Eighty impressions were made for 40 patients, using the open, then the closed impression techniques. There were 18 impressions in the maxillary and 22 in the mandibular arch; of these, 13 were in the anterior and 27 in the posterior region (Fig. 3).

A t test indicated no statistically significant difference between the open and closed tray techniques for intraoral horizontal measurements against the similar horizontal measurements on the master casts (Table 1). A normality line test, for intraoral and master casts readings (horizontal measurements), revealed that they were not normally distributed (Fig. 4).

The Wilcoxon signed-rank test revealed that there are no statistically significant differences between the open and closed tray implant impression techniques (Table 2). The Mann-Whitney U test also showed no statistically significant differences for maxillary and mandibular impressions in both the open and closed tray techniques (Table 3).

The Mann-Whitney U test evaluated impression accuracy in the horizontal measurements according to the arch. There were no statistically significant differences between maxillary and mandibular arches, for the open and closed tray technique (Table 4).

The Mann-Whitney U test also showed no statistically significant difference in the horizontal measurement between the anterior and posterior regions for the open tray impression technique (Table 5). However, statistically significant differences were detected in the horizontal measurements, between the anterior and posterior regions in the closed tray impression technique (Table 4).

The marginal discrepancy evaluation and percentages by arch and region are presented in Fig. 5. The chi-square test associated marginal discrepancies between maxillary and mandibular, and anterior and posterior regions. There were no statistically significant differences in the marginal discrepancy between maxillary and mandibular arches, and anterior and posterior in both open and closed tray impression techniques (Table 6).