The superior accuracy and the less variation of the FG protocol is most likely related to the control of all the drilling steps and the implant placement via sequential use of precision sleeves. This eliminated the manual orientation and handling of the drills at any stage of drilling or implant placement. In accordance with these observations, Noharet et al. reported a better accuracy of the FG protocol compared with the conventional surgical guide [5]. Likewise, Vermeulen found the FG protocol to be more accurate than the FH protocol [19]. Further, several clinical studies reported that the PG protocol is associated with approximately double the errors of the FG protocol [16, 20, 21]. On the contrary, the PG protocol inaccuracy seemed comparable to the FH protocol, which could be due to the execution of most of the drilling steps and implant placement without the use of guides, leading to inevitable deviation of the drills and implant placement. This is further accentuated in the hands of inexperienced operators who may not precisely control the subsequent drilling steps [14, 15]. While the actual difference between the FG and PG protocols in all the variables is minimal, and still within the recommended safety zone of 2 mm [17], it can still be of clinical significance in cases where the available bone is limited, surgical site is compromised, and the implant is in close proximity to natural teeth and vital anatomical structures [17]. Further, clinically, this will be accentuated with the self-taping abilities of implant threads and inhomogeneity of natural bone that can lead to more implant deviations [20, 21]. Thus, where great accuracy is desirable, the clinician should aim to complete all the drilling procedures and implant placement through the guide [7].

The observed accuracy of the FG protocol (approximate vertical deviation = 0.4 mm, neck deviation = 0.5 mm, apex deviation = 0.7 mm, and angle deviation = 2.5^o) confirms earlier studies [7, 17, 18] that reported neck deviation in the range of 0.4–0.9 mm [5, 19, 22], apex deviation of 0.5–1.2 mm [5, 19] and angle deviation of 0.3–4.0° [5, 21, 22]. Thus, despite the technological advancement, the FG protocol is still prone to error [4, 7, 9, 12, 18] that is an accumulation of deviations introduced from every step of the planning, guide fabrication, and implant placement procedures [17]. For example, the planning process involves scanning and segmentation of the oral and vital tissues, and any deficiency of the resolution will influence the accuracy of the virtually designed guides [23, 24]. Guides are produced from 3D printing or milling, and both fabrication techniques are susceptible to surface and dimensional errors [6, 25] that may affect the intraoral fit and sleeve orientation. A study that specifically evaluated the errors in the production of guides found that the sleeve centers deviated in the range of 0.07 mm to 0.38 mm, and the angle deviated in the range of 0.4°–3.3° [26]. Nevertheless, the greatest errors seem to occur during the surgical procedure. For example, improper seating of the guide and the deformation of the guide inside the mouth [9, 27]. The deformation of the guide from the present study and from previous studies seem to be more prominent on the buccolingual direction [4, 12, 21, 22], which does not have a rigid structure such as the teeth at the mesiodistal direction. The mechanical tolerance between the drills and the interchangeable sleeves can further contribute to implant deviation [23, 28, 29]. A recent study reported that the length of the sleeve and the drilling distance influenced the accuracy of guided surgery [18]. Further, the presence of debris within the osteotomy can prevent complete seating of the implant [14], which was observed in our study and another study, where the FG protocol implants were more coronal than the planned implants [12]. In clinical situations, more errors are anticipated from CBCT and 3D segmentation of the hard tissues prior to virtual implant planning [23, 24] and patient-related factors such as movement, limited visibility due to the presence of blood, and limited visual access [7, 8].