Automated chest X-ray reading for tuberculosis in the Philippines to improve case detection: a cohort study.

BACKGROUND: DetecTB (Diagnostic Enhanced Tools for Extra Cases of TB), an intensified tuberculosis (TB) case-finding programme targeting prisons and high-risk communities was implemented on Palawan Island, the Philippines.OBJECTIVE: To evaluate the performance of TB detection based on computerised chest radiography (CXR) readings.DESIGN: Data from 14 094 subjects were analysed from September 2012 to June 2014. All CXRs were read by a physician and by software. Individuals with TB symptoms or CXR abnormalities according to the physician underwent Xpert® MTB/RIF testing, the remaining persons were considered TB-negative (screening reference). A subset of 200 CXRs was read by an independent human reader (radiological reference). This reader also re-read a subset of the most abnormal cases as identified using the software but read as normal by the physician (discordant cases).RESULTS: A total of 10 755 individuals were included in the analysis, 2534 of whom had a positively assessed CXR; 298 cases were Xpert-positive. Using the screening reference, the area under the receiver operating characteristic curve for software readings was 0.93 (95%CI 0.92-0.94), with a sensitivity of 0.98 (95%CI 0.97-0.99) and a specificity of 0.69 (95%CI 0.40-0.98). Based on the radiological reference, the physician performed slightly worse than the software (sensitivity, 0.82, 95%CI 0.74-0.89 and specificity, 0.87, 95%CI 0.81-0.96 vs. sensitivity, 0.83, 95%CI 0.71-0.93 and specificity, 0.87, 95%CI 0.75-0.95), although this was not statistically significant. Of the 291 discordant cases, 70% were assessed as positive, resulting in a 22% increase in TB detection when extrapolated to the full cohort.CONCLUSION: The performance of automated CXR reading is comparable to that of the attending physicians in DetecTB, and its use as a second reader could increase TB detection.

Multidrug-resistant tuberculosis patients' views of interventions to reduce treatment loss to follow-up.

SETTING: Patients who initiated treatment for multidrug-resistant tuberculosis (MDR-TB) at 15 Programmatic Management of Drug-resistant Tuberculosis (PMDT) health facilities in the Philippines between July and December 2012. OBJECTIVES: To describe patients' views of current interventions, and suggest changes likely to reduce MDR-TB loss to follow-up. METHODS: In-depth interviews were conducted between April and July 2014 with MDR-TB patients who were undergoing treatment, had finished treatment at the time of the interview (controls), or had been lost to follow-up (LTFU). Responses were thematically analyzed. RESULTS: Interviews were conducted with 182 patients who were undergoing or had completed treatment and 91 LTFU patients. Views and suggestions could be thematically categorized as approaches to facilitate adherence or address barriers to adherence. The top themes were the need for transportation assistance or improvements to the current transportation assistance program, food assistance, and difficulties patients encountered related to their medications. These themes were addressed by respectively 63%, 60%, and 32% of the participants. CONCLUSIONS: A more patient-centered approach is needed to improve MDR-TB treatment adherence. Programs should strive to provide assistance that considers patient preferences, is adequate to cover actual costs or needs, and is delivered in a timely, uninterrupted manner.

Multidrug-resistant tuberculosis patients lost to follow-up: self-reported readiness to restart treatment.

SETTING: Multidrug-resistant tuberculosis (MDR-TB) patients lost to follow-up (LTFU) from Programmatic Management of Drug-resistant Tuberculosis facilities in the Philippines. OBJECTIVES: To gain insight into patients' readiness to return to treatment. METHODS: MDR-TB patients who initiated treatment and were categorized as LTFU were identified using TB registers, contacted, and asked to consent to an interview and medical record review. At the conclusion of the interview, patients' readiness to restart treatment was assessed and examined in relation to demographic, clinical, and interview data. Odds ratios were calculated. RESULTS: When asked if they would consider restarting MDR-TB treatment, 3% of the 89 participating patients reported that they had already restarted, 34% indicated that they wanted to restart, 33% had not considered restarting, 28% were undecided, and 2% had decided against restarting. Patients who wanted to restart treatment were more likely to report having borrowed money for TB-related expenses (OR 5.97, 95%CI 1.27-28.18), and were less likely to report being self-employed (OR 0.08, 95%CI 0.01-0.67), or perceive themselves at low or no risk for TB relapse (OR 0.30, 95%CI 0.08-0.96) than patients who did not indicate an interest in restarting treatment. CONCLUSIONS: Efforts to re-engage LTFU patients in care should consider financial barriers, knowledge gaps, and personal adherence challenges in patients.

