This report was presented at the 2nd Congress of the European College of Equine Internal Medicine (ECEIM), February 2007, Naas, Ireland. The abstract was published in J Vet Intern Med 2007;21:882.
Corresponding author: L. Armengou, DVM, Dipl ECEIM, Servei de Medicina Interna Equina, Departament de Medicina i Cirurgia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona, 08193, Bellaterra, Barcelona, Spain; e-mail: firstname.lastname@example.org
A 16-year-old Westfalian gelding was presented to the Equine Teaching Hospital of Barcelona for a 6-week history of a sudden, progressive bilateral abnormal gait in the pelvic limbs. Previous treatment with phenylbutazone and paralumbar corticosteroid (triamcinolone) infiltration had not resulted in improvement. The horse had lived for 2 years in the same pasture with other unaffected horses. None of the other horses developed clinical signs consistent with stringhalt. No history of recent trauma was reported. The horse did not have any previous medical problems, apart from degenerative joint disease of the left radiocarpal and carpometacarpal joints, both metacarpophalangeal joints and both metatarsophalangeal joints that was diagnosed and treated 1 year before presentation.
On admission, the horse was excited and reluctant to move, and sedation with xylazine (0.4 mg/kg IV) was needed to move the horse out of the van. When eventually moved, the animal showed a bunny-hopping gait with severe and exaggerated flexion movements, and knuckled on both pelvic limbs (supporting information Video S1). Clinical signs were bilateral, but the right pelvic limb seemed more severely affected than the left. The horse kept the right pelvic limb hyperflexed while standing for several minutes until it relaxed. In addition, severe skin abrasions were found on the dorsal aspect of both carpal areas, and on the cannon bone and fetlock areas of both pelvic limbs caused by the abnormal gait. Complete physical and neurological examinations did not identify any other clinically relevant findings. The clinical diagnosis was bilateral stringhalt grade V/V, according to the gradation scale of Huntington et al.1
Examination of the horse's pasture revealed very few dry plants, and no evidence of any toxic plant. The horse was fed festuca grass hay similar to other horses in the stable, and the food was in good condition. Results of CBC and serum biochemistry did not disclose any abnormalities. Lumbosacral cerebrospinal fluid collection was attempted, but a sample could not be obtained. Electromyographic (EMG) examination of the pelvic limb muscles including the long digital extensor was attempted with the horse awake to eliminate a peripheral neuropathy or primary myopathy. Unfortunately, the horse was extremely sensitive to any manipulation of the pelvic limbs and conclusive results could not be obtained.
To reach a definitive diagnosis, biopsies from the long digital extensor muscle and the superficial peroneal nerve were obtained 4 days after admission. The horse was premedicated with romifidine (0.03 mg/kg IV) and butorphanol (0.03 mg/kg IV). General anesthesia was induced with diazepam (0.05 mg/kg IV) and ketamine (2.2 mg/kg IV) and maintained with isofluorane in 100% oxygen with intermittent positive pressure ventilation. The horse was positioned in left lateral recumbency to obtain biopsies from the right pelvic limb. The lateral surface proximal to the tarsus was clipped and surgically prepared. An 8-cm longitudinal incision was made at the level of the lateral and long digital extensor muscle bellies. Superficial tissues were bluntly dissected until the superficial peroneal nerve was identified. The nerve was carefully separated from surrounding connective tissue using sharp dissection and avoiding the use of gauze or blunt dissection to prevent artifact. A 2.5-cm-long fascicular nerve biopsy then was obtained. The sampled nerve then was divided longitudinally into 2 pieces. Dissection was continued cranially until the long digital extensor muscle could be visualized. Two muscle samples were obtained, 1 of standard dimensions (0.5 × 0.5 × 1.0 cm) and a smaller sample. The skin was sutured in a discontinuous single pattern using 0 USP polypropylene. At the same time, a biopsy from the right middle gluteal muscle was obtained by a percutaneous punch biopsy procedure.2
One muscle and 1 nerve specimen were placed on a tongue depressor and immersed in 10% formalin. The 2nd specimens were wrapped in saline-moistened gauze sponges and placed into a dry, water-tight container. All samples were shipped overnight under refrigeration to the Comparative Neuromuscular Laboratory, University of California San Diego, La Jolla, California. Immediately upon receipt, unfixed, and chilled muscle and nerve specimens were flash frozen in isopentane precooled in liquid nitrogen, then stored at −80°C until further processing. Fixed tissues were either paraffin embedded (muscle) or embedded into resin blocks (nerve).
