Factors associated with inadequate treatment in a group of patients with multidrug-resistant tuberculosis

Abstracts

BACKGROUND: Multidrug resistant tuberculosis is a threat to tuberculosis control worldwide. In many studies, it has been suggested that inadequate treatment is a risk factor for the development of multidrug resistent tuberculosis. OBJECTIVE: To identify the factors associated with inadequate treatment in a group of patients with multidrug resistent tuberculosis. METHOD: The authors identified all drug sensitivity test for tuberculosis performed at the Central Laboratory of the state of Ceará from 1990 to 1999. Retrospective and prospective analysis was performed, comparing the characteristics of adequately treated patients with those of inadequately treated patients. Multidrug resistance was defined as resistance to 2 drugs (rifampin and isoniazid) or more, using the Cox proportional hazards model. RESULTS: Of the 1,500 sensitivity test performed at the Central Laboratory of Public Health of the state of Ceará, 266 revealed multidrug-resistant strains. Of those 266 patients, the authors were able to identify only 153, of whom 19 were excluded. Thus, our sampling consisted of 134 patients. Univariate analysis demonstrated that the significant factors associated with inadequate treatment were: noncompliance with the treatment, extreme poverty, drug intolerance, mistakes in the management of the patients, lack of medication being supplied by the health service, two or more previous treatments, pronounced lung cavities and bilateral lesions on chest X-rays. They found a further association (p < 0.0010) between alcoholism or smoking and noncompliance with the treatment. In the multivariate analysis, the factors associated with an outcome of inadequate treatment were: two or more previous treatments (p < 0.0001, OR = 5.9; CI 95%: 2.5-13.7), pronounced cavitation (p < 0.0217, OR = 2.7; CI 95%: 1.2-6.1) and bilateral chest X-ray lesions (p < 0.0226, OR = 3.2; CI 95%: 1.4-7.4) CONCLUSION: In the study, the authors observed that factors related to inadequate treatment are wide-ranging. An attempt at better control of the disease is warranted, especially in those patients with pronounced cavitation or bilateral lesions in chest X-rays. And those with 2 or more previous treatments.

Tuberculosis; multidrug-resistant; Risk factors

INTRODUÇÃO: A tuberculose multirresistente é uma ameaça ao controle da tuberculose em todo o mundo. Tratamento inadequado é freqüentemente apontado como fator de risco para tuberculose multirresistente. OBJETIVO: Identificar fatores associados ao tratamento inadequado em portadores de tuberculose multirresistente. MÉTODO: Foram levantados e identificados os testes de sensibilidade para tuberculose, realizados no Laboratório Central do Estado do Ceará de 1990 a 1999. Foi realizado um estudo retrospectivo e prospectivo, em grupo de portadores de tuberculose multirresistente, comparando as características dos pacientes com tratamento adequado com as dos com tratamentos inadequados. Foi considerado multirresistente o bacilo resistente a pelo menos rifampicina + isoniazida e utilizado o método das proporções. RESULTADOS: Dos 1.500 testes de sensibilidade realizados, 266 eram multirresistentes. Destes, identificaram-se apenas 153 pacientes, dos quais 19 foram excluídos, restando, no estudo, 134 pacientes. Pela análise univariada os fatores associados significativamente ao tratamento inadequado foram: não-adesão ao tratamento, pobreza extrema, intolerância medicamentosa, falha no atendimento, falta de medicação, dois ou mais tratamentos anteriores, lesões radiológicas bilaterais e grandes cavidades pulmonares. Foi encontrada ainda associação (p < 0,001) de alcoolismo e/ou tabagismo com a não-adesão. Na análise multivariada, permaneceram associados ao tratamento inadequado: dois ou mais tratamentos anteriores (p < 0,0001, OR = 5,9; IC 95%: 2,5-13,7), grandes cavidades pulmonares (p < 0,0217, OR = 2,7; IC 95%: 1,2-6,1) e lesões radiológicas bilaterais (p < 0,0226, OR = 3,2; IC 95%: 1,4-7,4). CONCLUSÃO: Neste estudo, observou-se que os fatores associados aos tratamentos inadequados são de grande abrangência. Deve existir uma tentativa para controlar melhor a doença, principalmente nos pacientes em retratamento e nos portadores de grandes cavidades e/ou lesões bilaterais na radiografia torácica.

