Blog

It was easy 9 months ago in mid 2023 when there was no whooping cough about. If you thought you might have a patient with it you could take a throat swab, pop it in a dry tube and send it to the lab asking for pertussis PCR if it was in the first two weeks of symptoms, or send blood for pertussis antibodies if it was more than 2 weeks. If blood was too problematic you could notify the Health Protection Team and they would send an oral fluid kit. You might then prescribe a macrolide antibiotic like azithromycin to kill any remaining Bordetella bugs and tell the patient to stay at home and not mix for 48 hours to avoid passing it on. Then you would spend 10 minutes explaining why the terrible cough was going to go on for at least another 4 to 5 weeks.

Now, in May 2024, in England and Wales, cases are running at about 800 per week. It sounds a lot but if you divide it into the number of GP practices, about 7,000, it works out at roughly one practice in ten is having a case confirmed every week. hardly anything to get excited about you would say, as over a year that is still 5 per year per practice. That is a minute proportion of the population and the same kind of numbers my practice was confirming in the 1990s when everyone except me (almost) thought it had gone away completely! There must have been as many everywhere else but they were unrecognised. We are now picking them up more readily because of publicity, self diagnosis and easy testing. (To understand the detail you will need to read about the Keyworth study). 

But average never happens does it? Whooping cough like so many infectious diseases occurs sporadically in little pockets like schools and causes  a few definite cases  and a lot more borderline ones that are deliberately kept from the attention of health services or drawn rapidly to their attention because someone is anxious to know if it is whooping cough or not. This may drag on in a school for several weeks and then settle, before the same thing happens in a nearby school.

We know most cases are in teens and adults and almost all of them will have been immunised in the past. They may suggest the diagnosis themselves if they have researched it or if not might have been telling you of their terrible cough with vomiting that won’t clear up and they think they need antibiotics or have lung cancer.

The first dilemma is whom to test

The official guidelines are not very helpful. They say notify the Health Protection Team as soon as you suspect it. If you are suspecting it because of the characteristic paroxysmal cough it is likely to be at least 2 weeks since the start of symptoms and this is the time when they are most infectious, and it will be at least a week before you know the outcome of the test by which time they are not considered an infection risk. What is gained is a firm diagnosis and a contribution to the official statistics for pertussis. Of more benefit is that doctor and patient can stop worrying it is something more serious.

But what if this is somebody with a slight cough who sits next in class to someone who now has confirmed pertussis? You suspect pertussis but the HPT team are not going help here in all probability. Ideally you could get throat swab for PCR and ask the lab to do it urgently to help you decide whether to give antibiotics. Good luck with that one!  

Then there is the 3 week old baby who is starting to cough who goes in the buggy to take big sister to school and back every day in company with random coughing children all around and mum did not have the pertussis booster when pregnant. You may have more luck with the lab here if have the time to talk to them. It could be life saver.

But you cannot possibly test everyone who might have whooping cough because that would be everyone with a cough. In the future, point of care rapid testing will definitely happen (just like covid), but until then the resources are not there.

The second dilemma is who gets antibiotics

The guidelines are wonderfully clear. No antibiotic means don’t mix until 3 weeks from the start of symptoms. If you have an appropriate antibiotic you can mix after 48 hours on it. What a difference! But not in the real world. Many cases will not get to see a doctor until 3 weeks have passed. Then there is the difficulty of knowing when it started anyway. Some kids are always coughing and nobody can quite say when it got worse. Or “He’s been starting to vape and that started him coughing and then he started vomiting with it”, so who knows when the possible whooping cough started?

On the other hand, given during incubation probably prevents it (but you will never know), and in the early stages may abort it (you will have to be very astute to know).

Where does the magic 3 weeks come from? Common sense tells us it cannot be as absolute as this. There must be a range. It probably comes from that fact that in the old days of detection via per-nasal swab cultures, you could hardly ever find a positive after 3 weeks. But in the old days before per-nasal swabs the method was a cough plate. The doctor would induce a coughing paroxysm (they had their methods!) while holding an agar plate in front of the mouth. The most expert of these people found that isolation tailed off after 3 weeks but could be positive up to 6. 

It is said to be passed on by cough droplets although direct transmission through oral mucus is believed to transmit it too. My own limited observations suggest that transmission propensity is positively related to the occurrence of paroxysms. That would imply that antiobiotics used to limit spread might be most effective in severe cases and vice versa.

What about resistance of B. pertussis to macrolides? It is said to be very common in the far east. It can only be a matter of time before we have the same problem. Then what? It is not as if there is a good alternative. Macrolides are pretty safe. The alternative, co-trimoxazole cannot be dished out so liberally.  Should we restrain prescribing for this reason at least?

Then there is the gut to think about now we know a little about its biome and the length of time it takes to recover from antibiotics. One might suppose that a narrow spectrum antibiotic might not be so detrimental, but who really knows? The seeds of doubt are firmly planted.

The third dilemma is do we use common sense or slavishly follow ‘perfect world’ guidelines

At what point do you stop using your day to day skills and common sense and start looking over your shoulder to see who might be criticising your actions for knowing there is an epidemic of whooping cough and not managing the situation according to NICE or UKHSA or other authorities.

Well at this point I am going to stick my neck out and say “never”, excepting when dealing with infants. Whooping cough and pertussis (they are not actually the same thing) are never going to be controllable or able to be managed in a uniform way. The nature of the infection means it is best handled in the same generic and holistic way as other general respiratory infections, probably including covid-19 too now. The whole whooping cough spectrum is too ubiquitous. Pertussis can be subclinical. We now know enough to at least get the terminology sorted.

Getting back to basics. Pertussis is a big killer in the undeveloped world when immunisation is absent. In developed societies the combination of primary pertussis immunisation with intrapartum boosting, if fully implemented will stop children dying of pertussis for practical purposes. With present vaccines no amount of boosting is likely to reduce this infection to a rarity and the spectrum of severity is so wide and testing and reporting so haphazard that official statistics are almost meaningless unless the circumstances  are tightly controlled or perfectly understood, which is almost never. We have to face the fact that until there is a better vaccine (and it is possibly round the corner), we will have to live with clinical whooping cough from time to time.

Very important is the fact that the vaccine gives little protection against B. parapertussis which can also cause bad whooping cough and seems to be becoming more common. Ten times more common recently in the USA with similar reports from Germany. Parapertussis does not produce pertussis toxin, which means it is not so bad for babies for whom this particular toxin causes the potentially lethal damage. Also of great significance is the fact that pertussis toxin is the basis for our serological tests so whooping cough caused by parapertussis will always test negative ( not easy to explain to patients). Of parallel importance is the fact that the UK Health Protection Agency does not consider B. parapertussis to require public health intervention.

So in practice it may be wisest to keep using common sense, clinical judgement, and treat the guidelines as guidelines. I bet this is what you are actually doing anyway.