The interaction between diamond bur wear and accuracy of dental restorations from subtractive manufacturing

Achieving optimal accuracy of fixed restorations produced by computer-aided design and computer-aided manufacturing (CAD/CAM) is a critical factor in long-term clinical success. Inadequate marginal or internal fit compromises prosthesis durability by hastening cement dissolution, thereby facilitating bacterial invasion and dental caries. Although the subtractive manufacturing technique (commonly known as “milling”) is extensively utilised in CAD/CAM workflow, current protocols often depend on static, manufacturer-defined estimates of rotary bur lifespan, neglecting the progressive effects of bur degradation. Maintaining consistent manufacturing quality over time is challenging due to worn burs. However, the specific relationship between progressive diamond bur degradation and the resulting dimensional accuracy of dental crowns remains insufficiently characterised, necessitating further investigation.

Lithium disilicate glass-ceramics are established as a restorative material in CAD/CAM dentistry, favoured for their superior aesthetic translucency and proven long-term clinical success. However, to address the inherent brittleness of ceramics, polymer-infiltrated ceramic network (PICN) materials were introduced. By infiltrating polymer resin into a porous ceramic network, this hybrid material aims to combine the durability and aesthetics of glass-ceramics with the elastic resilience of resin, thereby mimicking the physical behaviour of natural dentition.

The four objectives of the research described in this thesis were:

1. To undertake a critical comparison of digital measurement protocols, including trueness analysis, micro-computed tomography (micro-CT), and the triple-scan protocol, for evaluating restoration accuracy, as well as marginal and internal fit discrepancies.
2. To comprehensively characterise the extent of diamond bur degradation through a multidimensional assessment, integrating microscopic evaluation of surface integrity with quantitative measurements of roughness and dimensional loss.
3. To assess the specific effect of progressive bur degradation on the manufacturing accuracy of lithium disilicate and PICN restorations.
4. To evaluate the surface roughness of PICN restorations produced by subtractive manufacturing using various machines and to examine the impact of bur degradation on this parameter.

This thesis commences with a systematic review of the accuracy of subtractive manufacturing processes for lithium disilicate and PICN restorations (Chapter 2). This review reports that the quality of restorations is vulnerable to cumulative errors throughout the fabrication process, which are influenced substantially by data acquisition, digital design, material selection, and machine kinematics. A secondary systematic review was then conducted (Chapter 3), specifically examining the impact of diamond bur deterioration on the quality of ceramic restorations. It was observed that, although scanning electron microscopy (SEM) remains the predominant qualitative method, a notable gap exists in the literature regarding the reliability of manufacturer-recommended tool life algorithms as compared to the actual physical condition of the burs. Current quantitative metrics, such as weight and diameter loss, have been found to be susceptible to measurement inaccuracies, underscoring the need for more precise diagnostic techniques. Furthermore, despite the limited number of studies identified, repeated milling demonstrated a significant effect on the accuracy, adaptation, and surface roughness of ceramic crown specimens.

Based on the non-standardised measurement methodologies employed in previous studies (Chapters 2 and 3), there exists no consensus or direct comparison to determine which approach is superior or more reliable for evaluating the quality of CAD/CAM restorations in terms of the manufacturing process. Therefore, a comprehensive comparative analysis of digital measurement protocols was undertaken across four distinct subtractive manufacturing systems (Chapter 4). It was observed that, irrespective of the assessment methodology employed, notable performance disparities were present. Methodologically, the triple-scan protocol was validated as a reliable, non-radiological alternative to micro-CT; however, this method was only reliable to determine clinical suitability and not manufacturing reliability. Therefore, considering that the subsequent experiments (Chapters 5 and 6) focus on subtle variances arising from bur degradation, which are attributable to manufacturing factors, the adaptation measurement protocols were deemed inadequate in sensitivity and reliability. The utilisation of the surface comparison method for trueness analysis delivered detailed diagnostic insight while facilitating the direct detection and spatial mapping of specific manufacturing errors in relation to the reference design, thus giving insight into the manufacturing method.

To systematically address research gaps, a multidimensional analytical approach was employed to characterise the degradation of burs (Chapters 5 and 6) in two subtractive manufacturing systems, namely the PrograMill PM7 and inLab MC X5. It was determined that machine-reported tool life estimates are not reliable indicators of the actual condition of burs. Quantitative analysis demonstrated a non-linear degradation pattern marked by a rapid initial decline in bur surface roughness. Additionally, SEM analysis disclosed distinct machine-specific wear mechanisms: whereas the PrograMill PM7 burs exhibited superior durability, with diamond particles degrading primarily through brittle fracture, the inLab MC X5 burs experienced accelerated matrix failure and particle pull-out. Further experimentation indicated that these degradation patterns significantly reduced restoration accuracy and surface roughness due to different error mechanisms. As the burs deteriorated, the PrograMill PM7 exhibited a tendency toward overmilling, resulting in excessive material removal; conversely, the inLab MC X5 demonstrated a progressive undermilling, indicating insufficient material removal. The impact of the restorative material also varied depending on the system employed. For example, while the inLab MC X5 was highly sensitive to material hardness, it was capable of milling considerably more PICN than lithium disilicate units. In terms of surface quality, areal roughness parameters (Sa, Sq) were identified as superior to linear metrics (Ra, Rq) at detecting subtle, wear-induced changes in the texture of the restoration.

The research described in this thesis demonstrates that the trueness of subtractive manufacturing is not a fixed characteristic of the machinery but rather a dynamic variable influenced by the interaction of machine kinematics, material machinability, and tool condition. These findings provide evidence-based strategies for bur replacement and confirm a digital metrology framework capable of standardising quality assurance processes in high-throughput laboratory environments.