Outcomes of Category I and II regimens in mono- and polyresistant tuberculosis cases in the Philippines.

SETTING: National Tuberculosis Programme, the Philippines. OBJECTIVE: To compare treatment outcomes of Category I and Category II regimens among mono- and/or polyresistant tuberculosis (TB) cases under programme conditions. DESIGN: Retrospective cohort analysis of pulmonary TB patients from two data sets from the National Drug Resistance Survey and the Programmatic Management of Drug-resistant Tuberculosis by linking drug resistance patterns with treatment outcomes. RESULTS: Of 288 Category I patients, 193 were isoniazid (INH) resistant, 42 were either ethambutol (EMB) or streptomycin (SM) resistant, and 53 were resistant to a combination of two or all three TB drugs. Of 138 Category II patients, 92 were INH-resistant, 9 were either EMB- or SM-resistant, and 37 were poly-resistant. Respectively 206 (87.7%) and 41 (77.4%) mono- and poly-resistant patients treated with the Category I regimen achieved significantly higher successful treatment outcomes, in comparison to respectively 60 (59.4%) and 15 (40.5%) mono- and poly-resistant patients treated with the Category II regimen. CONCLUSION: The Category II regimen produced poor outcomes, whereas the Category I regimen achieved a treatment success rate of more than 85% among new patients with the same drug resistance patterns. The poor outcomes of the Category II regimen could be attributed to other factors such as patient behaviour and comorbidities, rather than drug resistance.

Market size and sales pattern of tuberculosis drugs in the Philippines.

OBJECTIVES: To identify the availability, types and quantity of anti-tuberculosis drugs in the public and private sectors from 2007 to 2011 in the Philippines. METHODS: Analysis of the procurement of and sales data on anti-tuberculosis drugs from both the public and private sectors from 2007 to 2011. RESULTS: Publicly procured anti-tuberculosis drugs were sufficient to treat all reported new tuberculosis (TB) cases from 2007 to 2011 in the Philippines. Nevertheless, the volume of anti-tuberculosis drugs in the private sector would have sufficed for the intensive phase of treatment for an additional 250 000 TB patients annually, assuming compliance with national treatment guidelines. Fixed-dose combination drugs comprised the main bulk (81%) of private market sales, while sales of loose drugs decreased over the years. Combining public and private sales in 2011, 484 725 new TB patients, i.e., 2.4 times the number of notified cases, could have been placed on treatment and treated for at least the intensive phase. Key second-line drugs are not available in the private market, making it impossible to design an adequate treatment regimen for multidrug-resistant TB (MDR-TB) in the private sector. CONCLUSION: An enormous quantity of anti-tuberculosis drugs was channelled through the private market outside the purview of the Philippine National Tuberculosis Control Program, suggesting significant out-of-pocket expenditure, severe underreporting of TB cases and/or misuse of drugs due to overdiagnosis and overtreatment.

Adverse events and development of tuberculosis after 4 months of rifampicin prophylaxis in a tuberculosis outbreak.

We screened tuberculosis (TB) contacts as an outbreak investigation with tuberculin skin test (TST) and interferon-gamma release assay (IGRA). We evaluated adverse events and TB incidence in all persons screened after rifampicin (RFP) prophylaxis, and specifically assessed the new TB cases in relation to initial TST and IGRA results. The 180 contacts were divided into four groups: TST+/IGRA+ (n = 101), TST+/IGRA- (n = 22), TST-/IGRA+ (n = 16), and TST-/IGRA- (n = 41). RFP treatment (4 months) was prescribed only to the TST+/IGRA+ group. Of 87 contacts who initiated prophylaxis, adverse events occurred in 21 contacts (24.1%) including hepatotoxicity (11.5%), flu-like syndrome (5.7%), and thrombocytopenia (3.4%). TB developed in two TST+/IGRA+ subjects after completion of prophylaxis, including one multidrug-resistant (MDR)-TB case during 21.8 months of follow-up. Adverse events were frequent, and development of TB including MDR-TB occurred after RFP prophylaxis.

Combined use of QuantiFERON-TB Gold assay and chest computed tomography in a tuberculosis outbreak.

