be observed for 3 years because they were terminally ill, aliens with temporary visas, migrant workers, and so forth. T h e 822 patients admitted to this trial were randomly assigned by hospital and extent of disease to one of the following regimens: (1) R450—450 mg of RIF plus 300 mg of INH given daily; (2) R600— 600 mg of RIF plus 300 mg of INH given daily; (3) R750—750 mg of RIF plus 300 mg of INH given daily. Initially, each patient was prescribed a 20-week regimen of RIF-INH* followed by daily administration
* T h e rifampin, in 150-mg and 300-mg capsules, was donated by CIBA-GEIGY Pharmaceutical Company, Summit, N.J. 07901. T h e isoniazid, in 300-mg tablets, was purchased from Mallinckrodt Chemical Works, St. Louis, Mo. 63160.
of 300 mg of INH and 15 mg of EMBf per kg of body weight for various lengths of time, depending on the randomly assigned duration of chemotherapy after bacteriologic conversion. Most of the patients began their initial therapy in the hospital, where their median length of stay was 14 weeks. After discharge from the hospital, patients were seen at an outpatient clinic at 4-week intervals while on medication. T h e number of days each patient took medications was estimated by counting the capsules and tablets returned to the clinic at each visit. Before treatment and at 4-week intervals throughout the initial 20 weeks of RIF-INH therapy, sputum specimens were obtained for smear, culture, and drug-susceptibility testing; posteroanterior chest roentgenograms, white blood cell counts, and concentrations of serum glutamic oxaloacetic transaminase (SGOT) were obtained; and patients were asked whether they had noted any changes in their condition (in an effort to elicit signs and symptoms of adverse reactions). Total bilirubin determinations were performed on all patients whose SGOT values were 50 Karmen units (KU) or more. Beginning in January 1971, platelet counts were also obtained at 4week intervals. (Blood for platelet counts was not drawn at any fixed interval after drug ingestion.) SGOT, total bilirubin, and mycobacterial drug susceptibility tests were done by the CDC Bureau of Laboratories, using standard procedures (3-5). All other laboratory tests were performed by local treatment facilities. At the start of the second phase of the trial (i.e., after 20 weeks of therapy with RIF-INH), patients from each, of the 3 RIF-INH regimens were assigned randomly to 2 subgroups that received the INHEMB regimen for different durations. Patients in one subgroup received INH-EMB for a length of time sufficient to give 52 weeks (12 months) of daily chemotherapy beyond the time their sputum cultures became negative (sputum conversion). Those in the other subgroup received enough INH-EMB to give 76 weeks (18 months) of chemotherapy after sputum conversion. A schematic diagram of the 12month regimen is shown in figure 1. Each bar represents patients whose sputum converted during the first, second, third, fourth, or fifth month of therapy. T h e unshaded section of each bar represents 20 weeks or 5 months of RIF-INH therapy, and the cross-hatched section of each bar represents INHEMB therapy. T h e part of each bar that falls between the 2 parallel dotted lines represents the 12
f T h e ethambutol, in 400-mg tablets, was purchased from Lederle Laboratories, Pearl River, N.Y. 10965. T h e dosage of 15 mg per kg was rounded to the nearest 200 mg and was based on each patient's body weight at the twentieth week of therapy.
USPHS COOPERATIVE TRIAL OF THREE RIFAMPIN-ISONIAZID REGIMENS
months of therapy after conversion. For example, patients whose sputum converted at the first month are represented by the bar at the top of the diagram. After sputum conversion, these patients received RIFINH for 4 more months, followed by 8 months of INH-EMB. Those whose sputum converted at the fifth month (represented by the bottom bar) received 12 months of INH-EMB after sputum conversion. A similar diagram could be drawn of the patients assigned to the 18-month, postconvers ion regimen by extending the cross-hatched section of each bar 6 months. These patients received INH-EMB for 6 more months after sputum conversion than did those assigned to the 12-month postconversion regimen. T h e patients within each RIF dosage regimen were assigned randomly to these duration regimens, but were stratified by age group, radiographic extent of disease, and admitted use of alcohol. All patients were followed for a total of 3 years after therapy was initiated. T h e number of months during which a patient was observed after completion of therapy varied by month of conversion and assigned duration of therapy. Patients whose sputum converted at one month and who were assigned to the 12-month regimen had the longest period of observation after completion of therapy—one year and 48 weeks. Those who converted at the fifth month and were assigned to the 18-month regimen were observed for the shortest period of time while not receiving drugs—one year and 8 weeks. Clinic visits and sputum examinations were made at 4-week intervals while the patients were receiving INH-EMB, and at 12-week intervals after therapy was completed. Chest roentgenograms were taken at 12-week intervals during the INH-EMB phase and at 24-week intervals after drugs were stopped. While taking INH-EMB, patients were asked about signs and symptoms of adverse reactions, but further examinations were performed only if an adverse reaction was suspected. All tests of significance were made using either a
Fig. 1. Diagram of regimen of 12-month duration.
