6

Addressing TB and Drug-Resistant
TB in Vulnerable Populations

Key Messages

  • The difficulty of obtaining positive cultures in children complicates the treatment of MDR TB and has obscured the extent of drug resistance in pediatric TB cases.
  • Many children are infected by the adults with whom they live, but in a significant fraction of cases, the infection comes from elsewhere.
  • In one study from Peru, children living in households with an MDR TB patient had 10 times the risk of infection of the general population.
  • Coinfection with HIV and drug-resistant TB is a serious threat to TB control.
  • Asia has a history of major refugee movements, and refugee status can lead to displacement and overcrowding, which lead in turn to the spread of infection.

Some vulnerable populations—including children, people coinfected with HIV, and refugees—are at higher risk of contracting TB and are more difficult to treat than others. Speakers at the workshop addressed each of these three populations, discussing MDR TB among children in India, Peru, and globally; data from India and elsewhere linking HIV infection to drug-resistant TB; the occurrence of drug-resistant TB among Tibetan refugees living in India; and case studies dealing with vulnerable populations (children in Cambodia and MDR TB patients in Ethiopia, many coinfected with



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 71
6 Addressing TB and Drug-Resistant TB in Vulnerable Populations Key Messages • he difficulty of obtaining positive cultures in children complicates T the treatment of MDR TB and has obscured the extent of drug resistance in pediatric TB cases. • any children are infected by the adults with whom they live, but in M a significant fraction of cases, the infection comes from elsewhere. • n one study from Peru, children living in households with an I MDR TB patient had 10 times the risk of infection of the general population. • oinfection with HIV and drug-resistant TB is a serious threat to C TB control. • sia has a history of major refugee movements, and refugee status A can lead to displacement and overcrowding, which lead in turn to the spread of infection. Some vulnerable populations—including children, people coinfected with HIV, and refugees—are at higher risk of contracting TB and are more difficult to treat than others. Speakers at the workshop addressed each of these three populations, discussing MDR TB among children in India, Peru, and globally; data from India and elsewhere linking HIV infection to drug- resistant TB; the occurrence of drug-resistant TB among Tibetan refugees living in India; and case studies dealing with vulnerable populations (chil- dren in Cambodia and MDR TB patients in Ethiopia, many coinfected with 71

OCR for page 71
72 DRUG-RESISTANT TUBERCULOSIS IN INDIA HIV). All of the speakers emphasized the particular difficulties of reaching vulnerable populations and the steps that must be taken to identify, diag- nose, and treat MDR TB among these groups. DRUG-RESISTANT TB IN PEDIATRIC POPULATIONS1 A major risk for pediatric TB is contact with an infected adult, observed Soumya Swaminathan, Head, Division of Clinical Research, National Insti- tute for Research in Tuberculosis. Rates of infection among adults aged 25–44 are highest in the African region, followed by Southeast Asia (Figure 6-1), and the risk of TB in children is likely to be correspondingly high in these regions. Other risk factors include large household size, severe mal- nutrition, exposure to household smoke, having a female index case, and in some cases, being a member of certain minorities. Like adults, children tend to go through several phases after infection with M.tb. After an initial phase marked by hypersensitivity responses and skin test conversion, which typically occur in the first 6–8 weeks, the pri- mary disease follows. Most of the disseminated disease tends to occur in the first 2–4 months after infection. Lymph node disease in younger children and pleural disease in older children can occur at 6–8 months. The adult form of the disease, which generally is seen in older children, can occur several years after infection. Swaminathan explained that more children than adults with TB are smear-negative, although this varies with the population under study. In one study of 1,098 children seen at the LRS Institute of Tuberculosis and Respiratory Diseases in New Delhi, 414 children were smear-positive, 404 were smear-negative, and sputum status was not known for 280 patients (Sharma et al., 2008). The smear-positivity rate was higher among older children—about 60 percent—but even among children younger than 6 years old, 30 percent were smear-positive. Unknown Burden of Pediatric MDR TB Data on MDR TB in children are virtually nonexistent. WHO does not include children in drug resistance surveys, and most countries have not collected these data systematically. A plan to gather data on children is urgently needed, said Swaminathan. 1 This section is based on the presentation of Soumya Swaminathan, who was Coordinator for Neglected Priorities Research with the WHO Special Programme for Research and Training in Tropical Diseases (TDR) at the time of the workshop. Since the workshop, Swaminathan has rejoined the National Institute for Research in Tuberculosis as Head, Division of Clinical Research.

