Post mortem survey of peripheral dental caries in 510 Swedish horses


  • I. GERE,

    1. Geres Veterinärpraktik, Västerslänt 142, 42435 Angered, Sweden; and Division of Veterinary Clinical Studies, The Royal (Dick) School of Veterinary Studies, University of Edinburgh, Easter Bush Veterinary Centre, Midlothian, Scotland, EH25 9RG, UK.
    Search for more papers by this author
  • P. M. DIXON

    1. Geres Veterinärpraktik, Västerslänt 142, 42435 Angered, Sweden; and Division of Veterinary Clinical Studies, The Royal (Dick) School of Veterinary Studies, University of Edinburgh, Easter Bush Veterinary Centre, Midlothian, Scotland, EH25 9RG, UK.
    Search for more papers by this author
    • *

      Author to whom correspondence should be addressed.


Reasons for performing study: Peripheral caries (PC) of equine teeth is a poorly described disorder that can cause serious clinical problems if it progresses.

Objectives: To assess the prevalence, sites and severity of PC in a population of Swedish horses.

Methods: A post mortem study of 510 equine skulls was performed in 2 Swedish equine abattoirs.

Results: PC only affected the cheek teeth (CT) and was present in 6.1% (31/510) of skulls. It affected mainly the peripheral cementum, and 87% of PC in the 29 affected mature horses occurred in the 3 caudal CT (Triadan 09-11). Concurrent infundibular caries involving most maxillary CT (mean 9.7/skull) was present in 32% of skulls affected with PC. Trotting horses (mean age 8.1 years) believed to be on a high concentrate and silage diet were preferentially affected with PC in this population. Food was usually tightly adherent to the PC lesions and this feature may have promoted the progression of the disease. Significantly increased levels of diastemata were present in PC-affected horses, and periodontal disease was present in areas adjacent to some PC lesions.

Conclusions: PC is a relatively common disorder of horses under certain management conditions that can progress to cause serious dental disorders, especially if concurrent, widespread infundibular caries is present.

Potential relevance: Equine clinicians should be aware of this significant dental disorder and research into its aetiopathogenesis, possible prevention and treatment are required.


Dental caries is defined as a bacterial disease of the calcified dental tissues, characterised by demineralisation of the inorganic and destruction of the organic dental structures (Soames and Southam 2005). Peripheral caries (PC), which usually affects the cheek teeth (CT), has been recorded briefly in horses (Baker 1979; Easley 1991; Dacre 2006; Dixon et al. 2010), with short descriptions of its pathology given by Dacre (2005) and du Toit et al. (2008a). Although sometimes termed peripheral cemental caries, progression of this disorder leads to caries of the underlying enamel and dentine (du Toit et al. 2008a) and so PC is a more suitable term for this disorder (Dixon et al. 2010). The anatomical location of caries on the periphery of affected CT differentiates PC from the well described disorder of infundibular caries that affects maxillary CT only (Colyer 1906; Honma et al. 1962; Baker 1974; Kilic et al. 1997a; Brigham and Duncanson 2000; du Toit et al. 2008a). Caries (of all the calcified dental tissues) can also occur in equine teeth secondary to dental fractures (Dacre et al. 2007), apical infections (Dacre et al. 2008a,b), and in incisors and canine teeth affected with equine odontoclastic tooth resorption and hypercementosis (EOTRH) syndrome (Staszyk et al. 2008).

In addition to the initial loss of cementum around the periphery of CT, destruction of peripheral cementum can also occur deep in cemental infoldings, thus contributing to increased occlusal wear in PC-affected CT (Dacre 2005). PC may also initiate the development of CT diastema by causing destruction of interproximal calcified tissues, and also instigates periodontal disease by damaging the normal tight junction between gingiva and peripheral cementum at the gingival margin (Dacre 2005). Destruction of peripheral cementum also allows the development of protruding areas of brittle enamel on the occlusal surface that may be prone to fracture (Dacre 2005). PC is anecdotally stated to preferentially affect the more caudally located CT in horses (Dacre 2006), which can be difficult to fully examine clinically, and therefore this disorder may be clinically underdiagnosed. Severe, widespread PC affecting groups of horses has been associated with feeding silage with high levels of added acid, and also with feeding high levels of processed maize foodstuffs (Dixon et al. 2010).

