ABSTRACT

Tuberculosis chemoprophylaxis is a therapeutic measure for the prevention of infection by Mycobacterium tuberculosis or to avoid development of the disease in individuals already infected with it. Isoniazid is the most commonly used therapy; however, the use of rifampicin and pyrazinamide has recently been introduced. The objectives of this study were to review the results of the principal studies evaluating the indications for chemoprophylaxis with isoniazid alone and in association with other drugs, its efficacy in the prevention of tuberculosis with respect to the different risk groups and the alternative regimens available. A systematic revision of the medical literature was carried out with particular emphasis on clinical trials and meta-analyses. Official records were also consulted. Those studies involving randomized clinical trials on the use of isoniazid, rifampicin or pyrazinamide in HIV-positive or negative patients were selected. Isoniazid continues to be effective for the prevention of tuberculosis in populations of both HIV-negative and HIV-positive individuals. The standard dose of 5-15 mg/kg/day has shown similar protection over treatment periods ranging from six to twelve months. The risk of developing hepatitis was less than 1%; however monitoring during treatment is recommended in patients over 35 years of age and in users of alcohol. Studies involving treatment regimens with other forms of medication were inconclusive and new studies would have to be performed to evaluate the efficacy of these regimens in populations at high risk of developing tuberculosis.

Keywords: Tuberculosis. Chemoprophylaxis. Isoniazid.

RESUMO

A quimioprofilaxia da tuberculose constitui-se numa medida terapêutica para a prevenção da infecção pelo Mycobacterium tuberculosis ou para evitar o desenvolvimento da doença nos indivíduos infectados. Geralmente baseia-se na administração de isoniazida. Entretanto, o uso de rifampicina e pirazinamida vem sendo recentemente introduzido. Este trabalho tem como objetivo revisar os resultados dos principais estudos que avaliaram as indicações da quimioprofilaxia com isoniazida e sua associação com outras drogas, sua efetividade na prevenção da tuberculose considerando os diversos grupos de risco, e as alternativas do uso de outros esquemas. Procedeu-se à revisão sistemática da literatura, com ênfase em ensaios clínicos e meta-análises. Foram consultados também os documentos oficiais. Foram selecionados aqueles estudos que envolveram ensaios clínicos ramdomizados com uso de isoniazida, rifampicina ou pirazinamida em pacientes HIV positivos ou negativos. Concluiu-se que a isoniazida continua sendo efetiva na prevenção da tuberculose na população de indivíduos HIV negativos e de HIV positivos. A dose padrão de 5 a 15 mg/kg/dia tem mostrado proteção similar para períodos de tratamento de seis e doze meses. O risco de desenvolver hepatite foi menor que 1%, sendo recomendada sua utilização com acompanhamento nos indivíduos com idade superior a 35 anos e usuários de álcool. Os estudos com esquemas de tratamento utilizando outros medicamentos não foram conclusivos, sendo necessária a realização de novos estudos para avaliação da efetividade desses esquemas em populações de alto risco de desenvolver tuberculose.

Palavras-chave: Tuberculose. Quimioprofilaxia. Isoniazida.

Introduction

Tuberculosis(TB) chemoprophylaxis consists of the administration of isoniazid (INH) toindividuals uninfected with Mycobacteriumtuberculosis (Mtb) in order to prevent infection (primary chemoprophylaxis)or to infected individuals in order to avoid the development of the disease(secondary prophylaxis). However, other drugs, such as rifampin (RIF) andpyrazinamide (PZA), have recently been introduced.(1)

Since1952, INH has been used for TB treatment in the USA and in Brazil. Since 1965,the American Thoracic Society (ATS) has recommended the use of INH for TBprophylaxis in individuals presenting positive reactions in tuberculin tests(tuberculin reactors). In the early 1970s, the use of INH was associated withsevere hepatotoxicity, which can be fatal. From 1974 on, the ATS restricted theuse of INH to tuberculin reactors younger than 35 and those older than 35 who presenteda high risk for TB reactivation, providing that they had not been diagnosedwith liver disease. Later, in 1983, it was suggested that patients in certainage brackets who were receiving INH be clinically monitored through periodicassessment of liver function. The use of INH should be discontinued ifaminotransferase levels reach three to five times higher than normal.(2-4)

The objective of this study was to analyze and review the principal studies on secondary chemoprophylaxis with INH and its effectiveness in TB prevention. We specifically intended to review questions regarding preventive treatment for TB, identifying the effectiveness of INH in the various risk groups, including those infected with human immunodeficiency virus (HIV), as well as identifying alternative regimens, determining duration of treatment, evaluating supporting evidence, and assessing adverse effects.