OCR for page 71
73 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS Incidence per 100,000 population 180 160 140 120 100 80 60 40 20 0 0–14 15–24 25–34 35–44 45–54 55–64 65+ African Region American Region Eastern Mediterranean Region European Region Southeast Asian Region Western Pacific Region FIGURE 6-1 TB incidence rates are highest in young adults in the African and Southeast Asian regions. SOURCE: Swaminathan, 2011. In India, an estimated 7 percent of the reported 1.3 million new TB cases annually are in children. This percentage ranges from 2 percent of new smear-positive cases to 15 percent of new cases of extrapulmonary TB. The percentage of drug-resistant TB is more difficult to estimate. In the Western Cape Province of South Africa, Schaaf and colleagues (2009) found isoniazid resistance to be 7.7 percent among pediatric TB cases and the MDR TB rate to be 6.7 percent, which represented an increase since the 1990s. These levels are higher than in adults in the general TB population in South Africa. Previously treated children had significantly higher rates of drug resistance than new TB cases, and HIV infection was not associated with drug resistance in children, which is also the case in India. In data from India, MDR TB rates among children with TB were found to be 2 percent about two decades ago (Ramachandran and Prabhakar, 1992).2 A few years later, a multicenter study on children with pulmonary TB found a rate of MDR TB of 3.5 percent. These rates depend on the population under study and whether patients are coming from the com- munity or a hospital. In general, said Swaminathan, drug-resistant TB in 2 These low pediatric MDR TB rates were observed prior to the implementation of DOTS in India.

OCR for page 71
74 DRUG-RESISTANT TUBERCULOSIS IN INDIA children mirrors that in the adult population because adults are the source of their infections. Challenges in the Diagnosis and Treatment of Pediatric Drug-Resistant TB Bacterial confirmation of MDR TB in children is not always possible. Instead, drug-resistant TB in a child must be suspected when the child has been in contact with a known case of drug-resistant TB, the child’s adult contact has been on chronic irregular treatment and continues to be sputum-positive, the adult contact dies after irregular treatment, or the child shows initial improvement with anti-TB treatment and then deterio- rates clinically and radiologically. Contact investigation is vital to detecting pediatric cases of TB and especially MDR TB. Diagnosis of drug-resistant TB in children requires specimens, such as gastric aspirate, induced sputum, nasopharyngeal aspirate, and extrapul- monary specimens, along with good specimen processing, transport, and testing. The yield of culture in various studies in children has ranged from 10 to 40 percent, depending on patient selection criteria and laboratory methods. The MODS method has been shown to be faster and more sensi- tive than Löwenstein-Jensen medium. Another study found that MODS has a slightly higher sensitivity than the MGIT method while also being faster (Ha et al., 2009). GeneXpert has the highest sensitivity in smear-positive culture-posi- tive adults, providing an assessment of rifampicin resistance within about 2 hours, but these data are not available for children. (See the section Case Studies in Cambodia and Ethiopia later in this chapter.) Preliminary data from South Africa show that the sensitivity of GeneXpert is prob- ably around 70-80 percent of culture-positive cases, which represents only 10–40 percent of all pediatric TB cases. More research will be necessary before recommendations can be made regarding the use of GeneXpert with different samples from children. The treatment approach in children is largely the same as that in adults. Factors to consider include the following: • The child should receive treatment that is consistent with that of the adult source case if no isolate is obtained from the child. • At least three or, preferably, four or more drugs to which the isolate is susceptible should be used. • The child’s growth and development need to be monitored, and drug dosages need to be adjusted for weight gain. • The caregiver needs to receive counseling about adherence, treat- ment duration, and adverse effects at every visit.

OCR for page 71
75 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS • It has been suggested that early primary pulmonary MDR TB, because of its paucibacillary nature, can be treated for 12–15 months rather than 18–24 months; however, this needs to be vali- dated in clinical trials. Microbiological monitoring is important, but follow-up cultures often are difficult to obtain and more often are negative. Clinical and chest radiographic monitoring during follow-up is helpful. In one study, the delay in initiating treatment for MDR TB in children was just 2 days if the MDR TB source case was taken into account, but the delay was 246 days if the source case was not considered (Schaaf et al., 2003). The correlation between the DST results for the child and adult was 68 percent in this study, although this correlation can vary greatly from place to place. Obtaining a detailed contact history is essential since a delay in initiating appropriate MDR TB treatment can have serious consequences. In very small cohorts from Peru, second-line TB treatment was well tolerated by children, even though they had high rates of malnutrition and anemia (Drobac et al., 2006). Even when children showed resistance to as many as five drugs, sputum conversion occurred in a majority of those receiving individualized treatment. However, significant residual sequelae and morbidity occurred, including 24 percent with airway obstruction and 40 percent with restrictive lung disease. Swaminathan highlighted several challenges in the diagnosis and treat- ment of MDR TB in children: • The definition of MDR TB in children is different from that in adults because a culture is not always available. • Rapid molecular tests need to be studied in the context of pediatric MDR TB. • More data are needed on the burden of disease in children, includ- ing drug resistance. • More information is needed early in the process of drug develop- ment on how new drugs work in children. • Shorter regimens are needed for MDR TB in children. • Preventive therapy is needed for MDR TB contacts of children. THE BURDEN OF PEDIATRIC TB IN HOUSEHOLDS OF PATIENTS WITH MDR TB3 An ongoing study in Peru, presented by Mercedes Becerra, Assistant Professor, Harvard Medical School, is examining the extent of MDR TB 3 This section is based on the presentation by Mercedes Becerra, Assistant Professor, Har- vard Medical School.