Despite the above reports, little objective information is available on the prevalence or degree of equine PC in different horse populations. The purpose of this study was to describe the prevalence and gross appearance of PC at post mortem examination in a population of horses examined in 2 Swedish horse abattoirs.

Materials and methods

Between December 2003 and April 2005, the skulls of 510 horses, that had passed health examinations prior to slaughter, were examined post mortem in 2 Swedish equine abattoirs. The abattoir records of each case included details of breed, gender and age. Information on the reason for slaughter and prior dental history were obtained by direct and telephone interviews with horseowners/trainers for 489 of the 510 cases. After being subjected to euthanasia by stunning and exsanguination, the heads were removed and the mandibles disarticulated to allow a complete visual examination of the oral cavity, with particular emphasis placed on documenting all instances of dental caries.

In this study, PC was classified using an adaptation of a human dental caries classification by Honma et al. (1962) for grading infundibular caries, as later modified by Dacre (2005) to also allow its use to classify equine PCs (Table 1).

Table 1. Grading of equine dental caries
Zero degree: No evidence of caries on a macroscopic level, but can include central infundibular cemental hypoplasia
First degree: Caries only affects cementum – subdivided into:
 Class 1 Small, pitting superficial focal lesions
 Class 2 Extensive destruction and loss of cementum
Second degree: Caries extends into adjacent enamel
Third degree: Caries extends into adjacent enamel and dentine
Fourth degree: Advanced caries affects the integrity of the tooth, predisposing to apical abscess or dental fracture
Fifth degree: Extensive caries results in tooth loss


The skulls of 510 horses (303 female and 207 male), mean age 12.7 (range 0.5–31) years, including 274 (53.7%) Warmblood trotters, 86 Swedish half-breds, 58 North Swedish horses, 24 ponies and 68 horses of 20 other breeds were examined. The reason for slaughter included: musculoskeletal problems (n = 162); poor racing performance or end of racing career (n = 122); old age (n = 68); lack of time/interest by owners (n = 53); behavioural problems (n = 23); infertility (n = 20); miscellaneous and multiple problems (n = 30); dental-related or perceived dental-related problems (n = 11). No information was available in 21 cases. Regular dental treatment was provided for 293 (60%) of horses, but only on an annual or more frequent basis in 112 (23%) and 6 owners stated this dental ‘care’ was given by farriers and 2 owners performed their own dental treatments.

Thirty-one skulls (6.1%) from 18 male and 13 female horses, mean age 8.1 years (range 1.5–15 years), including 29 mature horses and 2 immature horses (age 1.5 and 2.5 years) with incomplete dentition, had some CT affected with PC. No caries of incisors, canine or wolf teeth was found in any horse. The number of CT affected with PC varied from 7 to 23 (mean 11.8) per affected horse. Table 2 includes data on the 29 affected mature horses only. The 1.5-year-old horse had caries of all 12 deciduous CT (premolars) while in contrast the 2.5-year-old horse had caries of all its 8 permanent CT. The most affected CT had 1st degree (class 1 or 2) caries while more severe degrees of PC were uncommon (Table 2).

