From Volume 47, Issue 7, July 2020 | Pages 570-582
This article follows on from Part 1 which looked at the pathogenesis and diagnosis of cracked tooth syndrome (CTS). It combines a review of the available evidence with a discussion of established and more modern concepts of management, aiming to provide a clear, rational approach to the predictable management of CTS with the aid of case studies and a decision-tree.
There are no universally accepted guidelines for managing cracked tooth syndrome (CTS). Contemporary approaches are quite diverse and have included operative crack removal, mechanical encirclement of the tooth, adhesive intra-coronal splinting, or splinting with extra-coronal coverage retained either adhesively or mechanically. In addition, restorations have been fabricated either in the mouth (directly), or in the laboratory (indirectly).1,2,3,4,5 Understandably, choosing the best option can be confusing.
Treatment of CTS aims to predictably restore and maintain a functional asymptomatic tooth, with a vital pulp, by stabilizing the crack, or cracks.
The evidence base on management options for CTS is limited, with much experimental data missing.
A study involving 41 patients with CTS compared direct composite placed either intra-coronally, or overlaying the affected cusp following its reduction.2 In the overlay group (n = 21), 0% required pulpectomy and restoration survival was 100% at 7 years follow-up. In the intra-coronal group (n = 20), three teeth required pulpectomy (15%), two within 6 months and one at 7 years. Two of these teeth subsequently required extraction due to vertical fractures. Eighty-five percent of restorations survived in this group and the failures were non-catastrophic. Excluding the teeth requiring pulpectomy, a minority of teeth suffered from residual thermal sensitivity in both groups, but pain on biting resolved in all teeth.
Another study compared full occlusal coverage amalgams retained either mechanically with pins (n = 20), or adhesively with resin (n = 20) for the management of CTS.3 One tooth in the mechanically retained group (5%) required pulpectomy at 1 year and cold sensitivity remained similar to baseline levels. In the adhesive retention group, 0% required pulpectomy and cold sensitivity was reduced over the same period. Pain on biting resolved in all teeth and all restorations were intact at 1 year. Amalgam use has recently been phased-down, however and, with a complete phase-out on the horizon, it is likely to be unavailable for use and will not be considered further.
Seven percent of teeth diagnosed with CTS that were treated with indirect composite onlays (n = 43), following initial stabilization with intra-coronal restoration replacement with direct composite, required pulpectomy within 6 months. Restoration survival was 93% during the follow-up period of 4–6 years. A small minority had persistent thermal sensitivity.4
A study employing permanent or temporary crowns to manage CTS found that 21% of permanently crowned teeth required pulpectomy within 6 months of treatment (n = 127). The effects of crowning on CTS symptoms were not reported.6 In the temporary crown studies, the percentage of teeth requiring pulpectomy varied from 11% at 1 year (n = 28)1 to 29% (group size n = 38)7 and 43% (group size n = 21).8 The reporting of changes in symptoms was very limited in all of these studies, and the time frame from initiation of treatment to the requirement for pulpectomy was not specified in the latter two papers. Restoration survival was not reported in any of the studies.
Another study involved drilling out the visible cracks in teeth with CTS (n = 100), before placing a Ledermix lining and various glass ionomer cement-based temporary restorations.5 Fifteen percent required pulpectomy at this initial stage of stabilization and a further 5% required pulpectomy subsequently. The teeth were definitively restored with crowns or overlays, but the proportions were not reported, and there was a high level of participant dropout in the study. Restoration survival and post-treatment thermal sensitivity were not reported.
Minimally invasive management of CTS with a DCS was reported in an audit involving 151 patients.9 CTS was diagnosed by the disappearance of symptoms following application of a non-bonded DCS, as described in the first article of this series. The DCS was then adhesively bonded in supra-occlusion, resulting in disclusion of other teeth. Eleven patients fulfilling the inclusion criteria did not consent to bonding of the DCS, so had alternative management.