OCR for page 71
76 DRUG-RESISTANT TUBERCULOSIS IN INDIA in the household contacts, including children, of patients with MDR TB. Peru, with a population of almost 30 million, about 10 million of whom live in Lima, recently was removed from the list of the 22 countries with the highest TB burden. It now has an annual TB case notification rate of just over 100 per 100,000 population. Eighty percent of TB cases in the country are found in the capital, and the disease is concentrated in the capital’s poorer districts. In the last national drug resistance survey, which was conducted in 2006, MDR TB was found in 5 percent of new TB patients and in 23 percent of those with a history of TB treatment. HIV coinfection still is relatively rare in MDR TB patients, with an estimated seroprevalence of around 0.5 percent and a rate of about 1.4 percent. Study Environment and Design In 1996, Partners In Health, in collaboration with the government of Peru, launched a program to treat patients with confirmed MDR TB, lead- ing to a cure for about two-thirds (Mitnick et al., 2003, 2008). In 2004, the program conducted a retrospective cohort study to determine what had happened to the family members of these MDR TB patients, including chil- dren. First, each patient who had started the MDR TB treatment regimen between 1996 and 2002 was identified. A study team visited the households of these index patients and interviewed them and others in the household, asking specifically about TB treatment in any of the household members. The study team then reviewed the medical charts at public health centers of each household member who had reported TB treatment to obtain the dates of the regimen received and other details. With these data, the study team confirmed the list of individuals who had been living with patients when they started their MDR TB regimen. The team also identified those who had been treated for TB after the index patients started their regimen. In January 2011, the team published a first report about the TB disease burden in this cohort of almost 5,000 house- hold contacts (Becerra et al., 2011). Study Results and Findings Unpublished data being prepared for publication as of the time of the workshop cover 1,299 children in the households studied, 70 percent of whom had at least 4 years of follow-up from the time the index patient started the MDR TB regimen. Child household contacts were defined as those less than 15 years of age. Three key results emerged from these data. First, 67 children in the households were treated for TB disease during this 4-year retrospective observation period, so that pediatric TB accounted

OCR for page 71
77 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS for 20 percent (67/343) of all the TB cases reported in the household con- tacts. Thus, by the end of this period, more than 5 percent (67/1,299) of the children in the households had been treated for active disease. Furthermore, because 30 percent of the children had less than 4 years of follow-up, this proportion may be an underestimate. Second, of the 67 children treated for TB, only 8 had DST results in their medical chart, because in children it is difficult to obtain adequate sputum specimens for testing. Of these eight children, seven had MDR TB isolates. Six of the eight had TB isolates available for genotyping, and all six were found to be an identical genotypic match to the isolate from the index patient. They were most likely either infected directly by the patient or part of the same chain of transmission, so that both were infected by some other source patient. This result strongly suggests that the great majority of the observed TB disease in these children was due to MDR TB transmitted from the index case. Third, among this population of child contacts, the window to estimate the prevalence of treated TB was defined as the period up to 6 months before and up to 1 month after the date that the index patient started the MDR TB treatment regimen. The prevalence of treated TB was almost 1,800 per 100,000 children. This prevalence was highest among 1- and 2-year-olds—more than 2,500 cases per 100,000 children, which is 10 times the prevalence in the general population. The TB prevalence in 1- and 2-year-olds was similar to that observed in the group of adults. Similarly, the incidence rate of treated TB in all children during the first year of follow-up exceeded 2,200 per 100,000 child-years. “These disease rates are certainly alarming and should rightly give us pause,” said Becerra. “These rates among children [in the prevalence win- dow and in year 1 of follow-up] are squarely in the range of the TB case rates that were observed in the jails and prisons of Russia in the 1990s.” The only optimistic finding is that the high-risk window is within the 2-year period that is required to complete an MDR TB treatment regimen. If a patient is being visited by a health worker or treatment supporter dur- ing that period, that individual, with enough training and support, can observe others in the household. But programs need clear guidance about how to evaluate the household contacts efficiently over time. The results of this study are an important reminder, according to Becerra, that the household contacts of MDR TB patients, including chil- dren, are a likely source of more MDR TB cases. Children who are living with a patient starting MDR TB treatment are at high risk for having TB. In this study, the estimated prevalence of TB disease at baseline was about 2 percent, which means that at least 50 children would need to be screened to find each TB case. This amounts to roughly 10 times the TB case rates reported in the general population.

