Year : 2010 | Volume
: 54 | Issue : 3 | Page : 151--154
Is NDM-1 actually being imported to UK from India?
Shashi Kant1, Partha Haldar2,
1 Professor, Centre for Community Medicine, AIIMS, New Delhi, India
2 Public Health Specialist, New Delhi, India
Professor, Centre for Community Medicine, All India Institute of Medical Sciences (AIIMS), New Delhi
A recent article in Lancet Infectious Diseases suggested that Enterobacteriaceae containing New Delhi metallo-βlactamase (NDM-1) gene were being imported into UK from India. Since the study findings had widespread public health implications, it was necessary to scrutinise the adequacy of the evidence. The article was critically appraised on epidemiological, biological, and molecular evidence, the ethical principle of research, potential conflict of interest, and the justifiability of the recommendations. The study design was inappropriate to establish a causal chain between hospitalization in India and importation of NDM-1 in UK. Out of the total 29 NDM-1 positive UK patients, the NDM-1 gene was present in equal proportion among those who were hospitalized in India (44.3%) and those who were not (51.7%). Statistically significant strain relatedness between Indian and UK isolates could not be proved through dendrogram. There was a potential conflict of interest that was not disclosed. We found that the study findings did not support the authors«SQ» conclusion that India was a source for NDM-1 positive UK patients. Misplaced conclusions had the potential to cause unfounded scare and panic.
|How to cite this article:|
Kant S, Haldar P. Is NDM-1 actually being imported to UK from India?.Indian J Public Health 2010;54:151-154
|How to cite this URL:|
Kant S, Haldar P. Is NDM-1 actually being imported to UK from India?. Indian J Public Health [serial online] 2010 [cited 2019 Jul 20 ];54:151-154
Available from: http://www.ijph.in/text.asp?2010/54/3/151/75738
Reporting of a "Superbug" in the article "Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the UK: a molecular, biological, and epidemiological study"  published in Lancet Infectious Diseases has created a lot of heat and some light in Indian print and electronic media.
New Delhi metallo-β-lactamase (NDM-1) was first described in a non-resident Indian of Sweden by Yong et al. in 2009.  Anguish and hurt national pride led to questioning of the appropriateness of naming the responsible gene as blaNDM-1 (New Delhi metallo-β-lactamase), when it was first discovered. It was suggested, and correctly so, that though one could be certain of the origin of first reported isolate; one could not automatically assume that the resistance strain emerged at the same location as well. 
When the Government of India suggested that researchers did not comply with the existing national laws regarding shipment of Indian biological samples abroad, it was considered too bureaucratic that constituted "Red-tapism". Subtle hints were dropped that Indian medical fraternity was wrong to criticise the findings since "Lancet is the world's second most respectable medical journal in terms of impact." 
The article has a potential to create scare among professional as well as lay public, and likely to impact the antibiotic prescription pattern in India. It has already led to travel advisory by UK Health Protection Agency.  In United States of America, the Centers for Disease Control and Prevention (CDC) has advised clinicians to ascertain history of travel to India or Pakistan in last 6 months among patients where carbapenem-resistant Enterobacteriaceae are identified.  As a consequence one might expect travel advisory in future. Many more countries might emulate such a move. Therefore, although we acknowledge and respect the standing of Lancet among medical researchers, we do feel compelled to share our perspective on this article.
Title of the study
It suggested as if a novel mechanism of emergence of antibiotic resistance was to be described. However, in the concluding paragraph of the "Introduction" section the authors state that they "investigated importation of the resistant gene into the UK patients returning from the Indian subcontinent." On the basis of their findings, the authors felt that since "India also provides cosmetic surgery for other Europeans and Americans, and blaNDM-1 will likely spread worldwide." Thus the true thrust of the article was not reflected in the title. The focus of investigation was on India though Pakistan was also mentioned in the title. Except for a passing reference to travel history and number of samples collected from Pakistan, the authors have not utilized data from Pakistan in their analysis. Thus we feel that the title of the study was misleading.
Barring patients of road traffic accidents and cosmetic surgery, most of the UK patients that were admitted in Indian hospitals were suffering from serious medical conditions where immunity was likely to be compromised. These patients were not assessed for the presence of NDM-1 prior to their departure from UK. Without knowing the status prior to hospitalization in India, there can be no certainty that the NDM-1 was acquired in India. This descriptive research design may be considered adequate for hypothesis generation but would not be sufficient to establish a causal chain between hospitalization in India and importation of NDM-1 in UK.
The authors tried to buttress their argument that India was the source of NDM-1 to UK patients through three broad categories of evidence, i.e. epidemiological, biological, and molecular. Let us examine them individually.