Preventive use of isoniazid in risk groups

Based on the sensitivity and specificity of purified protein derivative (PPD) tuberculin tests and TB prevalence in the various risk groups, several cutoff points have been recommended. An induration 5 mm is considered positive in Mtb-infected individuals who are at high risk for developing active TB.(5) An induration 10 mm is considered positive in those who are likely to have had a recent infection or in those who present any type of clinical condition that increases their risk of developing the disease. In individuals at low risk, for whom tuberculin tests are not usually recommended, an induration 15 mm should be considered positive.(4,5)

Chemoprophylaxis in non-infected individuals (primary)

In Brazil, the Ministério da Saúde (Health Ministry) recommends that newborns cohabiting with active TB individuals be treated with INH during the first three months of life and subsequently be submitted to tuberculin testing. If the results are positive (induration 10 mm), chemoprophylaxis should be continued for an additional three months. If the results are negative, administration of the drug should be interrupted and the children vaccinated with the bacillus Calmette-Guérin (BCG) vaccine.(6) These recommendations are similar to those made by the ATS and BTS (British Thoracic Society).(4,7) However, American public health officials recommend that children born to mothers infected with HIV be submitted to annual tuberculin tests.(8)

Chemoprophylaxis in infected individuals (secondary)

Itis recommended that INH be administered to children younger than 15 who havenot vaccinated with BCG and have had contact with pulmonary TB patients,assuming that those children present no signs of active disease and have strongPPD reactions (induration = 10 mm). It is also recommended for thosechildren who were vaccinated with BCG and have a PPD reaction 15 mm.(1) This recommendation is in accordance with thefindings of a controlled clinical study on contact with TB patients in thehome, in which it was demonstrated that TB incidence was highest among childrenless than 5 years of age (12.9/1000). The occurrence of TB in this age group isalways considered a recent infection, and INH has demonstrated high efficacy(87%) against such infections.(9)

Thegeneral consensus in the literature is that children younger than 5, teenagersand young adult tuberculin reactors constitute high-risk groups and should beadvised to accept chemoprophylaxis.(5,7,10)

Inan observational study comprising 2494 children, the incidence of TB amongchildren receiving a 12-month course of INH chemoprophylaxis was 3.2/1000/year,compared with 30.2/1000/year in the placebo group. Administration of INHprovided 90% protection and no deaths were reported among those with pulmonaryTB. In addition, no disease reactivation was observed during adolescence, whichsuggests that the treatment provided long-term protection.(11)

Individuals who have recently (within the past 12 months) undergone tuberculin conversion (to 10 mm of induration) are also considered a high-risk group. Children younger than 4 who have positive reactions (10 mm) should be considered recent reactors.(4)

Thereis a high prevalence of active TB among indigenous populations. For example,among members of the Yanomami tribe, the rate is 6.4% of 625 individuals, whichis approximately 100 times higher than the prevalence in the state of Amazonasin general (68/100,000).(12,13) This calls forspecific prevention measures. In the USA, individuals from various ethnicgroups (African Americans, Hispanic Americans, and Native Americans) that havelimited access to health care services are considered high-risk groups, andpreventive therapy is therefore recommended whenever these individuals becometuberculin reactors, even if there are no other risk factors for TB.(4)

Chemoprophylaxis,under strict medical supervision, is also recommended for those who may nothave active TB but have severe reactions to tuberculin tests, as well as forthose have not previously undergone chemotherapy and those who present otherextraneous conditions that may be conducive to TB development. Such conditionsinclude alcoholism, insulin-dependent diabetesmellitus, silicosis, severe kidney disease, sarcoidosis, lymphoma, long-termcorticosteroid use (immunosuppression dose), cancer chemotherapy, use ofimmunosuppressants, and presenting radiographs compatible with latent TB. Thisrecommendation is based on the fact that chronic diseases such as diabetes,lymphoma, and severe kidney disease all result in some level ofimmunosuppression or require the use of corticosteroids as part of thetreatment, thereby increasing the risk of developing reactivation TB.Nevertheless, this risk has not been precisely determined. It is known that,due to its immunosuppression effect, a course of prednisone at doses higherthan 15 mg for 2 or more weeks substantially reduces tuberculin reaction,increasing the risk for the development of TB. There is no evidence thatchemoprophylaxis is recommended for individuals receiving less than 15 mg ofprednisone or its equivalents. It is necessary to evaluate patient risk forliver injury.(3-5)

Indeveloped countries, immigrants from regions where there is high prevalence ofTB are also considered a high-risk group. In the USA, the incidence of TB inthis population increased between 1986 (27.1/100,000) and 1993 (33.6/100,000),representing 21.6% of all TB cases in 1986 and 29.6% of all cases in 1993. Atthe end of an 8-year study, incidence among immigrants (30.6/100,000/year) wasdetermined to be 4 times higher than among non-immigrants (8.1/100,000/year).(14) The Pan American Health Organization and the BTS alsorecommend that this group receive preventive therapy.(15,16)

Treatmentfor latent infection with Mtb during pregnancy should be delayed until afterdelivery and should be administered with extreme caution in high-risk cases.(16) The ATS recommends its implementation in pregnantwomen infected with HIV and recently exposed to active TB or having previouslybeen tuberculin reactors.(4) National healthcare officials in the USA recommend the administration of INH after the firstthree months of pregnancy for women who are HIV-positive tuberculin reactorsand have contact with active-TB patients.(8)

Other groups that have been considered high-risk include individuals whose body mass is higher or lower than the ideal and those submitted to transplants, as well as residents of or workers at certain institutions - such as hospitals, correctional centers, elderly homes or centers for patients with acquired immunodeficiency syndrome (AIDS).(4,5,7)