Tooth contacts were then allowed to re-establish over time through relative axial movement prior to definitive restoration. The restorations varied in design, but a large majority were direct composite onlays. Management therefore involved two stages, but meant that no or minimal occlusal reduction was required, as space was created by intrusion and extrusion of teeth. Three percent of patients reported intolerance to the DCS and 10% required replacement or repair of the DCS, but all of these complications were evident within the first 2 weeks following bonding. Seven percent of the teeth went on to develop irreversible pulpitis and 3% suffered complete fracture, while 3% of patients did not re-establish occlusal contacts within 3 months but were asymptomatic.9 While the follow-up interval is short and the number of patients affected is low, the management of cases where contacts do not re-establish, and the occlusion is disrupted by uncontrolled tooth movements, may be complex and costly. Simply removing the DCS may not allow full re-establishment of the occlusion. Bonding a DCS was not advised in a number of situations highlighted in Table 1 (blue text).9 Restoration survival and post-treatment thermal sensitivity were not reported.
| Unable to tolerate non-bonded DCS |
| Antagonistic tooth vulnerable to fracture |
| Active periodontitis |
| TMD |
| Previous orthodontics |
| Reduced eruptive potential |
| Large horizontal RCP-ICP discrepancy |
| Already heavily restored tooth that would benefit from cuspal coverage even in the absence of a crack or cracks |
Only 85% of root canal treated permanent teeth with cracks survived for 2 years in a tertiary care setting (n=50). These all initially presented as CTS and were initially managed with a cemented orthodontic band in an attempt to stabilize the cracks. The teeth were restored with crowns, or amalgam cores, with retention of the cemented orthodontic bands. The number of teeth receiving each restoration was not specified, however, and the relationship to tooth survival was not reported.10 Ninety percent of cracked teeth receiving root canal treatment from an endodontist survived for 2 years when subsequently restored with crowns (n = 88). This also included teeth that were diagnosed as non-vital at presentation however, which is not CTS. Teeth with an initial probing depth >6 mm had a statistically significantly lower survival of 74% compared to 97% in the remainder of the teeth.7
Firm conclusions are difficult to make on the basis of limited data. Covering the affected cusp(s) with an adhesive restoration appears to be the most effective option for managing CTS in terms of resolving symptoms, reducing the need for pulpectomy and optimizing restoration survival in the short to medium term. Where specifically assessed, a minority of teeth with CTS showed persistent residual thermal sensitivity. Either attempting to drill out the visible cracks completely (see later discussion), or preparation of a tooth with CTS for a crown appears to increase the need for pulpectomy, over and above the generalized reported risk of loss of vitality associated with crowning teeth.11,12 Use of a bonded DCS to gain inter-occlusal space can be successful in resolving symptoms and minimizing the need for pulpectomy, but the technique can have complications and is not appropriate for everyone. Cracked teeth requiring root canal treatment appear to have a reduced prognosis compared to the reported general success of root canal treatment when managed by endodontists13 and the presence of a probing depth >6 mm is a negative prognostic factor.
A decision-tree has been developed to aid decision-making in the management of CTS (Figure 1). This will be explained through the use of successfully treated case studies and a discussion of the relevant concepts.
Effective management requires consideration of:
This involves decision-making on:
Classifying the restorative status of a restored tooth is difficult. For the purpose of managing a tooth with CTS, it is important to decide whether the existing restoration puts the tooth at an increased risk of fracture, even if the crack was not present. A tooth deemed to be at increased risk would be classified as ‘heavily restored’. The remaining restored teeth would be classified as ‘minimally restored’.
Defining which teeth are at risk of fracture, in the absence of a symptomatic crack, is very subjective, and a complex clinical judgement based on multiple factors. These include the presence of asymptomatic cracks, residual cusp dimensions, depth and configuration of the cavity, type and quality of dental substrate, restorative technique, position of the tooth in the arch, the nature and force of functional and potential parafunctional loading, periodontal support, alongside harder to define patient and dentist-related factors.14,15
Assessing fracture risk prior to removing a restoration involves unknowns, so is generally based on gauging whether any remaining cusp tips are likely to be supported by dentine, the depth of the restoration, and whether the marginal ridges are intact and supported by dentine.15
These factors are discussed in the case studies to help illustrate the decision-making process. Figures 2a, 4a, 6a and 7a show examples of where teeth were classified as being heavily restored, or minimally restored and the rationale explained.
When treating CTS, an unrestored, or minimally restored, tooth can be managed differently from a more heavily restored tooth by using a bonded DCS. This is not recommended in certain patient groups, however (Table 1) (see later).
