Letters to the Editor

Indian Pediatrics 2000;37: 1149-1154

Tuberculosis Control Without Protection from BCG

1. Implication of TB Epidemiology in Chingleput for Revised National Tuber-culosis Control Program (RNTCP): The author says that ‘...the case detection was active and passive...with 95% success in case detection. This community had a better TB control program, except for Directly Observed Treatment (DOT), than what was envisaged in the RNTCP...’.

We agree that the case finding was both active and passive in the area. However, the cases diagnosed were referred to the various health facilities implementing the control program for management. There was no additional input in this regard. It has been documented that the case holding under the program has been only about 30 to 40% with the 12 to 18 month regimens in use and 40 to 50% with the introduction of short course regimens. The 1992 review of the program has clearly brought out the lacunae (over emphasis on X-ray, inadequate drug supply, inconvenient timing of the health facilities, etc.) of the program. Thus, a case detection alone to the tune of 95% without an effort to improve the case holding or cure cannot be expected to have an epidemiological impact on the tuberculosis situation in the area.

The RNTCP addresses the issue of case holding and cure both from the provider and patient point of view. It ensures quality microscopy, prompt supply of short course chemotherapy drugs for any patient started on treatment by having patient-wise treatment boxes and a well-motivated treatment provider, identified both by the patient and the health care system together, ensuring that the patient consumes his medications regularly, thus avoiding defaulters and subsequent disasters. The Directly Observed Treatment Short Course Chemotherapy (DOTS) strategy of the RNTCP results in treatment completion rates exceeding 90% and also appears to be cost effective(2). It facilitates a very high bacterio-logic evidence of cure(3) and significant reductions in the frequency of primary drug resistance, acquired drug resistance and relapse(4). The best method, may be as Grzybowski points out, is "...not to try to add minor improvements to the existing program, but to start from "other end", as it were, by organizaing perfect, completely supervised treatment...and then start removing individual, more costly components of such an experi-mental program, while retaining excellent results..."(5).

Summarizing, the main reason for the tuberculosis situation not improving in Chingleput was because of the poor treatment program and active case detection. An ill-implemented program may cause a disaster by increasing the chronic excretor pool. Thus, it is necessary to study the epidemiological impact only after implementing the strategy of DOTS in the area, which has an objective of 86% cure and after achieving this objective, to go for 70% case detection. Part of the RNTCP study area is now being covered for active case finding by survey to measure the disease burden and a survey for the annual risk of infection (ARTI) also. TRC is undertaking this project at the same BCG trial area and we have to wait for results on ARTI and disease burden over the years to answer the queries raised.

2. The ethics of TB control methods and criteria–Improving sputum smear examination: The author suggests replace-ment of the existing Ziehl-Neelsen (ZN) microscopy with fluorescence microscopy (FM) to improve the sensitivity and quality of sputum examination(1).

FM has no added advantage over ZN method in terms of specificity and sensitivity and this has been confirmed after comparison between ZN and FM techniques by various workers(6,7). Even when the results of both methods were compared with that of their corresponding culture results, there was no difference in the detection of positive smears by both the methods (67.3% by FM and 66.1% by ZN method) and their false positivity was 3.3% and 3.1%, respectively. Thus, there is no additional advantage of FM over ZN. Moreover, with the decentralization of sputum microscopy under the RNTCP, the workload of an individual microscopy center is likely to be reduced.

ZN has the advantage that the microscope can be used for other investigations as well. The source of light is not a constraint since in the absence of electricity, it can perform equally well with sunlight.

The disadvantage of FM is that the microscope can be used only for sputum microscopy. Tuberculosis being an integrated program, the investment cost of Rs. 3.5 lakhs for the instrument alone as against Rs. 30,000 to 40,000 for light microscope cannot be justified. In addition, the maintenance cost is much higher for FM with the need for import of the spares. A continuous power supply without much voltage fluctuation is also essential for FM and this is difficult to achieve in most parts of the developing world.

As such the proposition of replacing ZN with FM microscope does not seem to be the need of the hour. If the 3-sputum examination is done for the properly selected chest symptomatics, we will be able to achieve good case finding using ZN microscopy itself.

3. Where do we go from Chingleput? Rethinking TB control–Technical adequacy: The author points out "... if we allow the infected persons to go through the natural history of infection, then some of them will develop adult type bacillary disease and disseminate the microbes in the environment resulting in new infections.... In the original NTP strategy, BCG was expected to reduce the risk of bacillary TB, but it did not do so. The only alternative method to reduce the risk of bacillary TB is to give preventive therapy to those who have been infected.... In other words only detection of early infection and preventive treatment will achieve what was originally expected of BCG..."(1).

