Semitendinosus myopathy and treatment with adipose-derived stem cells in working German shepherd police dogs

Abstract

Semitendinosus myopathy has been treated with numerous surgical and non-surgical therapies resulting in recurrence of lameness within 2 to 9 months. Eleven cases of semitendinosus myopathy diagnosed in 8 working police dogs that were treated with adipose-derived mesenchymal stem cells were retrospectively evaluated. At short-term follow-up < 6 mo, ultrasound and gait evaluations revealed a mean reduction in the overall intramuscular lesion size of 54.82% (SD +/− 18.02; range: 30.5% to 82.7%) and reduction in the Visual Assessment Score (VAS) of 1 to 3 points. At long-term follow-up > 1 y, in 8 cases the dogs had a normal gait and in 3 cases the dogs had an improved gait compared with initial examination, and all 8 dogs returned to active police work. Fisher’s exact test resulted in P = 0.000008 when comparing published historical reports and these 11 cases for resolution of lameness and return to active duty.

Introduction

Semitendinosus myopathy is an uncommon disease in companion animals, documented in the German shepherd, Belgian shepherd, Doberman pinscher, St. Bernard, and old English sheepdog breeds (1–4). The exact etiology of semi-tendinosus myopathy is unknown. A higher incidence occurs in working dogs which exert an excessive amount of acute force on the hamstring muscle group, such as dogs used for protection work, racing, or elite agility competition. This injury can be career ending for working dogs due to the secondary fibrosis and muscle contracture that occur following this type of injury (1–6).

The semitendinosus muscle “originates from the ischiatic tuberosity, passes deep to the gracilis muscle, and inserts on the medial aspect of the tibia distal to the insertions of the semimembranosus and gracilis muscles. This muscle functions to extend the hip, stifle, and tarsus and flex the stifle when in a non-weight-bearing position” (2). After this muscle is traumatized, lameness ensues that can be acute in onset, or progress over a period of weeks to months before reaching a plateau (2,5). The exact cause of the injury is not always identifiable. However, the secondary fibrosis that ensues results in a characteristic gait abnormality, in which “all affected dogs have a shortened stride with a rapid, elastic medial rotation of the paw, external rotation of the hock, and internal rotation of the stifle during the swing phase of the stride” (4). The fibrosis is generally palpable as a firm, taught band that extends from the ischiatic origin of the semitendinosus muscle to its insertion along the medial aspect of the tibia (1–5).

Historically, semitendinosus myopathy and secondary fibrosis have been treated by various methods including rest, antioxidants, non-steroidal anti-inflammatory drugs, steroids, physical therapy and a variety of surgical procedures (myotenectomy, myectomy). These therapies were unsuccessful at preventing the formation of fibrous tissue, and all affected dogs were unable to return to their full activity (1–4). Based on a published report of 2 cases (7), we hypothesized that by treating semitendinosus muscle tears with adipose-derived mesenchymal stem cells (MSC), the degree of scar tissue formation within the muscle would be reduced, and the affected dogs would be able to return to normal activity.

Materials and methods

Cases of pelvic limb lameness that were presented between July 2008 and May 2012 to the Veterinary Orthopedic Surgery Service in Fremont, California, USA were retrospectively evaluated. A total of 11 cases diagnosed with semitendinosus myopathy in 9 dogs were available for evaluation. Of the 9 dogs, 1 had concurrent bilateral semitendinosus muscle tears at the time of diagnosis, and another dog had semitendinosus muscle tears in each pelvic limb, approximately 10 mo apart. A single case of semitendinosus myopathy was missing the ultrasonographic report in the medical record, and was excluded from ultrasonographic analysis. Gait evaluation was reported in the medical record for this case, and these findings were included in the overall analysis.

Within approximately 1 wk of initial examination, all dogs underwent an initial ultrasound examination of the affected limb by the same board-certified ultrasonographer, David Dettweiler DVM, DACVR. Any lesions identified were measured in length and/or width and given a visual description in regard to disruption of fiber pattern and evidence of scar tissue formation (Table 1). Dogs were also graded on a Visual Assessment Score (VAS) (1 — normal, 2 — mildly affected, 3 — moderately affected, 4 — severely affected, and 5 — extremely affected) by a single examiner, S. Gary Brown DVM, DACVS (Table 2).