Cracks always have the potential to propagate, and cracks that communicate with the oral environment usually harbour biofilm,16 which could cause periodontal and pulpal pathology. It may therefore seem logical to remove cracks from teeth by cutting them out, yet this is rarely helpful, especially when a crack is vertically oriented (Figure 2b), or when a tooth is unrestored or minimally restored (Figure 6a). Cracked dentine does have an ability to self-limit crack propagation, and the complex dynamics of the host response to biofilm in the manifestation of pulpal and periodontal health or disease are not well understood.
Cracks can run sub-gingivally and centrally, with no reversible means of assessing the point of termination (Figure 2b). It is impossible to know if a crack has been completely removed when there is no external termination point, as the crack tip and micro-cracks will not be visible with clinically obtainable magnification.17 Attempts to remove cracks result in unnecessary tissue loss in areas critical to prognosis of the tooth, notably in proximity to the pulp, and in the peri-cervical region, which is biomechanically important.18 This also potentially places the restorative margin in dentine rather than enamel, where the adhesive bond will subsequently degrade more rapidly over time.19,20
Vertical cracks: Vertical cracks without caries should therefore never be ‘explored’. It is advised that full occlusal coverage (this does not mean a crown) should be provided for teeth with vertical cracks (Figures 2 b and c), acknowledging that such teeth have a guarded prognosis.
Oblique cracks: Oblique cracks associated with restorations can be managed differently from vertical cracks. If a crack in dentine is visible in the overlying enamel, it is advised that removal be attempted, provided this will not compromise the pulp, or the restorability of the tooth (Figures 3 a–c). This is based on clinical experience of seeing (very) occasional issues of recurrence of CTS at follow-up when retaining such cracks. Where complete removal of the crack is deemed appropriate and subsequently performed, the tooth has a favourable prognosis (Figure 3c).
If removal would compromise the pulp or restorability of the tooth, or if the dentine crack cannot be seen on the external tooth surface, the crack should not be ‘explored’ or removed. Instead, the affected cusp should be overlaid (Figures 4 b–d). In this situation, the prognosis is more favourable than a vertical crack, but still guarded in the longer term, as the crack has unknown parameters, including depth and direction (Figure 4e).
Occlusal coverage makes the tooth less susceptible to fracture by stabilizing the crack and changing the loading of the affected cusp(s) by converting damaging, primarily tensile stress into supportive compressive stress21 (Figure 5).
Space is required to provide occlusal coverage of the affected cusp or cusps. This can be gained through relative axial movement (tooth intrusion and extrusion), following the application of a DCS, or through cuspal reduction. Both are useful in different situations (Figure 5).
Unrestored or minimally restored teeth are excellent candidates for relative axial movement to provide space for occlusal coverage, as this means no tooth reduction is required (Figures 6 a–h). This maximizes retention of enamel, which is more predictable to bond to than dentine in the long term.19
In heavily restored teeth, thin cusps flex more, suffering tensile stress in function which puts them at increased risk of fracture.15,21 Though it may seem sensible to retain these cusps and gain space to cover them by relative axial movement, it is suggested that this is misguided.
When a tooth is heavily restored and unsupported enamel is present at remaining cusps tips (Figures 7 a and b) (whether crack affected or not, but especially in crack-affected cusps), leaving them unmodified, compared to reducing them, will increase flexure during definitive adhesive restoration. This is due to the less favourable cavity C-factor, and resultant increased development of interfacial contraction stress, which is unfavourable in an already cracked tooth.22,23
Thin cracked cusps quite commonly fall away during the removal of an existing restoration (Figure 8a), and retaining thin cusps where the crack in dentine is visible in the overlying enamel can be problematic, as described above. It is therefore often favourable to remove them, even if they remain. This would then lead to an increased height of restorative material being required to restore the tooth, if space has already been gained by relative axial movement. This is biomechanically unfavourable, resulting in increased relative cyclical fatigue at the restorative margin. Should the thin cusp be retained, there would be increased load transference to an internally initiated crack which, allied with the damaging contraction stress, could potentially lead to more rapid failure.24
It is therefore recommended that there is no benefit in gaining inter-occlusal space by relative axial movement in more heavily restored teeth, where it has been assessed that any cusp(s) would benefit from coverage, even if a fracture was not present. In these situations, it would be better to obtain inter-occlusal space by cuspal reduction (Figures 2a, 7a and b) or removal (Figures 3 a–c), as dictated by the presence of the dentine crack in the overlying enamel as previously discussed.
