4

Preventing Transmission of
Drug-Resistant TB

Key Messages

  • Global estimates indicate that approximately half of new MDR TB cases are occurring in people who have not been treated previously, indicating transmission of the disease (Nardell and Dharmadhikari, 2010; WHO, 2010c).
  • The ongoing development of effective infection control policies offers many opportunities to reduce the transmission of MDR TB in India.
  • Rapid diagnosis of MDR and XDR TB is key to initiating treatment as quickly as possible and isolating patients so they cannot infect others.
  • MDR TB patients can be treated in the community with confidence as long as they are receiving the appropriate therapy.
  • Drug-resistant strains of M.tb. can undergo mutations that make them less fit than drug-susceptible strains, but they also can undergo additional mutations that compensate for this lost fitness.
  • Unsuspected TB and MDR TB patients are the groups most likely to transmit infection to others. If DST could be completed quickly and early, patients could be treated immediately and transmission of disease reduced.
  • Measures to strengthen infection control efforts, including through reduced length of inpatient care and provision of community-based treatment, are key to reducing the transmission of MDR TB.


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 49
4 Preventing Transmission of Drug-Resistant TB Key Messages • lobal estimates indicate that approximately half of new MDR G TB cases are occurring in people who have not been treated previously, indicating transmission of the disease (Nardell and Dharmadhikari, 2010; WHO, 2010c). • he ongoing development of effective infection control policies of- T fers many opportunities to reduce the transmission of MDR TB in India. • apid diagnosis of MDR and XDR TB is key to initiating treatment R as quickly as possible and isolating patients so they cannot infect others. • DR TB patients can be treated in the community with confidence M as long as they are receiving the appropriate therapy. • rug-resistant strains of M.tb. can undergo mutations that make D them less fit than drug-susceptible strains, but they also can un- dergo additional mutations that compensate for this lost fitness. • nsuspected TB and MDR TB patients are the groups most likely U to transmit infection to others. If DST could be completed quickly and early, patients could be treated immediately and transmission of disease reduced. • easures to strengthen infection control efforts, including through M reduced length of inpatient care and provision of community-based treatment, are key to reducing the transmission of MDR TB. 49

OCR for page 49
50 DRUG-RESISTANT TUBERCULOSIS IN INDIA According to WHO estimates, approximately half of new cases of MDR TB appear to occur through transmission, not through inadequate management of preexisting infection (Nardell and Dharmadhikari, 2010; WHO, 2010c). Infection control is therefore a vital consideration in any national program to control drug-resistant TB. Presentations on this topic addressed some of the steps India is taking to prevent transmission of MDR TB, the potential for drug-resistant TB to spread both in health care institutions and in the community, the evolutionary forces that shape fit- ness in M.tb., and the contributions molecular epidemiology can make to understanding the spread of drug-resistant TB. INDIA’S PROGRAM EFFORTS TO PREVENT TRANSMISSION OF DRUG-RESISTANT TB1 India’s program efforts to prevent transmission of drug-resistant TB include the development of infection control guidelines, strengthening of laboratory capacity, and rational use of anti-TB drugs. Infection Control Guidelines TB infection control has historically been problematic in India, said Prahlad Kumar, Director, National Tuberculosis Institute-Bangalore. Guide- lines for airborne infection control have been lacking. Most health care facilities have been overcrowded, and hospital administrations generally have not been committed to infection control. Additional challenges have been the high TB burden, vulnerable populations, and the need for scale-up of response and coordination. At the same time, there are many opportunities to improve infection control in India, including the strengthening of the health system now under way, lessons learned in confronting pandemic influenza, growing awareness of the importance of infection control, and efforts by hospitals to gain accreditation. P. Kumar cited three examples of efforts to improve infection control in India: (1) the National Airborne Infection Control Committee has developed and pilot tested national guidelines for airborne infection control; (2) the RNTCP is upgrading infection control measures at DOTS-Plus indoor facilities; and (3) infection control measures are being implemented in the intermediate reference laboratories. The national guidelines developed by the National Airborne Infec- tion Control Committee provide up-to-date information on recommended methods for reducing the risk of airborne infections in health care facilities. 1 This section is based on the presentation of Prahlad Kumar, Director, National Tuberculosis Institute-Bangalore.

