Purpose: In the future, it is expected that treatment of Dupuytren disease (DD) may shift toward control of early disease. Ultrasound might be an accurate method to measure the outcome of such treatment. The aim of this study was to assess the reliability of sonographic measurement of palmar nodules. Methods: Fifty patients with nodules characteristic for early disease were assessed with ultrasound by 2 observers. Four different aspects of DD nodules were measured in the transversal and sagittal planes, width, depth, circumference, and area. The intra- and interobserver reliabilities were calculated using the intraclass correlation coefficient (ICC). The standard error of measurement (SEM) and the smallest detectable change (SDC) were also calculated for each aspect. Results: The intraobserver reliability was good (ICC, 0.724 [0.562–0.833] to 0.886 [0.808–0.934]), except for width in the sagittal direction (ICC, 0.671 [0.484–0.799]). The interobserver reliability was moderate (ICC, 0.385 [0.126–0.596] to 0.757 [0.538–0.869]). The intraobserver ICCs of area were highest (transverse, 0.847 [0.744–0.893]; sagittal, 0.886 [0.808–0.934]). The SEM and SDC of area were 6.1 and 16.9 mm2 in the transverse and 8.0 and 22.2 mm2 in the sagittal plane. Conclusions: The intraobserver reliability of sonographic assessment of DD nodules is good. The measurement of area is the most reliable and is, therefore, recommended for future studies. However, even single-observer measurements have a clear dispersion, and a change beyond 16.9 (61%) and 22.2 mm2 (79%) has to be observed in the transverse and sagittal planes, respectively, before it can be considered as regression or progression. Clinical relevance: Repeated ultrasonographic measurements in DD should ideally be done by a single observer, using area of the nodule in the sagittal plane. Change beyond 16.9 (transverse) and 22.2 (sagittal) mm2 can be considered as a real change in nodule size.