This Article Abstract Full Text (PDF) Submit a response View responses Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Chalmers, I. Search for Related Content PubMed PubMed Citation Articles by Chalmers, I. Related Collections Related AAP Red Book topics: Measles Social Bookmarking                         What's this?

PEDIATRICS Vol. 109 No. 2 February 2002, pp. 312-315

COMMENTARY

Why We Need to Know Whether Prophylactic Antibiotics Can Reduce Measles-Related Morbidity

ABSTRACT

In some parts of the world, measles vaccination has resulted in very substantial reductions in measles-related morbidity and mortality. Even so, it has been estimated that 30 million people still contract the disease every year, that nearly 1 million of these die, and that measles-related deaths account for about 10% of all deaths in children under the age of 5 in developing countries. Existing evidence from controlled trials suggests that antibiotic prophylaxis in measles can result in important reductions in measles-related morbidity, and, at a World Health Organization meeting convened in 1993 to decide on research priorities for the treatment of measles, highest priority was accorded to additional controlled trials of prophylactic antibiotics. As controlled trials of vitamin A in measles have made clear, such trials are feasible. Continued acquiescence in uncertainties about the effects of prophylactic antibiotics in a disease that continues to afflict so many children worldwide is unacceptable.

Key Words: measles • prophylaxis • antibiotics

Abbreviations: WHO, World Health Organization

MEASLES: STILL A DEVASTATING DISEASE

According to experts attending a recent meeting of the Pan American Health Organization’s Technical Advisory Group on Vaccine Preventable Diseases, the Western hemisphere is very close to eradicating measles.¹ Despite this encouraging prediction, measles continues to pose a severe threat to child health in other parts of the world. Worldwide, it has been estimated that there are nearly 1 million measles-related deaths every year.² Even without devastating epidemics like that in Afghanistan,³ measles-related deaths account for about 10% of all deaths in children under the age of 5 in developing countries.²

Of the 30 million people who contract the disease every year, a proportion suffer important and sometimes long-term complications. Pneumonia, croup, and diarrhea are the most common life-threatening complications. Other common complications include otitis media, severe stomatitis, malnutrition, blindness, and neurologic problems.⁴

Although effective vaccines against measles have been available for many years, global coverage appears to be declining² and serious epidemics can occur even in countries with high uptake of vaccination. Over 120 people died in a measles outbreak of about 55 000 cases in the United States between 1989 and 1991.² In the last 7 months of 1999 in the Netherlands, there were 1750 cases of measles (albeit, many among families that object to immunization on religious grounds). Forty of the children affected were admitted to the hospital with serious complications, and 3 of them died.⁵

THE IMPACT OF A CLINICAL IMPRESSION

It was while working for a couple of years in a Palestinian refugee camp in the Gaza Strip 30 years ago that I first became aware of just how devastating a disease measles can be. We had an immunization program, supervised by World Health Organization (WHO) staff, but measles was nevertheless common among the refugee children, many of whom were malnourished and in poor health in other ways, and complications and deaths were common.

It had been drummed into me at medical school in the early 1960s that antibiotics should never be prescribed for someone with a viral infection unless there was unambiguous evidence of bacterial superinfection. Accordingly, when a child was brought to me with early measles and I had convinced myself that there was no evidence of superinfection, I conserved our limited supply of antibiotics, prescribed symptomatic treatment, and asked the parents to return if the child deteriorated. The United Nations clinic in which I worked was only open 6 days a week, however, and military curfews at night and sometimes during the day often made it difficult for parents to go to the local hospital if their child’s condition worsened. Distressingly often, my child patients had died a few days after I had seen them.

My Palestinian medical colleague was seeing a very similar spectrum of patients with measles, but he seemed not to have a comparable experience. Toward the end of my first year working in the refugee camp, it was gently pointed out to me that this might be because he gave prophylactic antibiotics to children with measles, because, in his experience, rapid bacterial superinfection was very common in these vulnerable children. Having been convinced to change my practice, and doing exactly what I had been advised at medical school never to do, I had the impression that my child patients were less likely to die.

This clinical impression was very sobering. It made me wonder whether what I had been taught at medical school might have been lethally wrong, at least in the circumstances in which I was working, and precipitated a now incurable "scepticemia"⁶ about authoritarian therapeutic prescriptions and proscriptions unsupported by trustworthy empirical evidence.

