Cost-effectiveness of rifampin for 4 months and isoniazid for 6 months in the treatment of tuberculosis infection

OBJECTIVES

To assess the cost-effectiveness ratio of rifampin for 4 months and isoniazid for 6 months in contacts with latent tuberculosis infection.

METHODS

The cost was the sum of the cost of treatment with isoniazid for 6 months or with rifampin for 4 months of all contacts plus the cost of treatment of cases of tuberculosis not avoided. The effectiveness was the number of cases of tuberculosis avoided with isoniazid for 6 months or with rifampin for 4 months. When the cost with one schedule was found to be cheaper than the other and a greater number of tuberculosis cases were avoided, this schedule was considered dominant. The efficacy adopted was 90% for rifampin for 4 months and 69% for isoniazid for 6 months. A sensitivity analysis was made for efficacies of rifampin for 4 months of 80%, 69%, 60% and 50%.

RESULTS

Of the 1002 patients studied, 863 were treated with isoniazid for 6 months and 139 with rifampin for 4 months The cost-effectiveness ratio with isoniazid for 6 month was € 19759.48/avoided case of tuberculosis and € 8736.86/avoided case of tuberculosis with rifampin for 4 months. Rifampin for 4 months was dominant. In the sensitivity analysis, rifampin for 4 months was dominant for efficacies from 60%.

CONCLUSIONS

Rifampin for 4 months was more cost-effective than isoniazid for 6 months.

Is isoniazid for 6 months more cost-effective than isoniazid for 9 months?

SETTING

The optimal treatment for latent tuberculosis infection consists of isoniazid (H, INH) for 9-12 months. Although INH for 6 months (6H) is more cost-effective than the 12-month regimen, the cost-effectiveness of the 6H regimen and that of INH for 9 months (9H) have not been compared.

OBJECTIVE

To compare the cost-effectiveness of treatment with 6H and 9H.

METHODS

Cost-effectiveness was evaluated using the ratio of the cost of preventing one tuberculosis case using 6H vs. 9H. The cost was estimated as the product of the number of patients to be treated to prevent one case using 6H or 9H × the cost of 6H or 9H.

RESULTS

A total of 1039 patients were studied. The number of patients that needed to be treated to prevent one case was 33 (95%CI 21-83) using 6H and 26 (95%CI 18-50) using 9H. The cost of 6H and 9H was respectively €444.34 and €578.26, and the cost ratio of preventing one case with 6H/9H was 0.98 (95%CI 0.6-1.5).

CONCLUSIONS

The cost-effectiveness of treatment with 6H and 9H is similar.

[Excess mortality due to tuberculosis and factors associated to death in and annual cohort of patients diagnosed of tuberculosis]

OBJECTIVE

To calculate excess mortality in an annual cohort of tuberculosis patients and study the factors associated with death.

MATERIAL AND METHOD

Cases of tuberculosis reported in Catalonia (May 1996-April 1997). Patients were classified as completed treatment/cured (compliant), non-compliant, failures, transfers out and deaths. Excess mortality was defined as the ratio actual deaths/expected deaths (according to general mortality figures for Catalonia, May 1996-April 1997). Factors associated with death were determined by a comparative study of variables (demographic, substance abuse, comorbidity, tuberculosis-related disease) in deaths after diagnosis and survivors. Time from diagnosis to death was recorded.

RESULTS

Patients included: 2,085. Patients classified as: completed treatment/cured (compliant), 1,406 (67.43 %); noncompliant, 165 (7, 91%); failures, 5 (0.24%); transfers out, 25 (1.21%); deaths, 133 (6.38%), 28 of which occurred before diagnosis and 105 after diagnosis. Insufficient data in medical record for classification, 351 (16.83%) patients. Excess mortality: 5.98 (95% CI: 4.96-7.0). Factors associated with death: treatment with non-standardized guidelines, 46%; OR: 10.3 (6.2-17.4); HIV infection, 40%; OR: 13.0 (6.6-25.8); age greater than 64 years, 40%; OR: 14.6 (3.0-69.8); alcoholism, 25%; OR: 2.0 (1.1-3.6); neoplasm, 16%; OR: 3.9 (1.8-8.6; renal failure, 8%; OR: 10.1 (3.1-32.3). The shortest time from diagnosis to death was in patients with only one risk factor, except for HIV infection, where the time passed was the longest observed.