Cryostat sections (8 μm) of the long digital extensor muscle were evaluated by a standard panel of histochemical stains and reactions according to standard protocols.3 Histochemical stains and reactions included hematoxylin & eosin, modified Gomori trichrome, periodic acid-Schiff (PAS), myofibrillar ATPases at pH 9.8 and 4.3, esterase, nicotinamide adenine dinucleotide-tetrazolium reductase, acid phosphatase, alkaline phosphatase, and oil red O. Moderate variability in myofiber size was present with large groups of atrophic fibers having an anguloid shape and of both fiber types (not shown). Fiber type grouping was not observed. A few necrotic fibers and rare clusters of mononuclear cells were found. Intramuscular nerve branches were depleted of myelinated fibers. Myofiber size was appropriate within paraffin sections of the gluteus muscle with only rare foci of myonecrosis observed.
Semithin sections (1 μm) of the superficial peroneal nerve from the patient (Fig 1A,C) and a control horse (Fig 1B,D) were evaluated after toluidine blue staining. Although difficult to appreciate at low power (Fig 1A,B), large myelinated fiber loss, axonal degeneration (Fig 1C, white arrow) and myelin splitting, and degeneration (Fig 1C, dark arrows) were present in the nerve from the affected horse, but not in that from the control horse (Fig 1D). Regenerative changes were not found. The final diagnosis based on histopathologic study was mixed polyneuropathy with elements of both axonal degeneration and demyelination.
Treatment and Outcome
The horse remained hospitalized until the biopsy results were available. Initial treatment consisted of an anti-inflammatory drug (suxibuzone, 3.3 mg/kg PO q12h for 7 days), a free radical scavenger (dimethyl sulfoxide, 200 mg/kg transrectally q12h for 4 days), an antioxidant (vitamin E, 6,000 IU PO q24h for 7 days), and an antimicrobial (trimethoprim-sulfadiazine, 30 mg/kg PO q12h for 7 days); bandaging of the biopsy incision and wound care were added after biopsy procedure. When a final diagnosis was reached, the horse was discharged on vitamin E supplementation, moved to a different facility and kept in a stall during the 1st days for wound care. The horse improved rapidly from the 3rd day of hospitalization (grade III/V) and continued to improve after discharge. Three months later, the gait was considered normal and it has remained normal for 30 months after presentation.
Stringhalt, recently renamed as equine reflex hypertonia,4 is a syndrome characterized by sudden and involuntary hyperflexion of 1 or both pelvic limbs during attempted movement.5 Stringhalt is a clinical sign rather than a disease.6 Two clinical types or presentations have been reported. Idiopathic (classic or spontaneous) stringhalt is reported worldwide. This form is usually unilateral and clinical signs progress over years. Descriptions of specific nerve or muscle lesions related to this gait have not been reported. The 2nd form, Australian or acquired bilateral stringhalt, is a manifestation of distal axonopathy.7,8 Large myelinated fibers are selectively affected with a decrease in severity distal to proximal.
The Australian form of stringhalt usually appears in outbreaks during the late summer, has a sudden onset, usually affects both pelvic limbs, and is frequently related to the presence of one of the following plants in the pasture: Hypochaeris radicata (Australian dandelion), Taraxacum officinale (European dandelion), or Malva parviflora (mallow). Outbreaks related to the presence of toxic plants have been reported in Australia, New Zealand,1 North America,9 and Brazil.10 Cases clinically similar to those in Australia also have been reported in Italy, Chile, and Japan.7,11,12 However, as in this case, clinical signs can occur paradoxically in a sporadic or individual form with no exposure to toxic plants.
