The aim of this prospective study was to compare implant success rate and crestal bone loss around tilted and straight implants supporting immediate-loading full-arch rehabilitations. Twenty consecutive patients with edentulous jaws treated between June 2013 and July 2015 who satisfied all inclusion and exclusion criteria were included in the study. All patients were rehabilitated through a full-arch restoration supported by 4 or 6 immediately loaded implants. Clinical and radiographic examinations were scheduled every 12 months to evaluate implant success rates and crestal bone levels. Significant differences in crestal bone levels and success rates between straight and tilted implants were investigated by means of independent statistical analysis; differences were regarded as significant if P < .05. Seventy straight and 50 tilted implants were placed to rehabilitate 14 mandibles and 12 maxillae in 20 patients. After a follow-up of 12 to 36 months, survival rate was 97.1% for straight implants and 96.0% for tilted implants; while success rates were 94.3% and 94.0%, respectively. Success and survival rates were not significantly different (P > .05). Change in crestal bone level was 0.5 ± 0.4 mm for straight implants and 0.6 ± 0.4 mm for tilted implants (P > .05). Straight and tilted implants seemed to have similar behavior after immediate loading rehabilitations. After functional loading, straight and tilted implants did not differ significantly in clinical outcome.
Rehabilitation of the posterior maxilla is often challenging because of reduced bone volume caused by long-term edentulism, a condition that may prevent long-term successful results.1 Several techniques have been proposed to manage alveolar atrophy, such as performing bone grafting or other augmentation procedures and placing short or zygomatic implants. Nevertheless, each of these procedures presents biological and technical drawbacks, such as morbidity at the graft donor sites, postoperative discomfort, questionable predictability, prolonged treatment time, and surgical complexity.2,3 Further, they usually prevent the surgeon from loading the implants immediately. Malo et al introduced the use of tilted implants as an alternative approach to treat totally edentulous patients. These authors suggested placing 4 or 6 implants to support a full-arch fixed rehabilitation and named the solutions the “all-on-4” or “all-on-6” techniques. These procedures involve placing 2 or 4 straight implants in the anterior region and 2 implants, tilted at 35° to 45°, in the posterior region.4–8
Tilting the posterior implants allows for placement of longer implants. Thus, the contact area between the implant and the bone increases, enhancing the implant primary stability. The implant support is more distal, and the distance between implants is greater than when straight implants are placed, thus reducing or even eliminating the cantilever length. This provides a better load distribution and optimizes the anterior/posterior spread of the implants along the alveolar ridge.4,9
Agliardi et al10 suggested a similar technique, named “V-II-V,” which consists of rehabilitating the patient by delivering an immediately loaded full-arch bridge supported by 6 implants. This approach involves placing 2 distal implants, tilted at 30° to 45° relative to the occlusal plane, into the posterior wall of the maxillary sinus. Two other tilted implants engage the anterior sinus wall, and finally, 2 straight implants are inserted into the anterior maxilla.
Since these surgical approaches showed optimal results over short-term and long-term follow-up, they were regarded as reliable to carry out an immediate or delayed rehabilitation of the edentulous maxilla, avoiding bone grafting or other augmentation procedures.11–23
Yet concerns remain about the success rate and crestal bone loss around tilted implants compared with straight implants. Chrcanovic et al23 published a systematic review and a meta-analysis to evaluate implant failure rate, marginal bone loss, and postoperative infection for patients being rehabilitated by tilted or straight dental implants. They suggested that the differences in angulation of dental implants might not affect implant survival or marginal bone loss but underlined how the matter should be the subject of further investigation given the potential for biases and confounding factors that may affect this type of evaluation. The aim of the present study was to compare the implant success rate and crestal bone loss around tilted and straight implants supporting immediate-loading full-arch rehabilitation after a 1- to 3-year follow-up. The aim of this study was to answer the following question: Is there any differences regarding implant success rate and crestal bone loss around tilted or straight implants in patients rehabilitated by the all-on- 4 or all-on- 6 technique?
