Materials and Methods

Case selection—Medical records of Thoroughbred racehorses with a diagnosis of CSBI of the distal portion of MC3/MT3 at Rood and Riddle Equine Hospital between January 2000 and January 2009 were reviewed. To be included in this study, horses were required to have undergone lameness examination, diagnostic analgesia, nuclear scintigraphy, and MCPJ/MTPJ radiography to confirm the diagnosis. One clinician (LRB) examined all horses included in the study to reduce interobserver variability. Subjects were required to have definitive abolition of the lameness with diagnostic local anesthesia of the distal portion of MC3/MT3. Only subjects with no other concurrent lameness were included in the study to isolate the distal portion of MC3/MT3 for evaluation. Variables recorded included signalment, history, lameness examination details, nuclear scintigraphy findings, radiographic abnormalities, and treatment recommendations.

Lameness examination and diagnostic analgesia—Lameness examination was performed on an asphalt surface in a straight line and right and left circles in hand. The American Association of Equine Practitioners lameness scoring system on a scale of 0 to 5 (0 = no lameness; 5 = non—weight-bearing lameness) was used and modified to include 0.5 gradations.¹⁶ Metacarpophalangeal joint and MTPJ effusion was graded as mild, moderate, or severe on the basis of observation and palpation. Flexion tests of MCPJ/MTPJ were performed as previously described,¹⁷ and the response was graded on the same lameness scale.¹⁶ Diagnostic analgesia of the lateral condyle of MT3 and the lateral and medial condyles of MC3 was achieved via anesthesia of the lateral plantar metatarsal nerve and the medial and lateral palmar metacarpal nerves, respectively.^8,18

Nuclear scintigraphy—Horses were maintained in temperature-controlled stalls (15.5° to 23.8°C [60° to 75°F]) and injected IV with 200 to 250 mCi of technetium Tc 99m medronate via a jugular catheter at 2.5 to 3 hours before the imaging procedure. The distal portion of the limbs was bandaged at the time of injection to prevent contamination with isotope excreted in the urine during the incubation period between injection and imaging. Horses were sedated with detomidine hydrochloride (0.01 mg/kg [0.004 mg/lb], IV) during the examination. Scintigraphy was performed with a rectangular field-of-view camera.^a A standard protocol that included lateral and dorsopalmar or dorsoplantar views of the MCPJ/MTPJ was followed. Images were processed with proprietary software. Increased radiopharmaceutical uptake of the distal portion of MC3/MT3 was graded as mild, moderate, or intense.¹²

Radiographic examination—Computed radio-graphic examination^c of MCPJ/MTPJ consisted of a 30° dorsoproximal-palmarodistal (plantarodistal) oblique projection, a lateromedial projection (obtained with the horse standing), a flexed lateromedial projection (obtained with the joint flexed), a dorsoproximolateral-palmarodistomedial (plantarodistomedial) oblique projection (made at 20° proximal to the supporting surface and 30° lateral to the dorsopalmar [dorsoplantar] line), a dorsoproximomedial-palmarodistolateral (plantarodistolateral) oblique projection (made at 20° proximal to the supporting surface and 30° medial to the dorsopalmar [dorsoplantar] line), a nonconventional flexed 35° dorsodistal-palmaroproximal oblique projection (obtained with MCPJ flexed and the first phalanx perpendicular to the ground),¹⁹ and a nonconventional flexed 30° plantarodistal-dorsoproximal oblique projection (obtained with MTPJ flexed and the first phalanx perpendicular to the ground).^20,21

Radiographic findings were divided into 3 categories: no noteworthy abnormalities, trabecular bone remodeling represented by irregular sclerosis or radio-lucency of the condyles, and fracture including incomplete palmar (plantar) parasagittal condylar fracture distinguished by a linear radiolucent line < 10 mm in length oriented perpendicular to the joint surface. These fractures were visible only on the flexed non-conventional oblique projections. Semilunar articular fracture characterized by a crescent-shaped lucency with the long axis aligned parallel with the joint surface on the palmar (plantar) aspect of the condyle (Figure 1) was evident on the flexed 30° plantarodistal-dorsoproximal oblique radiographic view of the MTPJ or the flexed 35° dorsodistal-palmaroproximal oblique projection of the MCPJ. Horses with incomplete palmar (plantar) parasagittal condylar fractures > 10 mm in length or having fractures visible on both the standing 30° dorsoproximal-palmarodistal (plantarodistal) oblique radiographic view and the flexed nonconventional oblique projections were treated surgically and therefore were excluded from the study.

