Background Septic Arthritis (SA) is associated with considerable morbidity and mortality.
Objectives We describe the clinical features and outcomes of SA in patients admitted to a tertiary hospital in Singapore over 10 years.
Methods Adult patients admitted between 2005 and 2015 with diagnosis codes for SA were identified from electronic medical records. Detailed medical record review was performed to confirm the diagnosis of SA based on positive gram stain or cultures from the joint, or alternatively a clinical diagnosis of SA with positive cultures from blood or genital swabs. Patients with negative cultures, purely extra-articular infections and insufficient documentation were excluded. Data were collected by 2 independent researchers with standardization performed on 30 randomly selected patients.
Results 1067 patients were identified from diagnosis codes, of which 249 (23.3%) were confirmed to have culture positive SA. The median (IQR) age at diagnosis was 58 (46, 71) years and 64% were male. Sixty percent were Chinese, 22% Malay and 11% were Indian.
Clinical features: The most commonly affected joints were the knee (56.2%), shoulder (12.8%) and ankle (11.6%); 13.6% of patients had involvement of more than one joint. Fever was present in 43% patients and elevated white count (WC) and C-reactive protein (CRP) were observed in 62.6% and 96.8% respectively. Only 2.7% had both a normal CRP and WC. The most common organism was Staphylococcus aureus (53%), 29.6% of which (15.6% of all patients) were methicillin resistant Staphylococcus aureus (MRSA). Gram negative bacteria and Group B streptococcus (19.7% and 18.1%, respectively) accounted for most of the remaining cases.
Predisposing conditions: Pre-existing joint disease was identified in 59 (23.9%) patients (30 gout, 25 osteoarthritis, 5 rheumatoid arthritis and 1 systemic lupus erythematosus). Prosthetic joint infections accounted for 18.9% of cases. 3.6% patients were current intra-venous drug abusers and 27.7% had diabetes. Ten patients had potential preceding joint inoculation, including two with recent intra-articular corticosteroid injection and three with recent acupuncture to the index joint. Data on use of complementary medications and alcohol were not readily available in this retrospective study.
Treatment and prognosis: Eighty-three percent patients underwent surgical joint washout. Of those with prosthetic joint infection, 49% required removal of the prosthesis. The median (IQR) duration of intravenous antibiotics was 33 (18, 42) days and median (IQR) length of hospital stay was 21 (11, 35) days. 7.6% patients experienced a recurrence of infection in the same joint within a year. There were inpatient 12 deaths. Using univariate logistic regression, age ≥70 years, high WC, absence of fever and infection with MRSA were associated with higher mortality (Table).
Conclusions Staphylococcus aureus remains the most common cause of SA. Most patients in our dataset had at least one predisposing factor. There were more deaths in patients who were afebrile at presentation, possibly because the absence of fever led to a delay in diagnosing SA. Early aggressive treatment may be warranted in older patients, especially in those with higher WC or infection with MRSA.
Disclosure of Interest None declared