Anatomic Distribution and Clinical Findings of Intermuscular Lipomas in 17 Dogs (2005–2010)

Introduction

Lipomas are fatty tumors that are common in older dogs. They are mesenchymal in origin and are rarely associated with either local invasion or malignancy, unlike infiltrative lipomas and liposarcomas, respectively.^1–3 Infiltrative lipomas have a higher rate of recurrence (36%), and liposarcomas have a greater potential for metastasis compared to intermuscular lipomas.^1,2 Intermuscular lipomas (IMLs) in dogs have also been described.⁴ These lipomas reportedly occur in the caudal thigh, most often between the semimembranosus and semitendinosus muscles.⁴ Lipomas can be distinguished from liposarcomas via cytologic and/or histologic evaluation and can be differentiated from infiltrative lipomas either via advanced imaging or surgery.^2,4 Clinical signs of IMLs include a large, subcutaneous mass that presumably causes discomfort and may interfere with function, leading to lameness.⁴ Due to the size and firm nature of IMLs, patients are often referred to specialists with a tentative diagnosis of malignancy and a guarded prognosis.⁴ However, IMLs are associated with an excellent prognosis and are easily removed with blunt and finger dissection after separation of the muscular fascia surrounding the lipoma.⁴

There is a paucity of information regarding IMLs in the veterinary literature. The authors of a previous report concluded that IMLs were somewhat unique to the caudal thigh of the canine and that there did not seem to be a similar predilection for the forelimb.⁴ At the authors’ institution, however, multiple cases of axillary IMLs in dogs have been recognized. In some cases, referral was associated with a guarded prognosis due to the clinical signs, size, firm nature of the mass, and the apparent absence of information regarding IMLs of the thoracic limb in the veterinary literature.

The purpose of this study was to describe the anatomic location and clinical findings in a series of dogs that were diagnosed and treated for IMLs at a single referral institution. A secondary aim was to describe effective imaging techniques for preoperative assessment. The hypothesis was that IMLs of the thoracic limb are associated with a prevalence and prognosis similar to IMLs of the caudal thigh.

Materials and Methods

The medical records database of the Teaching Hospital of The Colorado State University was searched for all dogs that had surgery for lipoma, IML, or interfascial lipoma from 2005 to 2010. The term interfascial (i.e., between fascial planes) was included as IMLs were referred to as interfascial lipomas in some earlier cases. Only dogs surgically treated for IMLs with histopathologic confirmation were included in this report. Dogs were excluded if they were diagnosed with another form of lipoma, such as subcutaneous lipoma, infiltrative (intramuscular) lipoma, or liposarcoma. Data collected from the medical records included signalment, presenting complaint, duration of clinical signs prior to surgery, specific location of the mass, two-dimensional size of the mass, diagnostic tests performed, surgical procedure, surgical duration and complications, recurrence, and length of follow-up. Follow-up information was obtained either by physical examination or telephone communication with the owner/referring veterinarian. Data were stored and analyzed using standard statistical software^a.

Results

Sixteen dogs (8 castrated males and 8 spayed females) with 17 IMLs met the inclusion criteria. Breeds represented included Labrador retriever (n=4), golden retriever (n=1), border collie (n=1), Weimaraner (n=1), Irish setter (n=1), Australian shepherd (n=1), Jack Russell terrier (n=1), miniature pinscher (n=1), and mixed-breed (n=5). Mean age of all dogs was 9.6±2.0 yr and mean body weight was 31.2±10.4 kg. Clinical signs included a slowly enlarging mass (n=11), rapidly enlarging mass (n=6), lameness of the affected limb (n=6), thoracic limb abduction due to mechanical displacement (n=3), reluctance to ambulate (n=2), discomfort (n=2), licking at the mass (n=2), paresis (n=2), and ipsilateral Horner’s syndrome (n=1). Duration of clinical signs prior to surgery ranged from 2 wk to 3 yr (median, 16 wk). IMLs of the thoracic and pelvic limbs were identified in 7 and 10 dogs, respectively. IMLs of the thoracic limb were found between the superficial and deep pectoral muscles (n=3), the serratus and subscapular muscles (n=2), the omotransversarius and subscapularis muscles (n=1), and the antebrachial flexor muscles (n=1). All of the thoracic limb IMLs, except for the mass between the antebrachial muscles, were in the axilla. Pelvic limb IMLs were found between the semitendinosus and semimembranosus (n=8), the biceps femoris and vastus lateralis (n=1), and the biceps femoris and semitendinosus (n=1). Duration of growth was not different between thoracic (range, 6 wk to 2.5 yr; median, 3 mo) and pelvic limb IMLs (range, 2 wk to 3 yr; median, 5.5 mo; P=0.84). Thoracic limb IML sizes ranged from 24 cm² to 368 cm² (median, 100 cm²) and pelvic limb IML sizes ranged from 70 cm² to 300 cm² (median, 90 cm²). Tumor size was not different between the thoracic and pelvic limbs (P=0.52). There was no difference in prevalence of IMLs between the thoracic and pelvic limbs (P=0.49), and there was no predilection for the caudal thigh compared with the axilla (P=0.15).