A quarter of a century later, when I was called to give evidence to a British parliamentary inquiry into medical research, my comments to the committee reflected a state of mind that had originated in Gaza. I told the committee members that I thought that children had died unnecessarily while under my care because I had practiced what I had been taught at medical school. And I gave examples of the way that advice in some modern medical textbooks was lethal, specifically the failure of many of them, including the Oxford Textbook of Medicine, to acknowledge the strong evidence that thrombolytic drugs given during myocardial infarction reduce mortality.⁷ Rather sensational versions of my statements to the committee appeared the following weekend on the front page of The Sunday Times.⁸

EVIDENCE ON THE EFFECTS OF PROPHYLACTIC ANTIBIOTICS FOR MEASLES

Later that year, the most recent report of a study assessing the effects of prophylactic antibiotics in measles⁹ and a WHO report on clinical research on the treatment of measles⁴ were published. The study compared case fatality rates in a rural area of Senegal between 1983–1986 and 1987–1991. Immunization and antibiotics for prophylaxis and treatment had been introduced during the latter period. Age-adjusted case fatality rates during the intervention phase were less than half what they had been during the earlier period (mortality ratio: 0.41; 95% confidence interval: 0.21–0.81). In view of the study’s use of historical controls and co-interventions, the authors’ interpretation of this observation was that "prophylactic use of antibiotics in young children might (my emphasis) reduce respiratory complications in the latter part of the acute phase at a time when these are likely to be the result of bacterial superinfections." They concluded that "because there is a need for improved management of the many cases of measles infection seen at the primary care level in developing countries, it seems warranted to conduct randomized trials to evaluate the use of prophylactic antibiotics." This view was subsequently endorsed at a meeting called by the WHO in 1993, which accorded a controlled trial of prophylactic antibiotics highest priority among 4 controlled trials proposed in the report.⁹

In 1997, 2 years after the report of the WHO meeting had been published, a systematic review of the controlled trials of prophylactic antibiotics in children with measles¹⁰ was published in the British Medical Journal. Editors at the journal had been prompted by the article in The Sunday Times 2 years earlier to commission a pediatric intensive care specialist who had experience of working in developing countries to review the evidence from controlled trials. His report¹⁰ was accompanied by 2 commentaries.^11,12

The systematic review revealed the poverty of relevant evidence from controlled trials on which to base any firm conclusions about the effects of prophylactic antibiotics on mortality in measles. The author had considered 6 trials of variable quality, reported between 1939 and 1967. Out of a total of 1401 children involved in these studies, only 5 had died, so no meaningful estimate of the effects of the policy on mortality was possible, although the author of the systematic review noted that "the possibility of a reduced mortality from routine prophylaxis cannot be excluded." He also noted that "the studies suggest that routine antibiotic treatment might reduce the risk of developing pneumonia or sepsis (P = .0004)," but decided that this conclusion was unreliable after posthoc exclusion of 2 studies with control group complication rates which he felt were unusually high.

Unfortunately, none of the 6 reports in the review (Table 1) provide the detail that one would wish to have to assess whether biases are likely to have been controlled. In only 3^13–15 was it likely that allocation bias had been controlled in generating the comparison groups, and only 1 report mentioned measures to control biases in assessing outcomes.¹⁵ The clearest of the 6 reports was actually the first to be published (in 1939), and was a summary of a report to the Therapeutic Trials Committee of the British Medical Research Council.¹³ This trial found that prophylaxis with antibiotics was associated with statistically significant reductions in the incidence of serious complications of measles and the duration of bronchopneumonia. However, assessment of some of the outcomes may have been influenced by knowledge of the group assignment of the children being assessed.

View this table: [in this window] [in a new window]   TABLE 1. Assessment of the Likelihood That Allocation Bias Was Avoided in Six Studies Included in Reviews of Prophylactic Antibiotics in Measles (Shann 1997; Shann et al 2000).

 
In 1999, a version of the British Medical Journal review was published electronically in the Cochrane Database of Systematic Reviews. In contrast to the report of the WHO meeting and the articles in the British Medical Journal, the authors concluded that additional controlled trials should have a low priority.¹⁶ This judgment appeared to reflect a failure to distinguish ‘no evidence of an effect’ from ‘evidence of no effect,’ and an unwillingness to acknowledge that the size of any future controlled trials should take into account serious morbidity in children who survive measles, as well as mortality. I used the electronic comments and criticisms facility in the Cochrane Library to make these points to the authors, and to challenge their conclusion that additional controlled trials should not be seen as a priority.

