Short- and long-term results following standing fracture repair in 34 horses

Authors


email: polly.compston@rossdales.com

Summary

Reasons for performing study: Standing fracture repair in the horse is a recently described surgical procedure and currently there are few follow-up data. This case series contains 2 novel aspects in the standing horse: repair of incomplete sagittal fractures of the proximal phalanx and medial condylar repair from a lateral aspect.

Objectives: To describe outcome in a case series of horses that had lower limb fractures repaired under standing sedation at Rossdales Equine Hospital.

Method: Case records for all horses that had a fracture surgically repaired, by one surgeon at Rossdales Equine Hospital, under standing sedation and local anaesthesia up until June 2011, were retrieved. Hospital records, owner/trainer telephone questionnaire and the Racing Post website were used to evaluate follow-up.

Results: Thirty-four horses satisfied the inclusion criteria. Fracture sites included the proximal phalanx (incomplete sagittal fracture, n = 14); the third metacarpal bone (lateral condyle, n = 12, and medial condyle, n = 7); and the third metatarsal bone (lateral condyle, n = 1). One horse required euthanasia due to caecal rupture 10 days post operatively. Twenty horses (66.7% of those with available follow-up) have returned to racing. Where available, mean time from operation to return to racing was 226 days (range 143–433 days).

Conclusions: Standing fracture repair produced similar results to fracture repair under general anaesthesia in terms of both the number of horses that returned to racing and the time between surgery and race.

Potential relevance: Repair of lower limb fracture in the horse under standing sedation is a procedure that has the potential for tangible benefits, including avoidance of the inherent risks of general anaesthesia. The preliminary findings in this series of horses are encouraging and informative when discussing options available prior to fracture repair.

Introduction

The Thoroughbred racehorse population in the UK has an overall estimated fracture incidence of 1.15 fractures per 100 horse months [1]. Two common fracture sites in this racehorse population are third metacarpal/tarsal condylar fractures and sagittal fractures of the proximal phalanx, which constituted 14.5% and 10.4% of all racehorse injuries, respectively, in one cohort study [2]. Although these fractures are common and amenable to surgical fixation, fracture repair under general anaesthesia carries an increased risk of anaesthetic-related death compared with other elective surgical procedures [3]. Additionally, catastrophic fracture following fixation of medial metatarsal condylar fractures has been reported to be as high as 42% [4] due to proximal propagation of the fracture line. Medial condylar fractures in both fore- and hindlimbs often spiral proximally leading to a more unstable configuration.

Repair of condylar fracture under general anaesthesia has a good prognosis, with 56–83.5% of horses returning to race post operatively [5–9]. Lag screw fixation from the lateral side has been described for the repair of medial condylar fractures, with return to racing in 33–56% of horses [10,11], and a mean time from fracture to first race of 14.5 months in one study [10]. Return to racing following an incomplete sagittal fracture of the proximal phalanx has been reported at 61–88% [12–15].

Over the last decade there has been a substantial increase in the number and type of surgical procedures performed under standing sedation in the horse that would previously have been performed under general anaesthesia. Ocular enucleation [16], dorsal spinous process resection [17] and distal limb fracture repair [18,19] are all procedures recently described in the standing patient. There have been 2 previously published case series of horses that have had fractures repaired standing. Perez-Olmos et al. [18] reported results in 4 racehorses with spiral longitudinal metacarpal and metatarsal condylar fractures that were repaired under standing sedation and local anaesthesia. All 4 horses returned to racing, although their performance and the time taken to return to racing were not discussed. Russel and Maclean [19] described condylar fracture repair in 13 racehorses with propagating metacarpal and metatarsal condylar fractures using lag screw fixation through stab incisions. Short-term complications seen in that study were catastrophic fracture (one horse), and in one patient the drill bit snapped in the medullary cavity during surgery. Eight horses (61.5%) returned to racing, with average earnings/start increasing in 43%.

The current study is the first to report results following fixation of incomplete sagittal fractures of the proximal phalanx under standing sedation and local anaesthesia. It also reports repair of a medial condylar fracture from the lateral side in the standing patient. We report favourable short- and long-term follow-up results from a cohort of horses at Rossdales Equine Hospital that have had lower limb fractures repaired under standing sedation and local anaesthesia.

Methods

Inclusion criteria

All horses that had a lower limb fracture repaired, by R. J. P., under standing sedation and local anaesthesia at Rossdales Equine Hospital before 1 June 2011 were included in this study.

Surgical technique

Cases were groomed prior to surgery to decrease the risk of loose hair or dirt falling into the surgical field. Premedication (acepromazine) was given 30 min prior to surgery and a jugular catheter placed on the same side as the affected limb. Preoperative antimicrobials and analgesics were administered. Support bandaging and splinting were removed once the patient was in the anaesthetic induction room, where the surgery was performed. Continuous sedation was achieved with i.v. boluses of detomidine and butorphanol.