OBJECTIVE: To evaluate the value of the QuantiFERON-TB Gold (QFT-G) assay and chest computed tomography (CT), in addition to the conventional use of the tuberculin skin test (TST) and chest radiography (CXR), in a contact investigation of a tuberculosis (TB) outbreak. DESIGN: In a contact investigation of a TB outbreak in a high school, TST and CXR were performed on all 1044 employees and students. QFT-G was performed on TST-positive subjects, and CT on QFT-G-positive subjects and students with TST > or =20 mm. RESULTS: TST was positive in 388 subjects (37.2%), while QFT-G was positive in 7.6% (30/394). CXR showed abnormal findings suggestive of TB in 10 (1.0%) subjects, all of whom were TST-positive and six of whom were QFT-G-positive. Findings suggestive of active TB were noted in 17 (32.7%) of 52 subjects by CT. Collectively, among 21 (1.1%) TB patients, all were TST-positive, 12 (57.1%) were QFT-G-positive and active TB was diagnosed by CT, and not by CXR, in 11 subjects. CONCLUSION: Compared to the conventional approach, the additional use of QFT-G in TST-positive subjects and chest CT in subgroups with a high probability of infection was found to be more effective in the differentiation between active TB, latent TB and non-infected subjects in a contact investigation.

Protecting patients' rights, ensuring safety and quality assurance in tuberculosis prevalence surveys.

The conduct of biomedical studies is guided by statements of internationally recognised principles of human rights. The first principle of the Nuremberg Code was the centrality of voluntary participation of subjects with informed consent. All prevalence surveys should be reviewed by the appropriate ethics review committees. Each potential survey participant should be adequately informed of the aims, methods and sources of funding of the survey, any possible conflicts of interest, the institutional affiliations of the researchers, the anticipated benefits and potential risks of the study, and any discomfort it may entail. Attention should be paid to safety in each component of the survey. Test procedures that require particular attention are chest radiography (CXR) and bacteriological examination. Quality assurance should be applied to all aspects of research and, in particular, to any measurements undertaken, including CXR assessments, laboratory examinations and questionnaire and data management. Furthermore, to ensure comparability of data from different surveys, it is important to apply the same survey design and methodology and to use the same reporting format.

Tuberculosis prevalence surveys: rationale and cost.

This article is the first of the educational series 'Assessing tuberculosis (TB) prevalence through population-based surveys'. The series will give overall guidance in conducting cross-sectional surveys of pulmonary TB (PTB) disease. TB prevalence surveys are most valuable in areas where notification data obtained through routine surveillance are of unproven accuracy or incomplete, and in areas with an estimated prevalence of bacteriologically confirmed TB of more than 100 per 100,000 population. To embark on a TB prevalence survey requires commitment from the national TB programme, compliance in the study population, plus availability of trained staff and financial resources. The primary objective of TB prevalence surveys is to determine the prevalence of PTB in the general population aged >or=15 years. Limitations of TB prevalence surveys are their inability to assess regional or geographic differences in prevalence of TB, estimate the burden of childhood TB or estimate the prevalence of extra-pulmonary TB. The cost of a prevalence survey is typically US$ 4-15 per person surveyed, and up to US$ 25 per person with radiographic screening. A survey of 50,000 people, of limited precision, would typically cost US$ 200,000-1,250,000.

Impact of short-term exposure to fluoroquinolones on ofloxacin resistance in HIV-negative patients with tuberculosis.

SETTING: Seoul, Korea, a country with an intermediate tuberculosis (TB) burden and low prevalence of human immunodeficiency virus (HIV) infection. OBJECTIVES: To determine the frequency of ofloxacin (OFX) resistance in Mycobacterium tuberculosis, and to assess whether short-term use of fluoroquinolones (FQNs) induces ofloxacin-resistant M. tuberculosis. DESIGN: The subject cohort consisted of 2788 patients with culture-confirmed TB with drug susceptibility testing data; only four were HIV-positive. The patients were divided into two groups: those who were or were not recently exposed to FQNs. RESULTS: Of the 2788 isolates, the rates of OFX resistance were 1.1% and 8.5% in initially treated and retreated patients, respectively (P < 0.05). Of the 94 OFX-resistant isolates, 83 (88.3%) were multidrug-resistant (MDR). There was no difference in rates of OFX resistance throughout the study period, or between the FQN-exposed (1/39, 2.6%) and control groups (93/2749, 3.4%). The median duration of FQN treatment was 7 days (range 1-47 days). One OFX-resistant isolate in the FQN-exposed group was MDR. CONCLUSION: The rate of OFX-resistant M. tuberculosis was low and stationary throughout the study period in Korea. Most OFX resistance was accompanied by MDR, and the frequency of OFX-resistant M. tuberculosis was low in subjects taking short-term FQNs.

An Internet-based surveillance system for tuberculosis in Korea.

