Effect of Implant Angulation and Depth on the Accuracy of Casts Using the Open Tray Splinted Impression Technique

Rehabilitation of partially and completely edentulous patients using endosseous dental implants is a proven and successful treatment modality.^1,2  The importance of a passively fitting superstructure was first put forth by Branemark and colleagues.³  Deterrence of complications posttreatment is directly associated with minimizing the misfit of the prosthesis.⁴  This is one of the most important objectives of prosthodontic implant procedures. Though the causes of failure are multifactorial, an imprecise superstructure is one of the major contributing factors that result in mechanical and biologic consequences that disrupt the function of dental implants.^5,6 

The first step towards achieving an accurate, passively fitting prosthesis involves replicating the intraoral relationship of implants through an impression procedure. There are several clinical and laboratory variables that affect the accuracy of a working cast. Some of them include impression and pouring techniques, impression material and die stone properties, machining tolerance of prosthetic components, and implant angulation and depth.⁷  Inaccuracies in impression making are generally transferred onto the final framework, which will ultimately produce a prosthesis misfit.^8,9 

The entire procedure of making an impression for implant supported prosthesis includes many variables starting from the type of the tray chosen (stock or custom made) to the impression technique selected (open-tray or closed tray technique). The impression technique is particularly important in the fabrication of accurate working casts. To ensure maximum accuracy, Branemark et al¹⁰  affirmed the importance of splinting the impression copings together intraorally before making an impression.¹¹  Literature is rife with different implant impression techniques; however, open tray splinted impression technique has been proven to be the most accurate amongst all the techniques.^5,11 

Many in vitro studies have been conducted, which evaluated the implant impression accuracy in ideal conditions. The models that have been designed were either in the form of blocks or completely edentulous arches/arch form with little or no undercuts, which is not the scenario encountered clinically. Nonparallel implants are commonly encountered in implant prosthodontics. Distortion of the impression may occur due to an unfavorable path of removal of impression.⁸  Some amount of angulation of dental implants is unavoidably present keeping in mind factors like adjacent structures, curvature of the arch, morphology of the bone, and experience of the clinician. There are relatively few studies that have evaluated the effect of implant angulation on the impression in models, which closely simulate the clinical situation.

It can be quite a challenging task to record the 3-dimensional position of a dental implant when they have been placed more subgingival due to factors like esthetics and/or available bone.⁹  Literature is scant in providing information on the accuracy of master casts produced owing to the interplay of depth and angulation closely mimicking a partially edentulous clinical situation. This coupled with the presence of natural teeth creates an interplay of physical and mechanical factors that influence the making of an impression. Hence, this research work was undertaken to evaluate the effect of implant angulation as well as depth on the accuracy of the implant impressions for a partially edentulous situation.

Four partially edentulous mandibular master models with an edentulous span from mandibular left canine to left second premolar (22 to 20) were fabricated using a photopolymer resin (Anycubic photon DLP, ALL3DP GmbH). Implant analogs (Nobel Biocare) were placed in canine (22) and second premolar (20) region at a different depth and angulation for each group using a surveyor (Marathon-103 Surveyor). Ten impressions were made for each model, which led to the fabrication of 40 master casts. These master casts were divided into 4 groups.

• Group A (n = 10):

  Implant analogs placed at 2-mm depth with the 2 implants being parallel to each other.

• Group B (n = 10):

  Implant analogs placed at 2-mm depth with the posterior implant placed at an angle of 15 degrees.

• Group C (n = 10):

  Implant analogs placed at 4-mm depth with the 2 implants being parallel to each other.

• Group D (n = 10):

  Implant analogs placed at 4-mm depth with the posterior implant placed at an angle of 15 degrees.

The fabrication of master models (Figure 1) was done by making an impression of a completely dentulous subject. The impression was poured using type IV die stone (Ultrarock). Following retrieval, the area from mandibular left canine to left second premolar (22 to 20) was made edentulous by removing the teeth on the cast. Two impressions were made of this cast, one of which was modified to create a 2-mm and another with a 4-mm cutback. These modified impressions were poured using type IV die stone and 2 casts with the desired cutback were created. These models were scanned and converted into 4 partially edentulous models, 2 with a 2-mm cutback and the other 2 with a 4-mm cutback made of photopolymer resin with the help of 3D printing (Anycubic Photon DLP 3D printer).

A surveyor was used to position the implant analogs accurately in the master models. The master models were fixed to the surveyor platform, and the platform was calibrated to 0 degrees and 15 degrees using a digital inclinometer (Level Meter, Vtech communications). Implant analogs were then placed in the predetermined position and angle using a clear autopolymerizing resin stent (DPI) and guide pin of an impression coping fixed to the surveyor arm. Stents made of clear autopolymerizing resin (DPI) were used to make sure that the positioning of the analogs remains constant in all the models. All the laboratory analogs were placed flush with the cutback surface of the master models to achieve the variation in depth and were secured into position using clear autopolymerizing resin.

