Prospective or retrospective studies that derived, validated, or assessed the impact of clinical prediction rules were included. Each rule had to include three or more predictive variables from the history, physical examination, or simple diagnostic tests. Eligible studies had to be of children (birth to 18 years old) with suspected appendicitis of less than one week duration.
The included studies derived or validated six clinical prediction rules (Alvarado, Kharbanda, Lintula, modified Lindberg, Paediatric Appendicitis Score – PAS, and van den Broek). Study cohorts were from the USA, the UK, Canada, Finland, the Netherlands, Spain, or Italy. Two studies were multicentre. None of the studies reported the clinical impact of the prediction rule, five reported a derivation, and seven described validations. One study was conducted on a hospital ward, the others were in emergency departments. The most common predictors were migratory right iliac fossa pain, nausea or vomiting, right lower quadrant tenderness, rebound tenderness, and temperature (>38°C). Diagnosis was confirmed by pathology and follow-up, in most studies; some used pathology alone.
Two reviewers independently selected studies for inclusion and discrepancies were resolved by consensus.