Appr opriateness and Diagnostic Yield of Referrals for Oesophagogastroduodenoscopy at the Korle Bu Teaching Hospital Pertinence et rendement diagnostique de références pour oesophagogastroduodenoscopy à L'hôpital Korle Bu Enseignement

2011 
BACKGROUND: Increasing endoscopy workload in openaccess services necessitates adoption of appropriatene ss criteria to check abuse and improve yield. OBJECTIVE: To assess the appropriateness of referrals for oesophagogastroduodenoscopy (OGD) and its relationship to yield at Korle-Bu Teaching Hospital (KBTH), Accra. METHODS: Referrals, signs, and symptoms of 375 consecutive patients for diagnostic oesophagogastroduodenoscopy were evaluated over four months. Indications were categorized as appropriate or inappropriate using the American Society of Gastrointestinal Endoscopy (ASGE) guidelines and endoscopic findings (yield) categorized as positive or negative. The relationship between these was analyzed and the diagnostic accuracy of the guidelines determined. RESULTS: There were 209 (55.7%) females, and 316 (84.3%) open-access procedures. Mean age was 46 ± 17 years. Dyspepsia, 272 (72.5%) and epigastric tenderness, 192 (41.4 %) were the commonest symptom and sign respectively. Only 133 (35.5%) reported alarm symptoms. Appropriate referrals constituted 221(58.9%). Inappropriate referral rate was similar for endoscopists and non-endoscopists. Positive yield was 62.7%. Male sex, age > 45 years, haematemesis, persistent vomiting, gastroenterologists’ referrals and epigastric tenderness were the best predictors of positive yield. Gastritis, 121 (32.3%), duodenal ulcer, 48 (12.5%) and oesophagitis, 36 (9.6%) were the leading endoscopy diagnoses. Carcinomas were reported only after 45years and 18 (81.8%) of the cases had alarm symptoms. CONCLUSIONS: Inappropriate referral for OGD rate is high in Accra. Yield is impr oved by adher ence to the ASGE guidelines but its accuracy as a screening tool for OGD at Korle Bu Teaching Hospital is too low to recommend it for adoption. WAJM 2011; 30(3): 158–163.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    15
    References
    0
    Citations
    NaN
    KQI
    []