Representative flexed 30° plantarodistal-dorsoproximal oblique radiographic view of MTPJ of a horse. Notice the typical appearance of a semilunar articular fracture (arrowheads) on the plantar surface of the distal aspect of MT3.

Citation: Journal of the American Veterinary Medical Association 238, 10; 10.2460/javma.238.10.1316

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Representative flexed 30° plantarodistal-dorsoproximal oblique radiographic view of MTPJ of a horse. Notice the typical appearance of a semilunar articular fracture (arrowheads) on the plantar surface of the distal aspect of MT3.

Citation: Journal of the American Veterinary Medical Association 238, 10; 10.2460/javma.238.10.1316

• Download Figure
• Download figure as PowerPoint slide

Representative flexed 30° plantarodistal-dorsoproximal oblique radiographic view of MTPJ of a horse. Notice the typical appearance of a semilunar articular fracture (arrowheads) on the plantar surface of the distal aspect of MT3.

Citation: Journal of the American Veterinary Medical Association 238, 10; 10.2460/javma.238.10.1316

• Download Figure
• Download figure as PowerPoint slide

Treatment and follow-up—Horses were prescribed free-choice exercise in paddocks of at least 1 acre for an initial period of 60 days. Owners or trainers of 3 horses elected to forgo paddock turnout for varying reasons, and an exercise regime of multiple exercise periods each day with varying gaits involving more walking and trotting and less galloping was initially used with these horses; however, 2 of these horses were eventually given free-choice exercise in a paddock when the modified treatment routine failed to resolve the lameness within 30 days.

The 60-day reexamination included radiographic and lameness evaluation by the second author (LRB) at the Rood and Riddle Equine Hospital or by referring veterinarians. Radiographs that were deemed to have abnormal findings by referring veterinarians after the treatment period were provided to the second author (LRB) for assessment. If lameness or radiographic abnormalities persisted, the horses were prescribed continued turnout with reevaluation at 30-day intervals. When horses had resolution of radiographic abnormalities and were sound, they were returned to training.

Performance assessment—To determine the effect of CSBI of the distal portion of MC3/MT3 on performance, pre- and postinjury race starts, earnings, earnings per start, and racing class (maiden, claiming, allowance, handicap, and stakes races, in ascending order) were recorded from racing records obtained from the Jockey Club Information Systems.^d The number of days from the diagnosis to first start was also recorded. Loss to follow-up or other reasons for nonperformance were not allowed.

Statistical analysis—Because variables measured in dollar amounts were highly skewed to the right, these values were log transformed prior to statistical analysis. Paired t tests were used to compare continuous data with 2 observations for the same variable. A 1-way ANOVA for categorical variables or linear regression for continuous variables was used to determine the influence of age, sex, increased radiopharmaceutical uptake, joint effusion, joint effusion score, initial lame limb, lameness grade, limbs affected, flexion test score, radiographic findings, and racing class on performance and time from diagnosis to first start. The same variables were then compared with the postinjury performance values to determine whether any individual factor influenced outcome. A χ² test was used to determine the association between independent categorical variables and a diagnosis of CSBI of the distal portion of MC3/MT3. All analyses were performed with commercially available software,^e and a value of P ≤ 0.05 was considered significant.