Table 2. Prevalence and severity of peripheral caries (PC) lesions at individual Triadan positions of 367 affected permanent cheek teeth (CT) in 29 mature horses
Teeth affectedDegree of PC060708091011
Maxillary CT1st degree class 1411414114
1st degree class 2035355643
2nd degree000000
3rd degree000000
4th degree001000
Mandibular CT1st degree class 11081878
1st degree class 2026264848
2nd degree000000
3rd degree000000
4th degree002200
Total 563695112113

Food was usually firmly adhered to the sides of PC-affected teeth (Fig 1) and substantial brushing and rinsing with water was needed to remove it. In contrast, food was readily washed off unaffected teeth. Examples of (washed) teeth with 1st degree PC are shown in Figures 2 and 3.

Figure 1.

Mandible of a 5-year-old Warmblood trotter affected with peripheral caries, showing food bilaterally attached to the periphery of the caudal 3 cheek teeth (Triadan 09–11), with the rostral 3 cheek teeth (Triadan 06-08) largely unaffected by peripheral caries. Diastemata with periodontal food pocketing are present, including between 409 and 410 (arrow).

Figure 2.

Washed mandible of 5-year-old Warmblood trotter showing bilateral, 1st degree, class 2 peripheral caries of the caudal 3 cheek teeth, with marginal involvement of the caudo-lateral aspect of 408. Diastemata are also present between the caudal cheek teeth, most obviously in the left hemimandible (arrows).

Figure 3.

Close-up view of mandibular cheek teeth of a 1.5-year-old Warmblood trotter that are bilaterally affected with 1st degree, class 1 peripheral caries. In addition, note the irregular, slightly erythematous gingival margin adjacent to the carious cementum.

Data in Table 2 show that 87% (320) of the 367 PC-affected teeth in the 29 affected mature horses were in the Triadan 09-11 positions, with only 13% (47) located in the Triadan 06-08 positions. In particular, 97% (225 of 232) of the 2 most caudally located CT (Triadan 10 and 11) of affected horses had PC, while just 4.7% (11 of 232) of the Triadan 06 or 07 CT were affected in these horses. This caudal distribution of PC is very apparent in the specimens shown in Figures 1, 2 and 5. Table 2 also shows that more severe (in particular 1st degree, class 2) caries was 13.6-fold more common in the 3 caudal, as compared to the 3 rostral CT (Figs 2 and 7).

Figure 5.

The mandibular cheek teeth on left (410 of 13-year-old Warmblood trotter) has a fracture on its buccal aspect involving the 1st pulp horn (arrow) with a fissure fracture extending into the 2nd pulp horn, while the tooth on the right (409) has a fracture of the 2nd pulp horn (arrow), with a fissure fracture extending into the 1st pulp horn. The tooth on left has 1st degree, class 2 peripheral caries, with destruction of the underlying enamel also present at the fracture site.

Figure 7.

First degree, class 2 caries (almost total destruction of peripheral cementum) with discolouration of underlying enamel is confined to the caudal 3 maxillary cheek teeth (to left of image) in this 5-year-old Warmblood trotter. The more rostral tooth (208) has no peripheral caries on its lateral or medial margins. Concurrent 3rd degree infundibular caries is present in all 4 cheek teeth. Deep periodontal food pocketing was present in the diastema between 210 and 211 (arrow).

Dental fractures (4th degree caries) were found in 5 CT in 3 of the 31 PC-affected skulls (9.7%) (Fig 5), including in 4 mandibular and one maxillary CT. All 5 fractures were on the buccal aspect of CT and all ran through a single pulp horn, i.e. 1st or 2nd pulp horns, using the revised equine pulp numbering system of du Toit et al. (2008b), with fissure fractures extending to the adjacent ipsilateral pulp horn also present in some CT (Fig 5). No complete fracture planes ran through sites of normal anatomical weaknesses, e.g. sagittally through both the 1st and 2nd pulp horns, or through both infundibula, which are common fracture patterns in idiopathic fractures of equine CT (Dacre et al. 2007). Idiopathic fractures (including lateral ‘slab’ and midline sagittal fractures) were present in the non-PC-affected horses with 47 idiopathic CT fractures present in 43 of these 479 horses (10.9% prevalence) that did not significantly differ (P = 0.06; Fisher's exact probability test) from the prevalence in PC-affected horses.