Clinical evaluation of the use of isoniazid in chemoprophylaxis

Theeffectiveness of INH use (at a dose of 5 mg/kg, maximum 300 mg/day) inchemoprophylaxis has been established in double-blind, placebo-controlled,randomized clinical trials carried out since the 1960s. In a study conducted inAlaska, estimated effectiveness for one-year treatment was 75% in the first 4years of follow-up, and a long-lasting protective effect was shown: 70% after15 years and 50% after 19 years.(10,17,18) In the USA, aclinical trial showed a decrease in TB incidence among individuals havingcontact with TB in the home and treated with INH, which prevented primarypulmonary TB and extrapulmonary TB in uninfected children. The same studyshowed a decrease in the number of pulmonary TB cases among adults.(19) The authors of these studies reported neither INHresistance nor significant side effects. Another study showed that, in apopulation of war veterans in San Francisco, USA, 60% were protected against TBreactivation.(20) A recentmeta-analysis among HIV-negative individuals, including 11 controlled clinicaltrials, found a relative risk (RR) of 0.40 and a 95% confidence interval (95%CI) of 0.31-0.52. This corresponds to 60% protection in various known high-riskgroups. Such groups include those having had contact with active-TB patients,those living in areas with a high TB prevalence, those institutionalized forchronic psychiatric disorders, those recently converted to tuberculin reactors,those with no previous history of chemotherapy whose X-rays are compatible withlatent TB, those with pneumoconiosis due to silica exposure and those havinghad a kidney transplant. This finding is in accordance with those in theliterature if we consider that the level of protection is adequate forpopulations at low risk for TB. In those at high risk, considering the durationof treatment, this protection would have only a minor impact on the preventionof new cases.

Individualswith severe reactions to PPD (induration 10 mm), ranging from 28 to 65cases per 100,000 inhabitants, are known to be at high risk for contractingactive TB. However, this risk decreases with the passage of time.(18)

Randomizedcontrolled clinical trials involving patients with pulmonary fibrotic lesionshave been used to compare the effectiveness of INH regimens ranging from 3months to 2 years. These studies have shown that, after a follow-up period of 5years, a 3-month regimen decreased TB incidence by 21%, a 6-month regimen by65% and a 12-month regimen by 75%. In comparison to patients in the placebogroup, the ratio between benefit and risk was 1.2 for the first regimen, 2.6for the second and 2.1 for the third. In 2003, Smieja et al.,(21) in the previously mentioned meta-analysis, reportedno significant differences between 6-month and 12-month INH regimes (RR = 0.44and RR = 0.38, respectively; p =0.08). However, this difference may be significant, depending on the risk ofdeveloping active TB. For example, it has been estimated that it would benecessary to treat 179 individuals for 6 months in order to prevent 1 TB caseunder low-risk conditions (in adult tuberculin reactors with normal chestX-rays and risk of developing hepatitis lower than 0.5%). Using the 12-monthregimen, 161 individuals would have to be treated in order to prevent 1 case.In patients at high risk for TB infection (20%), it has been estimated that 1case is prevented for every 8 or 9 patients treated.(21) Another aspect to be considered is that the shorterthe duration of treatment, the higher the compliance, that is, more people willcomplete the treatment (78% in 6-month regimens and 68% in 12-month regimens).(22)

In a clinical trial in Canada, the effectiveness of INH alone and in combination with para-aminosalicylic acid (PAS) in the prevention of TB reactivation was estimated in individuals diagnosed with latent TB. After 18 months of treatment, the authors observed no benefits provided by the use of INH alone or in combination if the period of treatment was less than 6 months, with reactivation rates similar to those seen in the control group. The use of INH combined with PAS effected a 90% decrease in the reactivation rate, in comparison with a 70% decrease from the use of INH alone. They also reported an increase (from 30% to 60%) in the incidence of mild adverse effects with the combination of INH and PAS. These side effects caused noncompliance with treatment to increase from 19% to 42%, thereby limiting INH use.(23)

Chemoprophylaxis with other treatment regimens

Useof 6-month and 12-month INH regimes in TB chemoprophylaxis has been standardpractice in the USA for 30 years. Later, this recommendation was revised sincethe decrease in the incidence of TB was lower than expected, and mortality dueto liver failure began to be related to the use of INH. Moreover, poorcompliance with the treatment (due to its long duration) and the occurrence ofco-infection with HIV/AIDS has motivated studies on the effectiveness ofchemoprophylaxis using other drugs, which, of necessity, requires the study oftheir implementation. In 2000, the ATS released a series of recommendations onchemoprophylaxis, suggesting that a 2-month course of RIF or PZA be used as asubstitute for INH. These findings were based on clinical trials andexperiments in animals.(5)

In1998, the BTS recommended that TB chemoprophylaxis consist of the use of INHfor 6 months or the combination of INH and RIF for 3 months. Based oncontrolled clinical trials, the use of both drugs combined has shown similareffects to the use of INH alone, with no increase in the number of adverseeffects.(15) In 2000, it wassuggested that, when using both drugs in combination, the duration of treatmentshould be reduced to 2 months, and that INH alone should no longer be used forprevention since both regimens had proven to be similarly beneficial. Treatmentof high-risk, HIV-negative patients with the INH and RIF regimen for 3 monthsproved to be as effective as the INH regimen for 6 months. The RIF and PZAregimen for 2 months is less well tolerated than regimens using INH or RIFalone. The combination of INH and RIF is well tolerated by children for periodsequal to or greater than 3 months. However, the BTS still recommends the use ofRIF combined with INH for 3 months as an alternative to the 6-month INHregimen.(7)