To make sure that the thickness of the gingival mask around the implants remained constant, Aluwax (Morsa) was adapted around the implant analogs with the impression copings (Nobel Biocare) in place. The calibration of 2-mm and 4-mm depth was done with the help of a UNC 15 probe (GDC). A putty index of calibrated Aluwax was made to standardize the thickness accordingly to create the final gingival mask.

The specimens for all the groups were fabricated by making open tray splinted impressions of the master models using 40 special trays made by visible light cure resin (Individo-Lux) along with tissue stops for the orientation of the tray. The space for impression material for each tray was kept constant by using a putty spacer for all the specimens. The impression posts were splinted using dental floss (Thermoseal) and pattern resin (Pattern resin, GC) with constant thickness of 5.5 mm in all master models using a digital vernier calliper (Saafaseed). To compensate for polymerization shrinkage, the splint was split with a diamond bur (SR-12, dia-burs, Mani) to create a cut of 1.2 mm to maintain a constant gap and then rejoined.

Open tray splinted impressions were made using polyvinylsiloxane putty (Aquasil soft putty, Dentsply Detrey, GmbH) and light body impression material (Aquasil hydrophilic light body impression material, Dentsply Caulk, Dentsply) which was allowed to set for 10 minutes before removal. Separator was applied in the edentulous area of the impression and gingival mask (Gingifast elastic, Zhermack) was dispensed on the area. The impressions were poured using type IV die as per the manufacturer's instructions.

Following the fabrication, testing and analysis of the specimens were done using the portable coordinate measuring machine (Faro arm) in x, y, and z-axes. A probe was used to mark the 2 laboratory analogs so that linear distortion could be calculated with the help of points P1 and P2. The impression copings placed at points P1 and P2 represented the angles at which the laboratory analogs had been placed, which is denoted by C1 and C2 (Figure 2).

• P1: Implant analog placed in the position of 22. It represents “point 1” for the linear measurements in x, y, and z-axes in millimeters.

• P2: Implant analog placed in the position of 20. It represents “point 2” for the linear measurements in x, y, and z-axes in millimeters.

• C1: Represents the angle measured at point 1 (P1) in degrees.

• C2: Represents the angle measured at point 2 (P2) in degrees.

Laser scanning was done to scan the entire cast with the impression copings in place so that angular distortion could be measured and comparison of master models with their respective specimens could be done (Figure 3). Polyworks software (Innovmetric Software, Inc) was used to align the samples with their respective master models and to calculate the positional and angular deviation of the specimens with their respective master models. The alignment was done by matching the common points between the 2 images, which helped to align the images approximately, providing a prealignment after which a best-fit algorithm was used to compare the deviation in the specimens when compared with their respective master models. The linear deviation was calculated in the X, Y, and Z coordinates in millimeters and the angular deviation with respect to the X, Y, and Z axes in degrees.

The data were analyzed using SPSS version 23 (SPSS, Inc). Descriptive statistics and independent t test were used for the statistical analysis of the data. Comparisons were made separately for linear and angular distortion. Intergroup comparisons were made to compare the amount of distortion happening due to the factors of angulation and depth.

The results of this study show that there is an effect of implant angulation and depth on the accuracy of implant impressions. Independent T test was done for comparison between the groups (Tables 2 through 9) (Figures 4 through 11). Statistically significant differences (P < .05) were found in the linear distortion as well as angular distortion when groups with different depths were compared. However, when the groups at the same depth were compared, the results were not significant with respect to angular distortion.

The results reveal that implant angulation and implant depth have a negative effect on the accuracy of implant impressions when linear distortion is considered, but when angular distortion is considered, implant angulation has less effect than implant depth. In the models that had the implants placed at the same depth, the effect of implant angulation was not statistically significant. The null hypothesis was rejected based on statistical analysis as the P-value obtained was <.05.

Linear and angular measurements were made for each master model and their respective samples. Independent t test was done for intergroup comparison keeping 1 factor constant, to know the effect of the other factor. The results showed that when linear distortion was considered, a statistically significant difference was present for all the intergroup comparisons; that is, A–B, C–D, A–C, and B–D. This shows that the factors of depth and angulation create inaccuracies in impressions when linear distortion is considered. These findings are comparable with the results of other studies like Choi et al.¹¹ 

However, when angular distortion was evaluated, the results deviated from that of linear distortion. The intergroup comparisons showed a statistically significant difference (P < .05) when the comparison was made between the groups A–C and B–D, which shows that an increase in depth does affect the accuracy of the impressions. But when the comparisons were made between the groups A–B and C–D, the results were not statistically significant, which shows that angulation up to 15° between implants does not have a pronounced effect on the accuracy of the impressions for implants placed at same depth.

Impression making for dental implants is an extensively studied aspect of implant dentistry. Over the years, based on the results of the various studies done, open tray splinted impressions have been known to improve the accuracy of the impressions made.^12–15 According to studies conducted by Del'Acqua et al,¹⁶  Assif et al,¹⁷  and Mostafa et al,¹⁸  there has been a consensus that open tray impressions are more accurate in transferring the spatial orientation of implants than the closed tray impression technique.^16–18  One of the primary reasons for this can be less amount of distortion of the impression upon removal from the mouth.