Materials and Methods
Twenty consecutive patients were enrolled among those referred to the authors from June 2013 to July 2015 for a full-arch rehabilitation of the maxilla or mandible because of total or partial edentulism. All patients who satisfied the inclusion and exclusion criteria were included: patients presented moderate or advanced alveolar bone resorption in the posterior areas of the jaws and had an indication for a rehabilitation plan calling for placing straight and/or tilted implants and delivering an immediate-loaded full-arch prosthesis. The rehabilitation plan was discussed with each patient who provided informed written consent.
After enrollment, each patient received an informative sheet detailing the protocol and treatment (MN protocol code: Straight/Tilted implants). Inclusion criteria were as follows: good health; need for a full-arch implant-supported restoration; adequate bone volume; sufficient oral hygiene; nonsmoker or light smoker (≤10 cigarettes/day); and able to sign the informed consent, understand the study protocol, and participate for the whole duration. Patients were excluded if they presented insufficient bone volume in the anterior regions, showed bruxism and other parafunctions, smoked more than 10 cigarettes a day, abused alcohol, were concomitantly subjected to radiotherapy in the maxillofacial district, were undergoing chemotherapy, had uncontrolled systemic diseases or immunodepression, were pregnant, or presented insufficient oral hygiene.
Patients who met the inclusion criteria were informed about the objectives and conditions of the study. Each patient received written information and provided written informed consent before participating in any study-related procedure. The study was conducted in accordance with all relevant principles of the Declaration of Helsinki and STROBE guidelines.
All patients had a professional oral hygiene treatment before implant surgery and were then treated according to a previously established radiographic, surgical, and prosthetic protocol.
Two hours before surgery, antibiotic prophylaxis was begun by administering 2 g amoxicillin and clavulanic acid. The surgical area was anesthetized using articaine hydrochloride 4% with epinephrine 1:100 000. Teeth with an unfavorable long-term prognosis were extracted. A crestal incision was performed starting from the posterior region; full-thickness buccal and lingual flaps were raised, exposing the residual alveolar ridge, allowing the surgeon to locate the anterior and posterior walls of the maxillary sinus in the maxilla and the mental nerve in the mandible.
Each patient received 4 to 6 expanding tapered design implants that were placed straight or tilted in the mesial or distal direction at 30° to 45° relative to the alveolar ridge according to their position and the bone volume available at the placement site: (1) the 2 medial implants were placed straight in the positions corresponding to the lateral incisors or canines; (i2) 2 more straight or tilted distal implants were inserted into the anterior maxillary sinus wall (in the maxilla) or anterior to the mental nerve (in mandible); (3) the last 2 distal implants could be placed either straight, if in the mandible, or tilted mesially into the posterior sinus wall or the pterygoid plates in the maxilla. The tilting angle was calculated by subtracting 90° to the angle created by a line parallel to the occlusal plane and a line parallel to the long axis of the implants.24 Implant lengths were between 8 and 14 mm and diameters between 3.7 and 4.1mm. All implants integrated platform switching, self-tapping tapered design, and double-acid-etching surface. Postextraction sockets and peri-implant defects were grafted using autogenous bone chips harvested from drilling burs.
After implant placement, angulated multiunit abutments were connected on tilted implants; while standard abutments were connected to the straight implants. Finally, an impression was taken using a custom-made tray, and a plaster cast was fabricated.
Provisional prostheses were delivered within 72 hours (Figure 1). The patient was advised to rinse twice daily with chlorhexidine digluconate 0.2% during the first 2 weeks.
Definitive prostheses were delivered 4 to 12 months later, after achieving proper soft and hard tissue healing (Figure 2). All definitive restorations were placed in occlusion. The occlusal surface was thoroughly modeled, so that it was in contact with reduced areas during laterality and protrusion excursions, to reduce the dislocating vector components; several contacts were maintained in maximum intercuspidation. The patient was included in a maintenance program and underwent professional oral hygiene every 6 months. Clinical and radiographic examinations were scheduled every 12 months after functional loading (Figure 3a through c).