The author has based his recommendation on the fact that majority of the cases come from the already infected population as shown from the 15 years report of the Chingleput study. The proposition appears to be very sound theoretically. However, one has to consider the practical feasibility, both from operational and economic grounds, for giving chemo-prophylaxis to all those infected in a high endemic area (where the infected population is around half a billion) with poor economic resources.

Evidences available suggest that chemo-prophylaxis has rather limited scope in any tuberculosis program in developed countries and is of even lower importance in develop- ing countries with a high prevalence of infection(8). Giving chemoprophylaxis to all infected would demand that 40 to 50% of the population receives chemoprophylaxis. The drug regularity even among patients with active disease and symptoms has been only around 30 to 50% in our country. Those with tuberculosis infection alone are asymptomatic and to expect them to consume the drugs for a 6-months period as a prophylactic measure is a tall order. Styblo’s statement "...the two preventive measures, BCG vaccination and chemoprophylaxis have only a limited impact on the overall epidemiological situation, contrary to the various predictions made during the last 2 or 3 decades..."(9) holds good even today.

The one feasible proposition may however be, to give chemoprophylaxis to a selective population, e.g., children below 6 years of age who are contacts of open pulmonary tuberculosis patients and all individuals who are HIV positive. This is a population at a higher risk of contracting the infection and developing severe forms of the disease. Chemoprophylaxis could be useful only under these circumstances. The policy of chemo-prophylaxis for children below 6 years is already recommended in the RNTCP.

To sum up all the points discussed, it is clear that diagnosis by sputum microscopy of all reporting symptomatic cases using ZN method and treating all those diagnosed to cure with adequate chemotherapy using DOTS strategy, remain the back bone of an effective TB control program.

Tuberculosis Research Centre,
Mayor V.R. Ramanathan Road,
Chetput, Chennai 600 031, India.
E-mail: [email protected]

1. John TJ. Tuberculosis control without protec-tion from BCG. Indian Pediatr 2000; 37: 9-18.

2. Chaulk CP, Kayandjian VA. Directly observed therapy for treatment completion of pulmonary tuberculosis. Consensus statement of the public health tuberculosis guidelines panel. JAMA 1998; 279: 943-948.

3. Chaulk CP, Moore-Rice K, Rizzo R, Chaisson RE. Eleven years of community based directly observed therapy for tuberculosis. JAMA 1995; 274: 945-951.

4. Weis SE, Slucum PC, Blis FX, King B, Nunn M, Matney B, et al. The effect of directly observed therapy on the rates of drug resistance and relapse in tuberculosis. N Engl J Med 1994; 330: 1179-1184.

5. Grzybowski S. Drugs are not enough. Failure of short-course chemotherapy in a district in India. Tuberc Lung Dis 1993; 74: 145-146.

6. Toman K. Tuberculosis Case Finding and Chemotherapy: Questions and Answers. World Health Organization, Geneva, 1979; pp-25-27.

7. Holst E, Mitchison DA, Radhakrishna S. Examination of smear for tubercle bacilli by fluorescence microscopy. Indian J Med Res 1959; 47: 495-499.

8. Styblo K. Recent advances in epidemiological research in tuberculosis. Adv Tuberc Res 1980; 20: 1-63.

9. Styblo K. Epidemiology of tuberculosis: Selected papers. Royal Netherlands Tuber-culosis Association, 1991, Volume 24: p 89.

*This document was prepared by Dr. P.R. Narayanan, Director, Dr. T. Santha, Deputy Director (Sr. Gr.), Dr. P. Paul Kumaran, Senior Research Officer, Mrs. Fathima Rehman, Senior Research Officer, Dr. Rajeshwari Ramachandran, Deputy Director and Mrs. Sara Mathew, Senior Technical Officer of the Tuberculosis Research Centre (ICMR), Chennai, India.

I am very grateful to the Tuberculosis Research Centre at Chennai, for raising certain relevant issues regarding tuberculosis control, in response to the editorial on this subject(1). This illustrates the wider impact of Indian Pediatrics beyond the membership of the Indian Academy of Pediatrics. I shall first address the 3 points raised by Dr. Narayanan and then try to put the issues that seems to me to be crucial to the control of TB, in perspective.

Dr. Narayanan gives the unpublished information that in Chingleput only 40-50% success was achieved in case holding after introducing short course treatment. Therefore, directly observed treatment (DOT) has been built into the RNTCP. Even though DOT can be expected to give better results, if case detection is only 40% as was found in the review of National TB Control, the total success cannot exceed 40%. I am very much encouraged to read that in some areas under RNTCP they will have active case finding, in order to measure disease burden. If we must improve the rate of case-finding, the present method of mere passive case finding, confined to the government sector health care institutions, will not suffice. But can we conduct active case detection everywhere? If not, are there not alternatives to improve case-finding? This question needs to be urgently addressed by RNTCP. How can we make primary health care for all population groups user friendly, and make TB diagnosis and treatment available, with no bureaucratic red tape? The goal of case detection must be set at 100% and cure also aimed for 100%. Then we can measure our success against this gold standard. If we aim at 70-86% targets for case detection and cure, figures can always be fudged. We learnt that lessons from the Expanded Program on Immunization.

