ePoster
Talk Description
Institution: St John of God Midland Public and Private Hospitals - WA, Australia
Purpose
Negative appendicectomy (NA) is associated with similar morbidity to positive appendicectomy (PA) with greater diagnostic uncertainty in paediatric patients due to reliance on ultrasound. We aimed to determine the NA rate at our centre, identify predictors of NA and PA and assess predictive validity of combining Alvarado score with CRP,
Methodology
A retrospective cohort study was performed including all paediatric patients (12-16 years) undergoing emergency appendicectomy between June 2022 and October 2025 with a matched adult cohort randomly selected. Demographics, clinical features, investigations, Alvarado scores, histopathology, length of stay, and complications were collected. Comparative statistical analysis was performed.
Results
150 patients were included (76 paediatric). NA rate was 6.6% in paediatrics and 12.2% in adults. WCC discriminated PA from NA in paediatric (median 8.0 vs 13.1, p=0.003) and adult patients (median 7.8 vs 14.0, p=0.009), while CRP was discriminatory in paediatric patients only (median 1 vs 19, p=0.023). Alvarado scores were significantly higher in PA (paediatric median 3 vs 6, p=0.001; adult median 5 vs 6, p=0.031), with all paediatric NA confined to patients with scores <5. Ultrasound was commonly performed in paediatrics but did not discriminate NA from PA. CT use was infrequent in paediatrics but favoured PA in adults (p=0.006). In paediatric patients, combining Alvarado score ≥7 with CRP >5 demonstrated high specificity and positive predictive value (both 100%) but low sensitivity (29.6%), with wide confidence intervals reflecting limited case numbers. Length of stay and complication rates did not differ between NA and PA.
Conclusion
This study demonstrates a low paediatric NA rate. Inflammatory markers and Alvarado score provide useful discrimination, while ultrasound alone is insufficient. Combined Alvarado–CRP strategies may improve specificity, although larger cohorts are required to confirm predictive validity.
Speakers
Contributors
Authors
Dr David Singer - , Mr Warren Raymond - , Dr Christopher Lau - , Dr Ruwan Wijesuriya -