OCR for page 49
51 PREVENTING TRANSMISSION OF DRUG-RESISTANT TB The target audiences are health facility administrators and infection control focal points. The guidelines, which were adapted from the WHO infection control policy of 2009, cover managerial activities; administrative, envi- ronmental, and personal protective measures; special settings; household settings; airborne infection control risk assessment; and quarterly reporting systems. One objective of the pilot testing is to conduct systematic baseline assessments of infection control and administrative, environmental, and personal protective measures and practices at 35 selected health care facili- ties in three Indian states. Another objective is to offer state and district officials and administrators focal points for capacity building, specific rec- ommendations, and supportive supervision on a limited basis. The transla- tion of the pilot to practice will involve follow-up assessments, revision of national guidelines based on the feasibility and effectiveness of the measures implemented, integration of infection control into hospital accreditation and routine health system reporting, and integrated infection control train- ing materials for front-line health care workers. Strengthening of Laboratory Capacity Four national reference laboratories—the National Institute for Research in Tuberculosis, Chennai; the National Tuberculosis Institute in Bangalore; the LRS Institute in New Delhi; and the Central Jalma Insti- tute for Leprosy and Other Mycobacterial Diseases in Agra—are working closely with intermediate reference laboratories, medical college labora- tories, and private laboratories. The stakeholders and partners meet on a quarterly basis to share their experiences. Newer diagnostic tools also are being evaluated, with scale-up planned for those found to be effective. Rational Use of Anti-TB Drugs The rational use of anti-TB drugs has been a problem in India. A substantial quantity of first-line anti-TB drugs2 and almost 100 percent of second-line anti-TB drugs have been sold and used outside the RNTCP. Management of TB patients outside the RNTCP often is poor, leading to the risk of treatment failure and the development of drug resistance.3 The large and unregulated private sector has a conflict of interest with respect 2 According to 2006 data, the private sector accounted for the purchase of 74 percent ($69.7 million) of India’s total drug market for first-line anti-TB drugs; the remaining one-quarter of the drug market was accounted for by public-sector purchasing (TB Alliance, 2007). 3 A survey of 106 private practitioners in Mumbai found that only 6 of 106 practitioners wrote a correct prescription for the treatment of drug-susceptible TB, and only 3 of the 106 wrote an appropriate treatment regimen for MDR TB (Udwadia et al., 2010).

OCR for page 49
52 DRUG-RESISTANT TUBERCULOSIS IN INDIA to better TB management because it profits from the easy availability of anti-TB drugs. The Chennai Consensus Statement, developed by a broad range of stakeholders, called for all people involved in TB care to implement inter- national standards of care. In addition, the Indian Medical Association, on behalf of the RNTCP, collaborated with the Medical Council of India to draft guidelines for the rational use of anti-TB drugs. Discussions were conducted to establish additional prequalified drug manufacturers and the implementation of WHO standards. THE IMPACT OF TREATMENT ON MDR TB TRANSMISSION4 As noted above, more than half of new MDR TB cases are occurring in people who have not been treated previously. Many new cases therefore are resulting from transmission, not from poorly managed treatment. For that reason, treating MDR TB cases in hospitals can be a risk factor for transmission, and minimizing the time patients are in hospitals is desirable, said Edward Nardell, Harvard Medical School. Community-based MDR TB treatment programs, which now exist in many parts of the world (e.g., Cambodia, Ethiopia, Haiti, Lesotho, Paki- stan, Peru [Shin et al., 2004], Russia, South Africa [Heller et al., 2010], and Swaziland), may also appear to raise concern about transmission. But Nardell said that if community-based treatment is implemented well, trans- mission probably is not a major concern. For many years, 14 days was viewed as the necessary period of isola- tion for patients with drug-susceptible TB once treatment had been initi- ated. This time period was not based on patients having negative cultures, which can take much longer than 14 days to achieve. Rather, it appeared to be based on several studies dating back to 1960 regarding the effects of chemotherapy on transmission.5 Rouillon and colleagues (1976) concluded that smear positivity and culture positivity correlate with transmission before but not after therapy has begun. They went on to write, “There is an ever-increasing amount of evidence in support of the idea that abolition 4 This section is based on the presentation of Edward Nardell, Associate Professor of Medicine, Harvard Medical School. 5 For example, Andrews and colleagues (1960) showed that household conversions did not occur after the start of treatment, so people could be sent home rather than being kept in the hospital. Brooks and colleagues (1973) found 107 tuberculin skin test (TST)-negative subjects living with 19 patients with active TB, yet there were no TST conversions in the contacts after the beginning of treatment. A later review of these early studies found fault with many of them (Menzies, 1997). In some cases, for example, less infectious patients may have been selected for home treatment, or uninfected household contacts may have been less susceptible. Menzies concluded that smear-positive patients should still be considered infectious after 2 weeks.