Results

Subject details—Four hundred forty-seven Thoroughbred racehorses were diagnosed with CSBI of the distal portion of MC3/MT3 during the study period (58% [260/447] males and 42% [187/447] females). Fifty-five subjects met the inclusion criteria for the study, with the remaining horses excluded because of concurrent lameness of other sites identified with diagnostic local analgesia at the time of the lameness examination or because of incomplete study set data. Age distribution ranged from 2 to 6 years with 24% (13/55) 2 year olds, 47% (26/55) 3 year olds, 20% (11/55) 4 year olds, and 9% (5/55) 6 year olds. There were 75% (41/55) males and 25% (14/55) females included in the study. Compared with all 4,028 Thoroughbred racehorses admitted to our hospital for lameness examination during the study period (43% [1,725/4,028] males and 57% [2,303/4,028] females), study males (n = 41) were significantly (P < 0.001) more likely than were study females (14) to have CSBI of the distal portion of MC3/MT3 (odds ratio, 3.99; 95% confidence interval, 2.17 to 7.34]). The most common reason for evaluation at the time of admission was lameness, decreased performance, or both. All horses had trained, raced, or both on more than 1 surface at more than 1 location prior to hospital admission.

Lameness examination findings—Lameness examination findings were summarized (Table 1). All subjects had definitive (at least 90%) improvement in lameness after perineural anesthesia of the distal aspects of the lateral plantar metatarsal nerve or lateral and medial palmar metacarpal nerves of the affected limbs. Two of the 55 (4%) subjects received intra-articular anesthesia of the MTPJ that resulted in no reduction in lameness; the lameness was subsequently abolished with perineural anesthesia of the lateral plantar metatarsal nerve.

Nuclear scintigraphic findings—All patients had increased radiopharmaceutical uptake bilaterally or in all 4 limbs, ranging from mild to intense (Table 2). Fifteen percent (8/55) of the horses had mild increased radiopharmaceutical uptake, 51% (28/55) had moderate increased radiopharmaceutical uptake, and 35% (19/55) had intense increased radiopharmaceutical uptake. There were no complications or interference with examination findings in any horse with perineural anesthesia performed 24 to 48 hours before or after nuclear scintigraphic examination.

Radiographic examination—Radiographically, 69% (38/55) of patients had abnormalities, whereas 31% (17/55) had no identifiable osseous changes of the distal portion of MC3/MT3. Of the horses with radiographic abnormalities, 66% (25/38) had radiographic signs of subchondral bone or trabecular bone remodeling, while 34% (13/38) were diagnosed with fractures. Of the 13 horses with fractures, 10 had parasagittal fractures, with 3 horses bilaterally affected in both MCPJs. The remaining 3 horses had semilunar articular fracture, with 1 horse affected in both MCPJs.

Treatment—Ninety-five percent (52/55) of horses were treated immediately with 60 days of free-choice exercise in a paddock followed by reexamination. The remaining 5% (3/55) with owners who requested an alternate training program were advised to use a varying intensity-reduced exercise training program with multiple exercise periods per day with fewer long-distance gallops. However, 2 of these 3 horses were given free-choice exercise for 60 days after 30 days of the alternate training program as they remained lame.

Follow-up examination—Thirty-three percent (18/55) of horses were reexamined at the Rood and Riddle Equine Hospital, with 6 of these 18 horses requiring additional turnout for a mean of 50 days (median, 45 days; range, 30 to 90 days). The remaining horses were reexamined by referring veterinarians, and any radiographic abnormalities were reviewed by the second author (LRB). All horses with radiographic signs of fracture were reexamined by the second author. Of these 13 horses, 9 had resolution of radiographic abnormalities after 60 days of free exercise, whereas the remaining 4 needed an additional turnout. All horses had resolution of radiographic signs of fracture before returning to training, although 2 horses with semilunar articular fractures never raced after injury. Five horses with radiographic signs of bone remodeling were reevaluated at the author's practice; 2 of the 5 horses had improvement of radiographic signs of bone remodeling but required an additional 30 days of turnout for complete resolution of lameness. Radiographs from referring veterinarians were provided to the second author for an additional 8 horses with bone remodeling during the initial examination, with all 8 horses having resolution of abnormalities after the treatment period. The remaining 12 horses that were diagnosed with bone remodeling of the distal portion of MC3/MT3 were considered resolved by referring veterinarians after 60 days of free-choice exercise.

