Occlusal fissures of the equine cheek tooth: Prevalence, location and association with disease in 91 horses referred for dental investigation



Reason for performing study: Fissures of the occlusal surface of the equine cheek tooth are poorly understood and their association with dental disease is unknown.

Objective: To describe the prevalence and location of occlusal fissures in the cheek teeth (CT) of a group of horses referred for dental investigation/treatment, and determine association with intercurrent dental disease.

Methods: Digital video recordings of oral endoscopic examinations for all horses referred to the Rossdales Equine Hospital for dental investigation from November 2006 to June 2009 were reviewed. Location of occlusal fissures in relation to both Triadan tooth position and pulpar secondary dentine was recorded; direction of fissure and concurrent involvement of enamel was also documented. The CT location considered at the time of examination to be the primary site/s of disease was correlated with presence of fissures on these teeth.

Results: 91 cases meeting the inclusion criteria were identified. Occlusal fissures were documented in 58.2% (53/91) cases, with a total of 227 CT being affected. Fissures were most prevalent mid-arcade. The majority (92.1%) of fissures in maxillary CT were associated with the caudal palatal pulp horn. Fissures in mandibular CT were predominantly associated with the buccal pulp horns (95.7%). There was no significant difference in the median number of CT with fissures in relation to gender. There was no correlation between age (r2= 0.01) of horse and number of CT with fissures. A significantly greater number of CT with multiple occlusal fissures was found in mandibular compared to maxillary arcades. No correlation was found between presence of fissures and location of individual CT considered to be primarily responsible for presentation.

Conclusions: Occlusal fissures in this group of animals were common and not correlated to primary site of dental disease.

Potential relevance: In horses subjected to dental investigation, occlusal fissures of the cheek teeth should not be considered an indicator of tooth compromise. Location and direction of fissure propagation in most cases is inconsistent with occlusal fissures being causally implicated in slab fractures of cheek teeth, although site predilection may indicate a possible association with masticatory forces.


Advances in the imaging and understanding of equine dental disorders have led to greater recognition of the potential relationship between subtle abnormalities of the clinical crown of cheek teeth and concurrent dental disease. Defects of the secondary (pulpar) dentine and, to a lesser extent, advanced infundibular caries are now regarded as possible indicators of the presence of apical pulpitis (van den Enden and Dixon 2008; Casey and Tremaine 2009) and oral endoscopic imaging of the clinical crown and supporting structures has been shown to complement radiography in the detection of diseased teeth (Ramzan 2009). Fine occlusal fissures of the occlusal surface of equine cheek teeth have recently been described in a report of oral endoscopy findings (Simhofer et al. 2008) and, although recognised for some time (Becker 1962), an investigation of their prevalence, location and relation to concurrent dental disease has not been reported.

Materials and methods

Recorded digital video images for all horses subjected to endoscopic examination of the oral cavity at the Rossdales Equine Hospital from November 2006 to June 2009 were reviewed; case population consisted of both internal and external referrals. The period chosen for review coincided with the introduction of use at the facility of a rigid industrial borescope1, digital recording and a standardised protocol for intra-oral imaging; analogue recordings of examinations undertaken using a side-viewing videoendoscope2 in preceding years were considered to be of insufficient quality for inclusion. Cases were included for analysis only if high-definition images were available of the occlusal surface of all maxillary and mandibular cheek teeth.

Digital video recordings were processed using video editing software3 and pause/frame capture functions utilised to permit detailed examination of the occlusal surface of each tooth. Fissures were defined as any fine linear defect in the continuity of the occlusal dentine (+/− enamel), without perceptible separation of the occlusal surface. Overt fractures (+/− displacement of dental fragments) were not included for analysis. Fissures were identified and recorded for each case by tooth location (modified Triadan nomenclature) and by pulp horn location. For the purposes of statistical analysis pulp horns were recorded using a modified numerical system (du Toit et al. 2008a) but descriptive nomenclature is used for clarity in the discussion of findings. Orientation of fissure and apparent extension to peripheral or infundibular enamel was also recorded. Information on age and gender of horse was acquired from clinical records. Detailed information on previous prophylactic/therapeutic dental treatments was not available. Association between location of CT with identified occlusal fissures and primary site of dental pathology as determined at the time of original examination was investigated.

Statistical analysis

Complete information about dental status, age and gender was available. Results for age (years) were expressed as mean ± s.e. and counts were expressed as median plus IQR. A Mann-Whitney U was used to detect significant differences in relation to gender. The presence and position of fissures were compared using the Chi-squared test. A correlation was performed using Pearson product moment correlation. Groups were considered to be different when P<0.05. Statistical analysis was conducted with Statview4, version 4.02.