Tuberculose resistente a múltiplas drogas; Fatores de risco

ORIGINAL ARTICLE

Factors associated with inadequate treatment in a group of patients with multidrug-resistant tuberculosis

Elizabeth Clara Barroso^I; Rosa Mª Salani Mota^II; Mª Filomena Martiniano Morais^III; Creusa Lima Campelo^IV; Joana Brasileiro Barroso^V; Jorge Luis Nobre Rodrigues^VI

^IPulmonologist, with a Masters degree in Clinical Medicine

^IIProfessor of statistics, with a Masters in statistics

^IIISocial Services Agent with the Health Ministry

^IVBiochemical Pharmacologist

^VGraduate student in Psychology

^VIInfectologist, Parasitologist and Head of Research

^{Correspondence}

ABSTRACT

BACKGROUND: Multidrug-resistant tuberculosis is a threat to tuberculosis control worldwide. In many studies, it has been suggested that inadequate treatment is a risk factor for the development of multidrug-resistant tuberculosis.

OBJECTIVE: To identify the factors related to inadequate treatment in a group of patients with multidrug-resistant tuberculosis.

METHOD: The authors identified all drug sensitivity tests for tuberculosis performed at the Central Laboratory of the state of Ceará from 1990 to 1999. Retrospective and prospective analysis was performed, comparing the characteristics of adequately treated patients with those of inadequately treated patients. Multidrug resistance was defined as resistance to 2 drugs (rifampin and isoniazid) or more, using the Cox proportional hazards model.

RESULTS: Of the 1500 sensitivity tests performed at the Central Laboratory of Public Health of the state of Ceará, 266 revealed multidrug-resistant strains. Of those 266 patients, the authors were able to identify only 153, of whom 19 were excluded. Thus, our sampling consisted of 134 patients. Univariate analysis demonstrated that the significant factors related to inadequate treatment were: noncompliance with the treatment, extreme poverty, drug intolerance, errors in patient management, a shortage of medication supplied by the health service, 2 or more previous treatments, pronounced cavitation and bilateral lesions on chest X-rays. A further correlation (p < 0.0010) was found between alcoholism or smoking and noncompliance with the treatment. In the multivariate analysis, the factors related to negative treatment outcomes were: 2 or more previous treatments (p < 0.0001, OR = 5.9; CI 95%: 2.5-13.7), pronounced cavitation (p < 0.0217, OR = 2.7; CI 95%: 1.2-6.1) and bilateral chest X-ray lesions (p < 0.0226, OR = 3.2; CI 95%: 1.4-7.4)

CONCLUSION: In this study, the authors observed that factors related to inadequate treatment are wide-ranging. An attempt at better control of the disease is warranted, especially in those patients with pronounced cavitation or bilateral lesions in chest X-rays, as well as in those who have had 2 or more previous treatments.

Key words: Tuberculosis, multidrug-resistant. Risk factors.

Abbreviations used in this paper:

TB – Tuberculosis

MDR-TB – Multidrug-resistant Tuberculosis

INH – Isoniazid

RIF – Rifampin

HIV – Human Immunodeficiency Virus

CENLA – Ceará State Central Laboratory of Public Health

OR – Odds ratio

CDC – Centers for Disease Control

INTRODUCTION

The resistance of Koch’s bacillus to drugs is as old as antituberculosis treatments (approximately 56 years).^(1,2) In 1947, Pyle published the first study about bacilli resistant to streptomycin.⁽¹⁾ His study was followed by others by several different authors. In 1948, Crofton and Mitchenson⁽³⁾ showed that 12 of 13 serious bilateral tuberculosis (TB) cases developed resistance to streptomycin with a dosage of 2g/day over 4 or more months. In Brazil, the first study was published by Rosemberg et al. in 1953.⁽⁴⁾ He found primary resistance in 1 out of 8 (12.5%) of the bacilli isolated from patients without any previous treatment, and in 2 out of 10 (20%) of the bacilli isolated from patients during treatment with 200 mg/day of isoniazid (INH), after 42 days of treatment in 1 patient and 72 days in another.

