Objectives: This in vivo study evaluated the reliability of digital impressions from intraoral scans compared to digital scans of plaster casts of the edentulous maxilla of infants with cleft lip and palate (CLP). Materials and methods: To compare and evaluate the dimensional accuracy of digital and conventional impressions, 52 infants with CLP were selected. Conventional plaster and digital impressions were taken from each patient before surgical closure of the lips and/or palate. Plaster casts were digitized using the intraoral scanner. Eight linear distances were measured and 3D superimpositions were performed. Pearson’s correlation, mean difference and root mean square were calculated to compare the two impression methods. Results: Regardless of the form of CLP, the IOS and the plaster casts correlated very well. The measured linear dimensions of the IOS were greater than those of the digitized plaster casts. Although the linear measurements showed differences of up to 0.89 mm, superimpositions showed that the two impression methods did not differ in a clinically relevant way. Conclusions: No clinically relevant difference in accuracy was found between IOS and the plaster casts. When investigating the accuracy, 3D analysis should always be preferred to 2D analysis. Clinical relevance: This study confirms that IOS is a reliable and clinical valuable alternative for infants with CLP. IOS eliminates risks associated with traditional methods, providing a safer and more efficient workflow for PSIO. Despite minor differences in dimensional accuracy, these variations are clinically negligible and do not impact treatment planning or appliance fabrication.