OCR for page 49
53 PREVENTING TRANSMISSION OF DRUG-RESISTANT TB of the patient’s infectiousness—a different matter from ‘cure,’ which may take months, and from negative results of bacteriological examinations, direct and culture, which may take weeks—is very probably obtained after less than two weeks of treatment.” The 14-day period appears to be based on this assumption—that 2 weeks of treatment is sufficient to abolish infec- tiousness—said Nardell. This assumption led to the revision of guidelines for when patients could be discharged from the hospital, when they could go back to work, and when they were no longer a public health threat. In the classic experiments of Richard Riley in the 1950s, patients with strongly positive smears and cavitary TB in hospital rooms in Baltimore were connected by way of the ventilation system to hundreds of sentinel guinea pigs that occupied cages in the penthouse above the clinical ward (Riley et al., 1959). Only 3 of the 77 patients produced 35 of 48 (73 per- cent) of the guinea pig infections that were observed. The guinea pig infec- tions tended to occur when patients with drug-resistant TB on inadequate therapy were admitted to the ward. When drug-susceptible patients were admitted to the ward and started on treatment the same day, no guinea pig infections occurred. Compared with some patients who were left untreated for a time to gauge their infectiousness, treatment started the same day as admission was shown in other patients to be dramatically effective in preventing transmission to the guinea pigs. Drug-resistant patients were somewhat less infectious, suggesting the existence of a fitness deficit in their strains of M.tb. Riley concluded from these experiments that decreases in infectiousness resulting from treatment preceded the elimination of the organisms from the sputum (Riley et al., 1962). Later it was speculated that when people cough, large particles evaporate as they settle (Loudon et al., 1969). The drug in these secretions then is concentrated and inactivates the infectious agent. With sputum cultures, in contrast, there is no evaporation, and growth support is optimized. A recent similar study from Peru found that almost all the transmission that occurred on a TB ward was due to 9 unsuspected and inadequately treated MDR TB patients among 97 HIV-positive pulmonary TB patients, to whom 292 guinea pigs were exposed over 505 days (Escombe et al., 2008). Only 3 guinea pigs were infected by drug-susceptible patients, but all of those patients had undergone intermittent or delayed treatment. In Nardell’s similar experiments in South Africa, 360 guinea pigs were exposed over the course of 4 months to 109 smear-positive, cavitary, and coughing TB patients recently initiated on therapy. All of the infections in the guinea pigs were due to XDR TB patients who were not being treated for XDR TB. When none of the patients had XDR TB, virtually no trans- mission occurred. Nardell concluded that in that experiment, standard South African treatment for MDR TB starting on the day patients entered