881
Chi-Square Test or Fisher's Exact Test (6). P values of less than 0.05 were considered significant. Multiple-regression analysis was used to assess the effect of various factors on the risk of developing abnormal transaminase values during treatment (7). Relapse rates were estimated using a life table method (8). Results Initial Characteristics of Patients T h e initial characteristics (race, sex, e x t e n t of disease, age, alcohol use, a n d weight) of the 822 patients a d m i t t e d to the study are shown in table 1. T h e s e initial characteristics were similarly distributed a m o n g the p a t i e n t s assigned r a n d o m l y to each of the 3 R I F regimens. T h e n u m b e r of p a t i e n t s assigned to the R600 a n d R750 regimens was nearly equal, 324 a n d 331 patients, respectively. A p p r o x i m a t e l y onehalf as m a n y patients (167) were assigned to the R450 regimen because admission was discontinued w h e n p r e l i m i n a r y analysis indicated a slight lag in conversion r a t e for patients assigned to that regimen. Assessment of the Initial 20-Week Phase Deviations. Of the 822 p a t i e n t s a d m i t t e d to the study, 688 (81.3 per cent) were considered to have completed the initial phase of therapy because they received their assigned regimen of R I F - I N H for at least 80 p e r cent of the time (i.e., for at least 16 weeks d u r i n g the 20-week period, w i t h o u t missing 4 consecutive weeks of therapy). T h e 154 p a t i e n t s (18.7 p e r cent) w h o did not complete their initial therapy deviated from their assigned regimen for the various reasons shown in table 2. T h e most c o m m o n reason {or withdrawal was delinquency, i.e., failure to keep clinic a p p o i n t m e n t s (43 p a t i e n t s or 5.2 p e r cent). Twenty-seven patients (3.3 p e r cent) deviated because of adverse d r u g reactions, 23 (2.8 per cent) because of d e a t h u n r e l a t e d to tuberculosis, 18 (2.2 p e r cent) because of administrative errors, 9 (1.1 p e r cent) because of c o n c u r r e n t diseases that necessitated a change in chemotherapy, 8 (1.0 p e r cent) because of t r e a t m e n t failure, 7 (0.9 p e r cent) because n o n t u b e r c u l o u s mycobacterial disease was diagnosed, 6 (0.7 p e r cent) because of irregular ingestion of medication, a n d 5 patients (0.6 p e r cent) refused t r e a t m e n t . Eight a d d i t i o n a l p a t i e n t s were w i t h d r a w n from the R450 regimen w h e n a p r e l i m i n a r y analysis indicated a lag in conversion rates for the R450 regimen g r o u p . For several reasons, a larger pro-
882
LONG, SNIDER, AND FARER
3,500 cells per mm 3 . Eight asymptomatic patients demonstrated abnormal SGOT values ranging from 60 to 310 KU after 26 to 111 days of therapy. Two patients developed abnormal SGOT values and clinical jaundice without gastrointestinal symptoms after 17 and 140 days of therapy, respectively. Eight patients developed gastrointestinal symptoms and peak abnormal SGOT values of 73 to 700 KU during the first 3 weeks of therapy (average onset, 12 days). Two of these 8 patients did not develop jaundice, but six of them became jaundiced with peak total bilirubin values of 3.1 to 28.0 mg per 100 ml. One of the 27 patients with adverse reactions
portion of patients assigned to the R450 regimen were withdrawn as compared to the other 2 regimens, but none of the differences between regimens was significant. Treatment was changed or interrupted for 27 patients (3.3 per cent) because of "adverse reactions" (table 3). Five such reactions were attributed to INH, seven to RIF, and 15 to both drugs. There was no significant difference in the incidence of adverse reactions among the 3 RIFINH regimens. Two asymptomatic patients developed leukopenia, one after 58 and the other after 93 days of therapy. In neither case did the white blood cell count decrease to less than
TABLE 1 I N I T I A L CHARACTERISTICS OF PATIENTS R A N D O M L Y ASSIGNED TO R I F A M P I N REGIMENS Rifampin Isoniaz id Regimen Code Total
Patients Race and sex White male Other races, male White female Other races, female Extent of disease Minimal MA without cavity MA with cavity FA without cavity FA with cavity Age, years 65 Alcohol use None Infrequent Moderate Heavy Excessive Unknown Weight, kg < 50 50-59 60-67 68-76 > 11
R450
R600
R750
(no.)