Dogs were evaluated with cytology following fine-needle aspiration and diagnostic imaging of the tumor. Results of fine-needle aspiration and cytology revealed adipocytes, which was consistent with lipoma in all cases. Imaging of the masses was performed using computed tomography (CT, n=6), radiographs (n=4), ultrasound (n=2), and MRI (n=1). Four dogs had no imaging performed prior to surgical removal.

IMLs appeared as well-circumscribed, noninvasive, fat-dense masses with linear streaks of soft tissue-attenuating material on CT. Radiographically, IMLs imaged as fat-dense masses, but impressions regarding possible infiltration were not made. Ultrasound evaluation demonstrated echogenic, well-encapsulated, striated masses in all cases. The one IML imaged using MRI was hyperintense on T1- and T2-weighted images and hypointense on short T1 inversion recovery images. All imaging modalities accurately described the tissue density of the IMLs, but only CT, ultrasound, and MRI allowed concurrent evaluation of tissue borders and association with regional vasculature.

Mean surgical time was 58.3±16.3 min and 61.1±19.3 min for thoracic and pelvic limb IMLs, respectively (P=0.76). All masses were removed using blunt dissection following separation of fascial planes or muscular bellies. Closed-suction (Jackson-Pratt^b) active drains were placed in 10 dogs, and a passive Penrose drain^c was placed in one dog. Only two intraoperative complications were noted in the operative records. First, there was minor bleeding in one dog with both a subscapular and caudal thigh IML resected during the same anesthetic period. Postoperatively, seromas developed in two dogs and one dog developed an incisional infection within 5–7 days of surgery. Both seromas occurred in dogs with thoracic limb lipomas, and the incisional infection occurred in a dog with a pelvic limb lipoma. A Jackson-Pratt closed-suction drain was used in one of the seroma cases, as well as the case with the incisional infection, and all complications resolved with conservative treatment. Complications were not different between dogs with thoracic or pelvic limb lipomas (P=0.54).

Follow-up data were available for all dogs. Mean follow-up time was 13.5±8.5 mo (range, 2–31 mo; median, 12 mo). Recurrence was not reported in any of the cases in this study.

Discussion

To the authors’ knowledge, this is the first report of IMLs occurring in the thoracic limb. Moreover, the behavior, diagnostic findings, and surgical outcome of thoracic limb IMLs are similar to those reported for IMLs of the caudal thigh. Because these tumors may initially appear clinically ominous, it is vital that knowledge of their occurrence in the thoracic limb and associated excellent prognosis be documented in the veterinary literature.

Signalment of dogs reported in this study was consistent with previous reports that indicate that Labrador retrievers are over-represented.^1,4 The clinical signs noted in this study ranged from mild and expected (e.g., lameness and swelling) to menacing and unexpected (e.g., paresis and Horner’s syndrome). Many of the dogs included in this study were referred to the Clinical Oncology Service of the authors’ hospital with a guarded prognosis despite previous fine-needle aspirate cytology consistent with the diagnosis of a lipoma. Given the anatomic differences between the thoracic and pelvic limbs, the seemingly unique clinical signs noted in some dogs with IMLs of the thoracic limb were expected. Although there were not enough clinical signs reported for meaningful analysis, limb abduction and Horner’s syndrome were identified in dogs with thoracic limb IMLs. An IML should be considered as a differential diagnosis for any dog with those signs and an axillary mass.

