Pharyngotonsillitis in primary health care. Aetiology and clinical findings

Sammanfattning: Pharyngotonsillitis, or acute sore throat, is a common reason for attending primary health care and a common reason for antibiotic prescription. Group A Streptococcus (GAS) has long been considered the most important pathogen in pharyngotonsillitis, but a wide array of other bacteria and viruses have also been associated with this condition. However, few studies have used modern approaches for aetiological detection to evaluate the clinical symptoms associated with these other microorganisms.This thesis aims to learn more about which viruses and bacteria are present in patients seeking primary health care for acute sore throat and how these microorganisms are associated with the clinical course, complications and subsequent re-consultation for sore throat.The thesis is based on four observational studies in Swedish primary health care – three prospective cohort studies and one retrospective registry-based study. The prospective studies were performed with similar designs in two cohorts of 348 young adults and 111 children, respectively, and included both symptomatic patients attending primary health care for acute sore throat and healthy controls. All subjects were sampled and screened with PCR and culture for 20–29 different viruses and bacteria and followed up by diaries or a review of electronic medical records. In the registry-based study, all 14 024 patients in Region Kronoberg who were diagnosed with pharyngo¬tonsillitis between 2012 and 2016 and subjected to aetiological testing with a rapid antigen detection test for GAS or with a throat culture were selected to analyse the association between aetiology, antibiotic prescription and re-consultation for pharyngotonsillitis or a complication.The prospective studies showed that GAS was the most common finding in both children and young adults, and Streptococcus dysgalactiae subsp. equisimilis (SDSE) and Fusobacterium necrophorum were rare in children. Viruses were less prevalent than expected, especially in children. In children, the detection rate of viruses and bacteria was high also in healthy controls and did not differ significantly from the patients.Clinical signs and symptoms of viruses and bacteria overlapped extensively in both children and adults, so neither single nor combined symptoms were able to predict GAS or other aetiologies with a high probability. Cough and coryza have high negative predictive values for GAS but cannot readily be used to predict viruses. The Centor score was more predictive of any bacterial finding than of GAS specifically. The rapid antigen detection test (RADT) had an overall a high sensitivity and specificity for GAS but showed the best performance in patients with a Centor score of 3–4.In the follow-up of the prospective studies, young adult patients with GAS had a higher rate of re-consultation for a sore throat within a month than patients with other aetiologies, although not in a longer perspective of 2 years.In the registry-based study, antibiotic prescription was associated with a lower rate of return visits for pharyngotonsillitis in patients with a positive RADT for GAS. However, antibiotics were not associated with a lower incidence of purulent complications regardless of the aetiological finding.In conclusion, our findings suggest that GAS remains the most important pathogen in pharyngotonsillitis, both in children and adults. SDSE was rare in children and uncommon in young adults and did not distinguish itself as a significant cause of acute pharyngotonsillitis, recurrent infections, or complications. F. necrophorum was rare in children but commonly detected in young adults. Moreover, it was associated with a higher incidence of peritonsillitis in the registry-based study than were GAS and SDSE. The large prevalence of respiratory viruses and bacteria in healthy children makes it challenging to judge the diagnostic relevance of an aetiological finding in a patient. Clinical signs and symptoms of viruses and bacteria overlapped too much in both children and adults, so neither single nor combined symptoms helped determine aetiology. However, cough and coryza might be helpful to rule out GAS. The results of the registry-based study suggest that antibiotics offer some protection against re-consultation for a sore throat in patients with a positive RADT. In contrast, antibiotics did not seem to protect against purulent complications regardless of aetiology.