Out of the total 29 UK patients who were NDM-1 positive, 17 had history of travel to India/Pakistan in the last one year. Since the analysis and conclusions were attributed to India, it would have been better had the authors excluded those data sets that belonged to Pakistan. Notwithstanding, of the 17 that had travelled to India/Pakistan, 14 were hospitalized. Therefore, only 14 of the 29 patients (48.3%) had the history of exposure (hospital admission) compared to 15 of 29 (51.7%) who did not have the exposure but were also NDM-1 positive. The outcome of interest (presence of NDM-1 gene) was equally distributed among those that had the exposure (hospitalization in India) and those who did not have. Thus, there was no epidemiological evidence to prove that hospitalization in India was a risk factor for acquisition of this antibiotic resistance.
The mechanism of acquisition of NDM-1 by three UK travellers that had no history of hospitalization is unclear. If wide spread environmental presence of NDM-1 in India is postulated to be the source, then many more UK tourists who travelled to India (more than half a million each year) would be expected to develop the disease. Clearly this was not the case otherwise UK's Antibiotic Resistance Monitoring and Reference Laboratory would have identified them as well.
The risk factor included travel to India or links with India. The authors did not define what was meant by the word "links". Could having cousins, acquaintances, friends or even pen friends qualify as having 'links'?
The UK patients consisted of two distinct categories, i.e. those that travelled and also those, who did not travel to India. However, the analysis was performed by treating them as one single group while comparing with Indian isolates. Since travel to India had been postulated as a risk factor, the UK patients should not have been clubbed together. The current method of analysis did not allow us to ascertain whether or not travel to India was indeed a risk factor.
Dendrograms of strain relatedness did not prove statistically significant strain relatedness between Indian and UK isolates. Therefore, it is hard to support the notion that hospital admission in India was in anyway the source for antibiotic-resistant bacteria observed in UK patients.
Authors' observation, "there were many plasmids of identical size in isolates collected from India and UK" could not be constructed as the support for the causal link. We know that plasmids may lose or gain in size while getting transferred from one to another organism. Finding identical size plasmid could be purely coincidental. Had we known what exactly "many" translated into numerical value, one could hazard a guess if it was well beyond the realm of chance.
Moreover, suppose that the identical plasmid size was found between Indian isolates and those UK patients who had no history of travel/hospitalization in India, would it still be interpreted as evidence that both were linked? In the absence information about which the UK patient group had identical size plasmid, we are unable to interpret the finding.
Citing an Indian article  as a reference, the authors asserted "suggesting that bla NDM-1 is widespread in the environment." However, the cited article had concluded, "none of the food-borne isolates demonstrated the presence of any of the bla genes." There was no mention of NDM-1 at all in the reference quoted.
All Chennai and UK isolates of E. coli and Klebsiella had different pulse-field gel electrophoresis (PFGE) profiles, suggesting that the infections were community acquired. Since these were not hospital-acquired infections, the authors cannot conclude that hospitalization in India was a risk factor for NDM-1.
Apart from two study sites (Chennai and Haryana), nine other hospitals in India provided biological samples. Patients' clinical data might also have been accessed (e.g., to determine whether or not the infection was hospital acquired). The authors have not disclosed whether local institutional ethical clearances were obtained. Did the ethical clearance obligate the researchers to inform the patients from whom biological samples were collected regarding sensitivity of the organism to various antibiotics? Researchers needed to provide a more detailed description of ethical issues involved so as to make it meaningful to the readers.
Potential Conflicts of Interest
Only two drugs were found to be effective against the reported resistance namely, tigecycline and colistin. Colistin is known to be highly toxic antibiotic and therefore virtually unusable. This left tigecycline as the only drug which would be effective in NDM-1 cases. It would have been appropriate for the authors to disclose that the patent  for tigecycline was held by pharmaceutical company Wyeth (Tygacil  ), which was also one of the sponsors of this study.
Other Minor Issues
The legend of the table (number not given) refers Chennai as north India and Haryana as south India, whereas just the opposite is true. In the "Results" section it has been described that of the total of 3521 isolates, 141 were carbapenem resistant. Among the 141 carbapenem-resistant isolates, 44 were NDM-1 positive. This translates into 31.2% NDM-1 positivity rate among all resistant isolates. However, the corresponding figure quoted in the article was 1%. It appears that the authors used total numbers of isolates tested as denominator, which was certainly wrong. Such inattention to details did not speak well about the rigour with which scientific articles ought to be written for a reputed international journal like Lancet.
Overall, none of the evidence detailed under the three broad headings supported the hypothesis that the source of NDM-1 observed among UK patients was from India. Hence, the recommendation of the authors to avoid surgeries in India was not borne out of the study finding.
We agree with the study finding that NDM-1 was present in India, probably in much higher frequency than UK. The absolute numbers, especially when samples are chosen on non-representative manner, could identify but not quantify the magnitude of the problem. One must therefore refrain from over-interpreting the findings.
Population level generalization of study findings should not be done when the study design and method do not permit so. As much as possible, the resistance rates should be expressed in terms of cases within a defined human population in a given time period. The use of laboratory specimens and isolates as denominators (as has been done in this study) produces rates that are of limited epidemiological relevance.
