Digital Occlusal Splints

Computer-based production of fixed prosthodontics is now commonplace, with a large number of CAD/CAM systems available for producing all-ceramic restorations.1 Similar CAD/CAM methods have been applied to implant planning2 and the production of patient-specific implant abutments.3 In orthodontics, computer-based tooth setups are directly converted into solid models using rapid prototyping, which are then used to fabricate the sequential tooth positioners used in the Invisalign system (Align Technology).4


Interocclusal splints are routinely used to treat TMJ and masticatory disorders.5,6 Hard or soft removable acrylic appliances covering the teeth have been used to eliminate occlusal disharmonies7, prevent wear and mobility of teeth8, and treat derangements of the TMJ.9
Current laboratory methods for producing splints use several fabrication, articulation, and manual trimming methods. For unmounted cases, models have to be mounted on an articulator to allow the opening to be adjusted and to simulate the excursions needed to design centric relation splints. This can introduce error since a mechanical jig is typically used to “level” the lower occlusal plane. Placing the lower arch against a flat surface introduces a canting of the plane since the 3 highest points on the arch determine the plane. Also, whether the splint is produced from a wax-up, or directly by curing acrylic between the arches, the final contact surface is the result of a technician's subjective determination of smoothness and the absence of tooth impressions (Figure 1). This essentially guarantees that the same splint will never be made twice for the same case, even by the same technician.
This article describes the computer-based design and production of occlusal splints. The digital method provides the control and consistency needed to ensure that the exact same splint is produced for the same case every time.


Digital splint fabrication has 4 main steps: 1. model scanning, 2. articulation, 3. splint design, and 4. splint fabrication. Stone models are scanned to place them into the computer as 3-D objects. The case is then articulated in the computer and the splint is designed using custom software. Excess acrylic adapted to the model is machined down to the splint surface with an accuracy of less than 0.001 inch using a high-speed machine center.


Figure 1. Indexing divots are inherent to hand-trimmed splints.

Figure 2. A combination scan, or 3-D bite record, used to locate the upper model for unmounted cases. 

Upper and lower models are laser scanned using a Minolta VIVID camera system (Konica Minolta Sensing). Six individual scans taken 60° apart are combined into a single object. The scans are smoothed and calibrated so as to define the models with respect to the mounting plate. For mounted cases, this allows the models to be positioned in a computer exactly as they would be on a physical articulator.
For unmounted cases, an additional scan is taken. Create a 3-D digital bite record by taking a single laser scan (called a combination scan) of the upper and lower models together with the provided centric relation bite registration (Figure 2). This scan is used to locate the upper arch after using the lower to define a hinge axis.


All cases are articulated in the computer. Mounted cases are articulated by modeling new commercial articulators in software. The models are positioned in the computer exactly as they are on the mechanical articulator. For all cases, the condylar inclination angle, eminence curves, and the Bennet angle are digitally controlled. This allows the protrusive and lateral excursions required to design centric splints to be simulated in software.
For unmounted cases, the lower model is used to locate a hinge axis. The standard articulation method defines a lower occlusal plane and orients the plane 15° to a horizontal. A hinge axis is then located at a 100 mm axis-incisal distance and 50 mm vertical height. Since the occlusal plane angle, axis-incisal distance, and vertical height are controlled in software, patient-specific values can be used instead of average numbers.
After establishing a CR axis using the lower model, the upper is located at the bite position using the 3-D relationship captured in the combination scan. After a case is articulated, the splint can be designed.

Digital Splint Design

The model receiving the splint is designated the “splint” model, and the opposing arch is the “contact” model.
Splint design involves defining a set of parameters:

• the bite opening,
• the contact points,
• the width of the flat plane shelf,
• the location of any anterior and cuspid guidance ramps, and
• the perimeter or shape of the splint.

The first step in splint design is to adjust the opening to the desired setting. In general, one must be able to “look through” the occlusion to ensure a sufficient thickness of plastic as well as the absence of lateral interferences. A CR bite record close to the desired opening should be taken to avoid having to significantly change the opening. For cases with significant overbites, CR splints (with anterior and cuspid ramps) are frequently made instead of flat planes, to avoid excessively opening the bite.

Figure 3. Set of contact points and the smooth plane passed through the points.