Table 1.

Ultrasonographic measurements^a of intramuscular tears at diagnosis and short-term follow-up

Patient	Leg affected	Initial muscle tear size (mm)	Initial scar tissue	Muscle tear size (mm) < 6 months PO	Scar tissue < 6 months PO	% Reduction in muscle tear size
Cris	Left	30.8 × 8.1	None	15 × 11	None	33.8
Hunther	Left	41 × 18	Unknown	30 × 7	18 × 3	71.5
Jagob	Right	9 wide	Unknown	5.4 wide	Unknown	40.0
Jagob	Left	7 wide	Unknown	4.1 wide	Unknown	41.4
Brixb	Left	40 × 10	10 × 2	15 long	None	62.5
Brix	Right	24 × 6	Chronic	20 × 5	None	30.5
Sandrob	Left	2 wide	None	1 wide	None	50.0
Marco	Right	40 × 5	None	36 × 2.5	Unknown	55.0
Tom	Right	100 × 12	None	80 × 2.6	None	82.7
Arnyb	Left	52 long	Unknown	10 long	Unknown	80.8

^aA single case was excluded from this analysis due to missing ultrasonographic information from the medical record.

^bThe semitendinosus muscle tears were only measured in width ultrasonographically for these cases; the percent reduction was calculated by dividing the follow-up width measurement by the initial width measurement.

PO — post-implantation of MSC.

Table 2.

Visual Assessment Scores^a at presentation, short-term, and long-term follow-up

Patient	Leg affected	VAS at presentation	VAS < 6 months PO	VAS > 1 year PO
Cris	Left	3	2	1
Hunther	Left	4	2	1
Jago	Right	4	2	3
Jago	Left	1	1	1
Brix	Left	2	2	2
Brix	Right	4	1	2
Sandro	Left	3	2	Lost to follow-up
Marco	Right	2	1	1
Tom	Right	2	1	1
Arny	Left	3	2	1
Jim	Right	2	1	1

^aScore definitions: 1 — normal; 2 — mildly affected; 3 — moderately affected; 4 — severely affected; 5 — extremely affected;

PO — post-implantation of MSC.

Within 1 wk of the initial ultrasound, adipose tissue was harvested for treatment with MSC. Dogs were premedicated with a combination of an opioid [Buprenorphine; Par Pharmaceuticals, Spring Valley, New Jersey, USA, 0.01 to 0.02 mg/kg body weight (BW), IM; Butorphanol (Merck, Spring Valley), 0.08 mg/kg BW, IM; morphine (West-Ward, Eatontown, New Jersey, USA), 0.4 to 0.5 mg/kg BW, SC]; benzodiazepine (Diazepam; Hospira, Lake Forrest, Illinois, USA), 0.3 to 0.5 mg/kg BW, IV; and phenothiazine (Acepromazine; Henry Schein, Dublin, Ohio, USA), 0.02 to 0.06 mg/kg BW, IM. The dogs were then induced with intravenous Propofol (Zoetis, Kalamazoo, Michigan, USA), 0.3 to 2.6 mg/kg BW to effect, and maintained on Isoflurane gas anesthesia. Each dog received IV Lactated Ringers Solution (10 mL/kg BW per hour), and was monitored for non-invasive blood pressure (NIBP), oxygen saturation (SPO₂), end tidal cardon dioxide (EtCO₂), temperature and electrocardiogram (ECG) during the anesthetic procedure.

Approximately 230 g of falciform fat were aseptically obtained through a ventral midline laparotomy and placed in sterile transport centrifugation vials with 5 mL of a sterile buffered saline solution (8). The ventral midline incision was closed in a routine 3-layer fashion with absorbable suture (2-0 or 0 monofilament PDS; Ethicon, Cincinnati, Ohio, USA). The adipose tissue was shipped overnight to the Vet Stem processing facility, and returned within 48 h from the collection date for implantation (8).