Patients were questioned orally and in writing regarding name, age, sex, medical history, dental history, smoking habits (number of cigarettes per day), and number of professional oral hygiene procedures in the previous 3 years (number of hygiene procedures per year).
A clinical examination was conducted to evaluate the presence of chronic or aggressive periodontitis, presence of parafunctional habits (bruxism or clenching), and gingival biotype (thick or thin) and to integrate the treatment planning of patient.
For each implant, the authors recorded the following parameters at the time of surgery: jaw where the implant was placed (maxilla and mandible); site type (postextraction or healed); implant length (10, 12, or 14 mm); implant diameter (3.7, 4.1, or 4.8 mm); implant site (anterior, premolar, or molar); and insertion torque (IT). For each implant, the authors measured the crestal bone level (CBL) at baseline and at all follow-ups, calculating the corresponding peri-implant bone loss (PBL) with respect to baseline. We assessed CBL on intraoral radiographs collected for each placed implant using a parallel technique. Close attention was paid to proper positioning of the receptor and X-ray tube to obtain radiographs with the same field of view, the same projection and angulation, and the least possible amount of distortion/deformation. Moreover, if evidence of distortion, deformation, or other alterations was present, a new radiograph was taken to achieve an adequate image overlapping with previous images.
All radiographs were scanned, digitized in JPG format, converted to 600 dpi resolution TIFF images, and analyzed through an image analysis software (Image J, version 1.47, National Institutes of Health, Bethesda, Md) by 2 independent examiners. For each image, the software was calibrated using the known implant diameter at the most coronal portion of the implant neck. Them, CBL was measured considering the first contact point at the bone-implant interface: the distance from the implant-abutment interface to the most apical point of crestal bone observed to be in intimate contact with the implant was then measured to the nearest 0.1 mm at both the distal and mesial side of the implant, and the 2 values averaged. The PBL between 2 given time points was then calculated as the difference between the corresponding 2 crestal bone levels.
At all follow-ups, the authors also assessed implant osseointegration, success, and survival rates. Implant success rate was assessed according to the criteria described by Buser et al25 as modified by Albrektsson and Zarb26 including (1) absence of persistent pain, dysesthesia, or paresthesia in the implant area; (2) absence of peri-implant infection with or without suppuration; (3) absence of perceptible mobility of the implant; and (4) absence of persistent peri-implant bone resorption >1.5 mm during the first year of loading and 0.2 mm/year in subsequent years. Implants were considered successful when all the aforementioned conditions were met. Complications were also recorded, including soft tissue complications, significant bone loss, peri-implant radiolucency, and prosthetic complications.
Differences between tilted and straight implant concerning CBL, PBL, and success and survival rates were assessed considering the implant as the statistical unit of analysis. The reason for this assumption is that we considered the implants to be independent from each other as far as the treatment (implants inserted straight or tilted) was concerned. The whole analysis was performed by an independent statistician. Implants and corresponding data were divided into 2 groups, straight and tilted. To investigate if the average implant length was different between the 2 groups, implant lengths were averaged and compared by means of a t test for unpaired data. Differences, between the 2 groups, as far as implant primary stability and mean follow-up time were concerned, were assessed comparing the mean IT at insertion by means t tests for unpaired data.
To investigate the variation of CBL between the last follow-up and baseline, CBL values at the 2 time points were compared using t tests for paired data. Differences between straight and tilted implants concerning CBL and PBL were investigated by means of t tests for paired data. To investigate if CBL values varied differently at the mesial or distal side of tilted implants, they were compared at baseline and at the last follow-up by means of t tests for paired data.
Differences between straight and tilted implants in terms of survival and success rates were investigated using Fisher exact tests. Additionally, a correlation analysis was carried out to investigate the influence of sex (male/female), smoking habits (no smoker/smoking ≤10 cigarettes per day), a history of periodontitis (yes/no), jaw under treatment (upper/lower), the implant tilting (absent/present); placement site (healed/postextractive), implant length/diameter, IT on PBL, calculating the Spearman coefficients and corresponding significance considering the P value related to each R value.