Inan experimental study using guinea pigs vaccinated with BCG, the efficacy of a6-month course of INH was compared to that of 2-month courses of RIF alone, RIF+ PZA, and RIF + PZA + INH. Doses, in relation to serum levels, were equivalentto those used in human populations. Treatment started 2 weeks after infection.After 2 months of therapy using the various regimens, the study showed that thenumber of positive results in spleen culture was 100% in specimens from guineapigs treated with the first regimen, 50% with the second regimen, 0% with thethird, and 80% with the last. After the use of INH for 6 months, the percentageof positive spleen cultures was 38%. The study also tested the extremeeffectiveness of the RIF + PZA regimen compared to the RIF regimen for 3months. Using similar procedures, spleen cultures were 100% positive using theINH regimen for 6 months, compared to 20% using RIF for 3 months, 0% using RIF+ PZA for 2 months, and 80% using the combination of the 3 drugs for 2 months.Six months after the end of the treatment, the proportion of positive resultswere 100%, 60%, 56%, and 95%, respectively. The 2-month RIF + PZA regimen and the3-month RIF regimen proved to be more effective than the 6-month INH regimen.(24)

Inanother experimental study, the efficacy of RIF in various treatment regimens(RIF alone, RIF + PZA and RIF + INH + PZA) was evaluated. Infected guinea pigswere initially treated with INH + PZA for 7 weeks. Later, 4 groups comprising47 guinea pigs each, received the various treatment regimens for 6 weeks. Afifth group consisting of 10 guinea pigs was the control group and received INH+ PZA for 13 weeks. After a period of between 26 and 35 weeks, the animals weresacrificed so that spleen culture results could be evaluated. The percentage ofpositive results were 74% in the RIF-only group, 63% in the RIF + INH group,and 53% in the RIF + PZA + INH group. These differences were not statisticallysignificant, so efficacy of the various regimens was determined to be quitesimilar.(25)

Aplacebo-controlled, double-blind clinical trial conducted in Hong Kong andinvolving male patients diagnosed with silicosis compared four treatmentregimens (RIF for 3 months = 142 patients; INH + RIF for 3 months = 161patients; INH for 6 months = 123 patients; placebo = 133 patients) over afollow-up period of 5 years. The probability of developing TB was 10% when RIFwas used, 16% when the combination of INH and RIF was used, 17% when INH alonewas used, and 27% when a placebo was used. Regimens including RIF proved lesslikely to cause hepatotoxicity. Regimens using INH alone or in combinationshowed higher aminotransferase concentrations in the serum of patients duringthe period of treatment (p <0.001). The authors recommended that other studies be performed so that moreeffective and safer treatment regimens for patients diagnosed with silicosiscould be found.(26)

Ina placebo-controlled, double-blind randomized clinical trial carried out inSouth Africa, the effectiveness of two regimens in preventing TB in patientsdiagnosed with silicosis. Study subjects received either 600 mg of RIF alone or400 mg of INH combined with 1250 mg of PZA, the INH and PZA in accordance withrecommended doses. After a follow-up period of 4 years, TB was diagnosed in 11males who had received medication (annual incidence of 1480/100,000) and in 15males of the placebo group (p =0.40). The authors determined that the lack of treatment regimen efficacy wascorrelated with silicosis as well as with a high risk of reinfection due to thehigh incidence of TB in that community.(27)

Recently, the American entities the ATS and the CDC recommended that the combination of RIF and PZA not be used in any preventive TB treatment regimens. According to data from patient cohort studies conducted in the USA, high rates of severe liver impairment and mortality result from the use of the two drugs. Other treatment alternatives should be considered for TB prevention.(28)

Prevention of drug-resistant TB

Incases of INH resistance, the ATS recommends the use of RIF (10 mg/kg). Theyalso recommend the use of ethambutol (EMB, 15 mg/kg) when the strain is provento be susceptible. For both regimes, 6-month courses are recommended for adultsand 9-month courses for children. If there is evidence of contact withmultidrug-resistant TB, daily doses of the EMB + PZA combination (20 mg/kg and25 mg/kg, respectively) for 6 months are recommended. Although there arelimited data on the use of quinolones as therapeutic agents against TB whenthere is resistance to EMB, the use of PZA combined with a quinolone (400mg/day ofloxacin or 750 mg/day ciprofloxacin) for 6 months has been recommended.It is recommended that 6- to 9-month courses of RIF, alone or in combinationwith EMB, be given to children and immunosuppressed adults who have had contactwith patients infected with INH-resistant TB strains when strains aresusceptible to the latter drug(4,7)

It has been recommended that HIV-positive patients who have had contact with INH-resistant active-TB patients be evaluated on a case-by-case basis. For such patients, the decision to use other drugs should be based on the results of susceptibility testing of the Mtb strain isolated, as well as on the guidelines established by the respective health care authorities.(1,5)

Tuberculosis prevention in HIV-positive patients

Individualswho are HIV positive are at high risk for becoming infected with TB. Twospecific conditions should be taken into consideration for this group: reactionto tuberculin tests is poor in individuals with severe immunodeficiency, andchest X-rays of individuals with coinfection present atypical characteristics.These two aspects make it more difficult to distinguish between latentinfection and active disease. However, the International Union AgainstTuberculosis and Lung Disease and the World Health Organization recommendpreventive therapy for patients with coinfection.(29)