OCR for page 71
78 DRUG-RESISTANT TUBERCULOSIS IN INDIA Finally, the results underline the need for appropriate pediatric treat- ment regimens: in the small number of children for whom DST results were available, as in the larger number of adults tested, approximately 90 percent had MDR TB. To optimize their chance of being cured, children with MDR TB require regimens designed to treat drug-resistant disease. During the discussion period, Edward Nardell, Harvard Medical School, noted that inhaled drugs, particularly kanamycin and capreomy- cin, could offer promise in the treatment of children. Preliminary studies in guinea pigs reveal that inhaled capreomycin can achieve therapeutic levels in the blood and high levels in the lung (see, for example, Fiegel et al., 2008; Garcia-Contreras, 2007). Although very young children might have difficulty with inhalation, many children already receive asthma medication by that route. Becerra noted the possibility of using a shorter treatment regimen with children because their disease is being detected in an earlier phase relative to adults. Nardell also cited the possibility of treatment with experimental drugs because the drugs could be used under controlled circumstances to look for an effect in a very short time. There was some discussion about the use of prophylactic therapy in contacts of TB and MDR TB patients. One participant asked whether children should be considered a high-risk population and be offered DST even if they are not smear-positive, despite limited laboratory resources. Salmaan Keshavjee, Harvard Medical School, responded that this practice would represent a major policy change in the developing world, although it is seen in the developed world. DRUG RESISTANCE IN HIV-INFECTED POPULATIONS4 Coinfection with HIV and drug-resistant TB is a serious threat to TB control, said Digambar Behera, Director, LRS Institute of Tuberculosis and Respiratory Diseases. Kawai and colleagues (2006) found that more than 50 percent of HIV-MDR TB patients in Peru died within 2 months of diagnosis. Studies with longer follow-up observed death rates ranging from 72 to 89 percent (Coker, 2004). Authors of a study in the United Kingdom estimated that MDR TB patients who are immunocompromised are nine times more likely to die than those who are not immunocompro- mised (Drobniewski et al., 2002). In a 2005–2006 study in the province of KwaZulu-Natal, South Africa, 98 percent (52 of 53) of coinfected XDR TB and HIV patients died, with a median survival time of 16 days from the XDR TB diagnosis (Gandhi et al., 2006). A follow-up, retrospective, 4This section is based on the presentation of Digambar Behera, Director, LRS Institute of Tuberculosis and Respiratory Diseases.

OCR for page 71
79 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS observational study in South Africa conducted from 2005 to 2007 revealed that while 1-year mortality for MDR and XDR TB patients had improved, the majority of deaths—40 percent of MDR TB cases and 51 percent of XDR TB cases—still occurred in the first 30 days after diagnosis (Gandhi et al., 2010). Among the 272 diagnosed MDR TB patients and 382 XDR TB patients in this study, HIV coinfection rates were 90 and 98 percent, respectively (Gandhi et al., 2010). Although HIV infection has been associated with MDR TB outbreaks in institutional settings, such as hospitals and prisons, whether HIV infec- tion also is associated with MDR TB outbreaks in community settings remains unclear. In Thailand, data collected prospectively on pulmonary TB cases treated in public clinics showed that HIV is common among MDR TB patients but is not an independent risk factor for MDR TB (Akksilp et al., 2009). Nevertheless, populations at high risk for HIV—including young adults, men, and injection drug users—should be a priority for DST, said Behera. A systematic meta-analysis summarizing the evidence from 32 studies found no clear association between MDR TB and HIV infection across time and geographic location (Suchindran et al., 2009). Comparisons of MDR TB prevalence to HIV status ranged from 0.21 to 41.45. Assessment by geographic region or study period did not reveal noticeable patterns. The summary prevalence ratios for acquired and primary MDR TB were 1.17 and 2.72, respectively. While this meta-analysis could not demonstrate an overall association between MDR TB and HIV or acquired MDR TB and HIV, it does suggest that HIV infection is associated with primary MDR TB. In general, well-designed studies and surveillance in all regions of the world are needed to better clarify the relationship between HIV and MDR TB. Specifically with respect to India, Deivanayagam and colleagues (2002) found that about 60 percent of 1,000 TB patients were culture-positive, and 34 percent had MDR TB. The HIV seropositivity in the MDR TB group was around 4.4 percent. Swaminathan and colleagues (2005) found that the MDR TB rate in both new and previously treated TB cases was not substantially different in HIV-positive and HIV-negative patients. However, a study from Pune (Pereira et al., 2005) found that 10 percent of HIV-pos- itive patients and only 2.5 percent of HIV-negative patients had MDR TB. S. Singh and colleagues (2007) determined that of 54 patients with AIDS, 12 were resistant to first-line drugs, and 4 of these were also resistant to second-line drugs. Reports of XDR TB in India have been surfacing. However, most of these reports do not include information about HIV status (Table 6-1). With a population of about 1.21 billion, India has about 480 million people infected with TB (Figure 6-2). It also has an estimated 2.27 million people infected with HIV. People coinfected with TB and HIV are estimated