At the time of implantation, dogs were sedated with either an opioid alone (Butorphanol 0.1 to 0.2 mg/kg BW, IM) or a combination of an opioid (Butorphanol 0.08 mg/kg BW, IM), a phenothiazine derivative (Acepromazine 0.04 to 0.18 mg/kg BW, IM) and a benzodiazepine (Diazepam 0.3 to 0.4 mg/kg BW, IV). In 4 of the 11 cases the dogs were induced with Propofol (0.3 to 2 mg/kg BW, IV), maintained on Isoflurane gas anesthesia and given IV Lactated Ringers Solution (10 mL/kg BW per hour) during the procedure. The lesions within the semitendinosus muscles were identified, and 1 to 3 aliquots (0.6 mL pre-filled syringes) of adipose-derived MSC were injected intralesionally under ultrasound guidance. A second injection (5 to 6 mL pre-filled syringe) was administered intravenously over 2 min through a previously placed IV catheter. The dogs were discharged to their handlers the day of implantation with instructions for rest and no additional Schutzhund training for the following 12 wk.

Each case had short-term follow-up (< 6 mo) and long-term follow-up (> 1 y) for gait evaluation by the examiner who conducted the initial examination (Table 2). A follow-up ultrasound examination was also performed at the short-term follow-up (Table 1). On ultrasonography, each lesion was identified, measured in length and/or width and given a visual description in regard to disruption of fiber pattern and evidence of scar tissue formation. The same VAS grades were used to re-assess for any improvement or worsening of gait.

Results

All 11 cases were presented with a history of an acute to chronic onset of pelvic limb lameness. The lameness appeared in all dogs after lunging activities during Schutzhund training occurring several hours to 3 mo before presentation. At initial evaluation, all 11 cases were diagnosed with a semitendinosus myopathy. Each of the 11 cases of semitendinosus myopathy had a short stride in the affected limb, circumduction of the affected limb during ambulation, and a palpably enlarged and painful semi-tendinosus muscle. On VAS assessment of the 11 cases, 3 were severely affected, 3 were moderately affected, 4 were mildly affected, and 1 had a normal gait in the affected limb. In this case, the semitendinosus muscle tear was found incidentally on initial ultrasound examination when the patient was presented for contralateral pelvic limb lameness (Table 2).

In 8 of 10 cases, initial ultrasonographic findings of the semitendinosus muscle included hypervascularization of swollen areas near the origin of the semitendinosus muscle, hypoechoic fluid pockets, and disruption of the normal fiber pattern, indicative of an intramuscular tear (Figure 1). In 3 of 10 cases, there was evidence of an irregular-appearing fiber pattern (no longer parallel in orientation), indicative of early scar tissue formation.

Figure 1.

Ultrasonographic images of normal (left) and abnormal (right) semitendinosus muscles. A — Normal parallel muscle fiber pattern outlined in the white oval. B — Hypoechoic lesion and disruption of the normal fiber pattern indicative of a muscle tear outlined in the white oval.

Using a human ultrasonographic grading scheme (9), 8 of 10 cases were classified as a grade 3 muscle injury at the initial ultrasound examination, where there is “discontinuity of the muscle…and intermuscular, perifascial and subcutaneous fluid collections are common.” The remaining 2 cases were classified as a grade 2 muscle injury, where there is “> 5% but < 100% disruption of the cross-sectional area of the muscle… an intramuscular fluid collection may be seen surrounded by a hyperechoic halo.”

At the short-term follow-up examination < 6 mo after inoculation with MSC, the VAS had remained static at a score of 2 or less in 2 of the 11 cases. It had improved in 9 of the 11 cases, with a reduction in the VAS score by 1 to 3 points (Table 2). There was a mean reduction in the overall intramuscular lesion size of 54.82% (SD +/− 18.02; range: 30.5% to 82.7%) (Figure 2). In 2 of 10 cases, scar tissue formation that was detected at the initial evaluation had completely resolved. In 1 of 10 cases, no scar tissue formation was documented at the initial evaluation, but the subsequent evaluation revealed suspected scar tissue formation. The human ultrasonographic grading scheme of muscle tears (9) revealed a static grade in 6 of 10 cases, or a single grade of improvement from grade 3 to grade 2 in 4 of 10 cases.

Figure 2.