(%)
(no.)
(%)
(no.)
(no.)
(%)
822
100.0
167
100.0
324
100.0
331
100.0
294
35.8 42.8
53
131
39.5 40.4
113 146
34.1 44.1
61
7.4
75 15
31.7 44.9
128
352
9.0
23
7.1
23
6.9
115
14.0
24
14.4
42
13.0
49
14.8
13
1.6
0
0.0
8
2.5
5
1.5
56 204
6.8
17 42
10.2 25.1
16 82
4.9
24.8
25.3
23 80
24.2
26 523
3.2
3 105
1.8
15
4.5
62.9
8 210
2.5
63.6
64.8
208
62.8
3.7
5 78 141
19 191 330
2.3
2 37 66
1.2
12 76 123
266
23.2 40.1 32.4
60
22.2 39.5 35.9
16
1.9
2
1.2
135
16.4 20.1 30.3 21.9 10.9
31 56 34
18.6 20.4 33.5 20.4
12
7.2
0.4
0
0.0
37 1
37
110
22.1 32.1 27.5 13.4
48 19
22.2 35.3 28.7 11.4
40
4.9
4
2.4
165 249 180
90 3 182 264 226
34
59
(%)
6.9
1.5
108
23.5 38.0 33.3
98
23.6 42.6 29.6
5
1.5
9
2.7
54
16.7 21.0 30.9 19.8 11.4
50 63
41
15.1 19.0 28.1 24.8 12.4
0.3
2
0.6
22.8 27.8 29.0 14.5
71
90 94 47
115 84
44
21.5 34.7 25.4 13.3
19
5.9
17
5.1
68 100
64
74
93 82
Definitions of abbreviations: R450 = daily administration of 450 mg of rifampin plus 300 mg of isoniazid; R600 = daily administration of 600 mg of rifampin plus 300 mg of isoniazid; R750 = daily administration of 750 mg of rifampin plus 300 mg of isoniazid; MA = moderately advanced; FA = far advanced.
883
USPHS COOPERATIVE TRIAL OF THREE RIFAMPIN-ISONIAZID REGIMENS
TABLE 2 DEVIATIONS FROM ASSIGNED RIFAMPIN-ISONIAZID THERAPY Rifamp in-lsonizaid Regimen TotalI
R750
R600
R450
(no.)
(%)
(no.)
(%)
(no.)
(%)
(no.)
(%)
822
100.0 18.7
167
59
100.0 17.8
43 27
5.2 3.3
9 4
5.4
55 17
100.0 17.0
331
40
100.0 24.0
324
154
5.2
2.4
10
3.1
17 13
3.9
23
2.8
7
4.2
11
3.4
5
1.5
18 9
2.2
5
3.0
6 4
1.9
7
2.1
1.2
8 8
1.0
4
2.4
1
0.3
5 3
0.9
1.0
8
4.8
7
0.9 0.7
2
1.2
2
0.6
3
0.9
0.9
3
0.9
0.6
1
0.6
3 1
0.3
3
0.9
81.3
127
76.0
269
83.0
272
82.2
Patients admitted Deviations Delinquent Adverse reaction Deaths unrelated to tuberculosis Administrative error Concurrent illness Treatment failure Protocol modification Nontuberculous mycobacterial disease Irregular pill taker Refused treatment " C o m p l e t e d " rifampin-isoniazid therapy
6 5 668
1.1
5.1
1.5
For definitions of abbreviations, see table 1.