Additionally, to investigate if insertion in postextractive or healed sites influenced CBL and PBL, CBL values and PBL at the last follow-up of implants placed in the 2 types of sites were compared by means of t tests for unpaired data. Moreover, PBL values at different follow-up periods were compared between patients treated with 4 implants and patients treated with 6, as well as between lower and upper arches, by means of t tests for unpaired data. Multivariate linear regression models were estimated to study the relationship between possible independent variables and the PBL response variable.
Variable selection was performed through backward elimination fixing a significance level of 0.15 as a removal criterion and forcing the type variable as a regressor during backward selection, when appropriate. Finally, multivariate linear regression models stratified by the type variable were estimated to assess whether the effects on PBL differed in the straight and tilted groups, respectively. Beta coefficients are reported as estimates of the effect. P values were used to assess statistical significance. Finally, model R2 is reported as goodness-of-fit model measure.
Interrater reliability analysis was performed by estimating intraclass correlation coefficients (ICCs) measuring consistency for 2-way mixed-effects model average measures. Results are published as ICC estimate (95% confidence interval) comparing CBL-mesial baseline and CBL-mesial follow-up, CBL-distal baseline, and CBL-distal follow-up, for straight and tilted. Normality was tested on CBL-mesial baseline, CBL-mesial follow-up, CBL-distal baseline, and CBL-distal follow-up for the straight and tilted groups before ICCs were calculated. Finally, Spearman rho coefficients were estimated.
The statistical methodology and outcomes were reviewed by a second statistical expert. Differences were regarded as significant if P < .05. Statistical analyses were performed using the R Integration package for SPSS software (version 16; SPSS Inc, Chicago, Ill).
Patients and implants
Twenty consecutive patients (mean age = 67 ± 6 years) with totally or partially edentulous jaws were treated according to the previously described protocol. All patients were referred to the authors from private offices, and no patient dropped out during the follow-up period. Each patient received 4 or 6 implants that were immediately loaded with a full-arch prosthesis. A total of 120 implants were placed to rehabilitate 12 maxillae and 14 mandibles. Of these, 34 implants were placed in patients with a smoking habit, while 86 implants were placed in nonsmokers. Forty-six implants were inserted in patients who had chronic or aggressive periodontitis. In total, 70 Straight implants and 50 tilted implants were placed. Of all the implants, 68 were placed in healed sites, and 52 were placed in immediate postextraction sites. The mean IT for all implants was 61.2 ± 23.2 Ncm.
In 17 patients, implants were loaded within 72 hours as planned, while in 3 patients loading was carried out after 5 days for technical reasons. The mean follow-up after functional loading was 26.5 ± 8.0 months. Mean CBL was 0.5 ± 0.3 mm at baseline and 1.0 ± 0.4 mm at the last follow-up. For all implants, mean PBL at the last follow up was 0.6 ± 0.4 mm.
Of the 70 straight implants 22 were placed in patients with a smoking habit and 48 in nonsmokers. Twenty-eight implants were placed in patients with previously treated periodontitis.
Thirty-nine implants were inserted in postextraction sites and 31 in healed sites. Fifty-two implants were placed in the anterior region, 12 in the premolar region, and 6 in the molar region. Most of the implants were 12 mm (n = 38) or 10 mm (n = 21) long; 7 implants were 8 mm long and 4 implants were 14 mm long. Implants diameters were between 3.7, 4.1, and 4.8 mm.
Mean IT at insertion was 62.1 ± 22.7 Ncm. The average implant length in this group was 11.1 ± 1.5 mm.
At baseline, mean mesial CBL was 0.6 ± 0.3 mm while mean distal CBL was 0.5 ± 0.3 mm. The mean CBL value was 0.5 ± 0.3 mm. After follow-up, the corresponding values were 1.2 ± 0.5 mm (mesial), 0.9 ± 0.4 mm (distal), and 1.1 ± 0.4 mm (mean). Mean PBL at the last follow-up was 0.5 ± 0.4 mm.