Performance assessment—Follow-up data revealed that 95% (52/55) of the horses in this study with CSBI of the distal portion of MC3/MT3 raced after injury. In the horses receiving free exercise initially (n = 52), the mean number of days from diagnosis to first start was 230 (median, 194 days; range, 98 to 719 days). Eighty-four percent (46/55) of the patients had at least 1 start prior to a diagnosis of CSBI, while 82% (45/55) of horses raced both before and after injury. Of the 45 horses that raced both before and after a diagnosis of CSBI of the distal portion of MC3/MT3, 31% (14/45) had an increase in racing class, 31% (14/45) had no change in racing class, and 38% (17/45) had a decrease in racing class.

Following a diagnosis of CSBI in the distal aspect of the MC3/MT3, there was no significant difference in total earnings before injury, compared with after injury; horses had significantly more starts (P = 0.03) and decreased earnings per start (P = 0.02) after the injury.

Because many of the horses in the study were < 4 years of age and did not have the opportunity to participate in multiple races prior to diagnosis, the starts, earnings, and earnings per start of horses that had raced at least 4 times prior to and 4 times after the diagnosis were also examined. There was no significant difference in total earnings; although the earnings per start decreased after injury, compared with before injury, this difference was not significant (P = 0.06).

Discussion

This study was undertaken to document the postinjury performance of Thoroughbred racehorses with a definitive diagnosis of CSBI determined by use of lameness examination, perineural analgesia, nuclear scintigraphy, and radiography and treated with free-choice exercise in a paddock. All of the horses in this study were ≥ 2 years of age and participating in active race training and racing. Seventy-five percent (41/55) were males and 25% (14/55) were females, similar to a previous study.¹³ This could be related to multiple factors such as a larger proportion of young males in training, increased body weight, or higher training demands.²² There was a predilection for younger horses because 71% (39/55) of horses in this study were ≤ 3 years of age, similar to a previous report¹³ in which 78% of the subjects were ≤ 3 years of age, although this difference was not significant in our population and may just represent the distribution of training horses. Other stress-fatigue injuries including dorsal metacarpal disease, stress fractures of the tibia, and stress fractures of the humerus are predominantly found in younger horses and have been associated with the induction of bone modeling and remodeling in response to training.^23–25 Although most of the horses in this study were ≤ 3 years, 29% (16/55) of horses with CSBI were ≥ 4 years of age, documenting the occurrence of this disease in older Thoroughbred horses in agreement with prior studies.^9,13

Of the 447 horses that were diagnosed with CSBI of the distal portion of MC3/MT3, only 55 met the inclusion criteria, illustrating the frequency of multiple sites of lameness in the Thoroughbred racehorse and demonstrating the need for complete examination and diagnostic anesthesia to rule out other musculoskeletal diseases and guide treatment. The bilateral or quadrilateral distribution of this disease can make the identification of lameness difficult for some trainers and owners. This accounts for the frequent evaluation for decreased performance. The primary lameness in 55% (30/55) of horses in this study was the left hind limb, which may be the result of racing direction, much like the tendency of dorsal cortical stress fractures of MC3 to occur in the left limb.²⁶ Fifty-three percent (29/55) of the patients had bilateral lameness, with 90% (26/29) of those isolated to the hind limbs. This predilection for the plantarolateral aspect of MT3 involvement is likely related to an anatomic predisposition for the MTPJ as opposed to the gait and level of exercise intensity because this is also the most common finding in Standard-bred racehorses.¹² Although training styles vary widely within breeds, and especially between breeds, and because of their very different racing schedule, horses of both the Standardbred and Thoroughbred breeds can be affected with this condition when in high-speed training. Sixty percent (33/55) of the affected horses had a positive response to MCPJ/MTPJ flexion. Flexion of the MCPJ/MTPJ increases intra-articular pressure and subchondral intraosseous pressure and frequently elicits a pain response in horses with CSBI.²⁷ A small percentage (6/55 [11%]) of horses in this study had MCPJ/MTPJ effusion, and the 2 individuals that received intra-articular analgesia preceding local anesthesia of the distal aspect of MC3/MT3 failed to respond. Although the cause of joint effusion in these 2 horses was not determined definitively, they exemplify the importance of localizing the origin of pain.