CONCLUSIONS

We found substantial excess mortality in tuberculosis patients. Death was associated with the efficacy of treatment, HIV coinfection, advanced age, alcoholism and the coexistence of neoplasms or renal failure.

Ozone-induced foliar injury in saplings of Psidium guajava 'Paluma' in São Paulo, Brazil

Psidium guajava 'Paluma' was evaluated under field conditions as a tropical bioindicator species of ozone (O(3)). Three exposures of 90 days each were performed (June 21, 2004-March 19, 2005). In each one of them, saplings of 'Paluma' (n=30) were exposed to ambient air at a site in São Paulo (Brazil) with high ozone concentrations, and in a greenhouse with charcoal-filtered air. Ozone-like visible foliar injuries were observed during the winter, spring and summer exposures, when AOT40 reached 6166ppbh, 3504ppbh and 4828ppbh, respectively. No injuries were observed in the plants kept under filtered air. The injuries consisted in red stippling on adaxial leaf surfaces. They did not cover the veins and appeared first in older leaves, becoming more intense as the exposure period increased. Injury incidence, severity, and the cumulative exposure threshold at injury onset varied among the exposure periods. AOT40 explained partly the incidence, severity and leaf injury index LII (r(2)=0.52, 0.39, 0.38, respectively, p=0.002). The results confirm the potential of the species as an O(3)-sensitive bioindicator.

Characteristics of tuberculosis patients with positive sputum smear in Catalonia, Spain

BACKGROUND

Patients with positive sputum smears are those with the capacity to spread infection. The objective of this study was to describe the incidence of tuberculosis in Catalonia (an autonomous community in the northeast of Spain which includes Barcelona) and to determine risk factors associated to patients with positive sputum smear test.

METHODS

New cases of tuberculosis detected by active surveillance between May 1996 and April 1997 were studied. The study was analysed as a coincident cases and controls study. The rate of incidence was calculated per 100,000 persons-year. The association of the dependent variable--case of tuberculosis with positive sputum smear--with the remainder of independent variables was determined by odds ratio (OR) with a 95% confidence interval (CI).

RESULTS

A total of 2508 cases of tuberculosis were detected. The rate of incidence was 41.4 per 100,000 persons-year. Of these 19.4% (487/2508) were coinfected with HIV and 35.6% (893/2508) presented a positive sputum smear, which implies a rate of 14.7 per 100,000 persons-year. In an adjusted multivariate analysis, cases with positive smears were positively associated with the 15-24 (OR=1.9; 95% CI: 1.4-2.4), 25-34 (OR=2.1; 95% CI: 1.7-2.7) and 35-44 years (OR=1.7; 95% CI: 1.3-2.2) age compared with persons 45 years old and above; with males (OR=1.8; 95% CI: 1.5-2.2) and consumers of alcohol (OR=2.1; 95% CI: 1.7-2.7) and negatively with those under 15 years of age (OR=0.1; 95% CI: 0.1-0.2) and coinfection with HIV (OR=0.5; 95% CI: 0.3-0.7).

CONCLUSIONS

Measures to control tuberculosis transmission (prompt diagnosis, study of contacts and directly observed treatments) should be reinforced for male adults with excessive consumption of alcohol.

[Tuberculin test responses of tuberculosis patients]

OBJECTIVE

To determine the response of tuberculosis patients to tuberculin skin tests.

METHOD

The results of skin tests used for initial assessment of tuberculosis patients in Catalonia were reviewed (Multicenter Tuberculosis Research Project). Negative skin tests were those with indurations < 5 mm; positive tests were those with indurations measuring > or = 5 mm. Individuals were classed as having or not having risk factors for developing tuberculosis and those without risk factors were classified by age, location and extension of tuberculosis.