One implant failed osseointegration and one was removed because suppuration was present; 2 implants were considered failed because bone resorption exceeded the criteria for implant success. Consequently, the survival and success rates were 97.1% (68/70) and 94.3% (66/70), respectively. Details concerning failed implants are provided in Table 1.
Fifty implants were placed tilted with respect to the occlusal plane. Of these, 12 were placed in patients with a smoking habit. Eighteen implants were placed in patients who had had periodontitis. Thirteen implants were inserted in postextraction sites and 37 in healed sites. Thirty-two implants were placed in the premolar region and 18 in the molar region. Most of the implants were 14 mm (n = 30) or 12 mm (n = 20) long; no implants were ≤ 10 mm long. Implant diameters were between 3.7, 4.1, and 4.8 mm. The average IT at insertion was 62.7 ± 24.3 Ncm. The average implant length in the tilted group was 13.2 ± 1.0 mm.
At baseline, mean mesial CBL was 0.5 ± 0.3 and mean distal CBL was 0.2 ± 0.3 mm. Mean CBL was 0.4 ± 0.3 mm. After follow-up, the corresponding values were 1.1 ± 0.4 mm (mesial), 0.8 ± 0.5 mm (distal), and 1.0 ± 0.5 mm (mean). For these implants, mean PBL at the last follow-up was 0.6 ± 0.4 mm.
Two implants failed osseointegration and one implant was considered failed because of an excessive bone resorption. Consequently, the survival and success rates were 96.0% (48/50) and 94.0% (47/50), respectively. Details concerning failed implants are provided in Table 1.
The average implant length in the tilted group was significantly greater than that in the straight group (P < .001). Mean IT at insertion was not significantly different between the 2 groups (P = .71). Mean follow-up time was not significantly different between the 2 groups (P = .90).
The CBL at follow-up was significantly greater than that at baseline for all implants and for the 2 groups separately (P < .001 in all cases). This difference was also significant when stratifying the analysis for different follow-up periods (12, 24, and 36 months; P < .001 in all cases). At baseline, CBL was significantly smaller in the tilted than in the straight group (P = .002); no significant differences were observed at 12 (P = .91), 24 (P = .45), and 36 (P = .24) months of follow-up. The PBL at the last follow-up was not significantly different between the 2 groups (P = .44) (Table 2), even when considering the different follow-up periods separately (12 months, P = .46; 24 months, P = .62; 36 months, P = .81) (Table 3).
Concerning tilted implants, mean mesial and distal CBL values were significantly different at baseline and at 36 months of follow-up, but not at 12 and 24 months (12 months, P = .12; 24 months, P = .35; 36 months, P < .05), with mean distal CBL being smaller than mean mesial (Table 4).
When comparing patients treated with 4 or 6 implants, lower PBL was observed in the 6-implant group only at 12 months, whereas no significant differences were found at 24 or 36 months or without stratifying the analysis for the follow-up period (Table 5).
A further comparison of bone resorption between patients rehabilitated in the upper or lower arch was carried out, but no significant differences were observed (Table 6).
The interrater reliability analysis for the straight and tilted groups showed a moderate level of consistency agreement between CBL average measures at baseline and at follow-up (Table 7). Additionally, the estimation of ICCs revealed that baseline and follow-up CBL measures were highly correlated (Table 8), further supporting their reliability.
Success and survival rates were not significantly different between the 2 groups (P > .99 in both cases). Correlation analysis results are shown in Table 7. Correlations with the remaining variables were not significant and are not shown for sake of brevity. No significant correlation was observed between any of the variables assessed and the implant success and survival rate.