The international definition for multidrug-resistant tuberculosis (MDR-TB) is applied to cases with bacillus resistant to at least the combination of rifampin (RIF) and isoniazid (INH).⁽⁵⁾

Resistance is a phenomenon fundamentally connected to great numbers of bacteria found together. In human TB, the largest colonies of bacteria are found in areas of cavitation. It has been known for more than half a century that resistance is more common during treatment when there is such cavitation. The greater the number of bacteria found together, the greater the chance of finding resistant bacilli before the onset of treatment. The ratio of bacteria in areas of cavitation is approximately 10⁷ to 10⁸ bacilli, whereas in hardened, caseous lesions, that ratio is no more than 10² to 10⁴.⁽¹⁾ Therefore, in most nonpulmonary TB cases, the initial number of bacteria is much smaller than in cases presenting cavitation and the chance of pre-existing resistant mutant strains is therefore quite small. The number of bacilli is even smaller in the case of lingering TB infections, with only a slight chance of secondary resistance to INH.⁽¹⁾ This has been proven in studies of both HIV-positive and HIV-negative patients.^(7,8)

Some authors have reported transitional or false resistance that may occur when samples are taken during antituberculosis treatment. This is due to the fact that more time is required to eliminate the naturally resistant mutant strains than the rest of the M. Tuberculosis Bacilli. False resistance may also occur due to laboratory contamination. In the microepidemics of MDR-TB that occurred in the United States during the early 1990s, 221 patients were involved. The Center for Disease Control (CDC) of Atlanta, Georgia, found out that 57 (26%) of those patients did not suffer from MDR-TB, 55 did not present any TB symptoms and 2 patients were suffering from drug-sensitive TB. A review of the laboratory proceedings suggested that contamination occurred during the isolation of the bacillus as well as during the sensitivity test. Of the 57 contaminated specimens, 29 were evaluated using the BACTEC 460^® system after testing strongly positive for multidrug resistance, and 28 were cultured on the same day as strongly positive multidrug-resistant specimens. Thanks to the DNA pattern in the 37 cases available, the source of contamination was identified ⁽⁹⁾.

Resistance occurs due to the appearance of pre-existing, resistant mutant strains caused by antimicrobial pressure in the original group of bacilli. For example: if INH alone is prescribed (or if it is the only drug tolerated by a patient in a multi-drug regimen) against TB involving cavitation, organisms sensitive to INH will be eliminated, along with chance mutant strains resistant to other drugs, but those resistant to INH will survive. These will continue multiplying and will eventually become dominant. The probability that this will happen is influenced by the duration of the monotherapy: 25% among people receiving INH exclusively for 2 weeks, 60% among those receiving it for 6 months and 80% among those receiving it for 2 years.⁽¹⁰⁾ If RIF and INH are used in combination (RIF+INH), then, by the same mechanism, a cepaceous, multidrug-resistant bacillus (resistant to RIF+INH) will emerge.⁽¹⁰⁾

This classic theory of drug resistance in TB treatment states that a sequence of events cause TB to become drug-resistant in monotherapy patients. It does not explain how resistance arises due to irregularity in drug regimens other than monotherapy. Other mechanisms have been suggested to explain resistance in these circumstances. These mechanisms depend on various cycles of death (when drugs are taken) and bacteria growth (when they are suspended). In each cycle, selection occurs, favoring the resistant mutant strains to the detriment of the sensitive ones.⁽¹¹⁾ Resistance to one of the drugs used occurs first, followed by resistance to another, resulting in MDR-TB. This condition reflects the accumulation of the individual mutations of several independent genes and not the acquisition en masse of multidrug resistance.⁽¹²⁾

Inadequate treatment may be defined as direct or indirect monotherapy.^(13-15) This may be related to the health professional, the drug or the patient. Many researchers have reported that irregular treatment is a risk factor for the development of MDR-TB.^(14-18) The objective of the present study was to identify factors related to inadequate treatment of MDR-TB patients.

METHODS

A retrospective and prospective study was made with the aid of a standard questionnaire. The international definition of MDR-TB was adopted, according to which the bacillus studied must be resistant at least to RIF+INH.⁽⁵⁾

Cases were selected from a list of 1500 sensitivity tests performed between 1990 and 1999 at the Ceará State Public Health Central Laboratory. The Central Laboratory is the only laboratory conducting sensitivity tests in the state of Ceará and operates under the auspices of the Professor Helio Fraga Reference Center (national reference for the performance of sensitivity tests). Among the cases selected for study, only 25% were studied retrospectively. The remaining cases were evaluated at the time and the questionnaire filled out in the presence of the patient. In cases of death or loss of contact, some questionnaires were completed without the patient being present. All patients gave written informed consent and the Ethics Committee on Research of the Federal University of Ceará approved the study.