The authors’ responses (which were accompanied by personal disparagement and reference to The Sunday Times’ account of my views) were defensive. They noted that "over the last 20 years, WHO has put a large amount of effort into developing evidence-based guidelines" and these should not be "undermined on the basis of anecdotal evidence."^17,18 I noted, in response, that the WHO recommendations to which the reviewers had referred had not mentioned the issue of prophylactic antibiotics in measles, but that even if they had done so, that would not have been a reason for unquestioning deference to them: the trustworthiness of the WHO recommendations depends on the scientific quality of the process on which they are based.¹⁹ Sadly, the then editor-in-chief refused to publish my perceived challenge to authority!

WHAT SHOULD BE DONE?

Immunization and vitamin A supplementation can be expected to reduce the continuing toll of death and acute and chronic illness caused by measles.¹¹ However, the available evidence is compatible with prophylactic antibiotics halving serious morbidity, and reductions in mortality of the same order are not implausible. Against a background of 30 million cases of measles every year,² the WHO report identifying additional randomized trials as a priority remains as relevant today as it was when it was published in 1995.⁴ As far as I can ascertain, only 1 such trial has been conducted. A placebo-controlled trial of co-trimoxazole in children with measles was conducted between 1998 and 2000 by Aaby et al²⁰ in Guinea-Bissau. Unfortunately, the study was interrupted by the war in Bissau in 1989–1999. Although there were only 6 cases of pneumonia, 3 hospitalizations, and 2 deaths among 199 children recruited (Peter Aaby, personal communication, August 2001), these few additional data should be taken into account in updating the Cochrane review when the code has been broken and the study has been reported in full.

It has been more than 60 years since the first report of a clinical trial with concurrent controls addressing this question was published, and 35 years since the most recent was reported. By now, we should know more than we do. Although controlled trials may well present organizational challenges,⁹ the Guinea-Bissau study shows that these can sometimes be overcome. As the many recent controlled trials of vitamin A supplementation have made clear, such trials are entirely feasible if a will exists to organize them. Indeed, unanswered questions about the effects of prophylactic antibiotics in measles might have been settled long ago had some of the vitamin A trials used factorial designs to address these questions.

Concerns have been expressed that use of prophylactic antibiotics in measles will promote the development of bacteria resistant to antibiotics. Such as it is, the evidence that exists^21,22 does not suggest that use of antibiotics in the community (as distinct from in hospital) contributes significantly to the development of antibiotic resistance. Opposition to additional controlled trials of the use of antibiotics for this serious disease on these grounds would thus seem difficult to defend, particularly given the widespread use of antibiotics for far less defensible reasons. Indeed, additional controlled trials to assess the effects of prophylactic antibiotics in measles could be used to obtain estimates of possible effects on the development of resistant bacteria.

In today’s world of vitamin A supplements and better supportive care, the results of such trials may allow confident exclusion of important beneficial effects of giving prophylactic antibiotics to children with measles. If, however, important beneficial effects are identified, then the relative merits of this policy compared with other effective policies will need to be considered. This will involve taking into account factors such as the likelihood of measles-related complications, costs, the risks of promoting antibiotic resistance, and feasibility, including access to antibiotics²³ and other interventions relevant to reducing measles-related morbidity and mortality.

Reviewers of earlier drafts of this commentary have helped me to understand that I need to make quite clear what I am proposing here. I am not proposing the widespread promotion of prophylactic antibiotics in every child with measles on the basis of my clinical impressions in a refugee population quarter of a century ago and suggestive evidence from controlled trials conducted between 1939 and 1967. What I am proposing, like the 1995 WHO report on clinical research in measles,⁴ is that it is high time that research was conducted to obtain better estimates of the effects of prophylactic antibiotics in children with measles, taking into account their likely susceptibility to bacterial superinfection.

WHAT WOULD I DO TODAY, IN PRACTICE?

