Penicillin for acute sore throat in children: randomised, double blind trial

To the Editors:

The authors correctly state that sequelae of infections due to Streptotoccus pyogenes have become "[...] rare in affluent western communities. [...]" This good news, in my humble opinion, is both a consequence of the standard treatment of these infections, which includes Penicillin V for 10 (not seven!) days or another drug active against group A streptococci, and of changes in the relative distribution of different strains of those bacteria.

It has been known for decades that the signs and symptoms of streptococcal pharyngitis are hardly alleviated by treatment with antimicrobials: "[...] The course is shortened little by treatment, which is given primarily to prevent suppurative complications and rheumatic fever [...]"¹

The authors correctly point to the growing concerns about rising resistance to antimicrobials; S. pyogenes, however, has remained exceptionally sensitive against Penicillin G until today, despite the widespread use of this drug against streptococcal diseases.

The authors state that in their patient group, which included but 96 patients with diagnoses of S. pyogenes, suppurative complications which could have been prevented by antimicrobial therapy occurred rarely and
could safely be treated "[...] at the moment of their occurrence.[...]". One should keep in mind that ambulatory patients will not always have medical support available without delays. In addition, I would expect larger studies to support the tendency the present study has documented, i.e., patients deprived of antimicrobial therapy for group A streptococcal disease have a higher risk of suppurative sequelae.

The authors do not discuss nonsuppurative sequelae of infections by group A streptococci. The attack rates of poststreptococcal glomerulonephritis, Sydenham's chorea, erythema marginatum, and rheumatic fever vary widely and depend, among other factors, on the antigenic structure of the streptococcal strains involved. With attack rates of generally less than 3 percent, it does not come as a surprise that the authors have not seen such a complication in any of the 43 patients in their placebo group from which S. pyogenes has been isolated. With a carrier rate of 30 % reported by the authors, some of the 43 patients in the placebo group in which group A streptococci have been detected may have been mere carriers in which a sore throat has been caused by a viral infection, resulting in a still lower expected rate of nonsuppurative sequelae in this small group of patients. Of note, these sequelae will not resolve despite treatment in many of those hit by them. Nonsuppurative sequelae have been seen with varying rates; during the 1970s, they had almost disappeared in the United States, only to reappear in the mid and late 1980s².

The present study, in my opinion, does not present data to support the conclusions the authors draw with respect to antimicrobial treatment of group A streptococcal diseases. In my view, it can hardly be justified to deny antimicrobial treatment to patients suffering from diseases due to S. pyogenes; infections that might be caused by this bacterial species need diagnostic workup, and diseases caused by group A streptococci need antimicrobial treatment sufficient to eradicate the infection so as to prevent nonsuppurative and suppurative sequelae.

[1] Isselbacher et al.: Harrison' Principles of Internal Mecidine, 13th ed., MacGraw Hill, 1994, p. 618.

[2] Behrman et al. (eds.): Nelson Textbook of Pediatrics, 14th ed., W.B.Saunders, p. 640.

Competing interests:
None declared

Dear Editor

We read with interest the report of the randomised controlled trial
of penicillin for sore throat in children by Zwart and colleagues.[1] This
is an important study as little data exist to guide the treatment of
children and it demonstrated the lack of effect on outcomes important to
patients and parents, namely symptom duration, school attendance and
recurrence of infections. However, with regard to streptococcal sequelae,
the authors found an inverse relationship between children with sequelae
(14%, 4% and 2%) and increasing duration of penicillin (0, 3 and 7 days
respectively). The probability of this finding occurring by chance is 0.03
(chi-square test).