It is recommended that HIV-positive patients, regardless of age, with no evidence of active TB or previous history of TB treatment whose reaction to tuberculin test is equal to or greater than 5 mm and whose chest X-rays are normal should be treated for latent Mtb infection. Among this population, the risk of acquiring TB has been estimated at between 1.7 and 7.9/100 people/year(30), which is why the identification of coinfection and the administration of preventive treatment are highly relevant. Efficacy of this therapy has not been confirmed in individuals with negative tuberculin test results (anergic individuals). Some clinical trials involving this group have shown that patients do not markedly benefit from preventive treatment with INH. Problems with drug absorption and severe immunosuppression are likely to interfere with therapy efficacy. However, the use of INH for 12 months is recommended for HIV-positive patients, even for those whose PPD results are negative, if they have had recent contact with individuals diagnosed with infectious pulmonary TB. For this group of patients, decisions on the use of chemoprophylaxis should be considered on a case-by-case basis.(4,6,8,29)

Isoniazid and tuberculosis prevention in HIV-positive patients

InBrazil, the TB chemoprophylaxis treatment of choice for HIV-positive patientsis 5 to 10 mg/kg/day of INH (maximum dose: 300 mg/day) for 6 consecutivemonths.(6) In the USA, thedrug of choice is also INH (at the same dose levels), but is administeredeither daily or twice weekly for 9 months. The ATS recommends that adultsshould continue therapy for 12 months.(5,8) Nevertheless, the American Academy of Pediatrics recommends that, forchildren, the duration of such treatment be less than 9 months.(4)

Variouscontrolled, randomized clinical trials have shown the effectiveness of INH inadult populations coinfected with HIV and TB. In five of these studies, theeffectiveness of INH was assessed by comparing patients receiving INH to thosereceiving a placebo and to those receiving no treatment. Between 1986 and 1992,the incidence of TB in Haiti decreased 83% in tuberculin test reactorsreceiving daily doses of INH for 12 months. This protection continuedthroughout a four-year period of follow-up.(31)

A6-month course of daily administration of INH in tuberculin reactors andnonreactors has been evaluated in various studies, with varying results. In astudy carried out in Uganda, the authors reported 68% protection.(30) In Kenya, the rate of protection was lower (40%; 95%CI = 0.23 - 1.60).(33) In Zambia,protection provided by the use of INH twice a week for 6 months was consideredlow (38%; 95% CI = 0.38 - 0.99). Protection was higher (70%) in tuberculinreactors, but results were not statistically significant due to the smallnumber of individuals in this group.(34) Whalem et al., in 1997, and Gordin et al., in 2000, also reportedlittle protection provided by INH in anergic individuals.(32,35)

Ina meta-analysis encompassing seven studies that comprised a total of 2367people treated with INH and 2162 controls, the RR of developing TB was 0.58 inthe treated group (95% CI = 0.43 - 0.80), corresponding to 42% efficacy. The RRamong tuberculin reactors was 0.40 (95% CI = 0.24 - 0.65; 60% efficacy),compared with 0.84 among tuberculin nonreactors (95% CI = 0.54 - 1.30; 16%efficacy). Estimated RR for mortality prevention was 0.94 (95% CI = 0.83 -1.07), corresponding to 6% protection. Among tuberculin reactors, the RR formortality prevention was 0.79 (95% CI = 0.37 - 1.70) versus 1.0 for tuberculinnonreactors (95% CI = 0.90 - 1.17). The authors of this study concluded thatadministration of INH for 6 months in HIV-positive patients who are tuberculinreactors reduced TB incidence by 60%.(36)

In the USA, a study of the protective effect of a 24-month course of INH chemotherapy on the incidence of diseases caused by mycobacteria in HIV-positive, injection drug users showed that TB risk decreased in 83% of these individuals with the use of intermittent, twice-weekly doses.(35)

Use of other drugs for preventing active tuberculosis in HIV-positive individuals

Severalcontrolled randomized clinical trials have been carried out in order toevaluate proper duration of treatment and the effects of combining other drugswith INH. The combined use of INH and RIF for 3 months in daily doses has beenshown to provide 59% protection. It has also been shown that dailyadministration of a three-drug combination of INH, RIF and PZA provides 57%protection. These results were similar to those found for the use of INH alonefor 6 months.(5,32)

Thelevels of protection provided by twice-weekly, 3-month treatment regimens usingtwo drugs combined (INH + RIF) or three drugs combined (INH + RIF + PZA) were59% and 57%, respectively. These results were similar to that regimen using INHtwice a week for 6 months.(32) The combined useof RIF and PZA for 3 months provided 42% protection, similar to that providedby the use of INH alone, twice a week, for 6 months.(34) In Haiti, the level of protection provided by tworegimens (2 months of RIF + PZA and 6 months of INH twice a week) was similarafter a follow-up period of 12 months.(38) In another study carried out in the USA, Haiti, Brazil and Mexico, theauthors reported that daily administration of RIF combined with PZA for 2months provided the same level of protection against TB as the daily use of INHfor 12 months.(35) In Hong Kong, astudy with patients suffering from silicosis showed that daily administrationof RIF for 3 months provided protection (90%) similar to that of the INHregimen for 6 months (86%).(26)