Positioning of the horse so that the foot of the fractured limb is perpendicular to the floor is considered to be helpful. It is important that the floor is nonslip. Any movement once the support bandage has been removed may cause destabilisation of a nondisplaced fracture, and therefore efficient surgical technique and a fully competent surgical team is required to minimise surgical time and therefore the period during which the horse has to remain still.

The surgeon was positioned on the same side of the horse as the affected limb. The distal limb was clipped from the level of the coronary band to the carpus and a primary scrub performed. Either a high 4 (front leg: medial and lateral branches of both palmar and palmar metacarpal nerves) or 6 point (hind leg: medial and lateral branches of plantar, plantar metatarsal and dorsal metatarsal nerves) nerve block was performed. Additional local anaesthetic (mepivacaine) was injected to make the nerve block into a circumferential ring block at the desired level (typically at mid-cannon). The distal limb was then prepared for sterile surgery. Marker staples were placed to indicate the level of the joint on the medial and lateral aspects of the fetlock. Further staples were placed on the lateral surface of the limb at the estimated point of insertion of the lag screws, using routine palpable landmarks. The position of these staples was confirmed with dorsopalmar/plantar and lateromedial radiographs. A sterile cohesive bandage was placed from the top of the surgical field, over the carpus and up to the level of the proximal radius. Two impervious drapes were placed at the level of the second phalanx (Fig 1), creating a sterile field for the operation.

Figure 1.

Impervious drapes are placed around the horse's pastern to achieve a sterile field distally. Sterile cohesive bandaging is used to ‘drape-out’ the proximal part of the limb.

The primary surgeon had 2 assistants; one to pass instruments and one to hold the drill hose. Lag screw fixation using 4.5 mm cortical lag screws, through stab incisions, was performed as has been previously described under general anaesthesia [20]. Medial condylar fractures were repaired using a lateral approach. Intraoperative radiographs were taken to confirm screw positioning. Stab incisions were closed with skin staples, and an appropriate dressing/supportive bandage applied. For the majority of horses this consisted of a modified Robert-Jones bandage. Casting material was incorporated into the bandage when extended fracture lines were present.

This technique has been modified over the course of this surgeon's experience. Initially, a tourniquet was used proximally to control bleeding, as is routine practice in the authors' hospital. However, in the standing horse this was found to cause discomfort, resulting in movement. Following cessation of routine tourniquet use, horses tolerated the procedure much better, and minimal haemorrhage was seen.

Follow-up

Hospital records were examined and the following data recorded: horse breed, sex and age; date of surgery; site of fracture; affected limb; number of lag screws used; in-hospital perioperative complications; and relevant short- to long-term history.

The following parameters were recorded for Thoroughbred horses from race records on the Racing Post website (racingpost.com): if the horse had raced post operatively; and the time interval between surgery and first race. A telephone questionnaire was conducted to owners or trainers as relevant for all horses, and the following information was obtained: any post operative problems that were not attended by a vet; if the horse had become sound post operatively; if the horse had returned to training; for Arabs, or Thoroughbreds that were racing outside Racing Post records, if they had raced and if so had been placed or won post operatively; and the ultimate fate of each horse.

Records from Rossdales Equine Hospital were examined for horses with similar fracture configurations that had been operated on under general anaesthesia by the same surgeon (R. J. P.) between 1 January 2006 and 31 December 2010. Post operative medical and racing performance data were collated from hospital records and racingpost.com as described earlier. Statistical comparisons were performed in R [21].

Results

Thirty-four horses fitted the inclusion criteria. All were racehorses: the majority were racing Thoroughbreds (30/34, 88.2%) and the remainder racing Arabs (4/34, 11.8%). Preliminary results from some Thoroughbreds discussed here have been previously reported [22,23]. Incomplete sagittal fractures of the proximal phalanx were most common (14/34, 41.2%) followed by lateral condylar fractures of the third metacarpus (12/34, 35.3%), medial condylar fractures of the third metacarpus (7/34, 20.6%) and one lateral condylar fracture of the third metatarsus (2.9%) (Table 1). The majority of fractures (14, 41.2%) were repaired with one lag screw (Table 2). A summary of case background and follow-up for all cases is presented in Supplementary Table S1.