What I would do today if, despite efforts to protect vulnerable children from developing measles through vaccination, I had responsibility for caring for disadvantaged children who had developed the disease, and where the rate of measles-related morbidity attributable to bacterial superinfection was unacceptable? Such as it is, the evidence from controlled trials does nothing to relieve my sense of regret at having initially withheld prophylactic antibiotics from refugee child patients with measles 30 years ago, particularly as evidence about the effects of vitamin A²³ was not then available. On the basis of the existing evidence from controlled trials that prophylactic antibiotics are likely to reduce serious morbidity,¹⁶ I would prescribe them today for vulnerable children with measles. In addition, I would base my use of vitamin A on the evidence of its effects that are now available.²⁴

Would I encourage the participation of children with measles in additional controlled trials of prophylactic antibiotics? The 1995 WHO report suggested excluding children with severe protein-energy malnutrition from the trial proposed, presumably because it was felt that administration of prophylactic antibiotics should be the standard of care for such children. I too would find it difficult to withhold antibiotics from these and other children whose health was already compromised by malnutrition. However, there may be more important risk factors for mortality in measles, such as the intensity of exposure to the virus in overcrowded living conditions.²⁵

Because there is clearly no professional consensus on this important matter, however, substantive uncertainties about the merits and demerits of prophylactic antibiotics in measles should be addressed in controlled trials. Whereas I would feel unable to support the inclusion in future controlled trials of children who were malnourished or who had experienced intense exposure to the disease, others might feel that it would be ethically acceptable to include such children in such trials. Whatever the threshold deemed acceptable for giving or withholding prophylactic antibiotics in measles, continued acquiescence in unnecessary ignorance about their effects is unacceptable.

To reiterate: it has been estimated that 30 million people contract measles every year,⁴ that nearly 1 million of these die, and that measles-related deaths account for about 10% of all deaths in children under the age of 5 in developing countries.² As has been pointed out eloquently elsewhere, "the wrong research costs lives."²⁶ We owe it to the millions of anonymous children who suffer from complications of measles to try harder to reduce this substantial burden of morbidity.

Iain Chalmers, MD

UK Cochrane Centre
NHS Research and Development Programme
Summertown Pavilion
Middle Way
Oxford OX2 7LG, United Kingdom

-->

ACKNOWLEDGMENTS

I thank Frank Shann for sending me copies of the reports covered by his review; and to him, Peter Aaby, Jan Chalmers, Mike Clarke, Anthony Hamden, Greg Hussey, Tom Jefferson, Philippa Middleton, Patricia Priest, and 2 anonymous reviewers for comments on earlier drafts of this article.

FOOTNOTES

Received for publication May 10, 2000; Accepted Sep 6, 2001.

Reprint requests to (I.C.) UK Cochrane Centre, NHS Research and Development Programme, Summertown Pavilion, Middle Way, Oxford OX2 7LG, United Kingdom. E-mail: ichalmers{at}cochrane.co.uk

The views expressed in this commentary are strictly those of the author.