Every one would agree that there is an urgent need to build credible evidence on the level and trends of specific drug resistance by doing robust studies. Making recommendations from a weak study design with insufficient evidence could be counter-productive.
Thus, we believe that the best use of current finding would be to urge the Government of India to give a fresh look into the policy of antibiotic usage. Implementation of large-scale drug resistance monitoring system immediately may not be possible due to non-availability of trained manpower, logistical requirements for quality maintenance of sample transport and storage, and the absence of requisite accreditation of laboratories, etc. Study findings could be used as a advocacy tool for establishment of sentinel sites for monitoring the antibiotic resistance pattern and trend wherever such capacities might be available. However, it would be a self-fulfilling prophecy were the findings to be used to create scare and panic. We in India experienced this sort of panic created in media during the so-called plague outbreak in 1994. Our experience with recent H1N1 influenza is another reminder of inappropriate public health response under high voltage publicity in print/electronic media.
H1N1, the novel influenza virus, was believed to be of severity similar to the grade of 1918 influenza pandemic. Like the current case of NDM-1, without scrutinizing the credibility of the evidences, it was concluded that this was going to be the biggest ever influenza pandemic in the world. In a situation akin to the present NDM-1 situation, the pharmaceutical companies then were ready with their wonder drug. Vaccines against the H1N1 were hastily developed. The partial evidences coupled with incomplete understanding of the epidemic led many countries to hoard large quantities of vaccines and the wonder drug, as a 'critical' step for public health action.  The French government initially ordered 94 million doses of vaccine and later on cancelled the order for 50 million doses. Only five million people could be vaccinated prior to start of 2010. France was left with million of unnecessary doses.  Thus, even though there was a global economic slow down in 2009, the pharmaceutical companies were able to reap estimated profits of between 7 and 10 billion dollars.  We are afraid that the history might repeat itself. Governments might be forced by the public opinion, and not due to scientific evidence, to procure the only effective drug (tigecycline) at huge cost. Moreover, the health care providers in India may infer that in any suspected multi-drug resistant infection, it would be empirically correct to resort to administration of tigecycline. This would be truly tragic and the sole beneficiary would be the pharmaceutical industry. We are confident that none of the study authors would ever wish such a situation to arise.
|1||Kumaraswamy KK, Toleman MA, Walsh TR, Bagaria J, Butt F, Balakrsihnan R, et al. Emergence of a new antibiotic resistance mechanism in India, Pakistan and the UK: A molecular, biological and epidemiological study. Lancet Infect Dis 2010;10:597-602.|
|2||Yong D, Toleman MA, Giske CG, Cho HS, Sundman K, Lee K, et al. Characterization of a new metallo-beta-lactamase gene, blaNDM-1 and a novel erythromycin esterase gene carried on a unique genetic structure in Klebsiella pneumoniae sequence type 14 from India. Antimicrob Agents Chemother 2009;53:5046-54.|
|3||Reddy KS, Kakkar M. Science, sense and sensation. Indian Express (Aug 19, 2010).|
|4||Sharma MS. The brain-eating superbug. Indian Express (Aug 18, 2010).|
|5||Multidrug-resistant Enterobacteriacae (NDM1) in India, Pakistan and the United Kingdom. Available from: http://www.nathnac.org/pro/clinical_updates/ndm1_130810.htm [last accessed on 2010 Aug 25].|
|6||Centers for Disease Control and Prevention (CDC). Detection of Enterobacteriaceae isolates carrying metallo-β-lactamase - Unites States, 2010. MMWR Morb Mortal Wkly Rep 2010;59:750.|
|7||Shahid M, Malik A, Adil M, Jahan N, Malik R. Comparison of beta-lactamase genes in clinical and food bacterial isolates in India. J Infect Dev Ctries 2009;3:593-8.|
|8||Patent number: 20100035845. Available from: http://www.faqs.org/patents/app/20100035845 [last accessed on 2010 Sep 20].|
|9||Available from: http://www.wyeth.com/products_hcp?product=/wyeth_html/home/products/prescription/Tygacil®%20(tigecycline)%20for%20injection/Tygacil;%20(tigecycline)%20for%20injection_overview.html .|
|10||Godlee F. Conflicts of interest and pandemic flu. BMJ 2010;340:c2947. Available from: http://www.bmj.com/cgi/content/full/340/jun03_4/c2947 [last accessed on 2010 Aug 25].|
|11||European Centre for Disease Control and Prevention (ECDC): ECDC daily update on the 2009 influenza A (H1N1) pandemic.|
|12||Flynn P. Social, Health and Family Affairs Committee. Parliamentary Assembly of the Council of Europe; The handling of the H1N1 pandemic: More transparency needed, 2010. Available from: http://www.assembly.coe.int/CommitteeDocs/2010/20100329_MemorandumPandemie_E.pdf [last accessed on 2010 Aug 25].|