While I agree that fluorescence microscopy (FM) need not and must not replace Ziehl-Neelsen (ZN) staining and microscopy at all places, I do suggest the introduction of FM in the new RNTCP sputum smear examining centers. Where only a few smears are examined daily, ZN method is quite satisfactory. Recently, we found in one center a backlog of over 70 sputum smears waiting to be examined under the microscope. What the FM method offers is speed and efficiency, not necessarily sensitivity and specificity. So, at the level of even primary health care (not primary health center) sputum microscopy has to be made more efficient and timely reporting must be insisted upon. If a smear is not reported on within the same day or latest within one day, the ethics and efficiency of health care are being compromised. Before writing my suggestion, I had actually personally used both methods and have found the FM to be faster and less tiring. Under routine conditions, the technician error of missing AFB is reduced if fluorescing bacilli are searched for against a dark background. Since FM screening is done at low power, while ZN screening has to be at high power, much time will be saved by the former method. Let me reiterate that FM is appropriate only where the work load is high; in such centers it is well worth investing in fluorescent microscopes, which may be used for several other diagnostic tests including improved malaria screening. This issue needs a detailed discussion within and outside RNTCP. For example, what do the microbiologists of our country think?

The authors state that the "proposition appears to be very sound theoretically". The proposition is to detect infection in children (and adults) and give them preventive therapy. However, they caution that "one has to consider the practical feasibility, both from operational and economic grounds, for giving chemoprophylaxis to all those infected in a high endemic area (where the infected population is around half a billion) with poor economic resources".

If one ponders over the magnitude of TB problem in India and the operational and economic inputs necessary to control TB, we might get so discouraged that it would be very easy to despair and believe that RNTCP itself, while very sound theoretically, is not practical in India. On the other hand, I believe that the program must accept what is theoretically sound. How to incorporate detection of infection, and treatment in appropirate cases, do require considerable discussion and clear definition and guidelines. Not half a billion in one go; that is the public health thinking. But one by one, in each and every clinic where primary care is offered. that is primary health care thinking. That takes TB control into the arena of primary health care. Can we afford to miss the individual for the community? Can we not combine care with control?

Dr. Narayanan states that the policy of chemoprophylaxis for children below 6 years is already recommended in the RNTCP. Excellent; this is a good beginning. The prevalence of infection would be less than 5% at this age and over 15-17% at 14 years. If we agree on the principle of detection of infection and treatment in one age group, modalities and practicalities of expanding the age group can be worked out. The purview of pediatrics is up to 18 years.

If we explore what is taught in our medical colleges about the diagnosis and treatment of TB, then we will realize how RNTCP has failed to build collaboration even with the medical teachers. Students are not usually taught the diagnostic criteria of TB as required by RNTCP. Therefore, generations of doctors are being sent out every year without preparation to participate in RNTCP. Cases of TB diagnosed in many teaching hospitals do not find their way into the district TB records and into the DOT system of treatment and follow up. Thus, we must now build colla-boration with all medical colleges and also with the private health care sector, the microbiologists and laboratorians, the pediatricians and other relevant partners. These steps are essential for the success of TB control.

The second point I wish to enunciate is that good primary health care must be the backbone of TB control or any other disease control program. It is a gross error to think that just enough intervention for the sake of controlling the speed of spread of TB is what we can afford and what we can achieve. If we set our goal so low, we will be doing a dis-service to society. It also raises questions of ethics of distributive justice. TB diagnosis and correct treatment should be made accessible to each and every sick person, irrespective of in what institution the initial diagnosis is being made.

The third point is that good primary health care requires appropriate diagnostic tests, and tuberculin testing is one such appropriate step. When a young member of my family was detected with strong positive tuberculin reaction, preventive therapy was instituted immediately. Access to such good health care is a fundamental human right. A child or adolescent with positive tuberculin reaction (to be defined on evidence and by consensus) is a trigger for some epidemiological detective work. From whom did the child get infected? Who should investigate this? If RNTCP staff are not available for such basic epidemiologic search in the community, then the existing health care machinery must swing into action. But that requires integration between health care and public health and also between TB control and the District level public health. Unfortunately, there are unseen but rigid lines that the system has drawn, making TB the turf of the TB control department. This barrier must be broken down for the success of TB control.

T. Jacob John,
Emeritus Medical Scientist (ICMR),
2/91 E2, Kamalakshipuram,
Vellore 632 002, Tamil Nadu, India.

1. John TJ. Tuberculosis control without protec-tion from BCG. Indian Pediatr 2000; 37: 9-18.