Ninety-one cases meeting the inclusion criteria for oral endoscopic image quality and content were identified. Thirty cases were presented for investigation of suspected apical dental infection, 25 for oral discomfort related to diastema/periodontal disease, 15 for behavioural problems including headshaking, 12 for dental fracture, 8 for remedial dental work and one for a miscellaneous problem (oral haemorrhage arising from palatal tumour). In the study population there were 33 Thoroughbred (TB) or TB cross, 26 Warmbloods, 15 ponies, 4 Arabs and 13 horses of miscellaneous breed. There were 54 (59%) males and 37 (41%) females. Median number of CT with fissures per male and female horse was 1 (IQR = 0–3) and 2 (IQR = 0–3), respectively. There was no significant (P = 0.099, Mann-Whitney U) difference in the median number of affected teeth in relation to gender. Mean ± s.e. age was 12.2 ± 0.7 years (range 2–32 years). There was no correlation (r2= 0.01) between age and number of CT with fissures per horse.

At least one CT occlusal fissure was identified in 53/91 (58.2%) horses; and a total of 227 (135 maxillary, 92 mandibular) CT with fissures were identified. The majority (199/227) of affected teeth displayed a single fissure; 26 CT had 2 fissures and 2 CT displayed 3 fissures. Median number of fissures present per horse in the cases with at least one fissure was 3 (IQR 0–3, range 1–22). A significantly (P = 0.0001, Chi-squared test) greater number of CT with multiple occlusal fissures was found in mandibular compared to maxillary arcades. Twenty-four (86%) of the 28 CT with more than one fissure present were mandibular teeth.

All fissures were seen to be confluent at one end with the secondary dentine of a pulp horn. Secondary dentine was readily identified as a dark-brown stained feature of the occlusal surface overlying pulp horns, with good definition from the surrounding light-brown or tan-coloured primary dentine (Muylle et al. 2002; Dixon 2005). CT with more than one fissure typically had one fissure associated with each of 2 pulp horns (23/28 teeth with multiple fissures). In 3 CT (all mandibular) in 2 individuals, multiple fissures associated with a single pulp horn were documented. The predominant location of fissures on maxillary CT was related to the caudal palatal pulp horn (92.1%) (Figs 1 and 2), while that of mandibular teeth was the rostral buccal (57.6%), followed by the caudal buccal (38.1%) pulp horns (Fig 3). Occlusal fissures were found to be most common in the maxillary CT. Teeth in the mid-arcade (08s in maxillary arcades and 09s in mandibular arcades) were over-represented (Fig 4); few fissures were observed in the rostral and caudal CT of all arcades (06 and 11s).

Figure 1.

Occlusal fissure on maxillary tooth 108 associated with caudal palatal pulp horn (arrow). B: buccal, P: palatal, R: rostral, C: caudal.

Figure 2.

Occlusal fissure on maxillary tooth 109 (caudal palatal pulp horn) with extension through enamel to peripheral cementum (arrow).

Figure 3.

Graph showing total number of dental occlusal fissures (identified by pulp horn) in relation to arcades 1–4.

Figure 4.

Total number of affected CT in relation to Triadan position.

Fissure orientation was recorded as being predominantly between secondary dentine and the adjacent (buccal or palatal) margin of the tooth (198/227 CT; 87.2%). In only 4 maxillary CT were fissures seen to extend toward infundibular enamel. Approximately half of all fissures extended into or through the enamel layer (128/257 fissures, 49.8%) (Figs 5 and 6). Fissures extended in a mesio-distal direction (sometimes linking adjacent pulp horns) in 25 CT; these were predominantly (21/25; 84.0%) in mandibular teeth. In one of these cases a concurrent sagittal mandibular CT fracture was present and it was felt that the multiple occlusal fissures on adjacent teeth could be indicative of prodromal fracture pathology.

Figure 5.

Mandibular tooth 410: rostral buccal pulp horn affected, with caudobuccal fissure direction. Note fissure does not extend to peripheral enamel (arrow). B: buccal, L: lingual, R: rostral, C: caudal.

Figure 6.

Mandibular tooth 409 with fissure extending from rostral buccal pulp horn into peripheral enamel (arrow).