Patients were categorized and subcategorized by age (10 to 30; 31 to 50, and over 50), TB status (cured or not), treatment status (in or out), compliance or non-compliance with treatment regimens and intolerance to medication as defined by the Brazilian Guidelines on Tuberculosis, I.⁽¹⁹⁾ Level of education was also taken into account. Patients were considered "uneducated" if they were illiterate (unable to read and write, even if able to sign their names), and those who, though able to read and write, had completed less than half of elementary school. Those who had completed at least half of elementary school, junior high school, high school or college were considered "educated". The presence of running water and sanitation facilities in the home was also taken into consideration. In addition, patients were classified as either having or not having TB in the home and, in cases where there was TB in the home, classified by whether the TB was adequately treated. Adequate treatment of TB was defined as compliance with the treatment regimen, noncompliance for periods of less than 5 consecutive days, or noncompliance for less than 10 nonconsecutive days within the same month. Inadequate treatment of TB was defined as noncompliance for more than 5 consecutive days, or noncompliance for more than 10 nonconsecutive days within the same month. Cavitation was classified as serious if measuring more than 4 cm in a simple radiological examination, based on radiological examinations of other patients previously diagnosed with MDR-TB. The presence of diabetes was determined from the patient information or from the presence of blood sugar>127mg/dl. Drug addiction was noted when revealed by the patient, the family or other reliable source. Psychiatric problems were taken into consideration based on information provided by the patient or the family, or on information regarding psychiatric treatment or usage of controlled drugs. Inadequate treatment was defined as abandoning treatment or as irregularities in at least 1 of the first 3 treatments; noncompliance with the treatment regime, a shortage of proper medication and flaws in attendance, as well as intolerance to medication, all qualified as such irregularities. Another defining factor was extreme poverty, which was defined as being homeless, living alone without any means of subsistence or living on a low income with dependents. A shortage of medication was defined as no medication available at the treatment center. Inadequately treated patients were further defined as those who reported that their complaints were ignored, those who suffered due to errors made in the prescription of medications and those who did not comply with the treatment regimen. To analyze treatment quality, we studied the first 3 treatments given (some patients received as many as 12 treatments), because we concluded that these 3 would suffice to give us a good idea concerning the quality of the treatments. In fact, by the third treatment, treatment abandonment and noncompliance became evident. Serious (or at risk) alcoholism, as well as tobacco use, were evaluated using the criteria established by O’Connor⁽²⁰⁾ and Schottenfeldand Fahn et al.,⁽²¹⁾ respectively.

All Koch’s bacillus cultures were performed following the same protocol, according to the same biochemistry, throughout the same study period, at the Central Laboratory. The Lowenstein-Jensen culture media was used and the differentiation between M. Tuberculosis and other microorganisms was made through biochemical testing. For the sensitivity test, the proportion method on the Lowenstein-Jensen media was employed. Resistance was defined as growth of at least 1% in colonies in the critical concentrations of 0.2 µg/mL of INH, 2 µg/mL of ethambutol or 40 µg/mL of RIF. Strains were also considered resistant if there was growth of at least 10% in colonies in critical concentrations of 20 µg/mL of ethionamide, 100 µg/mL of pyrazinamide or 4 µg/mL of streptomycin.

In the univariate analysis, Pearson’s chi-square test and Fisher’s exact test were used. For the joint analysis of risk factors, multiple linear regression was employed according to the logistic regression model. Values of p < 0.05 were considered significant. Statistical analyses were elaborated with the computer programs SPSS, Microsoft Excel and Microsoft Word.

RESULTS

Of the 1500 sensitivity tests (all originating from patients within the state of Ceará) analyzed at the Central Laboratory during the 1990s, 266 strains were resistant to at least RIF+INH. After the duplicate tests were eliminated from the analysis, we were able to identify 153 of these patients and the standard questionnaire was filled out. Of these 153 patients, 5 were excluded because they carried an atypical microbe and 2 because they failed to meet the criteria for the MDR-TB definition. Of the 146 remaining cases, 12 had not received any previous treatment (primary MDR-TB cases) and were therefore also excluded. The final group consisted of 134 patients. Of these 134, 96 (71.6%) showed indications of irregular treatment in at least one of the 3 treatments analyzed.

The results of the univariate analysis of the characteristics relating to inadequate treatment can be found in Table 1. The factors that presented a correlation (p < 0.05) with inadequate treatment were: noncompliance with treatment, extreme poverty, intolerance to medication, faulty care, shortage of medication, a history of 2 or more previous treatments, bilateral radiological lesions and significant cavitation. Of the 134 patients with MDR-TB, 91 (67.9%) were tested for HIV, but none tested positive and this data is therefore not included in Table 1.