Astudy involving a cohort of HIV-positive tuberculin reactors was carried out inorder to determine the contribution of chemoprophylaxis using INH alone for 12months to the survival of these patients. The authors reported that thischemoprophylaxis regimen was related to decreased risk for active TB - annualincidence of 2% in the study group and 4.8% in the placebo group. These resultssuggest that chemoprophylaxis contributes to the survival of HIV-positivepatients in areas where there is high TB prevalence.(39)

Ina randomized clinical trial carried out in Spain and comprising 133 patients,the authors evaluated compliance with treatment and treatment tolerance, aswell as the efficacy of a 12-month course of chemoprophylaxis with INH alone incomparison to a 3-month course of RIF combined with INH. The authors reportedthat the incidence of active TB was 4.23/100 people/year in the INH-only group,and 2.08/100 people/year in the RIF + INH group. The RR of acquiring active TBwith the RIF + INH regimen was 0.51 (95% CI = 0.09 - 2.08) when compared to theuse of INH alone.(40)

Itis recommended that regimens targeting HIV-positive patients not include PZA.Severe liver damage reported in relation to daily administration of the RIF-PZAcombination to HIV-negative patients. In devising treatment strategies forlatent TB infection in HIV-positive patients exposed to strains resistant toINH or RIF, the RR of exposure to resistant strains must be taken intoconsideration. Therefore, each case must be evaluated in consultation withlocal health care authorities.(4,8)

In the USA, other drugs have been used. One such drug is rifabutin (RFB), which has been used to prevent the dissemination of infection with Mycobacterium avium. For HIV-positive patients, American health care officials currently recommend the use of RFB for a period of 4 months or RFB combined with PZA for 2 months. If there is resistance to INH or RIF, the use of RFB must be based on the risk of exposure to resistant strains, so each case must be considered separately. Since RFB interacts with protease inhibitors and with non-nucleoside reverse transcriptase analogs, it is recommended that doses be reduced by half (from 300 mg/day to 150 mg/day). Prophylaxis with RIF should not be used as a routine preventive treatment since resistance may occur.(8)

Toxic effects of isoniazid

Noadverse effects, such as hepatitis, were reported in the first studies of INH.(9) In 1972, the first evidence of adverse effects wasreported in patients submitted to therapy with INH in the USA - increasedtransaminase levels and fatal hepatitis. During therapy, 19 of the 2321patients treated (8.2 cases/1000 people/9 months) showed clinical signs ofliver disease and the deaths of two patients were attributed to the medication.(41) A clinical trial carried out by public health careagencies in the USA showed that, in a group of 13,831 patients receiving INH,1% developed hepatitis, and the incidence was higher (2.3%) in patients overthe age of 50. In addition to age, alcoholism has been shown to correlatesignificantly with a higher risk of developing hepatitis.(42) In a study conducted in Hong Kong, similar resultswere found for a group comprising 679 silicosis patients, 1% of which developedliver damage. Of the 9 patients diagnosed with liver injury, 4 had received theINH regimen, and 5 had been given a placebo. Only one patient receiving the INHregimen developed signs of hepatitis.(26) Other authors have reported rates of risk for acquiring hepatitis fromINH use ranging from 2% to 10%, with 5% to 10% estimated mortality rates.Clinical monitoring and cautious use of chemoprophylaxis are recommended forpatients who frequently use alcohol, have pre-existing liver or kidney injury,or are simultaneously receiving other drugs that are toxic to the liver orpancreas.(43)

Astudy carried out in Spain involving HIV-positive patients compared twodifferent treatment regimens (a 3-month course of 300 mg/day of INH + 600mg/day of RIF versus a 12-month course of 300 mg/day of INH). The authorsdemonstrated that only 18% of patients receiving INH + RIF showed signs ofhepatotoxicity, compared to 41% of patients receiving INH alone (RR = 2.22; 95%CI = 1.23 - 4.01).(38)

Thefact that the population under study presented a high prevalence of hepatitis Cmay explain the high levels of transaminases. However, between patients whopresented anti-hepatitis C antibodies and those who did not, there were nosignificant differences in relation to the incidence of adverse effects andhepatotoxicity with the regimen adopted. Duration of treatment correlated withthe appearance of toxic effects, independently of the type of drug employed.(40)

A study comprising 28,000 pulmonary fibrosis patients presenting positive tuberculin results, selected from 115 dispensaries in various countries, showed that the incidence of hepatitis in those submitted to chemoprophylaxis with INH for 12, 24 or 52 weeks was 0.5%, compared with 0.1% in those receiving a placebo. Reducing the duration of treatment has been found to reduce the risk of acquiring hepatitis in 1.6/1000 people submitted to the 6-month regimen and in 2.7/1000 people submitted to the 3-month regimen.(22) In 1979, Kopanoff et al. reported an incidence of 92 probable and 82 possible cases of hepatitis among 13,838 people using INH.(42) In 1997, Salpeter et al.(44) identified two deaths attributable to the use of INH within a group of approximately 200,000 patients treated for 6 months, concluding that the incidence of INH-related mortality is low.

Conclusions and recommendations

Thevarious studies reviewed have shown that INH continues to be effective in theprevention of active TB in both HIV-negative and HIV-positive populations. Ourresults are statistically significant and clinically relevant.