Splinting and nonsplinting of copings before impression making is a much debated topic. According to the studies conducted by various authors like Dorigatti et al and Zen et al,^19–21  it has been observed that splinting of the copings indeed produces more accurate impressions. Autopolymerizing resin is 1 of the most commonly used materials for splinting.^12,22,23  There are a few studies that have used light cure composite as well, but the difference in accuracy is not statistically significant, and a study done by Ongul et al²⁴  actually reported more accurate results when auto polymerizing resin was used in comparison with light cure composite.

The impression material used is known to have an effect on the accuracy of the impression made. Polyvinyl siloxane and polyether are commonly used materials for making implant impressions.^25,26  Though the difference in the accuracy of impressions is not statistically significant as seen in the study done by Pujari et al,²⁷  it has been seen that polyether works better in completely edentulous situations, but in partially edentulous situations, polyvinyl siloxane is known to distort less upon removal and hence leads to more accurate impressions.^27,28  Lee et al⁹  demonstrated that, in subgingival situations, polyvinyl siloxane leads to better accuracy compared with polyether.

Studies done by Sorrentino et al²⁹  and Mpikos et al³⁰  have stated that custom trays produce a more accurate impression. The use of custom impression trays leads to more accurate impressions because uniform thickness of material results in uniform polymerization shrinkage.^19,20,31 

Evaluation and comparison of the accuracy of implant impressions have been done by using many methods. Initial years of research used 2 dimensional methods for evaluation of accuracy, but with the advent of new technology, 3-dimensional evaluation of the impression accuracy can be done. This study makes use of a coordinate measuring machine, which is routinely used in other fields of engineering and for evaluation and comparison of accuracy between 2 identical models. This method has been used in a number of other studies as well.³²  Previous studies done on implant impressions have also made use of this particular technology because it helps us to know the amount of distortion happening spatially, which cannot be achieved with other methods like a profile projector or microscope.^6,11 

A lack of parallelism among the implants is a commonly encountered clinical situation. The studies that have assessed the influence of nonparallel implants on impression making have concluded that angulated implants have a negative influence on the accuracy of the impression.^{6,12,26,29,33}  It has been seen that the degree of angulation does play a role as well. Angulation up to 15° did not have a significant effect on the accuracy of implant impressions.^2,11,23  None of the studies have simultaneously studied the effect of angulation as well as change in depth, which is the situation usually encountered clinically.

The effect of implant depth on the accuracy of the implant impressions has not been delved into a lot. The few studies that have been done namely by Lee et al⁹  and Linkevicius et al³⁴  have stated that the subgingival position of implants affects the accuracy of impression.^9,24,34  This may be related to the length of the impression coping that is embedded in the impression.³⁵  It has also been found that the stability of the impression coping is less when dental implant is placed deeply, which will directly affect the recording of the implant position. To replicate the clinical situation in this study, a gingival mask has been used to simulate the soft tissues and give comparable results when the factor of implant depth was evaluated.

This study is an in vitro study that relies on models, which closely simulate the clinical situation; however, a clinical study would yield more accurate findings. With the advent of digital dentistry, it becomes necessary to study the accuracy of digital impressions versus conventional impressions. Recent scientific data reveal predictable and relatively accurate digital impression making in ideal situations. However, it would be worthwhile to evaluate the accuracy of digital impression making in an actual clinical situation or a well-designed simulation. The results obtained with respect to the few studies done are conflicting, which calls for a need for more research in this aspect.³⁶ 

This study has represented the clinical situation of an implant supported fixed dental prosthesis. Studies with more number of implants in a partially edentulous situation should be carried out to represent the variations that might be encountered clinically. This study evaluated the effect of angulation with a single variation in implant angulation (15°). A study involving more variations in angulations should be pursued for a better analysis of the effect of angulation.

The master casts were fabricated taking utmost care to follow standardized protocols; however, material properties may cause minute discrepancies, which are not entirely avoidable. However, to negate this limitation, a large sample size was selected. Similar in vitro studies in the past that deal with an interplay of implant depth have not been able to replicate the conditions presented by soft tissues in the oral cavity. In this research, a silicone gingival mask was used in all the master models and casts, which closely replicated the soft tissues.

Based on the findings and limitations of this study, statistically significant inaccuracies were observed in the master casts across all groups.

• It was observed that master models with laboratory analogs placed deeper within the gingival mask led to statistically significant differences in both linear and angular aspects.

• However, when the effect of angulation alone was observed, it was seen that the results were not statistically significant between parallel and angulated implants at the same depth. When linear changes were evaluated, the differences were statistically significant.

The authors thank Dr P. Mohan Raju (PGDAST, Applied Statistician, Statsmaster team) for his help with respect to the statistical analysis.

The author reports no conflicts of interest related to this study.