Cuspal reduction also offers advantages over bonding a DCS in that it reduces patient chair-time, treatment duration, and therefore cost, while avoiding risks of occlusal contacts failing to re-establish, among others (see later). Bonding a DCS should therefore be avoided in heavily restored teeth diagnosed with CTS (Figures 2a and 7a) (Table 1 green text).
Another important situation in which bonding a DCS should be avoided is in patients with a large horizontal slide from their retruded contact position (RCP) to the intercuspal position. This is because it is quite possible that a DCS could act to ‘deprogramme’ the muscles of mastication by taking their initial RCP guiding contact away, which in turn would allow the condyles to seat, putting the patient at risk of developing a large overjet and/or anterior open bite25 (Table 1 green text). Banerji et al9 also advise that there are a number of situations where bonding a DCS should be avoided. These seem sensible and are listed in Table 1 (blue text). Banerji et al9 also reported that bonding a DCS resulted in a number of other patient complaints. Table 2 summarizes these with a suggestion of information to discuss with a patient as part of the consent process prior to bonding a DCS.
|
|
Banerji et al also suggest that the tooth with bonded DCS should immediately disclude on the canine.9,26 If the DCS is bonded at the recommended 1.5 mm thickness in the posterior,26 this roughly equates to an opening in vertical dimension of 3–4 mm at the canine. This opening would likely necessitate composites being placed on both canines to achieve immediate disclusion of the DCS-protected tooth in all cases. This was only performed in 12.6% of cases, however. The author suggests that this is unnecessary. Immediate disclusion on the canines was not established in the case shown (Figure 6f), but the DCS was made as a flat surface with no deflective incline contacts (Figure 6e). This did not compromise the outcome for the patient (Figure 6m).
Defining which cusps are at risk of fracture is difficult and involves consideration of multiple factors, as previously discussed. Figures 2a, 3c, 4a and c, 6a and b and 7a show examples of where cusps, that were not undermined by a painful crack, were considered to be at increased risk of fracture, where they were not, and where other factors were important. This determined whether or not cusps were subsequently reduced. Whilst decision-making is very subjective, if in doubt, it may be better to reduce the cusp (Figure 7a), as this is a non-critical area of the tooth, as opposed to the peri-cervical region, which is more important from the perspective of reduced future fracture risk and prolonged lifespan of the tooth.27,28,29 Preferentially loading the restoration, by eliminating contacts in static and dynamic occlusion on retained cusps, may also be beneficial, aiming to move the eventual failure mode from the tooth to the restoration (Figure 4e).
The appropriate degree of reduction for coverage by direct composite has not been determined, as it is likely to be dependent on a number of variables considered previously. For example, if inter-occlusal space to overlay cusps in an unrestored tooth is created using a DCS, support for the restoration provided by a surface solely of enamel is likely to be much better than a previously heavily restored tooth, with a large, irregular volume of structure to replace and a preponderance of dentine to bond to.15 An in vitro study using a premolar model with a mesio-occluso-distal (MOD) preparation and missing cusp, compared restoration in composite with or without 1.5 mm reduction of the existing cusp. Overlaying the reduced cusp resulted in a much improved fracture resistance, but the mode of failure was slightly less favourable.30 In another laboratory study, direct composite onlay restoration following 2.5 mm cuspal reduction, showed a more favourable stress distribution than 1.5 mm reduction in premolar MOD cavity models.21 How cracks would modify these findings is unknown. Striking the optimal balance between modifying force distribution and maintaining tooth structure inevitably carries uncertainty, which makes translation to the clinical environment difficult. General advice would be to consider 1.5–2 mm of reduction (Figures 2b, 4c and 7 b–d).
If the remaining cusp is subsequently unsupported by dentine following reduction, consideration should be given to further reduction until it is supported. However, if this removal compromises the restorability of the tooth, it should not be performed.