A total of 111 CT were identified as primary sites of dental pathology related to presenting complaint. Of these, only 6% (7/111) were observed to have occlusal fissures: no predominant pathology type was noted in these teeth. The number of horses that had at least one occlusal fissure was not significantly (P = 0.3004, Chi-squared test) different to those without fissures regardless of whether a diseased CT was present. The depth of fissures could not be determined from oral endoscopic examination alone; as this was not a cadaveric study it was not possible to perform histological analysis of the lesions. Only one fissure was noted in association with a ‘pitted’ pulp typical of those seen in compromised teeth.


Recent histological and imaging studies of equine cheek teeth in health and disease have contributed significantly to the level of understanding of the aetiopathogenesis of equine dental infections (Dacre 2005, Dacre et al. 2007, 2008a,b,c,d; du Toit et al. 2008a; Windley et al. 2009). Advances in imaging have led to greater awareness of the potential importance of even small defects of the erupted crown and periodontium of CT in the diagnosis of clinical disease (Simhofer et al. 2008; van den Enden and Dixon 2008; Ramzan 2009). One feature of the occlusal surface of CT regularly observed by clinicians performing increasingly detailed oral examinations (Simhofer et al. 2008) is that of occlusal ‘fissures’. It has been suggested that these fissures may be implicated in the pathogenesis of some apical dental infections or crown fractures (van den Enden and Dixon 2008; Dacre et al. 2008c); however, to date they have been subjected to little investigation. The first step toward determining the significance of occlusal fissures must be the documentation of their prevalence, location and association with intercurrent dental disease.

The utilisation of oral endoscopic video recordings in this study permitted the detailed retrospective examination of the CT of cases submitted to a referral hospital for dental investigation or treatment. Inclusion criteria ensured image quality comparable to that of a cadaveric study, i.e. high definition images sufficient to identify even very small defects of the clinical crown of all CT in each case. The advantages of oral endoscopic examinations over those conducted with a dental mirror are well established (Tremaine 2005; Griss and Simhofer 2006; Simhofer et al. 2008; Ramzan 2009), and the attention paid by current workers in the field to dental/oral features until recently considered insignificant means that the findings of many previous cadaveric studies of dental disease should be viewed with caution (Dacre et al. 2008a).

The occlusal fissures observed in the present study were all of characteristic appearance: that of a fine, hairline ‘crack’ in the occlusal surface with one end invariably confluent with the secondary dentine overlying a pulp horn. The very fine nature of the fissures precluded the use of any probe to assess the depth of the defects. The point of origin of fissures along the pulpar secondary dentine varied between teeth. The fissures were usually stained similarly to the secondary dentine but on occasion were encountered free of pigment. All fissures appeared to radiate from the secondary dentine through primary dentine, and extended a variable distance towards, into and sometimes through the adjacent enamel. Only a small number of fissures extending into peripheral cementum were detected. The orientation of most fissures was between secondary dentine and the ipsilateral border of the tooth, with some mesio-distal variation. Only a small number of teeth were noted to have fissures directed in a sagittal (mesio-distal) plane, or axially towards infundibular enamel; of the CT with sagittal fissures, mandibular CT were over-represented. Mandibular teeth were also more likely to have multiple fissures than maxillary teeth.

Of considerable interest in the present study was the consistent appearance of fissures regardless of tooth, gender or age as well as the detection of predilection sites for fissure location in both maxillary and mandibular CT. The great majority (92.1%) of maxillary fissures occurred in association with the secondary dentine of the caudal palatal pulp horn (pulp horn 4) (du Toit et al. 2008a). Mandibular fissures were predominantly associated with the buccal pulp horns (95.7%). These predilection sites have not been documented before and may prove to be of some importance when considering possible causes of the condition.

The prevalence of occlusal fissures in the present population was high (58.2%) and in agreement with the results of recent work by Simhofer et al. (2008) who recorded fissures in over half (54.3%) of their study population. In the latter study, fissures were found to be most common in the mandibular CT, and these were typically considered to extend from the buccal enamel fold in towards the secondary dentine. This contrasts with the findings presented here, in which fissures were most common in the maxillary CT, and regardless of tooth location invariably had contact with pulpar secondary dentine (with variable extension to the periphery of the tooth). The authors therefore considered it probable that these fissures originated in secondary dentine, although actual direction of propagation could not be determined without ultrastructural analysis. It is possible that these differences in fissure description may reflect the techniques used for image capture or analysis in the studies, rather than any inherent difference between the studied populations.

