A comparison of the Berlin and Gent Nosologies in the diagnosis of Marfan syndrome: the NIH experience

Abstract

We reviewed clinical and radiographic data on 73 patients evaluated for possible Marfan syndrome (MFS) at the National Institutes of Health for the purpose of comparing the 1987 Berlin and 1996 Gent diagnostic criteria. All patients had a physical examination, all had echocardiograms or known aortic root dilatation, 64 (88%) had ophthalmologic exams or known ectopia lentis, and 35 (48%) have had MRI scans to screen for dural ectasia. Thirty-one had a first degree relative who had been diagnosed with MFS, and we directly confirmed this diagnosis in 22 cases. Forty-eight patients met diagnostic criteria under the Berlin criteria and 39 under the Gent criteria (kappa correlation coefficient K=0.75). No patient who met diagnostic criteria under the Gent criteria failed to meet the Berlin criteria. Of patients diagnosed under the Berlin criteria, all had skeletal features of MFS, 40 (83%) had aortic root dilatation, 19 of 27 examined (70%) had dural ectasia, 27 (56%) had an affected first degree relative, and 12 of 44 examined (27%) had ectopia lentis. Determination of dural ectasia established the diagnosis for 9 patients under the Gent criteria but had no effect on diagnoses made under the Berlin criteria (K=0.53 comparing Berlin to Gent without dural ectasia data). In summary, approximately 20 percent of patients diagnosed with Marfan syndrome using the Berlin criteria do not meet the Gent diagnostic standard. Long term follow-up or molecular diagnostic data are necessary to establish the relative sensitivity and specificity of these diagnostic tools.