Ultrasonographic images of the semitendinosus muscle pre- and post-treatment. A — At initial diagnosis there is disruption of the normal parallel fiber pattern with a large hypoechoic intramuscular lesion (outlined in the white oval), indicative of a grade 3 muscle injury (9). B — One month post-MSCs injection, there is a more parallel fiber pattern with a small hypoechoic intramuscular lesion (outlined in the white oval), indicative of a grade 2 muscle injury (9).

At the long-term follow-up examination > 1 y post-implantation of MSC compared with initial presentation, the VAS was normal in 7 of 10 cases. One case was lost to long-term follow-up. Two cases had improved VAS by 1 to 2 grades. A single case remained static in the VAS at the grade of 2. Ultrasonographic information was not obtained at the long-term follow-up examination.

Fisher’s exact test was used to evaluate the 11 cases herein compared with the 31 published historical cases for resolution of lameness and return to active duty (Table 3). One case in this study was lost to long-term follow-up; 2 of the published historical cases were lost to long-term follow-up. All 3 of these cases were excluded from analysis when conducting the Fisher’s exact test. There was a statistically significant difference (P = 0.000008) between the MSC-treated cases and the 29 historical cases that were not treated with MSC.

Table 3.

Fisher’s exact test comparing normal gait versus abnormal gait at long-term follow-up

	Normal gait > 1 year	Lameness > 1 year	Row totals
MSC treated cases	7	3	10
Historical published cases	0	29	29
Column totals	7	32	39 (Grand total)

Fisher’s exact test statistic value is 0.000008.

Discussion

Previous therapies for semitendinosus myopathy have included medical management (rest and a combination of anti-inflammatory medications) or surgical procedures (complete myotenectomy, segmental myectomy or myotenectomy, full thickness myectomy or myotenectomy, tenorrhaphy, tenotomy, myotenotomy) (1–4). Medical management has been unrewarding with limited improvement in overall function of the limb in the short and long term periods. Surgical therapy has resulted in resolution of the abnormal gait in the initial post-operative period, but all cases had documented recurrence of lameness within 2 to 9 mo after surgery (1–4).

A novel therapy for musculoskeletal injuries involves intral-esional or intravenous injections with mesenchymal stem cells. These stem cells have a high proliferative potential, ability to generate primary colony-forming unit fibroblasts (CFU-F) and the ability to differentiate into bone, fat, cartilage, skeletal muscle, tendon, and nerve tissues (10–15). They can be derived from a variety of tissues, including bone marrow, adipose tissue, muscle, tendon, dental pulp, periodontal ligament, umbilical cord blood, placenta, periosteum, liver, cartilage, synovium, synovial fluid, spleen, and thymus (10,16,17). Each of these tissues has limitations in regard to invasiveness for harvesting, time necessary for processing and culturing of stem cells, and overall yield. Kisiel et al (11) found periosteum to be a superior tissue source in providing the greatest number of MSCs per gram of tissue, followed by adipose tissue, muscle, and bone marrow.

Adipose tissue is an excellent source of MSC that can be harvested from various sites, including subcutaneous sites, falciform tissue, and omental tissue (8). Techniques for harvesting adipose tissue include liposuction under sedation (11), incisional techniques with local anesthetics (12), and intra-abdominal techniques under general anesthesia (8). Due to the ease of obtaining falciform fat under a short general anesthetic procedure, we elected to use this as the source of adipose MSC in this study. As the dogs were all working German shepherd police dogs, they required sedation or general anesthesia for ultrasound diagnosis, which allowed acquisition of the falciform adipose tissue under the same anesthetic procedure.

Several reports in canine, murine, equine, and non-human primate models have shown that MSC obtained from adipose tissue, muscle, or bone marrow have the ability to migrate to regions of inflammation/trauma, help restore normal architecture, and improve function in skeletal muscle and tendon tissues (13–15,17–23). Many of these reports evaluated functional outcomes, histological and/or immunological samples, or ultrasound examination to help determine the effectiveness of MSC therapy. Musculoskeletal ultrasound is a non-invasive method that is extensively used within the equine industry for evaluating muscular architecture for diagnosis of muscular contusions/tears and for evaluating changes after therapy (12,15,22,23).

