Prevalence of Beta-hemolytic Streptococcal Pharyngitis and McIsaac Scoring Among Children Attending a Primary Care Clinic in Malaysia

Background: Beta-hemolytic streptococci remained the important cause of acute bacterial pharyngitis. This study aims to determine the prevalence of betahemolytic Streptococcal pharyngitis and the correlation of positive throat culture with the clinical parameters. Materials and Methods: A prospective cross-sectional study was conducted involving 90 children from a primary health care clinic. Throat swab specimen was obtained from each child and sent to the microbiology laboratory for bacterial culture. McIsaac score was used as the scoring system for pharyngitis. Results: The majority of the children were boys (N=53, 58.9%) and most of the children belonged to the preschool age group (N=42, 46.7%). Throat culture was positive in 11 children. The prevalence of beta-hemolytic streptococcal pharyngitis was 12.2%. Group A and Group B Streptococci were isolated in 5 children each and one child had Group C streptococci. Positive throat culture was significantly correlated with the McIsaac score of at least 4 and the absence of cough (p<0.05). Conclusions: Non-GAS pharyngitis was slightly more common than GAS pharyngitis. McIsaac score of 4 and more is perhaps helpful to differentiate between bacterial and viral pharyngitis.


INTRODUCTION
Sore throat or acute pharyngitis remains among the most common reason patients visit to general practitioner clinic. It is mainly caused by viruses. However, bacterial infection particularly beta-hemolytic Streptococci either Group A or non-group A remains among the most important bacterial cause of acute pharyngitis that required proper antibiotic therapy. Group A Streptococci (GAS) can be cultured from the throat swab specimen. The incidence of swab positive GAS was reported around 13 per 100 child-years in children between ages 5 to 12 years (1).
As pharyngitis is commonly caused by a virus, prescription of antibiotics can increase the risk of unnecessary antibiotics prescriptions leading to resistance. Several clinical scoring systems were developed to help clinicians determine those children with acute pharyngitis that require antibiotic therapy. McIsaac scoring system is one of the most commonly used clinical criteria (2). It is perhaps important to highlight the important features that may be able to differentiate between bacterial and viral causes of acute pharyngitis. Thus, this study was conducted mainly to determine the prevalence and useful clinical features to determine beta-hemolytic streptococcal pharyngitis in a primary care setting.

Patient
A prospective cross-sectional study was conducted whereby the throat swab specimen was collected from children between ages 3 to 15 years old. The specimen was collected between 1 st September 2018 until 31 st March 2019. The written informed consent was taken from the parents. We excluded children whom we failed to take throat swab specimens because of poor cooperation and children who had recently received antibiotic therapy for at least two weeks before this presentation. This study was conducted in an urban primary care clinic in Kuala Lumpur, Malaysia. The children were categorized into three age groups; age between 3 to 6 years as preschool, 7 to 12 years as primary school and more than 12 years as a secondary school. This classification was based on the common age where the child enrolled in the school in Malaysia.

Data collection
Baseline epidemiological data was collected from each child. McIsaac scoring system was used for the determination of clinical scoring in all patients. The scoring system was as shown below: The data about antibiotic prescription was also acquired.

Specimen
We took the throat swab specimens in the primary care clinic during clinic consultation. Each child will see the attending doctor in the consultation room. Throat swab procedure was taken by well-trained staff nurses with more than 10 years of working experience. All swabs were then transported to Microbiology Laboratory in Amies transport medium.

Microbiology
The throat swab specimen was then cultured onto sheep blood agar and incubated aerobically at 37C in 5% CO2 for 24 hours. The bacitracin antibiotic disc was placed on the primary striking. The blood agar plate was then inspected for the growth of beta-hemolytic bacterial colonies. We identified the bacterial isolate as Streptococci spp. based on gram stain reaction of gram-positive cocci in the chain, negative catalase test and presence of a zone of inhibition around bacitracin antibiotic disc. The Lancefield grouping was determined using latex agglutination commercial kit (PathoDxtra TM Strep Grouping Kit; Thermo Fischer Scientific; Oxoid, Basingstoke, United Kingdom).

Statistical analysis
The prevalence of beta-hemolytic streptococci was calculated based on the number of positive throat swab cultures over total throat swab specimen cultured, then times 100. The categorical data were then analyzed by chi-square and a p-value of < 0.05 was taken as statistically significant.

