Scarlet fever is an exanthematous disease that occurs primarily in childhood. It is caused by group A beta-haemolytic Streptococcus strains that produce erythrogenic toxins (mainly A, B and C), which cause the characteristic exanthema through a delayed hypersensitivity reaction. In a previous study, in which we demonstrated that the pharyngotonsillitis associated with scarlet fever is not that similar to the typical streptococcal pharyngotonsillitis, we found a significant number of episodes of reinfection among our patients.1 Historically, the literature has described recurrent episodes as rare.2 A few recently published studies also reported low rates of reinfection in patients managed in paediatric emergency departments3,4 and in primary care.5 The availability of the rapid streptococcal antigen detection test (RST) in primary care clinics allows its diagnosis, and as we observed in the aforementioned study,1 cases of mild pharyngotonsillitis with compatible exanthema in young children are confirmed to be scarlet fever. A study conducted very recently in children that received the diagnosis in a primary care setting also found a high incidence of recurrent scarlet fever (up to 16.5% of cases).6

Based on the series of patients under study, whom we followed up until November 2015, we estimated the frequency of reinfection and the differences in its clinical presentation. We conducted a retrospective, observational and descriptive study of the cases of scarlet fever diagnosed in four paediatrics caseloads in one urban primary care centre between 2004 and 2015 in patients that had at least 2 episodes of scarlet fever. All episodes were confirmed microbiologically by means of a RST (84%) or culture (16%), and we collected data for the following variables: episode number, fever, cold symptoms, cough, cervical lymphadenopathy, tonsillar hyperaemia and/or exudate, petechiae on the palate, strawberry tongue, circumoral pallor, Pastia lines, age, sex, seasonality, time interval between episodes, antibiotherapy and Centor score (maximum of 4 points, based on the presence or absence of the following criteria: fever >38°C, absence of cough, exudate on tonsils and tender and swollen anterior cervical lymph nodes).

Of the 158 patients with microbiological confirmation of scarlet fever, 16 (10%) had more than one episode: 12 patients had two episodes, and 4 patients had three episodes. The mean age was 3.70±1.38 years for first episodes, 4.72±1.51 for second episodes and 5.34±1.69 for third episodes, and the minimum interval between the recorded episodes was 14 days, while the maximum was 5 years and 4 months. We found no differences based on sex. The clinical manifestations were attenuated in episodes of reinfection: fever of 38°C or higher in 53% of reinfections (95% confidence interval [CI], 29%–77%) compared to 71% of initial episodes (95% CI, 48%–95%); absence of cough in 66% reinfections (95% CI, 33%–79%) versus 77% of initial episodes (95% CI, 64%–100%); tonsillar exudate in 0% versus 20% (95% CI, 0%–40%), respectively; and lymphadenopathy in 89% (95% CI, 68%–100%) compared to 50% (95% CI, 15%–85%). The Centor score was higher than 3 in only one of the recurrent episodes (5%). The other classical signs of scarlet fever were observed in very few cases. The mean time interval between the first and the second episodes was of 1.02 years, while the mean interval between the second and third episodes was of 0.62 years. The prescribed duration of antibiotherapy was 9.63±1.02 days for first episodes and 9.95±123 days for reinfections. In the first episode as well as in reinfections, amoxicillin was prescribed significantly more frequently (81%; 95% CI, 68%–93%) than penicillin V (19%; 95% CI, 7%–32%).

In conclusion, now that we have the necessary diagnostic tools available in primary care, we are observing that the proportion of reinfections among cases of recurrent scarlet fever is in no way insignificant. Furthermore, the pharyngotonsillitis associated with scarlet fever in cases of reinfection also differs from the classical streptococcal pharyngitis,1 as in our study, the fever was low-grade in half of the cases, and none presented with tonsillar exudate. The average interval of time between episodes suggests that these are true reinfections, as opposed to relapses. At any rate, the 10% and 15% of children with recurrent episodes found when tools that allow an accurate diagnosis are available in primary care1,6 indicate that recurrent scarlet fever is a more common entity than previously described.