Table 1. Distribution of affected limbs in horses undergoing fracture repair under standing sedation and local anaesthesia
Limb affectedFracture typeTotal
Incomplete sagittal fracture of the proximal phalanxFracture of third metacarpal/tarsal bone
Lateral condyleMedial condyle
Left fore78520
Right fore74213
Right hind0101
Total 14 13 7 34
Table 2. Total number of screws used for each type of fracture repair
Fracture typeNumber of screws
1234
Incomplete sagittal fracture of the proximal phalanx7430
Fracture of third metacarpal/tarsal bone – lateral condylar5530
Fracture of third metacarpal bone – medial condylar2032
Total 14 9 9 2

Short-term follow-up

Thirty horses had no perioperative problems. One horse (Case 5) had a caecal impaction that was refractory to medical therapy and ruptured 10 days post operatively, necessitating euthanasia. This case was unusual compared with others in the series because it had an extensive work-up prior to surgery, with both scintigraphic and magnetic resonance imaging performed, and required prolonged periods of sedation as well as hospitalisation (a total of 7 days). The average (median) total hospitalisation period was 3 days. One horse (Case 6) had a colon impaction and one horse (Case 33) developed post operative colitis associated with Clostridium difficile infection; both recovered uneventfully with medical management. One horse (Case 34) developed mild bandage-associated skin sores that recovered uneventfully with medical wound care.

Mid-term follow-up

Thirty-three horses survived to discharge. Three horses (Cases 32, 33 and 34) had their operation after 1 January 2011; subsequent performance data have not been collected for them because at time of writing they are still in rehabilitation. Of the remaining 30 horses, all became sound at trot and 28 returned to training. Three horses (Cases 10, 15 and 20) had unresolved lameness of the fractured limb once training commenced and were subsequently retired. Four horses were retired subsequently to starting training due to lack of ability (Cases 6, 12, 21 and 22). One horse (Case 9) returned to training but was subsequently lost to follow-up.

Long-term follow-up

Cases 9, 32, 33 and 34 have been excluded from the following analysis of racing performance, due to loss of, or insufficient, follow-up. Twenty horses (66.7%, 95% confidence interval [CI] 49.8–83.5%) returned to racing (Fig 2): 10 horses with a proximal phalangeal fracture (71.4% of horses with this fracture configuration); 8 with a lateral condylar fracture (72.7% of horses with this fracture configuration); and 2 with a medial condylar fracture (40% of horses with this fracture configuration) (Table 3). Data on date of first race were available for 15 of these 20 horses; they returned to racing a mean 226 days (7.4 months, range 143–433 days; 95% CI 177–275 days) after fracture repair (Table 3). Of the 20 horses known to return to racing, post operative race performance was known for all (Fig 3), with 12 (60%) subsequently being placed or winning in at least one race. Owner/trainer satisfaction was assessed in the follow-up questionnaire with all responders agreeing that they would have other fractures in their horses repaired in this way.

Figure 2.

Number of horses that raced following fracture repair under standing sedation, or general anaesthesia, at Rossdales Equine Hospital (by R. J. P.), compared with previously published studies. ‘Standing’ and ‘general anaesthesia’ refer to procedures performed by the same surgeon in this study.

Table 3. Fracture type and post operative athletic performance
Method of repairFracture siteNumber of horses not includedHorses that did not raceHorses that did raceTotalMean time between fracture repair and first race (months)
  • *

    Case 33 was not included in this analysis as it was still recovering (surgical procedure <6 months prior to study date). Case 9 was not included as it was lost to follow-up.

  • #

    Cases 32 and 34 were not included in this analysis as they were still recovering (surgical procedure <6 months prior to study date).

Standing sedation and local anaesthesiaProximal phalanx (incomplete sagittal)0410149.1
Third metacarpal/tarsal (lateral condyle)2*38136.2
Third metacarpal (medial condyle)2#3275.2
General anaesthesiaProximal phalanx (incomplete sagittal)0811199.9
Third metacarpal/tarsal (lateral condyle)01061610.3
Third metacarpal/tarsal (medial condyle)022410.4
Figure 3.

Time between fracture repair and first race for horses operated on under standing sedation , or general anaesthesia, at Rossdales Equine Hospital (by R. J. P.), compared with previously published studies. ‘Standing’ and ‘general anaesthesia’ refer to procedures performed by the same surgeon in this study. Error bars represent standard deviation.

A total of 39 fractures with these 3 configurations were repaired by R.P. at Rossdales Equine Hospital under general anaesthesia between 1 January 2006 and 31 December 2010 (Table 3). Nineteen (48.7%, 95% CI 33.0–64.4%) returned to racing following surgery, a mean 308 days post operatively (10.1 months, range 135–574; 95% CI 256–359). The horses in this cohort returned to racing an average of 81 days earlier than horses that had a fracture repaired under general anaesthesia (P = 0.027).

Discussion

The results of this case series are comparable with previous studies reporting fracture repair for these configurations under general anaesthesia. Avoidance of general anaesthesia is an important incentive for the development of novel techniques in equine surgery that can be performed under standing sedation. To the authors' knowledge this is the first published report of repair of incomplete sagittal fractures of the proximal phalanx under standing sedation and local anaesthesia. In both previous reports of condylar repair in the standing horse [18,19] the authors reported the repair of medial condylar fractures from the medial side of the affected limb, with the surgeon positioned on the opposite side of the horse from the leg being operated on. Under general anaesthesia it is routine to approach repair of medial condylar fractures from the lateral side [10,11,20] and we report this technique in the standing horse.