The variables of age, gender, schooling, unsupervised compliance with TB treatment, plumbing and sewage facilities, severe alcoholism, smoking, diabetes, psychiatric disorders, drug abuse and HIV status were not related to inadequate treatment. Since we consider alcoholism and smoking of paramount importance, we researched these variables in connection with noncompliance to treatment and a statistically significant direct and indirect correlations were found (p < 0.0010) (Table 2).

In the multivariate analysis (Table 3), factors related to inadequate treatment up to a significance level of 25 percent were considered for the logistic regression model. The first 5 significant factors reported above were removed due to their unilateral relationships with inadequate treatment (no case with one of those characteristics was found in the group adequately treated) (Table 1). Presence of those variables would bring serious inconsistency to the model, preventing any conclusions, and that is why they were removed, therefore allowing better evaluation of the other variables. So, the variable "2 or more previous treatments" remained as the only one related to inadequate treatment (p < 0.0001), with an odds ratio (OR) of 5.9 and a 95% confidence interval (CI) of 2.5-13.7. However, due to its direct implication in irregular treatment, that factor was also excluded from the model. It should be emphasized that, in this case, the relationship was not unilateral. Of the 134 patients, 101 were submitted to 2 or more treatments and, of those, 19 (18.81%) followed regular treatment. Therefore, the associated variables, when incorporated into the model individually, were pronounced cavitation (p < 0.0217; OR = 2.7; 95% CI = 1.2-6.1) and bilateral radiological lesions (p < 0.0226; OR = 3.2; 95% CI = 1.4-7.4). These variables were included separately due to a high correlation between them, as proven by Fisher’s exact test (p < 0.0001).

DISCUSSION

Before commenting on the results, we should mention several limitations of the study. The first is that, of 259 MDR-TB patients listed in the records of the Central Laboratory, we were able to characterize the medical histories of only 145 (56%). We should also consider limitations inherent to retrospective studies and the biases that may result either from reviews of a patient medical histories, reliability of data recorded therein, or even the completion of the questionnaires – conducted in the presence of the patients – caused by lapses of memory. Nevertheless, only 25% of the patient data was generated from medical records. However, since these were chronic cases and the patients had been admitted numerous times, there was a great deal of information in their medical records. To prevent biases due to either lack of data or lack of credibility, we also looked for nurses and social workers, who, in addition to having been responsible for much of the information in the medical records, knew much about the patients through consultations and admissions. The other 75% had been monitored over the last decade and were present during the completion of the questionnaire. Another limitation of the present study is that our medical services do not perform sensitivity tests for antituberculosis drugs as a matter of course when TB is diagnosed for the first time. This fact may have minimized the percentage of primary MDR-TB found (only 8% against 92% of MDR-TB acquired). This is evidenced by the fact that 38 (28.36%) of the 134 MDR-TB patients did not present any signs of irregular treatment. It is probable that some of those patients were suffering from primary MDR-TB. Finally we should remember that 38% of the patients were not tested for HIV, since that test is not routinely performed, especially for outpatients (nearly all admitted patients were tested). It is possible that a few HIV-positive patients were not identified as such. However, it is unlikely that this had an influence on the results, since all were negative. Among those who refused to be tested, we did not find any evidence of high-HIV-risk lifestyles.

We declined to consider race as a factor in this study because, in the state of Ceará, miscegenation is quite significant, thus rendering impossible a satisfactory racial classification.

In an extensive review published in 1997, Natal⁽²²⁾ described the difficulty of identifying risk factors for noncompliance. In the several studies reviewed, employing the most diverse methods (longitudinal, prospective or retrospective cohort, transversal and case-study) the associated factors were: administrative and organizational, lack of schooling, positive bacilloscopy, hospitalization, alcoholism, psychiatric disorders, income lower than the minimum wage, inadequate care, living far from medical facilities, male gender, unmarried, living alone, living in a rural area, poorly-constructed and unsanitary housing, unemployment, age between 15 and 49, delay in seeking medical attention, multiple symptoms, lack of information, low social and financial status, intervals between consultations greater than 4 weeks, duration of treatment longer than 9 months, lack of nurse visits, sporadic employment, re-initiation of treatment, previous abandonment of treatment regime and a shortage of specific medication.