OCR for page 49
54 DRUG-RESISTANT TUBERCULOSIS IN INDIA the hospital, not 2 weeks before, completely suppressed transmission in the 27 patients studied. With respect to household contacts of MDR TB patients, at least half of those who contract the disease are infected with a strain that does not match that of the index case, suggesting that there are opportunities for transmission outside the home. In addition, only about one-third of pul- monary TB patients infect even close contacts, although household contacts would be the highest priority for active case finding. From these observations, Nardell concluded that unsuspected TB patients, whether drug-susceptible or drug-resistant, are the group most likely to be transmitting infection to others. Thus, if India performed DST only on people who had failed treatment, those patients could be in the hospital for 3–4 months transmitting TB before DST results showed that they were drug-resistant. Similarly, in MDR TB wards, unsuspected XDR TB patients would be the group transmitting the disease. Thus, rapid diagnosis of TB, MDR TB, and XDR TB is essential for patients to be treated so they cannot infect others. Those who are drug- susceptible or have MDR TB can be treated in the community with confi- dence as long as they are receiving the appropriate therapy. But if DST is not performed early and quickly, people may transmit for months before they are identified as drug-resistant. Nardell highlighted the need to find TB cases actively based on cough surveillance, perform rapid specific diagnosis using GeneXpert, separate cases safely while they are awaiting diagnosis and treatment, rule out XDR TB, and treat effectively based on DST and using a secure supply of quality-assured drugs. Prompt diagnosis of drug resistance and prompt therapy are critical, whether in the hospital or in the community. It is unclear whether the drugs available for XDR TB, which by defini- tion excludes the fluoroquinolones, are effective in halting transmission. It may be that XDR TB requires prolonged isolation while the patient is still coughing. Nardell noted that XDR TB patients are a far smaller group of patients to isolate than all TB and MDR TB patients, who can be treated safely in the community. THE GENETIC EVOLUTION OF M.tb.6 Evolutionary thinking is critical in considering the future of MDR and XDR TB, said Sébastien Gagneux, Unit Head and Assistant Professor, Swiss Tropical and Public Health Institute and University of Basel. The future of MDR and XDR TB depends largely on the fitness of the organisms that 6 This section is based on the presentation of Sébastien Gagneux, Unit Head and Assistant Professor, Swiss Tropical and Public Health Institute and University of Basel.

OCR for page 49
55 PREVENTING TRANSMISSION OF DRUG-RESISTANT TB cause the disease, which is determined by their virulence and transmissibil- ity. Biologists have long thought that drug-resistant strains of M.tb. are less transmissible and virulent than drug-susceptible strains, a conclusion dating back to studies done in the 1950s that measured the virulence of isoniazid- resistant strains in guinea pigs. Because at least some of the drug-resistant strains were less virulent, a large proportion of the TB community thought at the time that drug resistance was not a major public health problem. Fitness can be measured in many different ways, such as through viru- lence in guinea pigs or molecular epidemiological studies that measure the generation of secondary cases. Depending on the strain being studied and the definition of fitness used, the measured fitness of drug-resistant strains of M.tb. compared with that of drug-susceptible strains varies from 10 times more to 10 times less fit (Borrell and Gagneux, 2009). In other words, drug-resistant TB strains are not going away, Gagneux said. Experiments have shown that the development of resistance in most cases does cause a drop in fitness. Today, drug-resistant strains account for only 3 percent of 10 million new cases of TB each year, even though TB has been treated for at least 50 years with drugs. However, the less fit bac- teria can undergo additional mutations over time that compensate for the initial fitness defect. Also, not all mutations that cause drug resistance have an effect on fitness. Some strains can have a genetic background in which some mutations can confer drug resistance without reducing fitness. For example, the Beijing strain, which has been associated with drug resistance in many places, may be preadapted to deal with the potential fitness cost of resistance, Gagneux said, although this idea remains merely a reasonable hypothesis today. The interactions between mutations and genetic background are termed epistatic interactions, and they have been studied very little in M.tb. In E. coli, however, experiments using streptomycin, quinolones, and rifam- picin have shown that double mutants can have better fitness than single mutants (Trindade et al., 2009). This so-called positive epistasis suggests that one mutation can drive fitness down, but a second mutation can at least partly compensate for the defect while simultaneously conferring resistance to a second drug. In a study of laboratory and clinical strains of M.tb., Gagneux and colleagues (2006) found that mutations in the laboratory strains all had a statistically significant fitness defect, although the effect varied by strain and mutation. When drug-resistant strains from TB patients were compared with drug-susceptible counterparts from the same patients, however, at least some of the drug-resistant strains showed no fitness loss. The authors hypothesized that the clinical strains had undergone additional mutations over time that compensated for the initial fitness cost. Gagneux’s labora- tory is now testing this hypothesis using a combination of experimental