Lag screw fixation through stab incisions is identical in all of the fracture configurations described. Fractures that are amenable to standing repair are nondisplaced and in a location with minimal soft tissue coverage [19]. As previously discussed, familiarity with these surgical techniques in the anaesthetised horse is important prior to performing this technique in the standing animal, as is a competent surgical team [24]. The anaesthetic induction room was used for these operations. This has been proposed to allow drill vibration dispersion [18]; it had the added advantages of being a quiet room, which was close to theatre so that, if needed, induction of general anaesthesia and subsequent surgery could have been performed.

Three out of 4 horses (Cases 5, 6 and 33) with post operative complications had gastrointestinal signs. Gastrointestinal disease has been recognised as a complication following general anaesthesia in horses undergoing elective surgery [25,26]. This is especially relevant for a racehorse in training that is subject to large management changes following injury. Extended stabling and recent travel have been identified as risk factors for colonic obstruction and distension [27], and in the general horse population changes in management and routine are known to affect the incidence of colic signs [28]. Additionally, opioid and alpha-2 adrenergic drugs used to induce sedation are known to affect gastrointestinal motility [29,30]. Hospitalisation has also been reported to increase the frequency of C. difficile-associated diarrhoea in horses [31]. Problems associated with bandaging were observed in one horse (Case 34) but these resolved without any long-term problems. Fracture line propagation with subsequent re-fracture and breaking of the drill bit were both complications reported by Russel and Maclean [19], but neither occurred in this series.

Doubts have been raised as to whether appropriate sterility for orthopaedic surgery can be achieved in the standing horse [24]. However, there have been no reports of post operative infection in any other study describing standing fracture repair [18,19], nor in reports of arthroscopic removal of chip fractures from the metacarpophalangeal joint in 95 standing horses [32]. No reports of post operative infection were reported in this case series, despite using less draping than in previous reports. Concerns have also been raised over staff safety [24]; again no incidents have been reported in the course of this study, although the authors would like to stress that a competent and experienced hospital team is necessary to ensure that health and safety concerns do not become an issue. Nerve blocks were always performed by the operating surgeon as confidence in method is essential.

Horses returned to racing at a mean of 7.43 months following surgery. For all fracture configurations in this study, horses returned to racing within a similar time-frame following standing surgery as was recorded following repair under general anaesthesia at the authors' hospital and in previous studies [6–8,10,14,15,19]. The proportion of horses that returned to racing following surgery is also similar for horses in this study that had fractures repaired standing and under general anaesthesia, and those in other studies [6–12,14,15,19]. There appears to be no negative effect in terms of racing prognosis for standing fracture repair at the authors' hospital, as demonstrated by the overlapping confidence intervals. However, it is not possible to draw direct comparisons between this cohort of horses that have had fractures repaired under standing sedation and those that have fractures repaired under general anaesthesia at Rossdales Equine Hospital, or those previously reported in the literature. The main reason for this is that fractures that are amenable to repair in the standing horse are minimally or nondisplaced, and therefore have a different time-course of healing to more displaced fractures that necessitate repair under general anaesthesia. This is likely to account for the shorter recovery times seen in this study. Additionally, standing fracture repair has developed from a salvage procedure into the surgery of choice for this surgeon, resulting in a nonuniform cohort of horses in terms of racing ability.

Standing fracture repair has tangible advantages: the inherent risks of general anaesthesia can be avoided; it reduces the complexity and time requirements of the procedure; and this in turn reduces the cost compared with fracture repair under general anaesthesia. However, although no direct disadvantages were found with this procedure in the current study, it must be recognised that there are limitations to the technique. Some surgeons may struggle with the positioning and a successful procedure is dependent on having an amenable patient. The horse's temperament is particularly important as it must be completely stationary during surgery. There may be increased risk of injury to personnel and patient as well as damage to surgical equipment, despite this not being seen in this surgeon's experience. Some surgeons advocate routine endoscopy of the metacarpo/tarsophalangeal during lag screw fixation so that cartilage apposition can be visualised. This is not routine at Rossdales Equine Hospital for nondisplaced fracture configurations either standing or under general anaesthesia unless specifically indicated.

This study shows no disadvantages to repairing these nondisplaced fractures of the lower limb under standing sedation and local anaesthesia compared with under general anaesthesia. Facilities and experience are important when considering this technique.

Authors' declaration of interests

No conflicts of interest have been declared.

Source of funding

None.

Acknowledgements

Polly Compston is supported by the Margaret Giffen Trust. Thanks are extended to all staff at Rossdales Equine Hospital and to the owners and breeders involved in this study.

Ancillary