In 1989, publishing results in 1999, Natal⁽²³⁾ conducted a study case control nested within a TB treatment cohort, with the objective of determining risk factors involved when patients at the Raphael de Paula Souza hospital in Rio de Janeiro abandoned outpatient treatment. The risk factors detected in the multivariate analysis were: negative bacilloscopy at time of diagnosis, sporadic employment, daily consumption of alcoholic beverages, patient-reported lack of improvement during treatment and patient refusal of health services.

In 1997, Kritski et al.⁽²⁴⁾ conducted a retrospective case control study within the Phthisiology Department of the Federal University of Rio School of Medicine. The authors attempted to verify the occurrence of abandonment of antituberculosis treatment in a general hospital that was a reference for AIDS but had no structured program of TB control. They discovered that giving out return cards to patients, maintaining good doctor-patient relationships and taking patient blood pressure function as protection against patient abandonment of treatment.

From 1995 to 1997, another retrospective case control study was carried out, this time at the University of Sao Paulo.⁽²⁵⁾ The authors had the objective of identifying predictors of treatment abandonment by TB patients in Brazil. They found the following variables to be related to abandonment of treatment: male gender, smoking, alcoholism, drug abuse, presence of risk factors for HIV and previous hospitalization. As part of this study, the authors conducted a comparative study with a group of volunteers who participated in educational meetings. Although these patients exhibited characteristics similar to the group of studied patients as a whole, there was a lower incidence of treatment abandonment (p < 0.05). It is important to mention a fundamental difference between the group of volunteers as compared to the total number of patients studied. The fact that they volunteered to participate in educational meetings indicates that they were naturally more cooperative and inclined to follow the treatment as prescribed. This does not diminish the importance of these truly excellent educational meetings, which should be utilized more frequently in an effort to reach those patients who do not comply with their treatment regimes – since they are the ones who are the most likely to abandon treatment.

Noncompliance with treatment, much referred to, related to the use of medication in general and not only TB treatments and was confirmed in our study as a factor related to inadequate treatment.

Extreme poverty was a variable we created but which has been used in other studies, as has low income, inferior social and financial status, poorly-constructed and unsanitary housing, living alone, sporadic employment, unemployment, etc. These circumstances have been implicated in abandonment of TB treatment^(22,23), a finding confirmed in our study (Table 1).

Intolerance to medication is a common finding in the medical histories of MDR-TB patients. However, there have been no studies implicating drug intolerance in treatment abandonment, although adverse drug side effects have been shown to be risk factors in the development of MDR-TB. Handling intolerance to medication calls for a level of training and experience not always found among those who make up the medical service staff. Although treating TB is considered elementary in providing health care, as with any disease, there are always more complex cases. There is a tendency to consider TB easy to treat because less-highly-trained professionals can execute the ambulatory care protocols. Therefore, there is the possibility that, aside from the fact that, in some treatment centers, patient complaints will be ignored, there may be no alternative treatment center to which the case can be referred. Therefore, inadequate or inappropriate treatment may result and the patient will eventually abandon treatment and develop MDR-TB. In the present study, the correlation between intolerance to medication and inadequate treatment was quite significant (p < 0.0001) (Table 1).

Shortage of medication is a frequent problem in providing healthcare services. One tries to resolve this problem by referring patients to a different healthcare center or by asking patients to return at a later date. However, this inconvenience may cause patients to fall out of compliance with the treatment regimen or to abandon treatment altogether. A 1995 study by Lambregts-van Weezenbeek demonstrated the importance of maintaining a sufficient supply of medication in preventing MDR-TB.⁽¹³⁾ In 1993, Natal published a study pointing to medication shortage as a cause of noncompliance with treatment.⁽²⁷⁾ In the present study, a shortage of medication was also found to be a risk factor for inadequate treatment (Table 1).

It is even more difficult to deal with the subject of inadequate care, although this problem and its association with MDR-TB have been studied.^(15,26,28) In the national literature, variables designated as flawed attendance and no improvement (based on patient complaints) have been related to abandonment of treatment.^(23,24) In the present study, we found a correlation between flawed attendance and inadequate treatment (p < 0.0020).

Our research confirms that the previous treatment factor is related to abandonment of TB treatment, as previously shown.^(25,27)

From 1953 to the present, studies conducted in Brazil have shown that there is a relationship between cavitation and the development of drug-resistant strains of Mycobacterium tuberculosis. In 2000, Dalcolmo et al. found a statistically significant correlation between radiologically proven TB severity and risk of abandonment. Another national study, published in the journal Chest in l997, also showed such a correlation. The findings of these authors are confirmed in the present study.

