부산대학교 도서관

OBJECTIVE: Acromegaly, a disease caused by GH/IGF-I hypersecretion, is associated with a high mortality rate; early recognition is therefore necessary to ensure successful treatment and to avoid comorbidities. We have created a symptom/sign scoring tool (ACROSCORE) for physicians to use to identify acromegaly. DESIGN: To compare cases of acromegaly diagnosed between 1990 and 2014 against a control group affected by non-GH-secreting pituitary tumours to identify symptoms and signs that are most discriminative for acromegaly. PATIENTS: Confirmed acromegaly patients and patients affected by non-GH-secreting pituitary tumours. MEASUREMENTS: In all patients, signs, symptoms and comorbidities were recorded from medical records and collected using a specifically designed questionnaire. RESULTS: A total of 194 acromegaly patients [115 women; mean (SD) age 47·2 (14·2) years] and 243 patients affected by non-GH-secreting pituitary tumours [131 women; mean (SD) age 45·8 (15·8) years] were included. A strong association was observed for type 2/secondary diabetes [odds ratio (OR) 3·7], hyperhidrosis (OR 6·1), thyroid hyperplasia (OR 13·9), colorectal polyps (OR 10·4), spaced teeth (OR 25·4) and carpal tunnel syndrome (OR 4·3). Based on this information, a multivariable logistic model was built and a 14-point scoring system developed. A score of 0 excludes the risk of acromegaly [positive predictive value (PV) = 0·6%]; scores 1–5 comprise a grey area; scores >5 indicate that a diagnosis of acromegaly cannot be excluded (PV = 46·1%). CONCLUSIONS: Once validated in independent studies, ACROSCORE may represent a new tool for the clinical screening of acromegaly that can be used by general practitioners and nonendocrinology specialists.