OCR for page 71
80 DRUG-RESISTANT TUBERCULOSIS IN INDIA to number around 1 million. Of the approximately 2 million new TB cases annually, an estimated 100,000 will have both HIV and TB infection, given that about 5 percent of incident TB cases are estimated to be HIV-positive (Dewan et al., 2010). However, the TB epidemic in India is being driven pri- marily by the 400 million people with TB who are not coinfected with HIV. The proportion of registered TB patients who are HIV-positive is highly variable, ranging from less than 1 percent to more than 10 percent in differ- ent parts of India. Within some districts, as many as 46 percent of registered TB patients are HIV-positive. Overall, said Behera, studies in India have not demonstrated an asso- ciation between HIV infection and MDR TB, a finding that contrasts with results of studies conducted elsewhere. However, the National Laboratory Committee has decided that all future drug resistance surveys should cap- ture HIV status and that TB patients should routinely be referred for HIV testing. Great challenges remain in the areas of diagnosis and treatment of TB- HIV coinfection, said Behera. For example, DOTS-Plus does not include a separate program for HIV-positive patients, and drug interactions in coinfected patients can be difficult to manage. However, a TB-HIV col- laboration begun in 2001 is conducting joint training, intensified case finding, and HIV testing of TB patients with HIV risk factors. It has scaled up its activities to 14 states and has piloted routine referral of TB patients for HIV testing. A phased expansion of the TB-HIV initiative will cover the entire country by 2012. Operational guidelines for airborne infection control, refinement of treatment, and development of training materials all are under way. DRUG-RESISTANT TB IN MIGRANT AND REFUGEE POPULATIONS5 The United Nations (UN) High Commissioner for Refugees defines a refugee as a “person who owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable to or, owing to such fear, is unwilling to avail himself of the protec- tion of that country.” An internally displaced person is a person sharing the characteristics of refugees but displaced within the boundaries of a country. The UN estimates that in 2006 there were more than 32 million refugees, internally displaced people, or similarly vulnerable people. Asia has a history of major refugee movements. Since 1979, nearly 5This section is based on the presentation of Kunchok Dorjee, Director, Tibetan TB Control Programme, Delek Hospital.

OCR for page 71
TABLE 6-1 Profile of XDR TB in India No. of No. of MDR TB HIV-Positive Prevalence of Study Setting Cases Cases XDR TB (%) Reference Mondal and Jain, 2007 Tertiary care center, 68 Not reported 5 (7.4) Emerging Infectious Lucknow Diseases, 2007 Singh et al., 2007 Tertiary care center, 12 All HIV-infected 4 (33.3) AIDS, 2007 New Delhi Sharma et al., 2009 AIIMS, New Delhi, tertiary 211 All HIV-negative 5 (2.4) Indian Journal of Medical care hospital Research, 2009 Ramachandran et al., 2009 Gujarat, field study 216 Not reported 7 (3.1) International Journal of (all previously Tuberculosis and Lung treated cases) Disease, 2009 Myneedu et al., 2011 LRS Institute 223 Not reported 45 (20.17) International Journal of Tuberculosis and Lung Disease, 2011 SOURCE: Behera, 2011. 81

OCR for page 71
82 DRUG-RESISTANT TUBERCULOSIS IN INDIA Total population of 1.21 billion ≥0.9 million TB/HIV coinfected 400 million+ TB infected ~2.27 million HIV-infected 1.9 million new TB cases 4.85% of TB cases HIV-infected (~95,240 cases) FIGURE 6-2 The TB epidemic in India is being driven primarily by the approxi- mately 400 million people infected with TB who are not coinfected with HIV. Figure 6-2 SOURCE: Behera, 2011. 6 million Afghan refugees have moved in and out of Pakistan and Iran. The 1947 partition of India and Pakistan created the largest movement of people in history—15 million. And in 1959, when the Himalayan refugee crisis occurred, thousands of Tibetan refugees fled into India, Nepal, and Bhutan. Refugee status is a driver of TB, said Kunchok Dorjee, Director, Tibetan TB Control Programme, Delek Hospital, because it results in displacement, a scarcity of shelter, and overcrowding, which in turn lead to the spread of infections, including TB. Refugee status also can lead to delayed diagnosis as a result of such factors as financial and personal hardships, reluctance to visit a doctor for anything less than an urgent condition, language and cultural barriers, and a lack of health education. Delayed diagnosis of TB can in turn lead to increased spread in the community. These same factors can result in poor treatment adherence, default, and the emergence of drug- resistant strains. In 1959, thousands of Tibetans followed the Dalai Lama into exile. The government of India provides asylum to Tibetans who continue to flee