Figure 4. A case with a large curve of Spee shows arc-optimized points in red and standard plane-optimized points in green.

Figure 5. A complete set of contacts on a lower model. Red markers show the location of guidance ramps.

Figure 6. a) on the right A lower flat plane splint in place; b) Upper contacts touching the smooth splint surface.

Figure 7. A ramped flat plane splint. The splint has a 2 mm shelf extending lingually from the anterior contact points.

Contact points are then defined by clicking on the surface of the contact model. Software automatically relocates the points to optimized positions based on the occlusal plane of the model. The chosen point is either relocated to the point on the tooth closest to the occlusal plane, or to the point farthest from the plane (if the tooth extends above the plane). This ensures consistency and allows the designer to concentrate on which cusps should be contacting.
A circular “island” is then created at each point. Figure 3 shows a set of contact points and the circular islands. A best-fit plane is passed through the islands to form the flattest possible surface through all of the contact points. A horseshoe-shaped portion of this plane becomes the functional surface of the splint. For teeth associated with large curves of Spee, the contacts are optimized based upon the arc of closure. The red contact points in Figure 4 are arc-optimized, while the green contacts are optimized using the occlusal plane. The designing technician can also turn off the optimization and locate a contact anywhere desired.
When designing flat plane splints, the width of the flat surface anterior and posterior to the contacts can be specified. This is important for cases with large overjets to ensure sufficient length for protrusion. Centric relation splints can be designed with anterior and cuspid ramps to provide posterior disclusion when the patient protrudes or moves the jaw laterally. Figure 5 shows a complete set of contact points on a lower model. Four red markers indicate the location of the anterior and lateral ramps. The ramps are designed to provide a gentle rate of disclusion—only 5° past the angle needed for zero disclusion. As a final step, the perimeter or shape of the splint is defined by clicking a series of points on the splint model.
Figure 6a shows a lower flat plane splint in place. Figure 6b shows how the upper contact points hit the flat splint surface. Figure 7 is a closeup of a ramped flat plane splint showing anterior and cuspid ramps.

Splint Fabrication

Figure 8. CAM software is used to define machining toolpaths. Regions of the occlusal surface corresponding to contact points are selected (patches outlined in white) for finer machining.

Figure 9. Model covered with acrylic ready to cut.

Figure 10. Small tool used for finer details near contact points.

The first step in fabrication is importing the 3-D splint file into CAM software, which develops the toolpaths needed to cut each splint. Machining takes place using sequentially smaller tools. Regions of the splint surface with contact points are selected for finer machining. Figure 8 shows the selection of contact point regions in the CAM software (patches outlined in white).
The splint model is covered with the desired material and mounted in a Haas vertical machine center (Figure 9). The splints are produced by machining excess material down to the desired splint surface (Figure 10).
To withstand the machining forces, the model mounting must be sufficiently strong. Mounting stone must be used instead of plaster, which is relatively weak. By adjusting the toolpaths in the CAM software, the process also allows standard wire clasping to be used. The splint contact surface is accurate to less than 0.001 inch and does not require any further finishing.


Figures 11a and 11b. Equal magnification scanning electron micrographs show typical tooth impressions on a manually-produced splint (a). The digital splint (b) has 2 contact points on a smooth surface.

