The clinical detection of scleral icterus: observations of multiple examiners

Sixty-two medical observers at various levels of medical training examined six patients for the presence or absence of scleral icterus in a double-blind survey. At a total serum bilirubin of 42.8 mmol/l (2.5 mg/dl) and 53.0 mmol/l (3.1 mg/dl), 58% (95% confidence interval [CI] 33-80%) and 68% (95% CI 46-85%) of examiners detected the presence of scleral icterus, respectively. Level of training appeared to influence specificity, as 6 of 8 apparent false-positives (total serum bilirubin = 12.0 mmol/l [0.7 mg/dl] and the presence of scleral icterus) were attributable to medical students. Level of training did not appear to influence sensitivity, as 3 of 11 apparent false-negatives (total serum bilirubin = 66.7 mmol/l [3.9 mg/dl] and the absence of detection of scleral icterus) were not attributable to medical students (p = 0.278). Even at the often-quoted estimate of 42.8 mmol/l (2.5 mg/dl) at which at scleral icterus is detected, approximately one-third of medical examiners in our study did not detect scleral icterus.