The correlation between smoking or alcoholism and inadequate treatment is dependent upon the definitions of alcoholism and nicotine addiction employed and is therefore difficult to demonstrate. There have been few publications confirming this correlation.⁽²⁵⁾ We were intrigued by the fact that we did not find a correlation between inadequate treatment and smoking or alcoholism. Nevertheless, since we consider these variables very important in the present study, we studied their correlation with treatment noncompliance and found a significant correlation (p < 0.0010), and we must consider the possibility that this is related to inadequate treatment.

We conclude that the factors associated with inadequate treatment are quite complex and demand a complex approach to TB and MDR-TB control, especially in patients who are resubmitted to treatment and in those who present significant cavitation or bilateral lesions in chest radiographs. Several measures could be taken to minimize the impact of these factors. First, a strategy of directly controlled, supervised treatment could be adopted. In addition, social and financial support for patients could be improved. Furthermore, treatment for alcoholism and nicotine addiction could become an additional objective. Moreover, closer interaction could be fostered between patients and the medication distribution services.

ACKNOWLEDGMENTS

The authors hereby render thanks to Prof. Jose Rosemberg, to Prof. Marta Maria das Chagas Medeiros, Ph.D., and to Prof. Maria Grasiela Teixeira, Ph.D. for their guidance in the research, as well as in other aspects of the study. We would also like to thank the librarians Leonilha Maria Brasileiro and Norma de Carvalho Linhares for their invaluable help in compiling our bibliography.

REFERENCES

Submitted: 01/04/2003. Accepted, after revision: 13/08/2003.

* Research undertaken as part of a Masters degree program under the guidance of Prof. Dr. Jorge Luis Nobre Rodrigues. Study conducted at the Hospital de Maracanaú, the Ministério da Saúde (Health Ministry), the Hospital de Messejana and the Secretaria da Saúde (Health Department) of the state of Ceará.

