Whooping cough is quite common and can be diagnosed clinicallyBMJ 2006; 333 doi: https://doi.org/10.1136/bmj.333.7563.352 (Published 10 August 2006) Cite this as: BMJ 2006;333:352
All rapid responses
I welcome Professor Heininger's comments, and I agree that three
weeks of paroxysmal coughing is not sufficiently sensitive or specific for
a clinical diagnosis. As a clinician I am not too worried about
sensitivity as my patients with mild symptoms of B pertussis infection are
probably not missing out on useful medical advice. They are, of course,
vital for epidemiological modelling.
Specificity however, is something I am very concerned about, and I
have to admit to having been surprised by the results of testing my
clinically diagnosed patients serologically in the last few years. 17
consented to being tested, 15 were positive, one was negative and one was
lost. I know these are not big numbers, and the whole cohort was not
tested, but it suggests a very high specificity for my clinical diagnosis,
and I believe I know the reason why.
My clinical diagnosis is based on 3 weeks of paroxysmal coughing.
'Based on' is important. Paroxysmal coughing for this length of time can
be caused by many other infections and conditions as Professor Heiningen
points out. My observations indicate that B pertussis is the only one
likely to cause an EXCLUSIVELY paroxysmal cough for a minimum of 3 weeks.
The unique thing about whooping cough is the long intervals of hours
between paroxysms when there is not the slightest hint of a cough. Then a
paroxysm or series of them occur together, then another perfectly clear
interval of usually hours. This is a feature that the patients remember
very clearly, although they do not usually put it into words. I believe
that most other causes of prolonged paroxysmal coughing do not exhibit
this phenomenon. To put it as the paradox that it is, the marker of
whooping cough is the absence of coughing (except for paroxysms). A
history taking (with as much hindsight as possible) that elucidates this
feature will in my view considerably increase specificity.
It would be foolish of me to claim it is always as simple as that,
because it does not necessarily hold true if there is secondary infection,
or excessive sputum production, or in the early stages of the illness.
Nevertheless, I am sure it is sufficiently true to make clinical diagnosis
a realistic endeavour if this feature is sought.
Competing interests: No competing interests
Dr. Jenkinson is to be applauded for his continuous clinical work on
pertussis (or whooping cough). However, the title and content of his
current "letter to the editor" are misleading.
The clinical diagnosis of whooping cough is frequently difficult, for
the following reasons:
- Infections caused by Bordetella pertussis (vaccine preventable) can
not be reliably distinguished from those caused by Bordetella
parapertussis (not vaccine preventable)(1).
- Based on Dr. Jenkinson's clinical definition ("a minimum of three
weeks of paroxysmal coughing") only the tip of the iceberg will be
diagnosed. B. pertussis infections may be associated with shorter duration
and/or lack of paroxysmal coughing. This is mainly true in immunised
individuals and adolescents and adults (summarised in 2), but also in
unimmunised children (3).
- Even if cough with paroxysms (or whooping or posttussive vomiting)
is present for 3 weeks or longer, only 57 % of individuals in a vaccine
efficacy trial performed by our group had microbiological evidence of
Bordetella infection (4). Several other microorganisms can lead to similar
In conclusion, prolonged cough with accompanying symptoms suggestive
of pertussis is neither a sensitive nor a specific clinical diagnosis.
Specific laboratory tests (PCR, serology) are necessary if a reliable
diagnosis of pertussis is to be made.
1) Heininger U, Stehr K, Schmitt-Grohé S, Lorenz C, Rost R,
Christenson P, et al. Clinical characteristics of illness caused by
Bordetella parapertussis compared with illness caused by Bordetella
pertussis. Pediatr Infect Dis J 1994;13:306-9.
2) Heininger U, Cherry JD. Pertussis immunisation in adolescents and
epidemiology should guide vaccination recommendations.
Expert Opin Biol Ther 2006;6:685-97.
3) Heininger U, Klich K, Stehr K, Cherry JD. Clinical Findings in
Bordetella pertussis Infections: Results of a prospective multicenter
surveillance study. Pediatrics 1997;
4) Stehr K, Cherry JD, Heininger U, Schmitt-Grohé S, Überall MA,
Laussucq S, et al. A comparative efficacy trial in Germany in infants who
received either the Lederle/Takeda acellular pertussis component DTP
(DTaP) vaccine, the Lederle whole-cell component DTP vaccine or DT
vaccine. Pediatrics 1998;101:1-11.
5) Wirsing von König CH, Rott H, Bogaerts H, Schmitt HJ. A serologic
study of organisms possibly associated with pertussis-like coughing.
Pediatr Infect Dis J 1998;17:645-9.
I have served on advisory boards of several vaccine manufacturers including those who produce pertussis vaccines
Competing interests: No competing interests