REFERENCES

1. McCarthy M. Measles eradication may be near in the Americas. Lancet.2000; 356 :1247[Medline]
2. Measles, MMR, and autism: the confusion continues [editorial]. Lancet.2000; 355 :1379[Medline]
3. Ahmad K. "Alarming" measles epidemic kills hundreds in Afghanistan. Lancet.2000; 335 :998
4. World Health Organization. Clinical Research on Treatment of Measles: Report of a Meeting. Geneva, Switzerland: World Health Organization;1995 . WHO/CDR/95.15
5. Sheldon T. Netherlands faces measles epidemic. BMJ.2000; 320 :76
6. Skrabanek P, McCormick J. Follies and Fallacies in Medicine. Glasgow, United Kingdom: Taragon;1989
7. Antman EM, Lau J, Kupelnick B, Mosteller F, Chalmers TC. A comparison of results of meta-analyses of randomized control trials and recommendations of clinical experts. JAMA.1992; 268 :240 –248[Medline]
8. Lightfoot L, Rogers L. Hundreds killed by doctors relying on outdated manuals. The Sunday Times. February 5,1995 :1 –2
9. Samb B, Simondon F, Aaby P, Whittle H, Seck AM. Prophylactic use of antibiotics and reduced case fatality in measles infection. Pediatr Infect Dis J.1995; 14 :695 –696[Medline]
10. Shann F. Meta-analysis of trials of prophylactic antibiotics for children with measles: inadequate evidence. BMJ.1997; 314 :334 –336[Abstract/Full Text]
11. Hussey G. Managing measles. BMJ.1997; 314 :316 –317[Full Text]
12. Berlin JA. Summary statistics of poor quality studies must be treated cautiously. BMJ.1997; 314 :337[Full Text]
13. Anderson T. Sulphanilamide in the treatment of measles. BMJ.1939; i :716 –718
14. Karelitz S, King H, Curtis B, Wechsel M. Use of aureomycin and penicillin in the treatment of rubeola in the pre-eruptive and early eruptive phase. Pediatrics.1951; 7 :193 –199
15. Prasad R, Mathur GP, Trehan OP, Mehrotra ML, Dayal RS. A clinical and radiological study of measles. Indian Pediatr.1967; 4 :243 –250[Medline]
16. Shann F, D’Souza RM, D’Souza R. Antibiotics for preventing pneumonia in children with measles [Cochrane Review]. In: The Cochrane Library, Issue 2. Oxford, United Kingdom: Update Software;1999
17. Shann F, D’Souza RM, D’Souza R. Antibiotics for preventing pneumonia in children with measles [Cochrane Review]. In: The Cochrane Library, Issue 1. Oxford, United Kingdom: Update Software;2000
18. Shann F, D’Souza RM, D’Souza R. Antibiotics for preventing pneumonia in children with measles [Cochrane Review]. In: The Cochrane Library, Issue 2. Oxford, United Kingdom: Update Software;2000
19. Jefferson TG. What are the benefits of editorials and non-systematic reviews? BMJ.1999; 318 :135[Full Text]
20. Aaby P, Garly ML, Martins C, Balé C, Whittle H. Prophylactic use of antibiotics in measles infection in Guinea-Bissau: a randomized, double blind, placebo controlled trial. Unpublished protocol
21. Priest P. Antibacterial Use and Antibacterial Resistance in the Community [DPhil thesis]. Oxford, United Kingdom: University of Oxford;2001
22. Priest P, Yudkin P, McNulty C, Mant D. Antibacterial prescribing and antibacterial resistance in English general practice: a cross sectional study. BMJ.2001; 323 :1037 –1041[Abstract/Full Text]
23. Bauchner H, Wise PH. Antibiotics without prescription: "bacterial or medical resistance"? Lancet.2000; 355 :1480[Medline]
24. D’Souza RM, D’Souza R. Vitamin A for treating measles in children. In: Cochrane Library, Issue 3. Oxford, United Kingdom: Update Software;2001
25. Aaby P, Bukh J, Lisse IM, Smits AJ. Overcrowding and intensive exposure as determinants of measles mortality. Am J Epidemiol.1984; 120 :49 –63[Abstract]
26. Shann F. Uses of error: the wrong research costs lives. Lancet.2001; 357 :1691[Medline]

---

PEDIATRICS (ISSN 1098-4275). ©2002 by the American Academy of Pediatrics

CiteULike    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				M.-L. Garly, C. Bale, C. L. Martins, H. C Whittle, J. Nielsen, I. M Lisse, and P. Aaby Prophylactic antibiotics to prevent pneumonia and other complications after measles: community based randomised double blind placebo controlled trial in Guinea-Bissau BMJ, December 16, 2006; 333(7581): 1245 - 1245. [Abstract] [Full Text] [PDF]

				I. Chalmers Trying to do more Good than Harm in Policy and Practice: The Role of Rigorous, Transparent, Up-to-Date Evaluations The ANNALS of the American Academy of Political and Social Science, September 1, 2003; 589(1): 22 - 40. [Abstract] [PDF]

				F. J. Rosales Vitamin A Supplementation of Vitamin A Deficient Measles Patients Lowers the Risk of Measles-Related Pneumonia in Zambian Children J. Nutr., December 1, 2002; 132(12): 3700 - 3703. [Abstract] [Full Text] [PDF]

eLetters:

Read all eLetters

Prophylactic antibiotics for uncomplicated measles

Frank Shann

Pediatrics Online, 19 Mar 2002 [Full text]

WHO expert advisers need to promulgate consistent advice, based on evidence

Iain Chalmers

Pediatrics Online, 2 Apr 2002 [Full text]

This Article Abstract Full Text (PDF) Submit a response View responses Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Chalmers, I. Search for Related Content PubMed PubMed Citation Articles by Chalmers, I. Related Collections Related AAP Red Book topics: Measles Social Bookmarking                         What's this?