Inthe majority of studies, treatment regimes involving INH are typically 6 monthsor 12 months in duration, and the dosages used are 5-15 mg/kg/day (maximum, 300mg/day). Similar protection has been achieved with both treatment durations.Protection is significantly less when regimens shorter than 6 months areemployed, but there is no significant increase in protection if regimens longerthan 12 months are used. The 12-month regimen should be recommended for veryhigh-risk groups.

Inusing INH for TB chemoprophylaxis, the risk of toxic effects, especiallyhepatitis, appearing, although significant, is outweighed by the benefits ofits use. Age (over 35) and alcohol use have been associated with higher risksof developing hepatitis than that presented by INH use.

Furtherstudies are necessary in order to evaluate the effectiveness of these, andother, regimens that include INH when employed in populations at high risk ofdeveloping active TB.

References

1. Ministério da Saúde. Tuberculose Guia de Vigilância Epidemiológica. Brasilia: MS; 2002.
2. American Thoracic Society. Preventive Treatment of tuberculosis. Am Rev Respir Dis 1971;104: 460-463.
3. American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children. Am Rev Respir Dis 1986;134:355-63.
4. American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children. Am J Respir Crit Care Med 1994;149:1359-74.
5. Ministério da Saúde. Manual Técnico para o Controle da Tuberculose. Departamento de Atenção Básica. . Brasília: MS;2002.
6. Center for disease control and prevention.Targeted Tuberculin Testing and treatment of Latent Tuberculosis Infection. Recommendations and reports. M.M.W.R 2000;49(RR-6):1-54.
7. British Thoracic Society. Control and prevention of tuberculosis in the United Kingdom: code of practice 2000. Thorax 2000;55:887-901.
8. Masur H, Kaplan JE, Holmes K. Guidelines for preventing opportunistic infections among HIV-infected persons-2002. MMWR 2002;51(RR08);1-46
9. Ferebee SH, Mount FW. Tuberculosis morbidity in a controlled trial of the prophylactic use of isoniazid among household contact. Am Rev Respir Dis 1962; 85: 490-510.
10. Comstock GW, Ferebee SH, Baum C. Isoniazid prophylaxis among alaskan eskimos: a progress report. Am Rev Respir Dis 1974;110:195-7.
11. Hsu KHK. Thirty years after isoniazid. JAMA 1984;251(10):1283-5.
12. SousaAO, Salem JI, Lee FK, Verçosa MC, Cruaud P, Bloom BR, et al. An epidemic of tuberculosis with a high rate of tuberculin anergy among a population previously unexposed to tuberculosis, the Yanomami Indians of the Brazilian Amazon. Proc Natl Acad Sci1997;94:13227-32.
13. Escobar AL, Coimbra CEA, Camacho LA, Portela MC. Tuberculose em populações indígenas de Rondônia, Amazônia, Brasil. Cad. Saúde Pública 2001;17(2):285-8.
14. McKenna MT, McGray E, Onorato I. The epidemiology of tuberculosis among foreign-born persons in the United States, 1986 to 1993. N Engl J Med 1995;332(16):1071-6.
15. British Thoracic Society. Chemotherapy and management of tuberculosis in the United Kingdom: recommendations 1998. Thorax 1998;53: 536-548.
16. Organización Panamericana de la Salud. El control de las enfermedades transmisibles. Publicação Científica nº 581 2001;1- . ***
17. Comstock GW, Hammes LM, Pio A. Isoniazid prophylaxis in Alaskan boarding schools. Am Rev Respir Dis 1969;100(6):773-9.
18. Comstock GW, Baum C, Snider DEJ. Isoniazid prophylaxis among alaskan eskimos: a final report of the Bethel isoniazid studies. Am Rev Respir Dis 1979;119:827-31.
19. Ferebee SH, Mount FW, Murray FJ, Livesay VT. A controlled trial of isoniazid prophylaxis in mental institutions. Am Rev Respir Dis 1963;2:161-175.
20. Falk A, Fuchs GF. Prophylaxis with isoniazid in inative tuberculosis. Chest 1978;73:44-8.
21. Smieja M, Marchetti CA, Cook DJ, Smail FM. . Isoniazid for preventing tuberculosis in non-HIV infected persons. The cochrane library 2003. Website: http://cochranelibrary.com
22. International Union Against Tuberculosis Committee on Prophylaxis. "Efficacy of various durations of isoniazid preventive therapy for tuberculosis: five years of follow-up in the IUAT trial. Bull WHO 1982;60(4):555-64.
23. Grzybowski S, Ashley MJ, Pinkus G.Chemoprophylaxis in inactive tuberculosis: long-term evaluation of a Canadian trial. Can Med Assoc Journal 1976;114:607-11.
24. Lecoeur HF, Truffot-Pernot C, Grosset JH. Experimental short-course preventive therapy of tuberculosis with rifampin and pyrazinamide. Am Rev Respir Dis1989;140:1189-93.
25. Dhillon J, Dickinson JM, Sole K, Mitchison DA. . Preventive chemotheraphy of tuberculosis in Cornell model mice with combinations of rifampin, isoniazid and pyrazinamide. Antimicrob Agents Chemother 1996;40(3):552-5.
