This site is intended for health professionals only

At the heart of general practice since 1960

Myth: ‘We know what a normal heart rate is for a child’

Dr Edward Snelson discusses why heart rate and other values should not be overly relied on when assessing for sepsis

What is the myth and where did it come from?

Assessment of heart rate, respiratory rate and other relevant vital signs are an essential part of assessing an acutely unwell child. It’s just that there’s no such thing as normal and that the interpretation of the number is not as simple as the little tables of ‘normal values in children’ suggest.

So how did we get here? Clinicians have long recognised that physiological values are affected by various factors, including age. It was therefore seen as helpful to give some guidance as to what values to expect in well children at various stages of life. So a few people made up some numbers. This was done by about three different groups of people, all of whom had a really good idea of what those numbers should be, based on their own experience. Each of the given ranges was slightly different from the others and none of them gave the centiles for the upper and lower limits of the ranges.

Because the ranges were found to be useful in practice, they were just accepted as fact. As time went on, the existence of these ranges became so well established that most people assumed they were based on something other than expert opinion. Even the name used for them – ‘normal values’ – implied a robust statistical basis. So it was often the case that when a six-month-old baby had a heart rate of 159, everything was felt to be just fine. By contrast, a heart rate of 161 was a very serious matter.

What is the reality?

But surely someone has done this research? Lots of people have done lots of research on physiological parameters in children at different ages. There has been one large study1 and one systematic review2 on the heart rate and respiratory rate of children, both of which came to similar conclusions.

The first conclusion was that the numbers are close, but often off. Occasionally what is quoted as ‘normal’ strays outside the 1st or 99th centile. Age bandings have a significant effect on this because the changes in values are quite dramatic and a flat value that applies to a period such as two to five years old will inevitably be wrong at one end or the other. But the traditional ranges are usually well within the centiles, perhaps reflecting the cautious nature of any expert giving guidance as to when to treat a value as abnormal.

These studies have given us centile charts – so we do know what is normal. Except that we don’t. This brings us onto the second set of conclusions from these studies, which is about the clinical interpretation of the values. One major factor is the heterogeneous nature of the information that went into the analysis. It is impossible completely to account for the effects of temperature, pain, fear, activity, wakefulness or other factors on the data. Even if it were possible to take these into account, the necessary outcome would be a published set of norms that would only apply to well, sitting, happy, pain-free, unafraid children with a temperature of 36.75.

The other big spanner in the works is that no one knows which centile matters. Is it the 95th, 98th or 99.9th? At what point do you cross into the danger zone? This is the other issue with normal ranges. There is the implication of a binary element to any recorded value: if your value is within the limits all is well but if you have a number outside the limits it is time to be worried. The concern for GPs is that we are missing something or that in retrospect somebody might say that we did. In reality, any one piece of the jigsaw is exactly that. The idea that someone is able to tell in retrospect what was wrong with a child just by looking at the heart rate out of context is nonsense.

How does this change my practice?

I would imagine it’s highly unlikely that you have failed to notice the growing emphasis on early recognition of sepsis. Whether it is in the news, on social media, in the journals or the latest guideline, we are repeatedly told sepsis needs to recognised and treated earlier and better. General practice software now has sepsis warnings built in, which are triggered when a febrile tachycardic child is seen. As you look from the warning flashing on your screen to the child who is refusing to give back your stethoscope because it’s their new favourite toy, you might well wonder why things are this way.

The truth is that there is both validity and dysfunction in the recent changes in our approach to sepsis. There is good evidence that early recognition and treatment of sepsis does save lives. The treatment part is reasonably easy to improve on. When sepsis is diagnosed it is important to treat it with antibiotics, fluids and whatever else is needed, avoiding any unnecessary delay. We’re not talking here about possible sepsis, which includes just about every febrile child. We’re talking about those who are clinically septic, who need to be treated without delay.

No agreed definition of sepsis

The problem is there is no method of diagnosing sepsis in children that has been shown to be sensitive and specific. There is no reliable test or decision tool. There isn’t even a universally accepted definition of sepsis.3 As a result, the early recognition part is a bit broken. So the good news is, as you see the sepsis alert pop up on your computer, it’s not you that has lost the plot.

If we are unable to define or diagnose sepsis easily, then what are we doing? That’s an easy one to answer. We’re doing our best. None of what was written in the previous paragraph changes the fact that early recognition of sepsis will make a difference to the patient. Therefore we need to do everything we can to recognise the significantly unwell child and make sure that the possibility and probability of sepsis are carefully considered.