A significant negative correlation was observed between CBL and smoking habits, indicating that not smoking reduces, on average, the CBL; a significant positive correlation was also observed between CBL and the healing conditions of the extractive sites, indicating that implant insertion in postextraction sites is associated with a greater CBL; CBL also correlated significantly and positively with IT, indicating that a greater IT is associated with a greater CBL. As far as PBL was concerned, a significant correlation was only observed between PBL and sex (PBL showing an increase with men) and between PBL and diameter (a greater diameter was associated with a greater PBL). Even if significant, though, all correlations observed were weak. Remarkably, implant inclination (straight or tilted) did not correlate significantly with CBL or PBL. A multivariate analysis was also carried out to study the relationship between possible independent variables and the PBL response variable. The first multivariate linear regression model was adjusted for the variables type, length, and diameter and confirmed that implant type (straight versus tilted) has no influence on PBL. Instead, this first model revealed that an increase in implant length corresponded to a decrease in PBL, and that implants with a 4.1-mm diameter were associated with an increase in PBL (Table 9). When considering the straight and tilted groups separately, slightly different outcomes were observed. Indeed, in the straight group the association between 4.1-mm diameter and higher PBL was lost. Finally, partial edentulism was found to be related to a higher PBL (Table 9). In tilted implants, periodontitis and insertion in a postextraction socket were associated with an increase in PBL (Table 9).
The validity of statistical analysis was evaluated and confirmed by independent statistical consultants.
Bone resorption in edentulous patients and the consequent anatomic limitations, such as exposition of the alveolar nerve in the mandible or inferior sinus expansion in the maxilla, prompt clinicians to develop alternative surgical techniques to bone grafting or other bone augmentation procedures. Tilted instead of straight implants can be used to avoid anatomic structures, reducing patient discomfort and financial costs and shortening overall treatment time.17,25 In the maxilla, tilted implants can be placed mesially or distally to the maxillary sinus; in the mandible they can be inserted in intraforaminal regions.
Malò et al first proposed the all-on-4 concept. This involved inserting 2 implants parallel to the facial midline and 2 distal implants with a 35° to 40° angle. In the maxilla, the same authors suggested placing 6 instead of 4 implants because of the lower bone density and volume.6–8
The rationale for using tilted implants is that the vertical forces applied during function are supposed to cause more bone resorption than horizontal forces acting around tilted implants. The angulation of distal implants divides the occlusal forces in vertical and horizontal vector components, effectively reducing the distribution of load in the surrounding bone tissue.9 Furthermore, placing tilting implants in a reduced bone volume allows the surgeon to use longer implants engaging a greater quantity of residual bone, thus increasing implant stability.4,23
Concerning marginal bone loss, finite element analysis suggests that single tilted implants undergo a greater stress than straight implants, a condition that might increase the surrounding bone resorption.4,9,27–29 However, full-arch prostheses create a physical connection between straight and tilted implants that changes the distribution of loading forces, reducing crestal bone remodeling.4,9,13,24,27–32 In fact, implant splinting might limit implant micromotion favoring osteointegration.4,9 Further, Almeida et al33 showed that the loading stress around tilted implants is decreased compared with straight implants. Moreover, no statistically significant differences were reported considering the all-on-4 and all-on-6 protocols, confirming that the placement of a larger number of implants to support full-arch restorations is not strictly necessary for the successful rehabilitation of totally edentulous patients.9 Previous studies reported encouraging results and high implant success and survival rates for these procedures even compared with other surgical protocols.9–22 However, most studies lacked a follow-up longer than 3 years. As far as more recent results are concerned, Chrcanovic et al23 published a systematic review and a meta-analysis to determine survival rates, postoperative infection, and CBL of tilted and straight implants. Survival rates ranged from 88.1% to 100%, without significant differences between straight and tilted implants.23 Peñarrocha-Oltra et al26 described implant success rates varying from 91.3% to 100% for 666 straight implants and 92.1% to 100% for 782 tilted implants evaluating 319 patients.
Results of the present study are consistent with the previously cited studies concerning the lack of significant differences in peri-implant bone resorption and implant survival and success rates between straight and tilted implants. Further, results of the present study are consistent with those of Bellini et al34 concerning the immediate rehabilitation of the maxilla; they observed no significant differences in success rates between straight and tilted fixtures.
