Appendicitis, a common cause of abdominal pain in children, occurs more frequently in male children and peaks in incidence among older children and adolescents. Many children who have surgery for presumed appendicitis are found to have a normal appendix. Perforation is a common finding at surgery, especially in children younger than six years. Early and accurate recognition of appendicitis minimizes morbidity and mortality. Paris and Klein performed a systematic review of the literature concerning the value of the history, physical examination, and preliminary investigations in accurately diagnosing appendicitis in children.

Because no systematic reviews were available, the authors reviewed studies that compared diagnostic efforts with histologically proven appendicitis. The length of time that abdominal pain has been present is not clearly diagnostic of true appendicitis, but migration of the pain to the right lower quadrant increases the likelihood ratio for appendicitis. The accompanying symptom of vomiting increased the likelihood ratio for appendicitis.

Physical findings more commonly associated with appendicitis include right lower quadrant rebound tenderness and, to a lesser degree, percussion tenderness. The presence of guarding and rigidity also increased the likelihood ratio of the diagnosis. Important investigations include the diagnostic reliability of rising white blood cell (WBC) counts, especially a count of over 15,000 per mm³ (15 ×10⁹ per L) which increases the likelihood ratio of appendicitis to 7.0. A low WBC count makes this diagnosis unlikely.

Ultrasonography appears to be useful in children with appropriate symptoms in whom the diagnosis is uncertain. It does not appear useful in children with very weak or very strong evidence of appendicitis because of the significant possibility of false-positive results. Although computed tomographic (CT) scanning is accurate in predicting the presence or absence of appendicitis, its added value in children with very strong evidence for or against appendicitis is unclear.

The authors conclude that the presence of two or more characteristic predictors (vomiting, right lower quadrant pain, abdominal tenderness, and abdominal guarding) increases the likelihood ratio of appendicitis, with rebound tenderness increasing the likelihood ratio even more. The presence of an elevated WBC count adds to the suspicion of appendicitis.

Abdominal ultrasonography can help to identify children who should undergo surgery and children who can safely be observed. If more information is needed to make a treatment decision, a CT scan may provide more data, but it may cause delays in diagnosis and involves radiation exposure, hazards, rectal contrast, and possible sedation.

editor's note: Problems with accurate clinical diagnosis of appendicitis may be caused by an exaggerated emphasis on pain and tenderness. Rectal tenderness is not a highly sensitive or specific predictor of appendicitis. Other characteristics prominent with proven appendicitis include an inflammatory response, vomiting, and a prolonged duration of symptoms. Unfortunately, the use of CT, ultrasonography, and laparoscopy has had an equivocal effect on the misdiagnosis of appendicitis, leading to unnecessary appendectomy. Clinical algorithms that include characteristic symptoms with clinical findings are needed. The Ohmann score is a good example of a potentially useful decision tool. Diagnostic testing is most useful in select, equivocal cases and is highly operator dependent.—r.s.