Ethical statement
This study had been reviewed and approved by the Research and Ethical Committee of the Universiti Kebangsaan Malaysia Medical Centre with approval number FF-2018-122.

Demographic and clinical features of the patients.
Ninety children were included in this study period and predominantly boys than girls (53 versus 37, 58.9% versus 41.1%). Most of them were preschool groups (N=42, 46.7%). Fever was the most important clinical complaint in 49 children (54.4%) followed by the sore throat in 36 children (40%). The detail of the data was as shown in Table 1.

Correlation between positive throat culture and clinical parameters in McIsaac score.
A significant correlation was found between positive throat culture with overall McIsaac score of at least 4. When we further analyzed on each clinical parameter of the McIsaac Scoring System, absence of cough was noted to be significantly correlated with positive throat culture. We also noted that, significant proportion of patients with positive throat culture was prescribed with antibiotic than those with negative throat culture. The analysis was done by chi-square and the result was as shown in Table 2.

DISCUSSION
The current communication showed that the overall prevalence of betahemolytic streptococcal pharyngitis was 12.2% while group A Streptococci (GAS) was implicated in 5.5% of the cases. In contrary to previous studies, we demonstrated that non-GAS pharyngitis was more prevalent than GAS. A study among Egyptian children reported the prevalence of GAS pharyngitis was 42.2% while non-GAS was only 10.6% (3). Similarly, another study in Yemen, also reported that GAS pharyngitis was more prevalent than non-GAS with a prevalence of 35.3% and 1.8% respectively (4). However, we also noted of a similar finding of non-GAS was more common than GAS pharyngitis in a study from a tropical country, with the prevalence of non-GAS of 28.8 cases/100 childyear versus GAS of 14.7 cases/100 child-year (5). Among the non-GAS pharyngitis, group C Streptococci was among the most common type isolated as showed by previously (6,7). Interestingly, we found that group B Streptococci as the most common non-GAS among our population and its prevalence was similar to GAS pharyngitis. The clinical significance of group B streptococci is not well established but it was reported previously whereby they noted that patients with group B streptococcal pharyngitis were noted to be more likely had enlarged tonsils, exudate and tender enlarged anterior cervical lymph nodes (8). The GAS prevalence in this study was noted to be higher than another study in Taiwan (5.5% versus 4.1%) (9).
We noticed that the absence of cough was more likely to be associated with positive throat culture for beta-hemolytic streptococcal pharyngitis. This was also noted in several previous studies and this feature was described as an important clinical feature to be absent in GAS pharyngitis (4,6,9,10). However, other clinical features such as purulent tonsils, enlarged cervical glands, and duration of illness of less or equal to three days were also more commonly encountered in GAS than non-GAS pharyngitis (6). Clinical features like headache, fever, pharyngeal exudates and cervical lymphadenopathies were more commonly found in all streptococcal pharyngitis than in viral infection (6).
We noted that the McIsaac score of at least 4 was significantly associated with a positive throat culture. Ba-Saddik et al. previously described that the score of 4 and 5 markedly improved specificity (4). Perhaps, this criterion may also be able to differentiate between GAS and non-GAS pharyngitis because clinical features as the absence of cough, cervical lymphadenopathies, and enlarged tonsils were more likely to present in GAS than non-GAS (9). Thus, this contributed to a higher score in GAS pharyngitis. However, we noted that they were also patients with a McIsaac score of less than 4 that had positive throat culture and similarly those with McIsaac score of 4 and 5 but with a negative throat culture. We also noted that all patients with a McIsaac score of at least 4 received antibiotic therapy regardless of the throat culture result. The antibiotics given were mainly amoxicillin-clavulanic acid and ampicillin.
Several limitations were noted in this study that includes this study was conducted in a single center and in an urban area whereby it may not be necessarily reflecting the overall picture of GAS and non-GAS pharyngitis among Malaysian children. It was also noted that in some children we were not able to take the throat swab specimen either because of non-consented by the parents or uncooperative during throat swab taking that may affect our prevalence of GAS or non-GAS.

CONCLUSION
In conclusion, we showed that our prevalence of GAS pharyngitis was lowered than many of the previous reports and we also demonstrated that perhaps non-GAS pharyngitis was more prevalent than GAS. McIsaac score of at least 4 showed a significant correlation with positive throat culture thus initiating antibiotic therapy in this patient may be warranted.

Conflict of interest
No conflict of interest was declared by the authors.