Diagnostic tools for stringhalt are limited. Horses suffering from stringhalt have abnormal EMG recordings from the long digital extensor muscle, which can be useful for diagnosis as well as for follow-up and prognosis.1,12–14 However, the technique is not available in most clinical facilities and needs to be performed and interpreted by an experienced neurologist. In our case, no data could be obtained from the standing EMG study because of intolerance of the horse to EMG needle placement. EMG could have been performed with the horse under general anesthesia, but neurology personnel were not available at that time.
A definitive diagnosis of distal axonopathy has been obtained in several cases of Australian stringhalt after postmortem examination of selected muscles and nerves.8,15 The most severely affected nerves included the recurrent laryngeal and superficial and deep peroneal nerves, which showed mixed axonal degeneration and demyelination with concurrent atrophy of the innervated muscles. No alterations in the central nervous system have been found.15 In this case report, muscle and nerve biopsies were performed in an attempt to establish the diagnosis antemortem. Pathologic changes similar to those reported from postmortem examinations were present in the muscle and nerve biopsy specimens from the present case. Another diagnostic test frequently performed is laryngeal endoscopy to confirm the presence of recurrent laryngeal neuropathy, because some horses with bilateral stringhalt also have dysfunction of the recurrent laryngeal nerve.1,8,12,15 A distal axonopathy has been described in Leonberger dogs, in which a high-stepping pelvic limb gait with atrophy of distal limb muscles and laryngeal weakness are found.16 The horse of this case report showed no obvious clinical signs of laryngeal neuropathy, but endoscopy to eliminate this possibility was not performed.
There is no consensus regarding treatment for stringhalt. Australian cases are managed by limiting access to toxic plants and administering phenytoin.13 Limited joint movement caused by the disease may result in stress to affected horses and exacerbate signs of stiffness. Consequently, administration of central nervous system-acting myorelaxants and tranquilizers (eg, phenytoin, acepromazine) can be beneficial.12,13,17 Recovery of affected horses seems to depend on the degree of exposure to toxic substances and severity of axonal lesions. Palliative treatment consisting of myotenectomy of the lateral digital extensor tendon in the affected limb has been reported to be effective in 85% of cases.7 This treatment also has been performed in cases of spontaneous stringhalt with variable results.18
Several unknown etiologic factors probably are involved in stringhalt. Autralian stringhalt has been demonstrated to be the clinical manifestation of a distal axonopathy affecting the longest myelinated nerves. The case reported here also showed clinical manifestations of a distal axonopathy. However, it is still unknown whether all cases of stringhalt are because of neuropathy or not. Although different diagnostic tests have been proposed and used to eliminate or confirm the presence of axonopathy, biopsy of both the superficial peroneal nerve and the long digital extensor muscle is useful to establish a definitive diagnosis antemortem. Moreover, the same surgical approach might be used for the myotenectomy and for biopsy sampling (when myotenectomy is indicated, such as in chronic cases).
In the case reported here, an etiologic agent could not be found. A decision to perform medical treatment was made because of the clinical improvement observed after the 3rd day of hospitalization, and later because of the microscopic findings in the nerve and muscle biopsies.
The rapid clinical improvement seen in this case is unusual. The initial improvement could have been related to the administration of anti-inflammatory and antioxidant drugs, whereas the progressive recovery to a completely normal gait took longer (3 months) and could be attributed to axonal regeneration. Another possible explanation would be the effect of changing environment. Although a toxic agent was not found on the pasture, toxic agent withdrawal also could partly explain the rapid initial improvement after moving the horse to the hospital. A 3rd possible explanation would be the effect of rest during hospitalization. Finally, spontaneous recovery could have occurred without treatment because of time and environmental changes, thus the effectiveness of treatment might be questionable.
To our knowledge, this is the 1st case of Australian stringhalt reported in Europe with a confirmed diagnosis of distal axonopathy. Similar cases reported worldwide also could be manifestations of distal axonopathies, although no other cases have been previously confirmed as distal axonopathy outside of Australia or New Zealand. This case is clinically relevant because it demonstrates that diagnosis of stringhalt can be made antemortem rather than postmortem by histological examination of nerve and muscle biopsy specimens, thus avoiding euthanasia. In previous studies, diagnosis was only achieved after euthanasia.8,15 This case further demonstrates that horses can recover completely from this neuropathy.