Further, results of the present study seem to indicate that CBLs of distally tilted implants may decrease at a greater rate at the distal than the mesial side. This should be investigated in additional studies and over a longer follow-up. This could be explained by the placement of the tilted implants: the upper part of the neck is at the bone level and therefore the lower part is below the bone for most tilted implants, while all straight implants were placed at the crestal bone level. On the other hand, both tilted and straight implants are splinted together and have a similar functional behavior as functional units and not as isolated elements. As a consequence, it is not possible to apply the present results to single-unit implant behavior.
In the patients enrolled in the present investigation, mucositis and peri-implantitis prevalence rates observed during the follow-up period were lower than those observed by other authors who reported a prevalence of around 80% for mucositis and 10% for perimplantitis.35–39 A possible explanation might be that all patients in the present study were motivated to keep an elevated level of oral hygiene and were recalled every 6 months to have their prostheses removed for disinfection and the emergence profile corrected at all control visits.
As said, in the present study no significant differences were observed between tilted and straight implants concerning the end points of interest. These results, however, should not be generalized as many factors may influence the clinical outcome. These results, however, should not be generalized as many factors may influence the clinical outcome such as the limited number of patients, absence of sample size, and short follow-up. Due to the absence of sample size calculation, the statistical power of this study was less than 80%, and this reduces the capacity to detect significant differences between the study groups. Moreover, a small sample may not be representative of overall population of this kind of patient. In addition, although the main bone resorption occurs during the first year, and considering the rather high variability in the follow-up period observed in this study (± 8 months), a longer follow-up is recommended to evaluate differences over time. The absence of randomization was due to clinical reasons, such as the need to place tilted implants in posterior regions; as a consequence, a high variability in size and other characteristics was observed between test and control groups.
Furthermore, the presence of hopeless teeth and consequent postextraction sites at the time of implant placement introduced an adjunctive variable that may affect the results. In fact, CBL in postextraction and native sites were analyzed as influencing factors. Also, the different degree of angulation of the tilted implants, ranging from 30° to 45°, may influence CBL in the test group, but this variable was not measured. It will be interesting to evaluate the influence on CBL under functional loading over the time.
On the other hand, the study was carried out in private offices and so these results can be applied to clinical practice in the rehabilitation of edentulous patients. It is worth stressing that both techniques were tested in real clinical conditions, and patient inclusion criteria were particularly restrictive; therefore, the results of the present study could be generalized to a larger population with similar characteristics. In addition, other strengths related to the study protocol and design were the treatment of consecutive patients, use of the same surgical and prosthetic protocol, and the absence of dropouts and protocol deviation.
Moreover, the analyses were carried out by 2 independent examiners, and statistical analyses was performed by a independent statistical society to guarantee the rightness of data Since the examiners used implant length and diameter to calibrate the radiographs before measuring CBL, they cannot be considered totally blinded.
Results of the correlation analysis in the present study suggest that a weak positive correlation may exist between the decrease of the crestal bone levels and smoking, implant diameter, torque at insertion, and insertion in postextraction instead of healed sites. Additionally, multivariate linear regression analysis indicated that implant length, presence of periodontitis, and partial edentulism may also influence bone resorption. All of these parameters are known to affect peri-implant bone resorption; the extent of their effect on tilted implants should be the subject of further prospective studies on a greater number of implants and over a longer follow-up, consistent with conclusions reached by other authors.23,34 In addition, it could be interesting to analyze the different degree of inclination and its effect on CBL .
Within the limitations of this study, immediate full-arch rehabilitation supported by straight and tilted implants, in some cases, could be an alternative surgical and prosthetic approach for edentulous jaws. Straight and tilted implants seemed to have a similar behavior during functional loading. No statistically significant differences were observed between straight and tilted implants as far as their success and peri-implant bone loss rates are concerned. Further prospective comparative studies, including a greater number of implants and over a longer follow-up, are therefore mandatory.
crestal bone level
peri-implant bone loss
The authors have no financial or personal relationships with other people or organizations that could inappropriately influence (bias) their work.