While the findings of the present study differ from those of Simhofer et al. (2008) with regard to fissure description and location, it is clear from both data sets that many asymptomatic teeth have occlusal fissures. It follows that any causal association with dental disease is improbable. The location (primarily from palatal pulps in maxillary CT) and direction (primarily palatal) of fissures reported here are inconsistent with the configuration of most maxillary CT slab fractures: these typically run through buccal pulp horns in a sagittal (mesio-distal) direction (Dacre et al. 2007). It is possible that the significance of fissures in mandibular teeth may differ from that in maxillary teeth: in one case presented for treatment of a sagittal fracture of a mandibular tooth (410), multiple communicating fissures of the adjacent 409 were observed and it was considered that this might represent prodromal fracture pathology. It is possible that these mesio-distal fissures are of greater significance for future slab fracture propagation, and their greater incidence on mandibular CT may reflect a different aetiopathogenesis to maxillary fissures.

Although a link with sagittal CT fractures is unlikely in most animals, it is reasonable to presume that fissures could potentially contribute to pulpar compromise. Any breach of the occlusal secondary dentine has the potential to allow bacterial colonisation of vital pulp (Dacre et al. 2008c). In this study, maxillary fissures were predominantly located at the caudal palatal pulp horn; a recent study of pulpar exposure in apically infected cheek teeth determined that over 70% of exposed maxillary pulp horns were found at the caudal palatal site (van den Enden and Dixon 2008). In spite of this similarity in predilection site, the present study found no correlation between presence of fissures and location of diseased tooth when considering CT confirmed with apical infection; in addition to this only a single fissure was observed in association with suspected compromise to a pulp horn (Casey and Tremaine 2009). It is unlikely therefore that occlusal fissures are implicated in the pathogenesis of dental sepsis in the horse.

The origin of equine occlusal fissures is uncertain and until histological investigations of the condition are undertaken such basic information as depth of fissure from occlusal surface, relationship to vital pulp and transience or otherwise of the fissures in any individual will remain unknown. Regardless of this, the implication of the varying length of the fissures and extension to peripheral enamel in some teeth is that the feature does truly represent a ‘crack’ in the hard dental tissues rather than a developmental defect. Cracks in human teeth have long been recognised and can be found in symptomatic and asymptomatic teeth (Kahler 2008). Symptoms, when present, include pain on biting (or release from biting) and sensitivity to thermal changes and are determined by many factors including depth of crack and vitality or inflammation of pulp. Classification of these defects has varied between authors but essentially hinges on dental tissues involved, orientation and depth of crack (Roh and Lee 2006).

The predominance of palatal (medial) location of maxillary fissures, and buccal (lateral) mandibular fissures in this population is of particular interest and merits further investigation. Palatal primary dentine is significantly thicker than buccal primary dentine in maxillary CT in horses and donkeys (du Toit et al. 2008b), and both primary and secondary dentine are thicker on the medial aspect of maxillary pulp horns than the lateral aspect (Shaw et al. 2008). While this can probably be attributed to developmental rather than adaptive factors it has been suggested that the greatest lateral forces during equine mastication act on the palatal aspect of the maxillary teeth and the buccal aspect of the mandibular teeth (Bonin et al. 2007). It is plausible that these masticatory forces are a determinant of fissure location, and this is further supported by the predominance of centrally-located CT with fissures in the present study. There was no suggestion from the clinical case histories that any feature of previous prophylactic dental treatments was a factor in the development of the fissures, although insufficient information was available to permit statistical analysis of risk factors. In human subjects, cracks in teeth generally occur as a result of occlusal forces and iatrogenic procedures (Kahler 2008).

The findings presented here represent the first detailed description of prevalence, location and relationship to intercurrent dental disease of occlusal fissures of the equine cheek tooth. Occlusal fissures appear to be a common finding in asymptomatic teeth, and should not be considered an indicator of dental disease. Predilection sites for fissures exist in both the maxillary (caudal palatal pulp horn) and mandibular (buccal pulp horns) arcades and are rarely encountered in the rostral (Triadan 06) or caudal (Triadan 11) cheek teeth. It is postulated that masticatory forces contribute to development of fissures, however histological studies are required to fully define the condition.

Manufacturers' addresses

1 Everest VIT (UK) Ltd., Burford, Oxfordshire, UK.

2 Pentax (UK) Ltd., Langley, Slough, Berkshire, UK.

3 Apple, Cupertino, California, USA.

4 Statview, Abacus Concepts, Inc., Berkeley, California, USA.

Author contributions The initiation, conception, planning and pathology for this study were by P.H.L.R. Its execution and writing were by P.H.L.R. and L.P., with statistics by L.P.