Had a very interesting communication from a BMJ Case Reports author who was contacted by a patient’s family because the patient was suffering from a similar condition to that which the author had just published.

This raises interesting points about the role of the medical literature and the increasing ability of patients and their relatives to research the professional credibility of doctors. This is the modern world but it is it new? Before the Internet patients and relatives would contact by telephone or fax and before the telephone they would probably have written a letter. They always travel if they think the doctor has particular expertise.

There is a limitation of course since you may very well find a doctor with expertise but they may be in another continent and your selective research may have missed a world authority in your local hospital.

Another aspect of this is the networking between the authors themselves. Finding others with similar clinical interests and in publishing their work could lead to the exchange of information and collaboration.

This is a draft critical appraisal sheet for case reports. Would be interested in your views.

Critical Appraisal is used to glean scientific evidence from papers. Case reports or case series are not normally considered as part of this process. The reason is that there is usually no hypothesis testing or comparison within a case report and generalising the conclusions to other circumstances is difficult.

However, case reports have value which could be scientific or educational. When reading a case report it is important to be critical and judge the value of the report to the medical literature and to your own clinical practice. This checklist may help.

King Alfonso of Spain appears to be the first case reported in the newspapers in May 1918 followed by several cases in the medical press including a case series of fifty in July 1918 in the BMJ.

I’ve been looking for the earliest case reports of the flu in 1918 which seems to have first been reported in San Sebastian, Spain in February 1918 - hence the name Spanish flu. However, it probably originated somewhere else.

There are cases reported in Fort Funston, Kansas in March and on the Western Front in French and British soldiers in April¹. A Sergeant John Acker said that the troops called it “three day fever”².

There is much debate in the medical journals at this time of “trench fever” and “weak heart” and a good number of obituaries with influenza as a cause of death.

In June the BMJ is prompted to respond to the newspaper reports of King Alfonso’s illness - a third of Madrid are ill by this time. In the article it says, “We cannot help feeling that in the absence of any bacteriological proof, the extreme low mortality or its practical absence, and the possibility that the disease is gastro-intestinal influenza, render alarmist suggestions premature, and they do not seem to be countenanced by the medical profession in Spain” and went on to say that influenza was the probable cause. ³

The following week the BMJ updated the story saying, “The widespread epidemic of an acute catarrhal affection in Spain, which was stated in our last issue to be most probably influenza and attended by little or no mortality, is now reported to have caused 700 deaths in ten days, but if the number of cases has been as large as reported the case mortality must have been very low.”⁴

The first published cases in the BMJ come in the form of a case series of the first fifty cases at the Central Royal Air Force Hospital, Hampstead from July 1918. It contains useful advice on the presentation, comparison with physical signs from the 1889 pandemic, fever charts, differential diagnosis and the practical management of cases.⁵

What I find interesting is how the reporting seems to have mirrored the current pandemic. At the beginning there is doubt over the diagnosis, the debate on case fatality rates and whether it would be as severe as the previous pandemic, and later it focusses on the clinical management of cases and health policies. I hope the 2009 pandemic will not continue to mirror the 1918 one and become more deadly in late 2009 through 2010.

Remember to keep up to date with the current swine flu at the pandemic flu blog.

1) Tucker S (Editor). “Influenza Pandemic (1918 - 1920)” in “World War I a student encyclopedia”. ABC-CLIO, 2006.

2) Kolata G. Flu: the story of the great influenza pandemic of 1918 and the search for the virus that caused it. Touchstone, 2001.

3) No author listed. The reported epidemic in Spain. Br Med J 1918;1:627.

4) No author listed. The reported epidemic in Spain. Br Med J 1918;1:653.

5) Gotch OH, Whittingham HE. A report on the “influenza” epidemic of 1918. Br Med J 1918;2:82-85.

One criticism against case reports is that although they are widely read they are not often cited. I’ve therefore been looking for the case report or case series that has the highest number of academic citations.

At the moment the best I’ve found is:

James A.R. Nicoll, David Wilkinson, Clive Holmes, Phil Steart, Hannah Markham & Roy O. Weller. Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nature Medicine  9, 448 - 452 (2003)

with 641 citations listed by Google Scholar.

ISI’s Web of Knowledge has the ability of ranking searches by “Times Cited” but I haven’t been able to find one with more citations. PubMed and Highwire don’t seem to have a “sort by number of citations” feature.

I’m sure that some classic cases such as Sydenham’s account of gout or Hippocrates’ description of diabetes have been cited in many hundreds of textbooks alone but how should one judge the world record holder? Should it be the absolute number of citations - in which case Hippocrates has over 2000 years advantage - or the peak number of citations per year?

Arguments about the compatibility of different counts of citation numbers aside … any suggestions for other highly cited case reports would be most welcome.

Although there are some cases described by Mr Greenhow of the practical use of his fracture bed¹ (traction device for the treatment of leg and thigh fractures) in the very earliest issue of the Provincial Medical and Surgical Journal it is in the form of a letter.

I needed to go to the second issue, a week later, to find what appears to be the earliest case report available in the recently released BMJ Archive. In this article Mr Banner, Surgeon to the Liverpool North Hospital, describes the clinical presentation, operative findings and successful outcome of a lady with strangulated femoral hernia². It is not until the very last line of the report does Mr Banner reveal that she was 5 months pregnant with her 15th child at the time! He goes on to describe another case of osteosarcoma in a teenage girl.

The cases are well described but no attempt is made to generalise the findings or make learning points explicit. The style is more of a factual report presented in a way that a knowledgeable colleague could use for reference. This has the advantage of not over-interpreting the outcome or findings of a single case which is a common mistake in poorly written case reports and, clearly historically, part of the etiquette of publishing written case reports. Cautious conclusions are good but there is however, no patient confidentiality as the names of the two patients are published in the reports.

These early cases show the potential of well written case reports to communicate medical knowledge. Modern case reports are not greatly different. Of course there is more formality about the peer-review, citations to the literature and there is patient confidentiality. The educational or scientific value in modern case reports is also made more obvious to the reader but, in a good case report, not over generalised.

I’d be interested to know if anyone has found an earlier case report in the archive or any other worthy cases in the last 169 years.

1) Greenhow TW. Mr Greenhow’s Fracture Apparatus. Prov Med Surg J 1840;s1-1:8, doi: 10.1136/bmj.s1-1.1.8.
2) Banner. Case of Strangulated Femoral Hernia, &c. Prov Med Surg J 1840;s1-1:27, doi: 10.1136/bmj.s1-1.2.27