Ten (30%) of the 31 horses with PC also had infundibular caries that involved multiple maxillary CT in all 10 horses, with a total of 97 CT concurrently affected (mean 9.7 affected CT/horse) (Table 3; Fig 6). The horses without PC had a prevalence of infundibular caries of 16% (76/479) that was significantly lower (P = 0.02; χ21= 5.58, χ2 test) than that of PC-affected horses. Additionally, these 76 horses without PC had a total of 214 CT affected (mean affected 2.8 teeth/horse), with 83.6% of affected CT in the maxillary 09 position. In PC-affected horses, more severe (2nd and 3rd degree) infundibular caries was 3-fold more common in the caudal 3 CT (Triadan 09-11 positions) than in the rostral (Triadan 06-08) 3 CT. As noted, no midline sagittal fractures were found in CT with infundibular caries, despite extensive caries of both infundibulae being present in many of these CT (Fig 7). The extension of PC into peripheral enamel (2nd degree caries) was only present in CT with concurrent infundibular caries and the sites of enamel destruction at the rostral and caudal margins of CT, which was adjacent to areas of 3rd degree infundibular caries (Figs 6, 7) strongly indicated that an extension of infundibular caries, rather than of PC caused the enamel destruction in these maxillary CT.

Table 3. Triadan positions, degree of infundibular caries and numbers of maxillary cheek teeth having concurrent infundibular and peripheral dental caries
Triadan positionDegree of infundibular caries060708091011
 1st degree class 11612813109
1st degree class 2 21311
2nd degree  2223
3rd degree  2 55
4th degree      
Total number 161413181818
Figure 6.

In addition to peripheral caries affecting peripheral cement only, the tooth on left (Triadan 108 of 5-year-old North Swedish horse) has 3rd degree caries (i.e. involving cementum, enamel and dentine) of the caudal infundibulum (arrow) and less extensive 3rd degree caries of the rostral infundibulum.

There was a significantly higher prevalence of diastemata in horses with PC (64.5%; 20/31) compared to horses without PC (45.7%; 218/479) (P = 0.04; χ21= 4.23) with the more caudal inderdental spaces most commonly involved in both groups of horses.

Buccal mucosal lacerations, caused by sharp enamel points (all opposite maxillary 09s-11s), were present in 93.5% (28/31) of horses affected by PC (bilaterally in 22, unilaterally in 6) and dental-related lingual ulcerations were present in 2 heads. An 86.6% prevalence of buccal ulceration was found in the 479 heads without PC, which was slightly lower than the prevalence of buccal ulceration in PC-affected skulls (P = 0.02; Fisher's exact probability test). Low-grade gingivitis was commonly present adjacent to areas of PC (Figs 1, 3, 4), but only one horse had deep periodontal food pocketing, not associated with a diastema, adjacent to teeth affected with PC.

Figure 4.

The buccal (lateral) aspect of 2 mandibular cheek teeth (410 left; 409 right) of an 11-year-old Warmblood trotter affected with 1st degree, class 2, peripheral caries showing disruption of the gingival margin, especially adjacent to the interdental space. A periodontal pocket is apparent where the gingiva has been retracted.

A statistically significant difference in the prevalence of PC was found between breeds with 28 of the 31 PC-affected cases being Swedish trotters (10.2% prevalence in the 274 horses of this breed) while a 1.3% prevalence (3/236) were found among the other 236 horses of other breeds (P<0.001, χ21= 16.25). One of these latter 3 horses (a Swedish half-breed horse) lived in a trotting stable on the same diet as the trotting horses. No management information was available on the other 2 non-trotter horses affected with PC.