• 1. Canetti G. Present aspects of bacterial resistance in tuberculosis. Am Rev Respir Dis 1965;92:687-703.
• 2. Horsburgh CR Jr. The global problem of multidrug-resistant tuberculosis: the genie is out of the bottle. JAMA 2000;283:2575-6.
• 3. Crofton J, Mitchinson DA. Streptomycin resistance in pulmonary tuberculosis. Br Med J 1948;2:1009-15.
• 4. Rosemberg J, Brandi R, Carrijo LN. Resistência do bacilo de Koch à hidrazida do ácido isonicotínico. Rev Bras Tuberc 1953;21:63-6.
• 5. Pablos-Mendez A, Raviglione MC, Laszlo A, Binkin N, Rieder HL, Blustreo F, et al. Global surveillance for antituberculosis-drug resistance, 1994-1997. World Health Organization International Union Against Tuberculosis and Lung Disease Working Group on Anti-tuberculosis Drug Resistance Surveillance. N Engl J Med;1998;338:1641-9.
• 6. Ben-Dov I, Mason GR. Drug-resistant tuberculosis in Southern California Hospital: trends from 1969 to 1984. Am Rev Respir Dis 1987;135: 1307-10.
• 7. Peña JM, Ortega A. Estudio transversal multihospitalario de tuberculosis y resistencias en Madrid (octubre de 1993-abril de 1994). Med Clin (Barc) 1996;106:1-6.
• 8. Garcia Vazquez E, Esteban J, de Gorgolas, Fernández Guerrero ML. Infection by resistant Mycobacterium tuberculosis in a hospital population. A longitudinal study of incidental cases at the Fundation Jimenez Diaz. Rev Clin Esp 1999;199:564-8.
• 9. Valway S, Doolly S, Ikeda R, Jereb J, Kent J, Onorato I. False positive diagnosis of multidrug resistant tuberculosis due to laboratory contamination. Tuberc Lung Dis 1994;75:142.
• 10. Long R. Drug-resistant tuberculosis. CMAJ 2000;163:425-8.
• 11. Mitchinson DA. How drug resistance emerges as a result of poor compliance during short course chemotherapy for tuberculosis. Int J Tuberc Lung Dis 1998;2:10-5.
• 12. Riska PF, Jacobs WR, Alland D. Molecular determinants of drug resistance in tuberculosis. Int J Tuberc Lung Dis 2000;4:4-10.
• 13. Lambregts-van Weezembeek CSB, Veen J. Control of drug-resistant tuberculosis. Tuber Lung Dis 1995;76:455-9.
• 14. González Saldanha N, Neme Diaz GA, Macias M, Palau M, Levy A, de la Hoz R. Resistencia a los fármacos antituberculosos en América Latina. Rev Enfermedades Infecc Ped 1999;11:256-9.
• 15. Rao SN, Mookerjee AL, Obasanjo OO, Chaisson RE. Errors in the treatment of tuberculosis in Baltimore. Chest 2000;117:734-7.
• 16. Nunn P, Felten M. Surveillance of resistance to antituberculosis drugs in developing countries. Tuber Lung Dis 1994;75:163-7.
• 17. Bradford WZ, Martin JN, Reingold AL, Schecter GF, Hopewell PC, Small PM. The changing epidemiology of acquired drug-resistant tuberculosis in San Francisco, USA. Lancet 1996;348:928-31.
• 18. Barroso EC, Mota RMS, Santos RO, Souza ALO, Barroso JB, Rodrigues JLN. Fatores de risco para tuberculose multirresistente. J Pneumol 2003;29:89-97.
• 19. Coordenação Nacional de Pneumologia Sanitária. I Consenso Brasileiro de Tuberculose-1997. J Pneumol 1997;23:294-319.
• 20. O'Connor PG, Schottenfeld RS. Patients with alcohol problems. N Engl J Med 1998;338:592-600.
• 21. Fahn HJ, Wang LS, Kao SH, Chang SC, Huang MH, Wei YH. Smoking-associated mitochondrial DNA mutations and lipid peroxidation in human lung tissues. Am J Respir Cell Mol Biol 1998;19:901-9.
• 22. Natal S. Tratamento da tuberculose: causas da não-aderência. Bol Pneumol Sanit 1997;5:50-70.
• 23. Natal S, Valente J, Gerhardt G, Penna ML. Modelo de predição para o abandono do tratamento da tuberculose pulmonar. Bol Pneumol Sanit 1999;7:65-78.
• 24. Kritski AL, Gomes de Salles CL, Ferreira D, Conde MB, Nuner ZB, Sa L, et al. Taxa de abandono do tratamento anti-Tb, no Rio de Janeiro. Pulmão RJ 2002;11:9-15.
• 25. Ribeiro SA, Amado VM, Camelier AA, Fernandes MMA, Schenkman S. Estudo caso-controle de indicadores de abandono em doentes com tuberculose. J Pneumol 2000;26:291-6.
• 26. Barroso EC. Fatores de risco para tuberculose multirresistente [dissertação]. Fortaleza: Faculdade de Medicina, Universidade Federal do Ceará, 2001; 107p.
• 27. Natal S. Abandono do tratamento da tuberculose: aspectos do tratamento mal conduzido [dissertação]. Rio de Janeiro: Universidade do Estado do Rio de Janeiro, Instituto de Medicina Social, 1993; 140p.
• 28. Mahmoudi AMD, Michael D, Iseman MD. Pitfalls in the care of patients with tuberculosis: common errors and their association with the acquisition of drug resistance. JAMA 1993;270:65-8.
• 29. Rosemberg J, Waisbich E, Passos Filho MCR. Resistência às drogas standard em 1.759 casos de tuberculose pulmonar, bacilíferos, submetidos à quimioterapia por tempos diversos. Rev Serv Nac Tuberc 1967; 11:449-66.
• 30. Seiscento M, Melo FAF, Ide Neto J, Noronha AML, Afiune JB, Inomat T, et al. Tuberculose multirresistente (TBMR): aspectos clínico-laboratoriais, epidemiológicos e terapêuticos. J Pneumol 1997;23:237-44.
• 31. Dalcolmo MP, Fortes A, Melo FF, Motta R, Ide Netto J, Cardoso N, et al. Estudo de efetividade de esquemas alternativos para tratamento da tuberculose multirresistente no Brasil. J Pneumol 1999;25:70-7.
• 32. Dalcolmo MMP. Regime de curta duração, intermitente e parcialmente supervisionado, como estratégia de redução do abandono no tratamento da tuberculose no Brasil. Bol Pneumol Sanit 2000;8:60-1.
• 33. Kritski AL, Rodrigues de Jesus LS, Andrade MK, Werneck-Barroso E, Vieira MAMS, Haffner A, et al. Retreatment tuberculosis cases. Factors associated with drug resistance and adverse outcomes. Chest 1997; 111:1162-7.

Publication Dates

• Publication in this collection
  12 May 2004
• Date of issue
  Dec 2003

History

• Received
  01 Apr 2003
• Accepted
  13 Aug 2003