OCR for page 71
83 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS across the Himalayas into exile in India. Tibetans have resettled in various locations throughout India. The Tibetan Government in Exile is seated at Dharamshala, Himachal Pradesh. In the early years of exile, a large number of Tibetans died from TB. The Department of Health of the Tibetan Government in Exile estimates that the TB prevalence at that time was 30 percent of the entire exiled population, although the exact number is not known. The total population of Tibetan refugees is about 150,000, with the largest number residing in India. It is an extremely mobile population, with people coming into India and returning to Tibet annually. People move across the Indian, Nepalese, and Tibetan borders and within countries, which makes individual case management challenging. Many Tibetans live in closed and congregate settings, such as dormitories in schools, monas- teries, nunneries, and reception centers for newly arrived refugees. This situation makes community transmission very easy and delayed diagnosis very costly, said Dorjee. In the 1990s, the incidence of TB among Tibetan refugees in India was about 835 per 100,000 population (Nelson et al., 2005). A study of Tibetan immigrants from India and Nepal to Minnesota showed a positive tubercu- lin skin test (TST) rate of almost 98 percent (Truong et al., 1997). Among Tibetan refugee claimants in Toronto, almost 97 percent were TST-positive (Marras et al., 2003). The number of new TB cases detected in the Tibetan population in South India was relatively stable from 2006 to 2010, ranging from 223 to 291. Of these, between 14 and 23 were MDR TB cases, with a surge of cases, from 9 to 23, occurring in 2010. One reason for the surge may be that sputum cultures were done in 2010 for every smear-positive and relapsed case, leading to the detection of more MDR TB patients. At the Tibetan Delek Hospital in Dharamshala, the number of TB cases declined from 290 in 2007 to 171 in 2010. But the number of MDR TB patients stayed roughly stable, ranging between 33 and 43. Many of the MDR TB patients at the hospital are college students, who must withdraw from school to take their treatment for 2 years. These patients also include monks, nuns, businessmen, and the unemployed. Tibetans born in Tibet tend to have fewer cases of MDR TB than of drug-sensitive TB, even though acquired drug resistance is relatively high in Tibet, while Tibetans born in India have more. The number of women with MDR TB is slightly higher than the number of men, and the great majority are between 14 and 40 years old. Resistance to second-line drugs is variable, and some XDR TB strains are beginning to emerge in the Tibetan community. The size of the Tibetan population is only 6 million, so an epidemic of MDR and XDR TB in the community could be disastrous. Given Tibetans’ highly charged political situation, nonpolitical issues such as health care

OCR for page 71
84 DRUG-RESISTANT TUBERCULOSIS IN INDIA tend to be overlooked. “MDR and XDR are a risk to the entire genera- tion,” said Dorjee. CASE STUDIES IN CAMBODIA AND ETHIOPIA6 Projects carried out by the Global Health Committee/Cambodian Health Committee, which began working in Cambodia in 1994 and has cured approximately 25,000 people since then, illustrate some of the dif- ficulties of diagnosing and treating vulnerable populations. Novel Diagnostic Modalities Among Children in Cambodia The Global Health Committee/Cambodian Health Committee has been investigating novel approaches for TB diagnosis among children in Cam- bodia. As noted previously, diagnosis of TB is children is challenging. It is often difficult to obtain sputum specimens, and children frequently have paucibacillary disease, which makes microbiological diagnosis uncommon. With standard diagnostic criteria, diagnostic accuracy in children is poor. At the same time, validation of novel diagnostic technologies is per- formed primarily among adults. Data on pediatric prevalence and incidence are limited for both TB and drug-resistant TB. No TB drug trials in children are under way, and few pharmacokinetic studies have been conducted in pediatric populations. The result is that little information is available about the efficacy of control and prevention measures. The project in Cambodia (a partnership between the Global Health Committee/Cambodian Health Committee and the Aeras Global TB Foundation, with funding from the Annenberg Foundation), which was described by Anne Goldfeld, Cofounder, Global Health Committee/Cam- bodian Health Committee, and Professor of Medicine, Harvard Medical School, is designed to evaluate the performance of GeneXpert and the urine lipoarabinomannan (LAM) assay in a pediatric cohort, evaluate the diagnostic utility of stool specimens, and determine the prevalence of TB disease and latent TB infection in a cohort of Cambodian children. The study involves a cross-sectional survey in a province of eastern Cambodia abutting Vietnam where the Global Health Committee/Cambodian Health Committee has worked since 1994. The study enrolled household contacts of index patients with TB, children attending 28 outpatient health centers, and children admitted to 2 district hospitals, with enrollment being con- ducted from July 2010 to February 2011. 6 This section is based on the presentation of Anne Goldfeld, Cofounder, Global Health Committee/Cambodian Health Committee, and Professor of Medicine, Harvard Medical School.