Nuclear scintigraphy is described as being the most useful diagnostic modality for CSBI, and all horses in this study had increased radiopharmaceutical uptake of the distal portion of MC3/MT3.¹⁸ Eighty-five percent (47/55) of horses had increased radiopharmaceutical uptake subjectively graded as moderate or intense, representing increased bone remodeling and modeling activity in the distal condyles. In previous studies,^2,12 the presence of increased radiopharmaceutical uptake in the distal aspect of MT3 did not always correlate with clinical signs of lameness of the MTPJ, and nuclear scintigraphy was observed to lack specificity in determining normal bone remodeling from CSBI in horses receiving a controlled training regime. In agreement with this observation, there are a substantial number of horses at the authors' practice that have alternate sites of lameness as determined with diagnostic local anesthesia and concurrent increased radiopharmaceutical uptake at nonpainful sites of the distal condyles of MC3/MT3. Only horses with alleviation of the lameness in each affected limb by use of local analgesia of the distal portion of MC3/MT3 were included in this study to eliminate the possibility of including horses with increased radiopharmaceutical uptake that did not show signs of pain at that site. All horses with increased radiopharmaceutical uptake in the distal aspect of MT3 had focal uptake on the plantarolateral area, which is a common finding in both North American and European Thoroughbreds.^9,13,28 It has been hypothesized that this finding may be related to the nonperpendicular orientation of the lateral condyle to the long axis of the diaphysis in conjunction with increased load applied to the plantar cortex by force transfer from the suspensory apparatus via the lateral proximal sesamoid bone.^13,29 Anatomically, the lateral condyle of the distal portion of MT3 is also considerably smaller than the medial.³⁰ This may play a role in unequal force distribution. The bicondylar distribution of increased radiopharmaceutical uptake in the distal portion of many MC3s and the clinical experience with pain in either or both condyles necessitate diagnostic anesthesia of both the medial and lateral palmar metacarpal nerves to definitively isolate the origin of lameness in the forelimb.

Sixty-nine percent (38/55) of horses had abnormal radiographic findings. This is a higher percentage than results in previous reports^9,13 of 51% and 42%. All radiographic projections of horses in this study were performed by use of computed radiography, which provides higher resolution and greater detail than standard film-based radiography. Furthermore, abnormalities were identified most frequently on the flexed 35° dorsodistal-palmaroproximal oblique view (MCPJ) or flexed 30° plantarodistal-dorsoproximal oblique projection (MTPJ),¹⁹ which is not a standard radiographic projection in some examinations. Bone remodeling of the distal condyles on the palmar or plantar surface was the most common radiographic abnormality. Thirty-four percent (13/38) of horses with abnormal radiographs had incomplete fractures; 10 of the 13 incomplete fractures were short parasagittal fractures that were < 10 mm in length and seen only on the flexed 35° dorsodistal-palmaroproximal oblique. Three horses had semilunar articular fractures of the palmar or plantar aspect of the cortex consistent with gross findings of sclerosis and also had articular cartilage collapse similar to an induced model of overload arthrosis.³¹ Although there were few horses represented in this study with this particular fracture configuration, only 1 horse started a race after diagnosis, suggesting that semilunar articular fractures may be associated with a reduced prognosis for racing. Horses with more severe performance-limiting MCPJ/MTPJ abnormalities, such as marked degeneration on the palmar or plantar surface, and more significant parasagittal fractures were excluded by the stringent inclusion criteria because they would not have required scintigraphy or possibly diagnostic local anesthesia to diagnose.