26. Hong-KongChestservice/TuberculosisResearchCentre, Madras British. A double-blind placebo-controlled clinical trial of three antituberculosis chemoprophylaxis regimens in patients with silicosis in Hong Kong. Am Rev Respir Dis 1992;145:36-41.
27. Cowie RL. Short course chemoprophylaxis with rifampicin, isoniazid and pyrazinamide for tuberculosis evaluated in goldminers with chronic silicosis: a double-blind placebo controlled trial. Int J Tuberc Lung Dis 1996;77:239-43.
28. Center for disease control and prevention. Adverse event data revised American Thoracic Society/CDC. Recommendations against the use of rifampin and pyrazinamide for treatment of Latent Tuberculosis Infection - United States, 2003. Recommendations and reports. M.M.W.R 2003;52(RR-31):735.
29. International Union Against Tuberculosis and Lung Disease. Tuberculosis preventive therapy in HIV infected individuals; a joint statement of the IUATLD and GPA (WHO). Int J Tuberc Lung Dis 1994;75:96-8.
30. Center for disease control and prevention. Prevention and treatment of tuberculosis among patients infected with Human Immunodeficiency Virus: principles of therapy and revised recommendations. Recommendations and reports. M.M.W.R 1998;47(RR-20):1-51.
31. Pape JW, Jean SS, Jo JL, Hafner A, Jhonson WDJ. . Effect of isoniazid prophylaxis on incidence of active tuberculosis and progression of HIV infection. Lancet 1997;342:268-72.
32. Whalem CC, Johnson JL, Okwera A, Hom DL, Huebner R, Mugyenyi P, et al. . A trial of three regimens to prevent tuberculosis in Uganda adults infected with human immunodeficiency virus. N Engl J Med 1997;337(12):801-8.
33. Hawken MP, Meme HK, Elliott LC, Chakaya JM, Morris JS, Githui WA, et al. . Isoniazid preventive therapy for tuberculosis in HIV-1-infected adults: results of a ramdomized contolled trial. AIDS 1997;11:875-882.
34. Mwinga A, Hosp M, Godfrey-Faussett P, Quigley M, Mwaba P, Nyirenda O, et al. . Twice weekly tuberculosis preventive theraphy in HIV infection in Zambia. AIDS 1998;12:2447-57.
35. Gordin F, Chaisson RE, Matts JP, Miller C, Garcia ML, Hafner R, et al. . Rifampin and pyrazinamide vs isoniazid for prevention of tuberculosis in HIV-infected persons. JAMA 2000;283(11):1445-50.
36. Bucher HC, Griffith LE, Guyatt GH, Sudre P, Naef M, Sendi P, et al. . Isoniazid prophylaxis for tuberculosis in HIV infection: a meta-analysis of randomized controlled trials. AIDS 1999;13:501-7.
37. Graham NM, Galai N, Nelson KE, Astemborski J, Bonds M, Rizzo R, et al. . Effect of isoniazid chemoprophylaxis on HIV-related mycobacterial disease. Arch Intern Med 1996;156(8):889-94.
38. Halsey NA, Coberly JS, Desormeaux J, Losikoff P, Atkinson J, Moulton LH, et al. . Randomised trial of isoniazid versus rifampicin and pyrazinamide for prevention of tuberculosis in Hiv-1 infection. Lancet 1998;351:786-92.
39. Pinho A M F, Santoro-Lopes G, Harrison LH, Mauro Schechter. . Chemoprophylaxis for tuberculosis and survival of HIV-infected patients in Brazil. AIDS 2001;15:2129-35.
40. Alfaro E M, Cuadra F, Solera J, Maciá MA, Geijo P, Sánchez Martínez PA, et al. . Evaluación de dos pautas de quimioprofilaxis tuberculosa en pacientes infectados por el virus de la inmunodeficiencia humana. Med Clín Barcelona 2000;115:161-5.
41. Garibaldi RA, Drusin RE, Ferebee SH, Gregg MB. . Isoniazid-associated hepatitis. Am Rev Respir Dis 1972;106:357-65.
42. Kopanoff DE, Snider DEJ, Caras GJ. . Isoniazid related hepatitis. Am Rev Respir Dis 1979;117:991-1001.
43. Chuan-Fang J, Sable R. Isoniazid-induced acute hepatitis and acute pancreatitis ina patient during chemoprophylaxis. Clin Gastroenter 2002;35(1):100-1.
44. Salpenter SR, Sanders GD, Salpenter EE, Owens DK. Monitored isoniazid prophylaxis for low-risk tuberculin reactors older than 35 yr of age: a risk-benifit and cost-effectiveness analysis. Ann. Intern. Med. 1997;127:1051-61.

Abbreviations used in this paper:
95% CI - 95% confidence interval
AIDS - Acquired immunodeficiency syndrome
ATS - American Thoracic Society
BCG - Bacillus Calmette-Guérin
BTS - British Thoracic Society
EMB - Ethambutol
HIV - Human immunodeficiency virus
INH - Isoniazid
Mtb - Mycobacterium tuberculosis
PAS - Para-aminosalicylic acid
PPD - Purified protein derivative
PZA - Pyrazinamide
RFB - Rifabutin
RIF - Rifampin
RR - Relative risk
TB - Tuberculosis

Correspondence
Pedro Curi Hallal
Av. Duque de Caxias, 250
CEP: 96030-002 Fragata, Pelotas, RS, Brasil
Tel: 55 53 271 2442 Fax: 55 53 271 2645
E-mail: prchallal@terra.com.br