Predicting fatigue damage in intact and restored teeth Sam Evans Sam Smith School of Engineering, Cardiff University PO Box 925, The Parade, Cardiff CF24.

Predicting fatigue damage in intact and

restored teeth Sam EvansSam Smith

School of Engineering, Cardiff University

PO Box 925, The Parade, Cardiff CF24 0YF

Introduction•Tooth fracture or cracking is a

common cause of clinical failure •The cause of this problem is not

well understood•Cracks occur in the tooth due to

cavity preparation •Abfraction may involve fatigue

Typical tooth damage

Reproduced from:-http://www.umbc.edu/engineering/

me/bsms/dental.html

Abfraction

Typical non-

carious cervical lesions (Rees 1998)

Introduction

•The aim of this study was to model fatigue crack growth using computational fracture mechanics models

•This could provide insights into the mechanisms of abfraction and post- restoration cracking

The problem• A typical molar with an amalgam

restoration was modelled• A 114m crack was introduced at the

region of maximum stress, as found by Xu et al after preparation with a diamond burr1.

1. Xu, H. H. K., Kelly, J. R., Jahanmir, S., Thompson, V. P., Enamel subsurface damage due to tooth preparation with diamonds. J. Dent. Res. 76(10) (1997):1698-706.

Finite element model• A 2D finite element model was

developed, based on Arola et al2.• Modelled in plane strain, using

Franc2D (Cornell Fracture Group, www.cfg.cornell edu)

• Linear interface elements were used- mostly in compression

2. Arola, D., Huang, M. P. and Sultan, M. B., The failure of amalgam dental restorations due to cyclic fatigue crack growth

J. Mat. Sci.: Materials in Medicine 10(1999): 319-327.

Initial mesh, showing dentine, enamel and restoration

Initial 114mcrack

Crack regionremeshed

Crack propagated in 50m steps up to 1.6mm

Fatigue life prediction•Preliminary fatigue crack growth data by Arola et al3 was used to predict the crack growth rate

•A simple Paris Law model fits the data well

•Variable amplitude loading etc will affect crack growth in practice

3. www.enduratec.com/pdf/Appbrief-UMBC.PDF

Stiffness ofside muchreduced

Crack length vs time

0

0.5

1

1.5

0 2 4 6 8 10

Crack length (mm)

Time (years)

Discussion• Stresses in the tooth are in the

right range to cause clinical fractures in a typical clinical timescale

• Crack may become dormant due to load redistribution

• Tooth is then left vulnerable to unusual loads, decay etc

Abfraction

Crack propagation

Crack length

Figure The number of cycles for the crack to grow to a given length under a 10N load for the first and final models

Crack length (mm)

Cycles

50N load

20N load

Discussion• Propagation of cracks is likely at

typical physiological loads• Crack growth likely from small

(50µm) initial cracks• Possible formation of deep rounded

lesions inside the PDL• Interaction with erosion during

initiation and propagation

Conclusions

•These models predict crack propagation at relevant rates under typical physiological loads

•Fatigue seems likely to be a factor in abfraction damage

•Possible to avoid fatigue damage through improved restorations?

Acknowledgements

•The abfraction model was developed by Sam Smith

•The Franc software is provided by the Cornell Fracture Group http://www.cfg.cornell.edu