OCR for page 71
85 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS Children were admitted to Svay Rieng District Hospital, where stan- dardized and systematic TB screening data were collected, including med- ical history, physical examination, chest x-ray, symptom screening, TB exposure, HIV status, and HIV exposure. The laboratory workup included two gastric aspirates, one induced sputum sample, a stool specimen, and a urine specimen. The microbiological workup included acid-fast direct smear microscopy, Löwenstein-Jensen and MGIT culture, GeneXpert, and urine LAM. Among 876 children enrolled, the TB prevalence was 16 percent. But the yield of microbiological confirmation using routine culture methods in a well-established laboratory at the Institute of Cambodia in Phnom Penh was low—just 1 percent. The TB in these children was being detected very early, which probably increased the rate of overdiagnosis. GeneXpert could be used for gastric aspirates and provided rapid diagnosis of TB, but it yielded no incremental results compared with the culture method. The results were obtained in 2 hours as opposed to weeks or months, said Goldfeld, but the results demonstrate that other approaches to diagnosing children with TB are needed. During the discussion period, a workshop participant pointed out that the advantage of GeneXpert and other diagnostic technologies is not necessarily greater sensitivity but the decentralization of analysis. A major difficulty with diagnosis is collecting quality specimens and transporting them to a laboratory, and this difficulty is even greater with children and other high-risk groups. Automated technologies can decentralize diagnosis to such populations, the participant pointed out. Filling the MDR TB Treatment Gap in Ethiopia Ethiopia has the world’s fifteenth highest burden of MDR TB and is one of the poorest countries in the world. In 2009, the Global Health Committee/Cambodian Health Committee, working under the name Global Health Committee in Ethiopia, initiated the countrywide MDR TB treat- ment program in partnership with the Jolie-Pitt Foundation and the Ethio- pian Ministry of Health. Among a population of 80 million people, an estimated 130,000 new TB cases occur each year in Ethiopia, including about 6,000 new MDR TB cases, and these numbers are likely to be underestimates, said Goldfeld. Before 2009, a major effort to build laboratory capacity, with assistance from FIND, resulted in 221 MDR TB cases being documented by DST. In August 2008, these 221 patients were waiting in the Addis Ababa area for treatment. A GLC application initiated in 2007 had been approved for the first 45 courses of treatment starting in October 2008. The Global Health Committee/Cambodian Health Committee brought a team to St. Peter’s Hospital in Addis Ababa to help initiate the Ethiopian

OCR for page 71
86 DRUG-RESISTANT TUBERCULOSIS IN INDIA MDR TB program. It also brought the Ethiopian MDR TB team to Cam- bodia to provide didactic and hands-on treatment. As of December 2008, however, there still were no anti-MDR TB drugs in Ethiopia. Furthermore, many other problems challenged the delivery of MDR TB treatment. Isola- tion beds were not available, and the construction of a new ward at the hospital had been delayed. Human resources were limited, and only partial laboratory testing was available. A pharmacy for second-line drugs had not been established, and an outpatient system did not exist. Using a supply of capreomycin donated by Eli Lilly & Co. and funding from the Jolie-Pitt Foundation, the Global Health Committee/Cambodian Health Committee initiated MDR TB care, in partnership with the Ethio- pian Ministry of Health, at St. Peter’s hospital in February 2009. Between then and the arrival of the GLC drugs in September 2009, three cohorts totaling 37 patients began treatment. When the 45 courses of treatment arrived, the treatment team was already assembled, and other issues, such as ancillary medications for side effects and management issues, were under control. By the time the new MDR TB ward at the hospital was completed in June 2010, five more cohorts of patients were receiving treatment. After reviewing the project, the GLC sent another 245 treatment courses to Ethiopia. Another program in Gondar, in northern Ethiopia, began with three patients in August 2010. As of the time of the workshop, 213 patients had been initiated on therapy, including 183 in Addis Ababa and 17 in Gondar. Seven patients had completed treatment. Eighteen had died, six within the first 30 days, an indication of how sick this group of people was. One of the deceased was a suspected XDR TB case. A total of 188 patients were currently on active treatment as of the workshop—125 outpatients and 46 inpatients in Addis Ababa and 17 in Gondar. Three patients who were presumed to have XDR TB were being treated for it. Only one patient had interrupted treatment. Of the 221 backlogged cases of MDR TB confirmed by DST, 66 had started on therapy. Twenty percent of these cases were confirmed dead in a house-to-house search, and 50 percent, many of whom presumably had died, could not be located. Of the 18 deaths among the 213 people treated, the mean time to death was 79 days, with a range of 1 to 298 days. The mean age was 31.5, with a range of 20 to 58. The mean number of prior treatments was 2.24. Comorbidities included severe malnutrition (44.4 percent), HIV infection (33.3 percent), diabetes (16.7 percent), cor pulmonale (16.7 percent), and cirrhosis (5.5 percent). Most people died as a result of respiratory decom- pensation associated with end-stage TB, including two with probable ten- sion pneumothorax. One had TB pericarditis, another had worsening chest x-rays, another died suddenly with cor pulmonale, and others had prob- able superimposed pneumonias. At the time the six patients who had been