No incisors, canine or wolf teeth (Triadan 05) were affected with PC in this study. Horses affected with PC had multiple CT involved (mean 12.3 affected CT/horse) and the more caudally located CT were more commonly affected by PC, with 87% of affected teeth situated in Triadan positions 09–11 and only 13% in the 06–08 positions (Table 1). The more caudal affected CT also had more severe PC than rostrally situated CT (Table 2). Combined, these findings indicate that the environment in the caudal aspect of the oral cavity (likely a low pH and perhaps abnormal bacterial proliferation) is more liable to induce PC in susceptible horses. As the equine salivary ducts drain into the rostral aspect of the oral cavity, decreased salivary pH buffering may be present in the caudal aspect of the oral cavity (C. Staszyk, personal communication).

Horses suffering from the classical form of infundibular caries usually have a low number of affected CT, and usually those in the Triadan 09 position (Baker 1974; Kilic et al. 1997b; Crabill and Schumacher 1998; Dixon et al. 2010; Fitzgibbons et al. 2010). Similarly, this study found that 83.6% of CT with infundibular caries in the non-PC-affected skulls were in the Triadan 09 position, with a mean of 2.8 affected CT per non-PC-affected horse. In contrast, the 10 PC-affected horses that had infundibular caries, had most of their maxillary CT (mean 9.7 CT per horse) affected with infundibular caries, indicating a major difference between these latter cases and the classical form of infundibular caries. The caudal distribution of infundibular caries, especially of more severe infundibular caries (Table 3), further supports that the caudal aspect of the oral cavity in the current PC-affected cases had a caries-inducing environment. In this study, where the mean age of all horses was 12.7 years, PC tended to affect younger horses (mean age 8.1 years). This finding is also in contrast to classical infundibular caries, which increases in prevalence in older horses (Honma et al. 1962; Baker 1974; Crabill and Schumacher 1998). However, this young age likely just reflects the at-risk population in this study, i.e. Swedish trotting horses that were subjected to euthanasia towards the end of their normal racing career or earlier, if racing poorly or if suffering intercurrent diseases.

Teeth affected with PC were consistently found to have food tightly attached to their periphery that could only be removed with difficulty. This finding may be associated with the irregular and porous nature of peripheral cementum affected by PC that allowed tight adherence of food, in contrast to the normal smooth surface of peripheral CT cementum (Fig 2). It is likely that such adherence of food to the periphery of teeth in the early stages of PC would propagate the disease by preventing the normal cleaning action of food movements during mastication and the pH buffering action of saliva on the teeth, likely allowing prolonged and higher levels of caries-inducing bacteria and food-derived acids to remain in direct contact with the periphery of affected teeth. However, this mechanism would not explain the widespread presence of infundibular caries concurrently present in the maxillary CT of 30% of PC-affected horses.

It was believed that a reduction in the protective (‘crack stopper’) effect of peripheral cementum on adjacent (brittle) enamel in some PC-affected teeth, may have contributed to the development of fractures, as the fractured enamel did not appear to be eroded by caries, but being unsupported by normal peripheral cementum, it may have fractured under the high pressures of normal equine mastication. However, there was no significant difference in prevalence of idiopathic CT fractures between PC-affected horses (9.7% had one or more fractured CT) and non-PC-affected horses (10.4% had idiopathic CT fractures). The pattern of a single (buccally situated) pulp involvement in the CT fracture plane (albeit with fissure fractures to adjacent buccally situated pulps also present in some CT) is possibly a characteristic of fractured teeth concurrently affected with PC, with further examples of this fracture pattern in PC-affected CT illustrated by Dacre (2005) and Dixon and Dacre (2005). Lacerations on the buccal and lingual mucous membranes caused by sharp enamel points on CT are the most common equid dental disorder and are certainly not specific to PC, and a similar prevalence of buccal ulceration was found in PC-affected (93.5%) and non-PC-affected (86.6%) horses in this study.