We wish to raise three issues regarding this finding. The first is
that ‘imminent quinsy’ accounted for 82% of sequelae. However, the authors
do not clearly define what they meant by this term and so the implications
for clinical practice remain uncertain. Second, although the authors’
conclude that they had insufficient power to draw firm conclusions, and we
acknowledge their restraint in not overstating a secondary result, we feel
this is an important finding that deserves replication. Furthermore, if we
are to rationalise the use of antibiotics in primary care, it would be
useful to know if there are any clinical factors, for example fever[2,3]
or vomiting,[3] associated with the development of such sequelae. Finally,
what should the message be to parents and children? If presented with
these data, including the lack of increased side effects associated with
penicillin, we think many would opt for a three or seven day course of
antibiotics to prevent the pain, discomfort and inconvenience of ‘imminent
quinsy’. Given that the decision to prescribe antibiotics in primary care
is often finely balanced and frequently influenced by patient
expectations,[4] trials of antibiotics need to include the full complement
of outcomes important to prescribing. These are symptomatic outcomes, side
effects, complications and, importantly, antibiotic resistance.[5] Only
then will clinicians and patients be able to engage in fully informed and
balanced decision-making.

Reference List

1.Zwart S, Rovers MM, de Melker RA, Hoes AW. Penicillin for acute
sore throat in children: randomised, double blind trial. BMJ 2003;327:1324
-0.

2.Hay AD, Fahey T, Peters TJ, Wilson AD. Predicting complications
from acute cough in pre-school children in primary care: a prospective
cohort study. Br J Gen Pract 2004;(in press).

3.Little P, Gould C, Moore M, Warner G, Dunleavey J, Williamson I et
al. Predictors of poor outcome and benefits from antibiotics in children
with acute otitis media: pragmatic randomised trial. BMJ 2002;325:22.

4.Macfarlane J, Holmes W, Macfarlane R, Britten N. Influence of
patients' expectations on antibiotic management of acute lower respiratory
tract illness in general practice: questionnaire study. BMJ 1997;315:1211-
4.

5.Schrag SJ, Pena C, Fernandez J, Sanchez J, Gomez V, Perez E et al.
Effect of short-course, high-dose amoxicillin therapy on resistant
pneumococcal carriage: a randomized trial. JAMA 2001;286:49-56.

Competing interests:
None declared

Editor – Zwart et al has presented a well designed study on a common
problem in general practice; the efficacy of penicillin in children with
sore throat (1). In his comment Paul Little has discussed the possible
problems of this study and its conclusions: How valid are the Centor
criteriae and secondly; do the criteria predict benefit from antibiotic
treatment (2)?

As I see it in line with Little, the main problem is to find the
subgroup with a clinically significant GAS-infection. The strep A test is
not good enough since it is so far not capable of distinguishing between
carriers of GAS and patients with an immunological significant GAS-
infection. However, in Scandinavian general practice, the CRP rapid test
is widely used in daily practice to distinguish viral from bacterial
infections. The question is whether this test could be able to distinguish
carriers of GAS from immunological significant infections that probably
would benefit from treatment with penicillin.

We have performed a diagnostic study (unpublished data) from
Norwegian general practice on patients with sore throat. Among the
children with sore throat, less than half of the patients had GAS
detected in the throat. Among the children with GAS, around 50% had a CRP-
value of more than 50, which is often used as a cut-off for demonstrating
immunologically significant infection. We will go further to see whether
we will find an association between elevated CRP and a serological
response against GAS.

Another related question that was raised by the first article of
Zwart et al, was whether group C and G streptococci were found in the
cultures. If so, did they give significant infections in children as was
demonstrated in adults?

1) Zwart S., Rovers MM., et al. Penicillin for acute sore throat in
children: randomised, double blind trial. BMJ 2003;327:1324-6.

2) Paul Little. Commentary: More valid criteria may be needed. BMJ
2003;327:1327-8.

Competing interests:
None declared

To the Editors:

I disagree with Zwart et al. in their recommending against the
routine use of antibiotics for streptococcal pharyngitis.
It appears that conclusions were drawn before study was conducted. When I
saw that in their study, even 7 day course of Penicillin was not
successful in eradicating Streptococcus in 38% patients, I was dismayed.
More Straightforward conclusion would be that a more effective antibiotics
regimen is warranted than a 7 day course of Penicillin (as was done after
breaking the code for those with sequelae).
Comparing an ineffective Antibiotics regimen with placebo is beyond my
understanding.

Competing interests:
None declared