SETTING: The Korea Tuberculosis Surveillance (KTBS) network includes 248 health centres throughout the country, as well as other public and private health institutions. OBJECTIVE: To develop a web-based surveillance system for tuberculosis (TB) and to monitor implementation of the National TB Control Programme (NTP) on an ongoing basis. DESIGN: A TB notification form was developed with new case definitions, and standardised to obtain uniform essential information of the cases with ease and speed. Data collection, compilation, analysis and feedback were made available at every level of the health authority via the Internet without restrictions of time and space. RESULTS: The Internet-based surveillance system was successfully implemented across the country, providing real-time national figures of TB using different variables-patient, time, area, site and type of disease--and facilitating on-line evaluation of NTP implementation. CONCLUSION: The web-based surveillance system has been well established within the existing health infrastructure, providing real-time figures on the TB burden. However, it requires continued improvement of the quality of information and of case reporting activities.

The treatment of tuberculosis in South Korea.

South Korea's complex system of tuberculosis control has never been fully described. The prevalence of tuberculosis has dropped dramatically since 1965, partly because of farsighted governmental policy that provided low-cost, accessible tuberculosis treatment to the entire population. Within the tuberculosis control system, public and private sector entities provide a wide variety of treatment options. The National Tuberculosis Program focuses on improving cure rates for new cases, while the private sector has taken more of a role in the treatment of drug-resistant tuberculosis and other types of complicated cases. There has been a decrease in drug-resistant tuberculosis since 1980 for multiple reasons, including increased cure rates from the introduction of rifampin-based regimens, improved nutrition and living standards, and the treatment of drug-resistant cases in the private sector. Multidrug-resistant tuberculosis, however, still poses a significant threat to public health. The limited outcomes data that exist in South Korea for multidrug-resistant tuberculosis treatment suggest that cure rates are low and failure and abandonment rates are high. New public health measures are needed to improve the control of multidrug-resistant tuberculosis.

Transmission of Mycobacterium tuberculosis among high school students in Korea.

SETTING: Screening of pulmonary tuberculosis (PTB) patients from high school students by chest radiography and sputum examination. OBJECTIVE: To investigate transmission of Mycobacterium tuberculosis among high school students. DESIGN: The restriction fragment length polymorphism (RFLP) profiles of M. tuberculosis isolated from PTB patients of high school students were analysed. RESULTS: Clustering cases were found in PTB patients screened among high school students. During this study, a single strain with a particular RFLP profile was found most frequently. This particular strain was also the one that was found to be most prevalent among Korean clinical isolates of M. tuberculosis during a nationwide TB prevalence survey previously done in 1995. This strain seems to be the most widely distributed one in Korea, and has thus been designated K1. In addition, by close examination of the RFLP profiles, it was also found that there existed 34 isolates that had RFLP profiles that were similar to the K1 strain. Therefore, K1 and K1-related strains (18.4% of strains, designated as the K family) seem to comprise the most dominant M. tuberculosis strains present in Korea. CONCLUSION: Clustering cases were found in PTB patients screened among high school students in Korea. The K family was found to be most prevalent among the clinical isolates of M. tuberculosis found.

Incidence of pulmonary tuberculosis in Korean civil servants: second study, 1992-1994.

SETTING: A survey based on biennial medical examinations for civil servants, including school teachers. OBJECTIVE: To estimate the incidence of pulmonary tuberculosis (PTB) for the general population from a sample survey, and to compare it with the previous results from 1988-1990. DESIGN: A retrospective follow-up study which started in April 1996 and ended in December 1997. Development of PTB among all South Korean civil servants who were tuberculosis-free in the 1992 biennial medical examination was followed up to the 1994 medical examination. The survey was based on newly-developed cases by 1) the findings of the biennial medical examination, 2) scrutiny of the medical records of all those claiming health insurance for tuberculosis, and 3) checking of the TB registers in all health centres during the period. Civil servants constituted 2.6% of the Korean population aged 20 to 64. RESULTS: Of 958137 subjects who were found to be TB-free in the 1992 medical examination, 957216 were eligible for the study. Radiologically active PTB was found in 4146 cases over the 2-year period of the study. Bacteriological examinations were performed in 3306 patients: there were 1551 bacteriologically proven cases (46.9%), including 1049 smear-positives (31.7%) and 502 smear-negative culture-positives (15.2%). The age-sex adjusted incidence rate for the general population was 202/100000 in radiologically active PTB, 81/100000 in bacteriologically proven PTB, and 54/100000 in smear-positive PTB. Overall PTB incidence rates were higher in males (240/100000) than in females (163/100000), and in the age group 20-24 years (280/ 100 000) than in the other age groups. CONCLUSION: Comparing the 1992-1994 PTB incidence with that observed in the 1988-1990 survey, the highest incidence was observed among young adults in both surveys, and it had not changed over time.