The lower age of horses affected with PC (8.1 years) as compared to the whole population studied (mean 12.7 years) is as noted, likely related to the at-risk population that are often subjected to euthanasia at a young age, i.e. 90.3% of horses affected with PC were Warmblood trotters that comprised just 53.7% of the population studied. Additionally, one of the 3 nontrotter horses affected with PC lived in a stable with the same management as trotting horses. It seems more likely that environmental factors, particularly diet, anecdotally the feeding of higher concentrate levels and silage (haylage), rather than breed susceptibility, predisposes to the development of caries in affected horses.

Regardless of affected site or initiating factors, caries is a disease of the calcified dental tissues resulting from the local production of acids by micro-organisms fermenting carbohydrates as originally proposed by Miller (1891) over a century ago. Caries of human dentine and enamel starts with demineralisation (due to low pH) of these tissues prior to bacterial infection, whereas in caries of cementum, demineralisation and bacterial infection occur simultaneously (Frank 1990). Recent ultrastructural studies of donkey teeth with cemental caries supports these findings in equids also (du Toit et al. 2008a). There is limited published information on the normal bacteriology of the equine mouth and even less on the bacteria that incite dental caries formation and this is an area that greatly needs investigation. Because destruction of dental tissues can also be due to nonbacterial causes, such as chemical erosions (Soames and Southam 2005), it could be argued that the dental lesions detected in this study are in fact chemical dental erosions caused fully by low dietary pH rather than bacterial-derived acids, but the gross similarity of the current lesions to the PC lesions described briefly by Dacre (2005) and du Toit et al. (2008a), and also to infundibular caries lesions and the concurrent presence of very widespread infundibular caries in some of the current cases, strongly indicate that the described lesions are true caries lesions. Detailed histological and ultrastructural studies are also needed for this disorder.

As noted, dietary mechanisms may play a part in the pathogenesis of PC, with affected horses believed to be fed higher levels of concentrates than unaffected horses. Such diets are higher in energy, lower in cellulose and need less mastication than the natural diet of horses (grass) or traditional forage diets such as hay. The reduced masticatory movements and reduced time spent cycling forage and saliva around their oral cavities could lead to increased bacterial plaque accumulating on the organic pellicle that has been shown to cover normal equine teeth (Kilic et al. 1997b). Additionally, the increased levels of simpler carbohydrates in high concentrate diets would also be more likely to be fermented into organic acids and so lower the oral pH. The widespread occurrence of severe and often generalised PC of equine and donkey mouths in Mexico has been attributed to feeding almost a solely processed maize diet (Dixon et al. 2010).

Silage (haylage) is increasingly used instead of hay as horse forage worldwide. When harvested early, silage has higher levels of simpler carbohydrates (and lower cellulose and lignin levels) than traditionally saved hay that in turn, produces high levels of organic acids during the ensilage process. Furthermore, the addition of high levels of extraneous acids to preserve silage has been associated with the development of severe and widespread PC in all horses fed such acid-treated silage (Dixon et al. 2010). There is increased use of silage feeding in Swedish horses, especially in larger stables where silage making and handling can be mechanised and we hypothesise that the feeding of silage may also play a role in the pathogenesis of PC in the current cases, by reducing oral pH. In support of this hypothesis, 2 Swedish dental surveys performed prior to the introduction of widespread silage feeding in horses (Lundström and Pettersson 1988, 1990) found just a 0.9% prevalence of PC in a total of 335 Swedish horses.

In conclusion, this survey showed that equine PC is a disease of CT that preferentially affects the more caudal CT, and that widespread infundibular caries is often concurrently present in affected horses. In the population examined, the disease preferentially affected younger trotting horses that were believed to be fed a high concentrate and silage diet. Most peripheral CT carious lesions had tightly adherent food that may play a part in the pathogenesis of the disease.


We thank Carsten Staszyk and Nicole du Toit for help with the manuscript and Nicole du Toit for statistical assistance.

Author contributions Both authors contributed to all aspects of this study.