developed arthritis and anemia after 46 days of therapy, and another patient had a febrile reaction after 5 days of therapy. Five patients developed rashes after 30 to 101 days of therapy, which ranged from a minimal skin reaction with pruritis to an erythema multiforme-type reaction. Of the 8 patients who deviated from their assigned regimen because of "treatment failure"
(table 3), five were judged to be clinically worse, although their sputum cultures remained negative. Three additional patients died of tuberculosis; two of them received the R750 regimen for only 2 or 3 days, and the other received the R450 regimen for only 10 days. Hematologic monitoring. During RIF-INH therapy, 6 patients had a white blood cell count of less than 3,000 cells per mm 3 : one received
TABLE 3 PATIENTS WITH ADVERSE REACTIONS OR TREATMENT FAILURES DURING RIFAMPIN-ISONIAZID THERAPY Rifa mpin-lsoniazi id Regimen
Adverse reactions Asymptomatic leukopenia Abnormal SGOT* concentration w i t h o u t symptoms Abnormal SGOT cbncentration with jaundice G l * symptoms w i t h o u t jaundice* * Gl symptoms w i t h jaundice Arthritis and anemia Fever Rash Treatment failures Clinically worse Death f r o m tuberculosis
Total
R450
R600
R750
27 2
4
10
13 2
8
1
5
2
2 2
1
6
1 1
1
3
3 1
1 1
1
3
5
2
8
4
1
5
3 1
1
3
Serum glutamic oxaloacetic transaminase. * Gastrointestinal. * *Abnormal SGOT concentration in addition to Gl symptoms. For definitions of other abbreviations, see table 1.
3 1 2
884
LONG, SNIDER, AND FARER
R450, 3 received R600, and 2 received R750. None of these patients had 2 or more counts of less than 3,000 cells per mm^, and RIF-INH therapy was continued in all 6 patients without subsequent adverse effects. During RIF-INH therapy, 84 patients (14.0 per cent) had one platelet count of less than 150,000 cells per mm*, and 12 (2.0 per cent) had 2 or more counts of less than 150,000 cells per mm 3 on consecutive monthly measurements. The incidence of these persistently low counts was 1.9, 1.2, and 2.7 per cent in the R450, R600, and R750 regimens, respectively. Eight patients (three on R600 and five on R750) had a single platelet count of less than 100,000 cells per mm 3 during therapy, and one R750 patient had 2 consecutive counts of this magnitude. Although there was a trend toward a dose-related effect of RIF on platelet counts, the differences among the regimens did not achieve significance. None of the patients with low white blood cell or platelet counts had clinical signs or symptoms attributable to the abnormal laboratory finding. SGOT monitoring. Of the 822 patients admitted to the study, 804 had serum for SGOT determination shipped to the Center for Disease Control for at least one of the 5 specified intervals during the 20 weeks of RIF-INH therapy. In table 4, these patients are listed by regimen according to the highest SGOT value obtained. Before the initiation of RIF-INH therapy, 233 of 751 patients (31 per cent) had increased SGOT values (> 40 KU). During therapy, 56 per cent of the 804 patients had SGOT values that never exceeded 40 KU (level 0).
Thirty-four per cent demonstrated SGOT values between 41 and 99 KU (level I); 8 per cent had SGOT values between 100 and 249 KU (level II); and the remaining 2 per cent (16 patients) had SGOT values equal to or greater than 250 KU (level III). These data were analyzed by stepwise multiple-regression analysis, using the highest value of SGOT observed at 4-week intervals during treatment as the dependent variable. The set of independent variables included RIF regimen, dosage of RIF per body weight (mg per kg), pretreatment weight, age, sex, race, admitted use of alcohol, extent of disease, presence of concurrent diagnoses at initiation of therapy, SGOT value before treatment, and presence of signs or symptoms of hepatic dysfunction during treatment. Of these independent variables, only race, use of alcohol, extent of disease, SGOT value before treatment, and presence of signs or symptoms were significantly related to the highest SGOT value. Nonwhite patients had higher SGOT values during treatment than did whites. As expected, the SGOT value before treatment, admitted use of alcohol, and presence of signs and symptoms of hepatic dysfunction were correlated positively with the highest SGOT value. Somewhat surprisingly, the extent of disease was inversely related to the highest SGOT value; i.e., patients with more extensive disease before treatment tended to have lower SGOT values during therapy. Although the preceding relationships were significant, only approximately 7.5 per cent of the variation in the dependent variable (highest SGOT value during treatment) was explained by this par-
TAB LE 4 NUMBER A N D PERCENTAGE OF PATIENTS WHOSE HIGHEST SGOT V A L U E OCCURRED A T SPECIFIED LEVELS D U R I N G 20 WEEKS OF R I F A M P I N - I S O N I A Z I D T H E R A P Y BY R E G I M E N Rifampin!Isoniazid Regi men Level of Highest SGOT Value Patients with SGOT data* 0: S G O T t < 4 0 K U * * 1: SGOT 4 1 - 9 9 KU II: SGOT 1 0 0 - 2 4 9 KU III: SGOT > 2 5 0 KU
Total
(no.) 804 451 270 67 16
R450
R750
R600
(%)
(no.)
(%)
(no.)
(%)
(no.)