OCR for page 49
56 DRUG-RESISTANT TUBERCULOSIS IN INDIA evolution and whole genome sequencing. In addition, epistatis as has been observed in E. coli needs to undergo much further study in M.tb., said Gagneux. The phenomenon is not straightforward, but it is important to understanding the future of drug resistance around the world. THE MOLECULAR EPIDEMIOLOGY OF M.tb.7 According to S. Siva Kumar, Technical Research Assistant, National Institute for Research in Tuberculosis, molecular epidemiological studies are essential to efforts to curtail the epidemic of TB and MDR TB. Molecular epidemiology combines molecular biology, clinical medicine, statistics, and epidemiology. It focuses on the role of genetic and environmental risk fac- tors at the molecular, cellular, and biochemical levels in the etiology and distribution of a disease or pathogen. In the case of M.tb., molecular epidemiology uses a multidisciplinary approach to analyze the dynamics of transmission, mainly by contact trac- ing. It also is useful in dividing recurrent TB into two types. The first is endogenous reactivation, in which latent TB reinfects an individual and leads to disease after a previous episode of TB has been clinically cured. The second is exogenous reinfection, in which a person becomes reinfected with a new strain of M.tb. In addition, molecular epidemiology studies are important in detecting laboratory cross-contamination, identifying hyper- virulent strains circulating in a population, investigating the evolution of M.tb., evaluating TB-control programs, and monitoring the transmission of drug resistance. Three methods are widely used for molecular epidemiology focused on TB. IS6110 restriction fragment length polymorphism typing detects the variable occurrence of a 1,335 base pair genomic repeat in different strains of the bacterium. Spoligotyping measures the occurrence of a 36 base pair repeat and spacers between those repeats in a particular region of the bacterium’s genome. Finally, MIRU-VNTR typing examines the variable number of tandem repeats (VNTR) within mycobacterial inter- spersed repetitive units (MIRU). Other methods used for typing M.tb. are polymorphic GC-rich sequence analysis, genomic deletion analysis, use of strain-specific markers for rapid diagnosis of the strain prevalent in an area, and single-nucleotide polymorphism analysis. Using IS6110 RFLP analysis and spoligotyping, Shanmugam and col- leagues (2011) investigated the distribution of different genotypes of M.tb. in the Tiruvallur region of South India and their association with drug resistance. They divided 1,650 samples into two broad categories: the East 7This section is based on the presentation of S. Siva Kumar, Technical Research Assistant, National Institute for Research in Tuberculosis.

OCR for page 49
57 PREVENTING TRANSMISSION OF DRUG-RESISTANT TB African-Indian (EAI) group and the non-EAI group. The EIA group can be divided into six subgroups, with EAI 3 and EAI 5 being common in this area of South India. Other strains found in South India are Beijing, Central Asian (CAS), and Haarlem. In South India, the EAI group predominates, whereas in North India, the CAS and Beijing strains predominate. Among the MDR TB strains, the Beijing and CAS strains are most abundant at 12.8 and 11.4 percent, respectively. The EAI strains in South India are less prone to drug resistance than the other strains. When spoligotypes were compared with treatment regimens, no dif- ference was found in the strain distribution among patients undergoing category I, II, and III treatment. In the category II cases, however, a higher percentage of drug resistance was seen. In a second study, Narayanan and colleagues (2010) looked at the type of TB recurrence among HIV-infected and HIV-uninfected patients. Using IS6110, spoligotyping, and MIRU-VNTR, they found that exogenous reinfection was 88 percent and endogenous reactivation was 12 percent in HIV-infected patients. In contrast, in HIV-uninfected patients, endogenous reactivation was 91 percent, compared with an exogenous reinfection rate of 9 percent. Understanding these rates of recurrence and reinfection is essential to preventing the spread of MDR TB, said Siva Kumar. 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: • Evolutionary thinking is critical in considering the future of MDR and XDR TB. The future of MDR and XDR TB depends largely on the fitness of the organisms that cause the disease, which is determined by their virulence and transmissibility. • There is a need to find TB cases actively based on cough surveil- lance, perform rapid specific diagnosis using GeneXpert, separate cases, rule out XDR TB, and treat effectively based on DST and using a secure supply of quality-assured drugs.

OCR for page 49