We aimed to evaluate the feasibility and tolerability of fundus examination in preterm newborns after implementing a premedication with intranasal midazolam as a quality improvement project in a neonatal intensive care unit (NICU). Prospective examination of all fundus examinations between January and June 2022, before 0 (T0) and after 1 (T1), 15 (T15), and 60 (T60) minutes from intranasal midazolam administration. The EDIN (Echelle de Douleur et Incomfort du Nouveau-né [newborn pain and discomfort score-in French]) pain score was calculated and various physiologic parameters were recorded. Data were analyzed with repeated measures ANOVA (analysis of variance). A total of 36 fundus examinations were performed in 27 noninvasively ventilated patients. The mean EDIN score went from 0.8 ± 1 at T0 to 0.72 ± 1.05 at T1, then to 0.22 ± 0.59 at T15, and to 0 ± 0 at T60 (p < 0.001). Thus, it remained below the threshold of 5, which is associated with significant pain. The mean heart rate decreased from 164 ± 16 at T0, to 161 ± 16 at T1, then to 154 ± 14 at T15, and to 153 ± 12 at T60 (p < 0.001). There was no significant change in the oxygen saturation/inspired oxygen fraction ratio across the four time points (p = 0.202) and the mean arterial pressure was similar before and after the administration of midazolam (60 ± 10 vs. 59 ± 10, p = 0.571). Intranasal midazolam provides good comfort and normal physiological parameters; therefore, the implementation of a NICU protocol to improve quality of retinal examinations seems feasible and well tolerated. · Fundus examinations did not cause pain after premedication with intranasal midazolam.. · No respiratory or circulatory adverse effect was reported.. · Implementing such a premedication protocol in the NICU seems feasible, helpful, and well tolerated..