(%)
100.0 56.1 33.6 8.3 2.0
162 89 60 7 6
100.0 54.9 37.0 4.3 3.7
318 175 111 29 3
100.0 55.0 34.9 9.1 0.9
324 187 99 31 7
100.0 57.7 30.6 9.6 2.2
Eighteen of 822 patients were excluded because no SGOT determinations were available at 4, 8, 12, 16, or 20 weeks of therapy. * Serum glutamic oxaloacetic transaminase. Karmen Units, 32° C. For definitions of other abbreviations, see table 1.
885
USPHS COOPERATIVE TRIAL OF THREE RIFAMPIN-ISONIAZID REGIMENS
ticular set of independent variables. Thus, other unknown factors must have contributed most of the observed variation. Furthermore, in this multivariate analysis, significant interrelationships among the "independent variables" cannot be excluded. Of the 353 patients who demonstrated increased SGOT during RIF-INH therapy, 121 (34.3 per cent) developed abnormal SGOT values during the first 4 weeks, 187 (53.0 per cent) during the first 8 weeks, and 254 (72.0 per cent) had abnormal SGOT values during the first 12 weeks of therapy. Supplemental hepatic dysfunction data. One hundred three patients were reported to have "suspected hepatic disease" because they had routine SGOT values of 100 KU or more, total bilirubin values of 2.5 mg per 100 ml or more, or jaundice reported during therapy. To evaluate these patients further, supplemental data were obtained from their hospital and clinic records. Only 32 of these 103 patients (31.1 per cent) developed overt signs or symptoms of an adverse reaction. Twenty-one became clinically jaundiced, and 18 developed gastrointestinal symptoms. Two of the 32 patients had a pruritic rash. Nearly all developed signs or symptoms during the first month of therapy. Slightly more than one half (56 or 54.3 per cent) of the 103 selected patients had other factors identified that may have caused or contributed to the hepatic dysfunction. Forty of them abused alcohol, 14 were addicted to heroin, and two were taking hepatotoxic medications in addition to RIF-INH. Sputum conversion during the initial phase. Of the 668 patients who completed initial therapy, 116 were excluded from this analysis of efficacy for the following reasons: (1) the pretreatment sputum cultures from 39 patients contained Mycobacterium tuberculosis organisms that were resistant to EMB, INH, and/or RIF; # 33 patients had negative cultures; nine had contaminated cultures; and 26 had pretreatment cultures containing nontuberculous mycobacteria. The remaining 9 patients were excluded because they had previously received antituberculosis drugs for more than 14 days. Thus, the
analysis of efficacy, as measured by the rate of bacteriologic conversion, was based on 552 patients. Approximately two thirds of the patients admitted to each regimen were included in this efficacy analysis; their initial characteristics had a distribution similar to that shown in table 1. Shown in figure 2 is the per cent of cultures that contained M. tuberculosis organisms at 4week intervals among the patients in the 3 RIF regimens. At 12 weeks, 12 per cent of those on the R450 regimen, 7 per cent of those on the R600 regimen, and 6 per cent of those on the R750 regimen had positive cultures. By 20 weeks, 7.7 per cent of those on the R450 regimen but only 0.5 per cent on the R600 and 0.9 per cent on the R750 regimen had positive cultures. There was no significant difference in the percentages between the R600 and R750 regimens at any 4-week interval. However, at 16 and 20 weeks, the percentages of positive cultures for the R450 regimen were significantly greater than were those for the R600 and R750 regimens (P < 0.01). The per cent of cultures positive for M. tuberculosis among patients who received RIF, based on dosage, in mg per kg, is shown in figure 3. At 12 weeks, 9.5 per cent of those who received less than 9.0 mg per kg of body weight, 7.6 per cent of those who received 9.0 to 12.9 mg per kg, and 5.4 per cent of those who received 13.0 mg or more per kg had positive cultures. By 20 weeks, 4.7 per cent of those who received less than 9.0 mg per kg, 1.4 per cent of those who re-
8
* Of the 39 patients who had resistant organisms in the pretreatment culture, 20 showed resistance to INH alone, seven to RIF alone, one to EMB alone, three to RIF and INH, six to INH and EMB, and two to all 3 drugs.
12
16
20
WEEKS
Fig. 2. Per cent of cultures containing Mycobacterium tuberculosis organisms among patients completing rifampin-isoniazid (RIF-INH) therapy, by week and regimen.
886
LONG, SNIDER, AND FARER
RIFAMPIN mg/kg
signed to the R450 regimen, but only 19 per cent of those assigned to R600 and 2 per cent of those assigned to R750 received less than 9.0 mg per kg per day.
