Although the application of ultrasounds in endodontic surgery allows for effective debridement of the root canal, incorrect device setting or inefficient tips seem to generate cracks during root-end retropreparation. The primary aim of this in vitro study was to establish the presence, or absence, of a correlation between ultrasonic root-end preparation and the formation of cracks. The present study was conducted on human teeth, extracted for periodontal reasons. After root canal treatment, roots were resected 3 mm from the anatomical apex by using a high-speed handpiece and carbide burs. The resected teeth were retroprepared by using an ultrasonic tip (R1D, Piezomed, W&H, Bürmoos, Austria), setting the piezoelectric device at maximum power available for the tip. Time required for the retropreparation was recorded. Before and after retropreparation, all roots were photographed under a stereomicroscope and analyzed by two different operators to evaluate: (a) the presence and extension of dentinal cracks and (b) the morphology of root-end preparation. Finally, piezoelectric tips were analyzed by scanning electron microscopy (SEM) to evaluate morphologic changes after use. A total of 43 single roots (33 with one root canal, 10 with two root canals) were treated. Average preparation time was 1 minute and 54 seconds. None of the roots without initial cracks developed new cracks after retropreparation. Quality of the preparation margins was fairly equal among the prepared specimens. None of the piezoelectric tips broke during instrumentation, and SEM analysis showed minimal surface wear of the tips after performing 11 retropreparations. Within the limits of the present study, the tested piezoelectric system does not seem to represent a major cause for root crack formation. Pre-existing cracks may expand after ultrasound root-end preparation.