Article Text

Download PDFPDF

Molecular epidemiology unmasks the tubercle bacillus: new techniques reveal new aspects of virulence
  1. P D O Davies
  1. Correspondence to:
    P D O Davies
    Cardiothoracic Centre, Thomas Drive, Liverpool L14 3PE;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Molecular methods show that TB is being transmitted with surprising efficiency

Two papers in this month’s edition of Thorax show how the use of strain typing of Mycobacterium tuberculosis can reveal new aspects of tuberculosis (TB) transmission.1,2 The study by Ruddy et al,1 based around an outbreak of isoniazid resistant TB in north London, shows that a single outbreak can extend over many years and affect a broad range of individuals including medical staff. The second study by Hernández-Garduño et al2 suggests that sputum smear negative disease can have an appreciable transmission rate. Although molecular epidemiology can unmask the problem, solutions may be more difficult to develop.


Using IS6110 restriction fragment length polymorphism (RFLP) analysis, at the time of writing Ruddy et al had identified over 70 cases, ominously adding that, on epidemiological modelling, the peak had not yet been reached. Initial estimates by the end of December 2003 suggest that the outbreak may already have reached 132 (Dr Helen Maguire, personal communication).

This is the biggest outbreak ever identified in the UK, although not from a single point source, as was the more spectacular Leicester outbreak.3 The outbreak of isoniazid resistant TB in north London reported by Ruddy and colleagues seems to have been largely among groups of young adults of mixed ethnic backgrounds, particularly black Caribbean and white, although some were of business or professional backgrounds. Drug misuse and/or prison detention were common to many cases. Although the association between prison and TB has been well established in countries such as Russia and the USA,4,5 this is the first prison outbreak to be documented in the UK. We have previously been quite proud of our record of keeping TB out of prisons.6 Now that record has been spoilt and may never be restored.

The Home Office has sometimes been reticent in joining the fight against TB but, as has been demonstrated elsewhere, prison is a continuum of the free society outside the prison walls and infections are neither confined within or without.7,8 Greater continuity of health care between prison and community will probably be achieved now that primary care trusts rather than the Home Office are responsible for prison health.

Drug misuse and poor adherence to treatment frequently go hand in hand, and a point made poignantly by Ruddy and colleagues is that almost a quarter of the cases had been unreliable in taking treatment, despite frequent follow up visits by specialised TB nurses. Two important methods of helping patients to comply with treatment arise from the paper but are not specifically referred to—namely, directly observed therapy (DOT) and special hospices or hostels for patients while on treatment.

New York, faced with a situation of similar proportions, invested huge amounts of money in setting up new facilities for TB treatment which included dedicated DOT workers at virtually one per patient and inpatient facilities, even using an island off the shore of New York for compulsory detention.9,10 But the cost of controlling the epidemic was enormous, estimated at several billion dollars or approximately $20 000 per case cured. DOT workers need not be expensive if friends or former patients can be employed. One aspect of health care that we have lost in the last generation of hospital staff is patient centred user friendly TB wards not dissimilar to hospices. The enormous sums we have spent on negative pressure rooms for the treatment of drug resistant disease in our hospitals has made us forget the need to treat patients in a friendlier environment. The current recommendation for treatment of isoniazid resistant TB is for a year.11 No one would wish to stay behind a double door for so long, but a bungalow in a pleasant setting may persuade a reluctant patient to stay for some or all of the duration. They would be receiving free food and, as long as the law was not overtly broken, staff might be persuaded to turn a blind eye to some of the less medically safe habits in which patients might indulge, as long as they were present at the once daily drug round. We once had such a ward in Liverpool until it was closed by the management without discussion while I was on holiday.

In the atmosphere of politically motivated competition between hospitals, one wonders whether one hospital in (say) four in a city such as London could be persuaded to open such a ward. The presence of this type of ward may completely dispense with the need to consider compulsory detention. As the authors point out, there is no clear and rapid system for bringing extra resources to bear on such a situation. It must be hoped that the new government initiative “Getting Ahead of the Curve” can be implemented.12

At present this ongoing outbreak represents a worrying development in the breakdown of UK public health services. Ruddy et al suggest that treatment should be supplied free of charge. If patients could be given the drugs free of charge at the clinic when they attend, the potential for breakdown in drug taking between the clinic and the patient’s home would be eliminated.

Two messages come clearly from this paper. Firstly, a relatively new virulent strain of M tuberculosis is being transmitted across ethnic, cultural, and financial boundaries. Secondly, those who are battling to control the infection have insufficient resources to do so.


The paper by Hernández-Garduño et al2 from Western Canada poses another worrying, if less immediate, problem. Traditionally we have regarded smear negative TB as posing almost no risk of infectivity. Historical studies have suggested that there is very little risk of co-householders of such a case developing disease.13 The tendency over the last decade has been to downgrade contact tracing as cases of TB have been declining in most areas of the country with the notable exception of London, concentrating instead on contacts of smear positive cases only.14–16 Using the same IS6110 insertion sequence to identify the strain type, the authors suggest that one sixth of 791 patients identified with culture positive pulmonary and/or extrapulmonary TB received infection from a sputum smear negative case. Unlike an earlier study by Small et al from San Francisco17 which found that 17% of TB cases were due to smear negative transmission, Hernández-Garduño et al have included cases with extrapulmonary disease. They hypothesise that patients who appear to have extrapulmonary disease alone could be transmitting tubercle bacilli by previously undetected sputum smear negative transmission.

The methods used to ensure that apparent smear negative transmission could not have been caused by smear positive transmission appear rigorous. One theoretical confounding factor which the authors do not seem to have considered is the possibility that a smear negative patient at the time of diagnosis may have been smear positive earlier on in the disease. As the historical data suggest that 25–50% of untreated patients with pulmonary TB healed spontaneously, this remains a possibility. The finding that one sixth of the cases were due to smear negative transmission is remarkably similar to that of the earlier San Francisco study.17 The fact that half of all patients with TB have never, to their knowledge, been in contact with a case of TB (so called “casual transmission”) perhaps adds some weight to this evidence.18

If this is true, what are the implications for TB control? Firstly, it means that it is going to be much harder to eliminate TB in low prevalence settings than we had hoped. Secondly, we may have to revise our contact tracing procedures to include more extensive screening of contacts of smear negative cases, particularly if these may be immunocompromised in any way. Thirdly, the implication for the provision of adequate resources for TB control in low prevalence settings is made clearly in the paper by Ruddy et al.


The use of molecular methods for studying the epidemiology of TB is proving to be a two edged sword.19 Unlike the dilemma of Pooh who found that the more he looked for Piglet in Piglet’s house without finding him the more Piglet wasn’t there,20 the more we look at TB with this methodology the more we find it is there or, at least, is being transmitted with surprising efficiency. The implications for resources to improve TB control are evident. Unless we can convince our political masters that this is the case, we will have to stand by and watch as things get worse.

Molecular methods show that TB is being transmitted with surprising efficiency


Linked Articles

  • airwaves
    Wisia Wedzicha