OCR for page 71
87 TB AND DRUG-RESISTANT TB IN VULNERABLE POPULATIONS on therapy longer than 6 months died, three had culture-converted, and three were persistently positive. None of the deaths was the direct result of adverse events attributable to MDR TB. Among all the patients, the mean age was 30.17 (with a range of 8 to 76). The mean number of prior treatments was 2.65, with a range of 1 to 8, and the HIV coinfection rate was 23 percent. The mean time to culture conversion was 38 days. The very low default rate has been achieved by engaging family and other social resources in supporting the patient, said Goldfeld. All patients are visited at least once a month in their homes and once a month in the clinic. If their discharge orders include daily injections, they are seen every day in the health center. Health care providers make sure patients are taking their medications and deliver medications to patients who cannot obtain them. Patients sign a contract stating that they will complete treat- ment. Providers also bring food baskets to patients to support them and help them deal with the gastrointestinal difficulties associated with MDR TB treatment. The direct collaboration that has been achieved among Ethiopian and Cambodian physicians has been described as a “south-to-south transfer,” or the sharing of expertise and best practices from one resource-limited setting to another. Goldfeld explained that the approach of integrating hospital- and community-based treatment has filled the gap in Ethiopia and provides a model for expansion. A key challenge is to forecast the need for drugs and make them available. Goldfeld contrasted the difficulties faced in obtaining needed drugs with the speed and efficiency of retail systems, noting that “we can get flowers from Holland to Winnipeg in 12 hours. Why can’t we get life-saving drugs from drug stocks controlled by the GDF to countries where patients are dying due to their lack in a timely fashion?” Funds also are needed for clinical care, ancillary medications, basic labora- tory tests, staff support, food, and outpatient monitoring. In Ethiopia, these funds come from private sources, which Goldfeld said is surprising since the government sets funds aside for treatment. However, in Ethiopia, it has not been possible to directly access U.S. Agency for International Develop- ment (USAID) funds to support treatment of MDR TB, as the substantial funds allocated are directed to historical USAID partners not doing direct care. Similarly, in Cambodia, the initiation and expansion of MDR TB care in the country done by the Global Health Committee/Cambodian Health Committee has been supported by a private donation from the Annenberg Foundation and not by the USAID grants awarded to that country.

OCR for page 71
88 DRUG-RESISTANT TUBERCULOSIS IN INDIA POTENTIAL INNOVATIONS AND ACTION ITEMS Through the presentations provided in this session and the subsequent discussions, individual workshop speakers and participants noted key inno- vations and action items. They include the following: • More research to obtain more data on children with MDR TB is urgently needed. Such data could assist those attempting to develop better diagnostics and treatment regimes for children. • Expansion of ongoing efforts to address the challenges associated with diagnosis and treatment of TB-HIV coinfected patients is key to controlling the spread of MDR TB. • “South-to-south transfer,” or the sharing of expertise and best practices from one resource-limited setting to another, offers an opportunity to learn from relevant experiences elsewhere in the world. Finally, since the workshop, a research network on pediatric drug- resistant TB, The Sentinel Project on Pediatric Drug-Resistant TB, has been launched by two workshop participants.7 7 Since the workshop, Mercedes Becerra, Assistant Professor, Harvard Medical School, and Soumya Swaminathan, Head, Division of Clinical Research, National Institute for Research in Tuberculosis, collaborated to launch a research network on pediatric drug-resistant TB. As of April 2012, more than 140 individuals from more than 30 countries had come together to collaborate on joint projects through the network, titled The Sentinel Project on Pediatric Drug-Resistant Tuberculosis. More information is available at http://sentinel-project.org/ (accessed April 30, 2012).