Ambulatory treatment of multidrug-resistant pulmonary tuberculosis patients at a chest clinic.

SETTING: Retrospective cohort analysis of multidrug-resistant tuberculosis (MDR-TB) patients treated at a Korean National Tuberculosis Association out-patient chest clinic. OBJECTIVE: To evaluate treatment outcomes and contributing factors. DESIGN: A review of clinical records of 1011 pulmonary MDR-TB patients retreated with individualised regimens selected on the basis of previous chemotherapy and drug susceptibility testing from 1988 to 1996. RESULTS: The patients (mean age 38.6 years) had resistant organisms to an average of 3.7 drugs and were retreated with an average of 4.2 drugs which they had previously not taken and to which they were susceptible. Treatment outcomes were as follows: 487 cases (48.2%) cured, 82 (8.1%) failed, 394 (39.0%) defaulted, 45 (4.5%) transferred out, and three (0.3%) died. The treatment efficacy among those who completed chemotherapy was 85.6%. In a multivariate analysis favourable response was significantly associated with a greater number of newly prescribed drugs in the regimen to which they were susceptible (odds ratio [OR] 3.6; 95% confidence interval [CI] 1.3-9.5), younger age (OR 2.0; 95%CI 1.1-3.9), and a lower number of drugs to which they were resistant (OR 1.8; 95%CI 1.1-3.1). The case fatality rate, including the follow-up period, was 1.7% (17 cases). CONCLUSION: The cure rate of MDR-TB patients treated at an out-patient clinic was 48.2% due to a high defaulter rate (39.0%). However, 85.6% of those who completed treatment were cured.

Twenty-year trend of chronic excretors of tubercle bacilli based on the nationwide tuberculosis prevalence surveys in Korea, 1975-1995.

SETTING: A study of chronic excretors of tubercle bacilli (chronic cases) based on the nationwide random sample surveys of tuberculosis prevalence conducted in Korea from 1975 through 1995. OBJECTIVE: To investigate the temporal trend of the prevalence of chronic cases, and to match these with treatment outcomes and drug resistance rates. DESIGN: Bacillary cases were classified by history of chemotherapy into new (those who denied a history of chemotherapy), non-chronic (those who had taken chemotherapy for less than 2 years) and chronic cases (those who had taken chemotherapy for more than 2 years). RESULTS: Chronic cases decreased from 107 to 12 per 100000 population (annual rate of reduction [ARR] 11.89%) over the 20-year period. The ARR of chronic cases was significantly greater than that of new cases, and accelerated from 1985 (ARR 15.83%), after the application of short course chemotherapy. Rates of overall drug resistance rates increased up to 1980, and those of multidrug resistance up to 1985, followed by a decrease thereafter. A reduction in chronic cases was observed even during the period of increase in drug resistance (including multidrug resistance). CONCLUSION: The prevalence of chronic tuberculosis cases has decreased due to improvements in overall treatment outcome.

Treatment of bacillary pulmonary tuberculosis at the chest clinics in the private sector in Korea, 1993.

SETTING: Cohort study of bacillary pulmonary tuberculosis patients treated at private sector chest clinics in Korea. OBJECTIVE: To assess the treatment behaviour of physicians in private chest clinics and the treatment outcomes of their patients. DESIGN: 1) A retrospective analysis of a cohort of patients admitted from July through October in 1993, and 2) comparison with results from health centres under the National Tuberculosis Programme (NTP). RESULTS: Nine hundred and sixty bacillary patients (507 newly diagnosed--'new', and 453 retreatment--'old') were admitted to the study. Initial smears and cultures were not performed in 7% and 21%, and follow-up smears and cultures not done in 19% and 28%, respectively. The regimens prescribed were variable: 23 in 'new' and 72 in 'old' patients, 86 in total. Six-month short-course treatment using HRZE was prescribed for 26.2% of 'new' patients. In many instances, the planned treatment duration was excessive. The success rates (cured plus completed) for 'new' and 'old' patients were 74% and 51%, respectively. The failure rates were less than 1% in 'new' and 9% in 'old' patients. CONCLUSION: Prescribed regimens were variable in terms of drug combinations and treatment duration. Overall treatment outcome was inferior to that of the health centres under the NTP.

Tuberculin PPD RT23: has it lost some of its potency?

PPD RT23 is a tuberculin that is used worldwide. Korea has been using 1TU RT23 for its nationwide tuberculosis prevalence surveys at five-yearly intervals since 1965, and found a drop in its potency after the 1975 survey. This finding draws attention to the interpretation of tuberculin survey data observed with RT23 at different time periods.