The recommended treatment for horses in this study was free-choice exercise in a paddock to facilitate the bone modeling and remodeling process by increasing the hours per day that the horse was engaged in submaximal loading and unloading of the bone, compared with the confinement required of horses in training. Sixty days was chosen as the minimum time recommended because prior to the study, a 30-day period was attempted but did not consistently allow enough time for horses to become sound or achieve resolution of radiographic abnormalities. In the authors' opinion, the most important aspect of treatment is free-choice exercise in a paddock and 60 days is suggested because many horses will respond favorably to this period, although some need more time to heal and remodel. All of the horses in the study were reexamined by the authors or referring veterinarians and judged to be sound and radiographically normal prior to resuming training. The most objective and important measure of success is return to athletic performance documented by race records. The results of treatment surpass any previous reports in the number and quality of return to performance. This would indicate that free-choice exercise in a paddock is a successful method of restoration of athletic soundness. A period of at least 60 days, and perhaps as much as 120 days, is necessary to allow the necessary bone remodeling and healing.

The median number of days from diagnosis to the first race after injury was 194, which represents the time to soundness plus the time to retraining to fitness. This time could be substantially influenced by trainer or owner wishes, where the horses were stabled, and the racing calendar. This is the likely reason there was no significant effect of the time from diagnosis to first start on postinjury performance variables, as this period may overestimate the amount of time for resolution of lameness and account more for the time to reentry of the horse into the normal racing calendar.

Previous studies^9,13 report that 40% to 54% of horses with subchondral bone injury to the distal portion of MC3/MT3 are able to return to racing at equal or higher levels of performance. In this study, 95% (52/55) of horses with CSBI of MC3/MT3 were able to resume racing and 62% (28/45) of horses that raced before and after diagnosis equaled or improved their racing class. When comparing preinjury with postinjury performance variables, the earnings per start were significantly (P = 0.03) decreased, the number of starts was significantly (P = 0.02) higher, and there was no significant (P = 0.20) difference in total earnings, supporting our hypothesis. This increase in the number of starts is likely related to the frequent occurrence of this disease early in a horse's career, as 71% of horses were ≤ 3 years of age at the time of diagnosis, but it also indicates that once resolved, the CSBI did not appear to be a career-limiting factor. When examining horses with 4 starts before and after diagnosis, earnings per start decreased but not significantly (P = 0.06). Although mean earnings per start declined in both of these groups, this is documented to be a normal occurrence for Standardbred racehorses in the United States and Canada during their careers.^32,33 In the authors' experience, the same is true for Thoroughbreds because as they age, the number of entry-limited and purse-supplemented races available for each horse declines.

In evaluating each clinical variable recorded individually, there was no single factor that influenced postinjury performance and could aid clinicians in predicting response to treatment. This could be because of the relatively small sample size or related to our strict inclusion criteria. Joint effusion could indicate soft tissue injury, synovitis, or articular damage that could not be documented at the time of diagnosis. It is the authors' opinion that effusion likely indicates articular cartilage damage due to the progression of the disease to the articular surface. With an extremely small number of horses that had joint effusion, it is unknown if this finding would influence outcome and further investigation is needed. Horses that receive a diagnosis of CSBI at an older age may have more accumulated damage to the distal condyles of MC3/MT3, decreased athletic ability, or fewer years of active racing in their career.

Cumulative stress-induced bone injury of the distal portion of MC3/MT3 is an important disease of Thoroughbred racehorses that can be difficult to identify. Localization of the lameness via diagnostic local analgesia is the only definitive criterion. Lameness examination findings are beneficially supplemented with radiography, including the flexed 35° dorsodistal-palmaroproximal oblique projection and flexed 30° plantarodistal-dorsoproximal oblique projection and nuclear scintigraphy to assess the distal portion of MC3/MT3 for bone modeling and remodeling activity.

Free-choice exercise in a paddock appears to be a successful treatment strategy. Reexamination to ensure soundness is advisable, as some horses will take longer for abnormalities to resolve. Contrary to previous reports, CSBI of the distal portion of MC3/MT3 in our population treated with free-choice exercise carried a favorable prognosis. Ninety-five percent of the horses were able to race after diagnosis, had significantly increased numbers of starts, and had no decrease in total earnings, and 62% were able to maintain their racing class or improve.