Philipson et al have provided more evidence on subclinical Bordetella pertussis infection.1 After reading the article one could think that B. Pertussis was just another, impossible to distinguish, cause of prolonged coughing, that only a laboratory test will illuminate. I think it needs to be pointed out that whooping cough is a real syndrome, with a largely forgotten, but unique characteristic that makes clinical diagnosis possible, and that we now realise, co-exists with subclinical infection.
I have studied 740 cases of clinically diagnosed whooping cough in the Keyworth Practice since 1977.2 The characteristic that distinguishes clinical pertussis is not the ‘whoop’, but the very long intervals (can be hours) without coughing, contrasting with the severe choking paroxysms that occur on average every 2 hours. Patients do not volunteer this information, indeed very few are aware of it until they have thought about it.
It is possible, but I think unlikely, that none of the oral fluid positive patients in Philipson’s study had clinically diagnosable pertussis. If the right questions had been asked, the software may have learnt something, and very likely improved on the average clinician.
Pertussis is diagnosable if the characteristic symptoms are known and the right questions asked, or if the clinician hears the sound of a real whooping cough paroxysm and learns the tune, which few have had the opportunity to do, since the cough is inconveniently intermittent.
There is probably more danger from cases missed through lack of diagnostic skill than there is from the unknown number of subclinical cases, which, as opposed to missed cases, are not very important in the transmission of this disease that is still killing babies.
- © British Journal of General Practice 2013