An important feature of digital splints is the smoothness of the contact surface. Conventionally produced splints have residual indexing impressions left by the contact model. These divots tend to lock patients in place and inhibit the free movement needed to deprogram muscles. Digital splints have perfectly flat surfaces and provide a skating rink effect that allows free movement of the teeth over the splint surface. Figures 11a and 11b are equal magnification (20x) scanning electron micrographs of typical indexing marks in a manually-produced splint and a digitally-produced splint, respectively. As a result of the smooth surface, contacts are clinically visualized as fine markings or points. Also, any required adjustments are made at the precise contact points, preventing the removal of unnecessary material. The overall required chair time for adjustment has been significantly reduced compared with conventional appliances.
All materials commonly used to make splints are readily machined, including: cold-cure acrylic, hard thermoformable materials, hard/soft materials, heat-softening acrylics, and light cure materials. The Great Lakes Orthodontics laboratory routinely applies acrylic on top of a thermoformed base of 1.5 mm clear PETG (Splint Biocryl) to produce splints with good interproximal detail and a moderate amount of flexibility. Digital splints can also be produced using ethylene vinyl acetate (EVA) over a wide a range of durometers generally reserved for mouth-guard applications.
Software uses a standard 20° condylar inclination angle and average eminence curves found in fully adjustable articulators. For cases in which there has been significant change in condylar angle or joint remodeling, the clinician needs to give more design information to the laboratory to construct a proper ramp and minimize the disclusion angle, eg, delta bruxers.
Another clinically important issue is the fit of the appliance over the teeth. It is critical that the splint readily seats over the teeth before beginning equilibration procedures. The design of the splint should consider the dental anatomy to ensure good seating. For example, flared anteriors dictate that the splint should seat passively over the anterior teeth and rely upon the posterior teeth for retention.
The arc of closure influences the location of initial contacts in an anterior-posterior direction. If the splint is designed with an axis-incisal distance that is too short, the initial contacts occur posteriorly on the splint. When the distance is too long, initial contacts occur anteriorly on the ramp.
Perhaps the most important clinical factor in producing an accurate splint is the bite record. Using a Lucia jig, one can readily control the posterior interocclusal distance and wind up with a stable bite position for injecting a silicone bite material. This has the additional advantage of not having to change the bite opening when designing the splint, which is particularly important for unmounted cases, since rotation is occurring on an arbitrary hinge. When taking wax bites, it is more difficult to control the posterior interocclusal space.
While digitally produced splints afford distinct advantages over conventional splints, accurate recordings from the clinician remain paramount for a clinically successful prosthesis.


A new digital process for producing occlusal splints is described. The process mirrors conventional restorative CAD/CAM systems consisting of scanning, customized CAD design, and machining. The method provides precise and consistent digital control over articulation and design parameters, and is suitable for mounted and unmounted cases. With over 2,500 beta cases to date, an overall reduction in adjustment time has been generally reported.



    1. Raigrodski AJ. Contemporary materials and technologies for all-ceramic fixed partial dentures: a review of the literature. J Prosthet Dent. 2004;92:557-562.
    2. Drago CJ. Two new clinical/laboratory protocols for CAD/CAM implant restorations. J Am Dent Assoc. 2006;137:794-800.
    3. Priest G. Virtual-designed and computer-milled implant abutments. J Oral Maxillofac Surg. 2005;63(suppl 2):22-32.
    4. Miller RJ, Derakhshan M. Three-dimensional tech-nology improves the range of orthodontic treatment with esthetic and removable aligners. World J Orthod. 2004;5:242-249.
    5. Boero RP. The physiology of splint therapy: a literature review. Angle Orthod. 1989;59:165-180.
    6. Laskin DM, Greene CS, Hylander WL, eds. Temporomandibular Disorders: An Evidenced-based Approach to Diagnosis and Treatment. Chicago, IL: Quintessence; 2006:377-390.
    7. Dawson PE. Functional Occlusion: From TMJ to Smile Design. St Louis, MO: Mosby; 2006:380-392.
    8. Pavone BW. Bruxism and its effect on the natural teeth. J Prosthet Dent. 1985;53:692-696.
    9. Lundh H, Westesson PL, Kopp S, Tillström B.  An-terior repositioning splint in the treatment of temporomandibular joints with reciprocal clicking: comparison with a flat occlusal splint and an untreated control group. Oral Surg Oral Med Oral Pathol. 1985;60(2):131-136.

Mr. Lauren is Director of Research at Great Lakes Orthodontics, responsible for new products and technology at Great Lakes. A chemical engineer, Mr. Lauren worked in the cardiovascular device industry developing vascular grafts and heart valves for 20 years before moving to Great Lakes. Mr. Lauren’s interests include polymers, computer modeling, and digital manufacturing methods.  He can be reached at (800) 828-7626 or via e-mail at This e-mail address is being protected from spambots. You need JavaScript enabled to view it .


Disclosure: Mr. Lauren receives no compensation associated with the publication of this article.

Dr. McIntyre serves as clinical professor of restorative dentistry, University of Buffalo School of Dental Medicine. He is a diplomat of the American Board of Prosthodontics, and a member of the American College of Prosthodontics, International College of Dentists, and American Academy of Fixed Prosthodontics. He can be reached at (716) 481-2560.

Hide comment form



1000 Characters left

Antispam Refresh image Case sensitive