Comparison of clinical and laboratory test results of septic and non-septic cases in childhood acute monoarticular arthritis

Septic arthritis in children is an emergency condition that requires prompt and effective treatment, otherwise it may result in severe sequelae. Other reasons mimicking the clinical and laboratory findings of septic arthritis may complicate the differentiation between septic and non-septic arthritis. Aim: Aims of the present study were: (1) to obtain diagnostic threshold values in differentation between septic and non-septic acute monoarticular arthritis by comparing clinical and laboratuary findings associated with septic arthritis, and then (2) to investigate the diagnostic values of these laboratory parameters in the differentation of septic arthritis from non-septic arthritis. Materials and Methods: A total of 117 children aged 2 months to 16 years who presented with acute monoarticular arthritis to our hospital between 2005 and 2019 were enrolled in the present study to be evaluated in terms of clinical and laboratory findings at the time of admission. After obtaining the relevant data, all patients were divided into two groups: septic arthritis group (group SA; 57 patients) and non-septic arthritis group (group SO; 60 patients). Several clinical and laboratory findings at the time of the presentation were recorded. Clinical findings included the time from onset of symptoms to diagnosis, fever and restriction of range of motion. Laboratory parameters consisted of complete blood count, C-reactive protein (CRP) level and sedimentation rate. The diagnostic values were determined and the threshold values for the parameters that were found to be significant were determined by ROC analysis. Multivariate logistic regression analysis was performed to determine the predictive criteria and the probability of septic arthritis. Results: Each group had a median age of 88 months; the age range was from 2 months and 10 days to 186 months in group SA and 17 months to 189 months in group SO. There were 18 females (31,6%) and 39 males (68,4%) in group SA, and 30 males (50%) and 30 females (50%) in group SO. In both groups, the most commonly involved joint was the knee. No significantdifference was observed between the two groups in the time from onset of symptoms to diagnosis (p=0.52). Body temperature was significantly higher in group SA than in group SO (37,6±0,9°C vs. 36,72±0,65°C) (p=0.001; p<0.01). All the patients in group A and 60% in group SO had developed restriction of range of movement (p=0.001; p<0.01). In terms of laboratory parameters, CRP and SED were significantly higher in group SA (mean CRP: 90±48 mg/L, mean SED: 70±30 mm/h) than in group SO (mean CRP: 41±37 mg/L, mean SED: 35±24 mm/h). Similarly, leucocyte and neutrophil counts in addition to the percentage of neutrophils were higher in group SA (mean leucocyte count: 14365±6035/mm3, mean neutrophil count: 9787±4489/mm3, mean percentage of neutrophils: %66±9) than in group SO (mean leucocyte count: 9877±3652/mm3, mean neutrophil count: 5563±2991/mm3, mean percentage of neutrophils: %54±13) (p=0.001; p<0.01). In contrast, there was no significant difference in platelet counts (mean platelet count: 345±95 x109/l in group SA, mean platelet count: 348±111 x109/l in group SO; p=0,86). Possible threshold values for distinguishing septic arthritis from non-septic arthritis are as follows: 1. Body tempreature >37.1°C (Odds ratio 10.092, %95 CI: 4.316-23.599) 2. CRP ≥ 63 mg/L (Odds ratio 11.200, %95 CI: 4.717-26.592) 3. SED ≥ 53 mm/h (Odds ratio 11.104; %95 CI: 4.695-26.262) 4. Leucocyte count ≥ 12100/mm3. (Odds ratio: 6.857; %95 CI: 2.988-15.734) 5. Neutrophil count ≥ 6300/mm3 (Odds ratio: 10.578; %95 CI: 4.454-25.119) 6. The percentage of neutrophils ≥%65 (Odds ratio: 8.507; %95 CI: 3.686-19.630). The predicted probability of septic arthritis for these six predictors was determined as 4% for zero predictor, 11% for one predictor, 27% for two predictors, 52% for three predictors, 75% for four predictors, 90% for five predictors, and 96% for six predictors. Conclusion: Among all variables tested in the study, CRP is the best independent predictor of septic arthritis in children with high sensitivity and specificity. In children with acute monoarticular arthritis, the presence of 3 or more criteria tested in the study may predict the diagnosis of septic arthritis with the possibility of 50% or greater.