With adipose-derived MSC therapy, the 11 cases in this study showed improvement in gait analysis (VAS scores) in the short- and/or long-term follow-up period compared with initial evaluation. Within 3 mo of diagnosis and adipose-derived MSC intralesional injections, all of the affected dogs had returned to full training level and competition or work. At long-term follow-up, 1 y or more after treatment, the dogs in all 10 cases were still on active duty with improved (7 of 10) or static (2 of 10) gaits on visual assessment scores. A single case did not have a long-term follow-up examination, but the dog was on active duty as reported by the police department canine force.

The ultrasonographic grade applied to each of the 10 cases (9) revealed a static grade in 6 of 10 and a single grade improvement in 4 of 10 at the short-term follow-up examination. To the best of the authors’ knowledge, there is no comparable grading scheme that has been validated in the veterinary literature.

At the time of writing (2015), 5 of the 9 dogs had been euthanized in previous years for causes unrelated to injury or MSC therapy (hepatic neoplasia, chronic renal failure, intervertebral disc herniation, unknown neoplasia, unknown cause). These deaths occurred between 2 to 5 y after adipose-derived MSC treatment. The remaining 4 dogs were alive and doing well (Figure 3). One dog was still on active duty with the police force, and the other 3 were retired. There was no documented recurrence of lameness in the affected pelvic limb in any of the 11 cases in this study, unlike historical controls that showed recurrence of lameness within 2 to 9 months in all cases (1–4). This demonstrates that adipose-derived MSCs may be a novel and beneficial treatment option for dogs with semitendinosus myopathy.

Figure 3.

Post-MSC treated German shepherd dog, Brix, during active Schutzhund training.

There are a number of limitations to this study, including being a retrospective analysis of current semitendinosus myopathy cases compared to previously published reports. Since these dogs were all police force dogs on active duty at the time of diagnosis, histological and immunohistochemical analysis were not performed to further evaluate muscle architecture. Postmortem histopathologic samples could not be obtained; those dogs that expired were cremated elsewhere. Long-term follow-up ultrasonographic examinations were not performed in this study. This may account for the difference between the 60% static ultrasonographic grade at short-term follow-up compared to the 70% normal VAS score at long-term follow-up. In addition, with semitendinosus myopathy being an uncommon disease limited mainly to working or highly athletic dogs, there is a small number of cases presented here for evaluation.

Despite these limitations, there are several important features of this study. A single examiner was used to diagnose, make all visual assessment scores, and obtain the falciform adipose tissue for all cases. The same board-certified radiologist was used for initial and post-treatment evaluation of the semitendinosus muscle, as well as ultrasound-guided injections in all 11 cases of semitendinosus myopathy. Controlling these 2 major aspects helped to eliminate intra-observer variability in diagnosis, initial evaluation, treatment, and post-therapeutic evaluation. The same processing facility was used for all 11 cases, which helped to control environmental, procedural, or packaging variables that may have altered the MSC viability.

Future studies evaluating semitendinosus myopathy would be strengthened by using histopathological evaluation of the muscle at initial diagnosis and post-MSC therapy. Histopathological evidence would be able to document disruption of the myofiber orientation and/or presence of scar tissue formation after the injury. It would also allow documentation of whether the normal parallel orientation of the myofibers is repaired post-MSC therapy or is replaced by scar tissue formation.

Obtaining a biopsy sample pre- and post-MSC therapy may be difficult to achieve in a controlled manner. Naturally occurring semitendinosus myopathy is uncommon, and it is difficult to replicate this pathology or create a model for evaluation. Several equine studies have used collagenase to create artificial lesions in tendinous structures as a model to determine the value of MSCs (12,15,22,23), but this has not been evaluated in equine muscle tissue to the best of the authors’ knowledge.

In conclusion, this study demonstrated that semitendinosus myopathy treated with adipose-derived MSCs can improve, or help prevent progression, of the career-ending fibrosis and muscle contracture that occurs with this pathology. Initial diagnosis of semitendinosus myopathy is easily confirmed by ultrasound examination, and falciform adipose tissue can be collected under a short anesthetic procedure at the time of initial ultrasound examination. All 9 affected dogs returned to active police duty with no restrictions and continued on active duty until their retirement or demise from unrelated diseases 2 to 5 y after MSC implantation. This is a vast improvement from the reported historical controls, in which there was recurrence within 2 to 9 mo in all cases.