Those who write guidelines have made this look simple but it is not. Febrile, tachycardic children are commonplace in primary care, while sepsis is rare. Despite the fact that sepsis is difficult to differentiate from uncomplicated viral illness, correct decisions about whether a child is septic are the norm and ‘missed sepsis’ is rare. Often, the second visit by the parents is made to an emergency department not because the GP missed sepsis at the initial consultation but because sepsis has developed in the interim.

Global assessment

While there is no such thing as a normal heart rate in a child, pulse, respiratory rate and capillary refill are all important elements in the assessment of the unwell child. Activities, behaviour and progression of symptoms are also key elements.

Your global assessment is also important. All GPs will have seen lots of unwell children so they are likely to have developed a gut feel for what is normal. That is not the same as a presumption that everything will be fine (because it usually is). The gut feel is not about false reassurance, it is about knowing that the pop-up message on your computer screen can safely be closed because you have considered the possibility that the child has sepsis.

Indeed, the way the guidelines have changed will definitely make clinicians feel that they should lower their threshold for referring any cases of possible sepsis.

So it’s important to avoid getting overly caught up in the significance of any one number. Look instead at the whole picture and look at the child. They will usually give you the answer, even if they won’t give you back your stethoscope.

Dr Edward Snelson is a consultant paediatrician specialising in paediatric emergency medicine at the Sheffield Children’s Hospital. He is the author of ‘The Essential Clinical Handbook of Common Paediatric Cases’ and of gppaedstips.blogspot.co.uk - an internet site for GPs to keep up to date with paediatrics in primary care. Before becoming a paediatrician, he practised as a GP for more than five years, and now uses that experience to fuel his educational work in primary care. Twitter: @sailordoctor

References

  1. O’Leary F et al. Defining normal ranges and centiles for heart and respiratory rates in infants and children: a cross-sectional study of patients attending an Australian tertiary hospital paediatric emergency department Archives of Disease in Childhood 2015;100:733-737
  2. Fleming S et al. Normal ranges of heart rate and respiratory rate in children from birth to 18 years of age: a systematic review of observational studies, The Lancet 2011,19;377(9770):1011-8
  3. Snelson E. Paediatric Sepsis - the facts, the myths, how we got here and where we need to go next. GPpaedsTips blog. tinyurl.com/GPpaeds-sepsis

Rate this article  (4.76 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Readers' comments (7)

  •  Dr Kon Conrad

    Well reasoned and common sense article with great significance for the day to day work of any GP.

    Unsuitable or offensive? Report this comment

  • Excellent article. 'Expert witnesses' might benefit from reading it.

    Unsuitable or offensive? Report this comment

  • Thought provoking article that hightlights the importance of global assessment rather than focusing purely on numbers.

    Unsuitable or offensive? Report this comment

  • Dr B Garba did mention in her reflection that on her second visit patient globally appeared better as he was sitting up etc etc
    I am not against global out look approach but having a low threshold for diagnosing sepsis and then pressurising secondary care doctors does not help GP mind set!

    Unsuitable or offensive? Report this comment

  • Playing devil's advocate here -

    Sometimes we need to put our "clinical experience" to one side, and just rely on the algorithm.

    I had a child with fever and tachycardic in my consulting room a while ago. Child looked ok. But no focus (other than pink eardrums, but what child doesn't have pink drums after a bit of crying),

    I was so tempted to say,
    "They'll probably be fine".

    Instead, I said, "they'll probably be fine. However, look at this traffic light triage system".

    I pulled up the NICE traffic lights for febrile children on the screen, and showed mum.

    "The heart rate is red. That means that even though your child looks ok, a few children like this may become seriously not ok in the next few hours. Could you take them to hospital?".

    Child had meningitis, sepais, antibiotics, lumbar puncture, PCR demonstrating viral meningitis, etc. Required lots of inotropes and supportive care to survive.

    It taught me that sometimes I should trust my gut. But sometimes, I should ignore my gut, and defer to the algorithm.

    I saw the child again recently, for nothing too serious, burst into tears after they left, and felt on top of the world for the rest of the day.

    Unsuitable or offensive? Report this comment

  • Curb Your Enthusiasm

    well done Patrick McNally

    Unsuitable or offensive? Report this comment

  • Edoardo Cervoni

    Well written article. In my experience, the best guidance is in Dr Snelsons' statement:"Activities, behaviour and progression of symptoms are also key elements", this compounding with the outcomes of general observations.

    Unsuitable or offensive? Report this comment

Have your say