AbstractIn monogenic disorders, correlation between genotype and phenotype is a premise for predicting prognosis in affected patients. Predictive genetic testing may enable prophylaxis and promote clinical follow‐up. Although Marfan syndrome (MFS) is known as a monogenic disorder, according to the present diagnostic criteria a mutation in the gene FBN1 is not sufficient for the diagnosis, which also depends on the presence of a number of clinical, radiological, and other findings. The fact that MFS patient cohorts only infrequently have been examined for all relevant phenotypic manifestations may have contributed to inconsistent reports of genotype–phenotype correlations. In the Norwegian Study of Marfan syndrome, all participants were examined for all findings contained in the Ghent nosology by the same investigators. Mutation identification was carried out by robot‐assisted direct sequencing of the entire FBN1 coding sequence and MLPA analysis. A total of 46 mutations were identified in 44 unrelated patients, all fulfilling Ghent criteria. Although no statistically significant correlation could be obtained, the data indicate associations between missense or splice site mutations and ocular manifestations. While mutations in TGF‐domains were associated with the fulfillment of few major criteria, severe affection was indicated in two cases with C‐terminal mutations. Intrafamilial phenotypic variation among carriers of the same mutation, suggesting the influence of epigenetic facors, complicates genetic counseling. The usefulness of predictive genetic testing in FBN1 mutations requires further investigation. © 2007 Wiley‐Liss, Inc.