RESULTS

Negative skin tests were seen in 1,566 patients (23%). Negative tests were more common in patients with risk factors, significantly so in those undergoing immunosuppressant therapy (50%) or with HIV infection (61%). Negative tests were less prevalent in patients with no risk factors (13%) and, among them, in children (1%), in patients between 15 and 29 years of age (10%) or in those with non-pulmonary forms (10%). Negative tests were significantly more prevalent among patients 60 to 74 years of age (27%), those over 74 (44%), and those whose disease was pulmonary and extrapulmonary (26%) or disseminated (64%). No significant differences in induration size of positive skin tests were observed for patients with and without risk factors (including HIV infection or non-infection). A normal distribution of induration size was observed in all groups.

CONCLUSIONS

A negative tuberculin skin test for initial assessment should be interpreted in function of the presence or absence of risk factors, age, location or extension of tuberculosis. When a skin test is positive, the response will be similar whether or not an immunodepressant factor is present.

[Booster effect in elderly patients living in geriatric institutions]

BACKGROUND

Tuberculinic response is weaker in elderly patients. To know the real tuberculinic reactivity in elderly patients living in geriatric institutions in Catalonia, Spain, and to reduce the probability of false negatives this study was carried out to detect the booster effect.

METHODS

Three hundred forty-three individuals with a mean age of 81 +/- 6.5 years, residing in four geriatric centers within the Central Health Care Region of Catalonia, were studied. After performing the first Mantoux intradermoreaction this was repeated at one week in case of negativity. A third test was carried out after one more week on persistence of negativity. Qualitative study was carried out on cell immunity in those patients whose first Mantoux test was negative.

RESULTS

The first Mantoux test was positive in 163 patients (47.5%) and the performance of the second and third test in cases of negativity of the previous test increased the number of positivities by 12.5% and 4.1%, respectively, up to a total of 220 patients (64%). In the patients who did not react to the first intradermoreaction and who preserved cell immunity a booster effect was obtained in 57 (52.8%). The prevalence of tuberculous infection in this group of elderly patients calculated in the group with preserved cell immunity was 81.2% while the morbidity of tuberculous disease in the positive cases in the Mantoux test was 1.8%.

CONCLUSIONS

Given the high frequency of the booster effect in elderly people, systematic testing is recommended to investigate tuberculinic reactivity in patients who are initially negative on Mantoux intradermoreaction.

Case-finding of pulmonary tuberculosis on admission to a penitentiary centre

SETTING

The prison population is considered to be at high risk of tuberculosis infection and illness. Therefore, a tuberculosis prevention and control program was put into effect upon the opening of a penitentiary center in Barcelona whose inmates suffered a high prevalence of intravenous drug use (IVDU) (48.4%) and HIV-1 infection (36.0%).

OBJECTIVE

To ascertain the prevalence of tuberculosis infection and pulmonary tuberculosis detected on admission to jail among those prisoners who had no history of tuberculosis; and to study as predictors of tuberculosis infection and illness the variables age, IVDU and HIV-1 infection.

DESIGN

Cross-sectional study.

RESULTS

Of the 729 prisoners studied, 56.2% were considered infected with Mycobacterium tuberculosis. Age was the only variable associated with the infection, with an Odds Ratio (OR) of 1.017 per year (Confidence Interval [CI] 95% = 1.006-1.029) (P = 0.004). The rates of HIV-1 and M. tuberculosis co-infection, which may run between 18.9% and 21.7%, reached 42.8% in the IVDU. A 2.7% prevalence of pulmonary tuberculosis was observed. The most accurate logistic regression model for pulmonary tuberculosis diagnosis (P = 0.09) includes the variables tuberculosis infection (OR = 13.00; CI 95% = 1.7-98), IVDU (OR = 2.60; CI 95% = 0.66-7.81) and age (OR = 1.05; CI 95% = 0.99-1.11).

CONCLUSION

Given the high prevalences observed, we propose the activation and maintenance of programs designed to seek out and identify cases of M. tuberculosis infection and illness within the prison population.