Cavity preparation results in reduced fatigue resistance of dentine, and particle abrasion cut dentine fares better than bur cut dentine in this respect.31 Deeper dentine is more brittle, compounding the effect of structural compromise in deeper cavities.32 Finalizing cavity preparation with particle abrasion may therefore be beneficial to the compromised dentine, whilst also aiding bonding.33 Particle abrasion can also help to visualize retained crack remnants, facilitating complete removal, as even when a cracked portion of tooth splinters away during operative exploration, the crack is not always completely removed (Figures 8 a and b).
Resin-bonded restorations result in contraction stress at the interface with the cavity, as previously discussed.34 It is unclear how this might affect a dentine crack in terms of propagation, but it has been shown in vitro, that teeth with an occlusal cavity that are restored with various composite materials and techniques (bulk-fill composites topped with a conventional composite versus a layered approach with conventional composite) developed cracks in the overlying enamel where none existed beforehand, whilst also resulting in extension of pre-existing cracks.34 Bulk-fill composites developed less contraction stress at the restorative interface than a layered approach, but all tested materials varied. The teeth restored with bulk-fill composites all showed increased fracture resistance compared to those restored with a conventional layered composite.34 It is as yet unclear, however, how a more compromised tooth may behave over time, when restored with some of the flowable bulk-fill materials which have a reduced modulus of elasticity (rigidity),35 though more rigid, paste-like bulk-fill composites are available.36
The rigidity of direct composite is also lower than indirect composite, metal or ceramic, and some commentators would suggest that a more rigid material, mechanically or adhesively retained, is appropriate to prevent crack propagation.37 One study, however, showed comparable fatigue resistance of cusp replacement restorations of ceramic, direct and indirect composite in premolars.38 They might also justifiably consider that the interfacial contraction stress is reduced when placing indirect restorations,39 which may reduce the tendency to propagate a fracture. Conversely, it has been suggested that internal development of stress could contribute to the fracture resistance of cusp-replacing restorations,40 though the stress does reduce over time, primarily through hygroscopic expansion.41
Currently, there is no compelling clinical evidence to support the importance of these contentions, so direct composite can be recommended, based on all clinically relevant outcome parameters discussed, alongside the reduced clinical time required, the absence of need for a non-bonded temporary and the consequent economic benefit. However, that is not to say that bonded indirect onlay restorations are inappropriate, especially for a tooth with a stabilized crack.
Optimization of the bonding procedure is advisable, so a suggested protocol involves use of rubber dam,42 particle abrasion of the preparation,33 use of a three-step bonding system (4th generation bonding agent or 5th generation bonding agent with selective enamel etch procedure) to optimize bonding,43,44,45,46,47 which helps to resist bond breakdown long-term, especially with dentine margins,19,43 and use of a bulk-fill composite material in combination with a conventional composite34 (Figures 7 d–h).
Even if DCS bonding is not going to be used to gain inter-occlusal space, it can be performed as an immediate short-term temporary measure in an attempt to prevent propagation of a crack (for example whilst the patient awaits an appointment for definitive restoration). Alternative immediate management strategies include cementation of a copper ring or orthodontic band,26 to minimize separation of any cracks, or lowering of cusps and reduction of cuspal inclines to minimize lateral loading,48 again while the patient awaits an appointment for definitive restoration.
Evidence on the management of CTS is very limited. Direct and indirect bonded composite restorations overlaying affected cusps show good survival, success in resolving symptoms, and maintaining pulp vitality in CTS-affected teeth in the medium term.
While it is clear that a DCS is an excellent tool for diagnosing CTS, and bonding a DCS can be very successful, in many situations it is not appropriate (Table 1). It may unnecessarily complicate and prolong treatment for patients with heavily restored affected teeth and expose them to risks that they need not face. Each case should be assessed on its merits and managed appropriately, with input from the patient.
Crowning CTS-affected teeth leads to significant need for pulpectomy in the short term and should be avoided where possible.
Decision-making will always involve numerous variables, and sometimes the cost of sacrificing non-critical tissue with dubious structural integrity is a worthwhile price to pay for improved clinical predictability and patient-centred outcomes. Given that the orientation of a crack in restored teeth is generally unknown prior to restoration removal, the patient should be informed of the guarded prognosis prior to any operative intervention. The potential for persistent thermal sensitivity should also be discussed.
