EDITOR – I have read the case reports on the 2 year-old child with a
rash and fever with interest.(1) Last week I thought about the problem in
two ways.
Firstly, as a GP I came up with the following differential diagnosis:
1. Very likely a viral infection.
2. Reasonable probability of streptococcal pharyngitis.
3. Possibly Kawasaki syndrome.
Viral illnesses that give this pattern of symptoms and signs are very
common. Streptococcal pharyngitis is fairly common, but Kawasaki syndrome
is rare.
Secondly, as a reader of the article I came up with the following
differential diagnosis:
1. Very likely - Kawasaki disease.
2. Reasonable probability of streptococcal pharyngitis.
3. Very remote possibility of viral illness.
This was meant to be an interesting case report. The presentation is
compatible with early Kawasaki disease, and failure to diagnose it early
enough results in serious sequelae. Streptococcal pharyngitis is much more
common, but is also less interesting as a case report. Viral illnesses are
very common and very uninteresting, so it goes at the bottom of the list.
After this week’s update, ‘viral illness’ has dropped of my
differential diagnosis. As a GP, Kawasaki disease is clearly the most
likely. Streptococcal pharyngitis remains a possibility with the spiking
fever heralding the development of rheumatic fever. However, this is rare,
especially in a child so young.
As a reader, I put streptococcal pharyngitis at the top of the list.
The clinical picture is now very obviously Kawasaki disease, and yet there
is still another case report to appear next week. The development of an
atypical rheumatic fever would be a thrilling finish to the series.
This case report like many others illustrates two important points
relating to risk as illustrated in this issue. Firstly, a collection of
symptoms and signs may be typical and have a have a high sensitivity (e.g.
for Kawasaki disease) and yet have a very low positive predictive value
because of the rarity of the disease(2) (an incidence of about 1:10,000 in
the UK)(3). Secondly, the risk or probability of an outcome is highly
dependent on the reference class.(4) In this case, as a GP, my reference
class is ‘all cases of 2 year-old children presenting in Primary Care with
a rash and fever. As a reader the class is ‘all case reports appearing in
this journal’. The context makes an enormous difference to the accuracy
of our diagnoses.
Reference List
(1) Sharma R, Boon A, Harnden A. A 2 year old child with rash and
fever. BMJ 2003; 327(7417):720.
(2) Loong T. Understanding sensitivity and specificity with the
right side of the brain. BMJ 2003; 327(7417):716-719.
(3) Harnden A, Alves B, Sheikh A. Rising incidence of Kawasaki
disease in England: analysis of hospital admission data. BMJ 2002;
324(7351):1424-1425.
(4) Gigerenzer G, Edwards A. Simple tools for understanding risks:
from innumeracy to insight. BMJ 2003; 327(7417):741-748.
Competing interests:
None declared
Competing interests:
No competing interests
29 September 2003
Christopher J Martin
GP and lead researcher, Laindon Health Centre Primary Care Research Team
Rapid Response:
Reference class: journal case report.
EDITOR – I have read the case reports on the 2 year-old child with a
rash and fever with interest.(1) Last week I thought about the problem in
two ways.
Firstly, as a GP I came up with the following differential diagnosis:
1. Very likely a viral infection.
2. Reasonable probability of streptococcal pharyngitis.
3. Possibly Kawasaki syndrome.
Viral illnesses that give this pattern of symptoms and signs are very
common. Streptococcal pharyngitis is fairly common, but Kawasaki syndrome
is rare.
Secondly, as a reader of the article I came up with the following
differential diagnosis:
1. Very likely - Kawasaki disease.
2. Reasonable probability of streptococcal pharyngitis.
3. Very remote possibility of viral illness.
This was meant to be an interesting case report. The presentation is
compatible with early Kawasaki disease, and failure to diagnose it early
enough results in serious sequelae. Streptococcal pharyngitis is much more
common, but is also less interesting as a case report. Viral illnesses are
very common and very uninteresting, so it goes at the bottom of the list.
After this week’s update, ‘viral illness’ has dropped of my
differential diagnosis. As a GP, Kawasaki disease is clearly the most
likely. Streptococcal pharyngitis remains a possibility with the spiking
fever heralding the development of rheumatic fever. However, this is rare,
especially in a child so young.
As a reader, I put streptococcal pharyngitis at the top of the list.
The clinical picture is now very obviously Kawasaki disease, and yet there
is still another case report to appear next week. The development of an
atypical rheumatic fever would be a thrilling finish to the series.
This case report like many others illustrates two important points
relating to risk as illustrated in this issue. Firstly, a collection of
symptoms and signs may be typical and have a have a high sensitivity (e.g.
for Kawasaki disease) and yet have a very low positive predictive value
because of the rarity of the disease(2) (an incidence of about 1:10,000 in
the UK)(3). Secondly, the risk or probability of an outcome is highly
dependent on the reference class.(4) In this case, as a GP, my reference
class is ‘all cases of 2 year-old children presenting in Primary Care with
a rash and fever. As a reader the class is ‘all case reports appearing in
this journal’. The context makes an enormous difference to the accuracy
of our diagnoses.
Reference List
(1) Sharma R, Boon A, Harnden A. A 2 year old child with rash and
fever. BMJ 2003; 327(7417):720.
(2) Loong T. Understanding sensitivity and specificity with the
right side of the brain. BMJ 2003; 327(7417):716-719.
(3) Harnden A, Alves B, Sheikh A. Rising incidence of Kawasaki
disease in England: analysis of hospital admission data. BMJ 2002;
324(7351):1424-1425.
(4) Gigerenzer G, Edwards A. Simple tools for understanding risks:
from innumeracy to insight. BMJ 2003; 327(7417):741-748.
Competing interests:
None declared
Competing interests: No competing interests