Reported growth rate and size were not different between thoracic and pelvic limb IMLs in this study. This is not an unexpected finding given the benign biologic behavior of IMLs previously reported.⁴

The previously suspected predilection of IML for the caudal thigh was not observed in this study. However, 80% of the pelvic limb IMLs occurred between the semitendinosus and semimembranosus muscles. In contrast, thoracic limb IMLs were more evenly distributed among the muscles of the thoracic limb, with 43% occurring between the pectoral muscles and 43% occurring between the scapular muscles. The reasons for these differences in distribution of the masses are not known; however, it is likely that anatomic differences between the tissues surrounding the scapulohumeral and coxofemoral joints likely play a role. IMLs of the antebrachium appear to be rare given that only one case was identified in this study. To the authors’ knowledge, antebrachial IMLs have not been reported as a distinct entity in the veterinary literature. This is in contrast to the human literature where these tumors have been reported to compress the median nerve, causing paresis and pain.^5,6 The dog in this study with an antebrachial IML also had compression of the median nerve, which was identified at the time of surgery. This dog reportedly had chronic licking of the antebrachium over the mass and was lame and painful on physical examination. Those clinical signs resolved following removal of the IML, which is consistent with human reports.^5,6

Diagnostic work-up was similar for all dogs, exception for the specific imaging modality used. The choice of advanced imaging in the majority of cases is notable as it likely reflects the concerns of malignancy and invasion of not only the referring veterinarian and pet owner, but also the surgeon. Caudal thigh IMLs were the only lipomas imaged with radiographs only. This is likely due to the relatively uncomplicated anatomy in this region compared with the axilla and because IMLs of the caudal thigh are relatively well-documented in the veterinary literature.⁴ CT was the most common imaging modality used in this study. In all cases, a well-circumscribed noninvasive mass was identified. Previous veterinary reports have suggested variability in the accuracy of CT for determining the extent of infiltration associated with infiltrative lipomas.^1,7 In people, however, advanced imaging is considered helpful in differentiating IMLs from other lipomas.⁸ It was not within the scope of this study to assess the ability of CT to discern between infiltrative lipomas and IMLs because dogs with infiltrative lipomas were excluded from the study. However, all of the masses in this report were characterized by CT as being well-circumscribed and noninvasive. The ultrasonographic imaging characteristics of lipomas have previously been reported and include the following: a striped appearance, sharp edges, and a thin hyperechoic capsule.⁹ Ultrasound evaluation of two IMLs in the current study demonstrated these characteristics and allowed the identification of important vessels in the area, such as the axillary artery and vein (Figure 1). This identification allowed careful surgical planning and likely imparted a margin of safety to the procedure. In the authors’ opinion, ultrasound should be considered as an important diagnostic aid in the preoperative work-up of dogs with suspected IMLs. However, the ability of ultrasound to discern between infiltrative and intermuscular lipoma variants was not evaluated in this study. Further investigation would be needed to compare the accuracy of CT and ultrasound for distinguishing between infiltrative and intermuscular lipomas.

View Figure

• Download as Powerpoint
• Download Figure

MRI was used in a single dog in this study. That dog had an axillary IML causing Horner’s syndrome due to compression of the vagosympathetic trunk (Figure 2). Surgical removal of the tumor was performed easily after separation of the deep and superficial pectoral fascial planes. Resolution of clinical signs occurred 2 wk following surgery. MRI proved useful in this case; however, MRI may not be necessary for the assessment of the majority of IMLs in dogs.

View Figure

• Download as Powerpoint
• Download Figure

It might be expected that surgical resection of IML of the axilla would be more difficult than those in the caudal thigh; however, surgical time was ∼1 hr for IMLs of both the caudal thigh and axilla. Additionally, the only intraoperative complication noted was minor bleeding that was easily controlled with monopolar cautery. These findings further demonstrate the lack of invasiveness of IMLs, and veterinarians should consider this information when counseling clients about the risks of surgery.

Postoperative complications were rare, limited to a single incisional infection and seromas in two dogs. All complications resolved with minimal intervention and did not affect long-term outcome. This finding is consistent with those of a previous report of caudal thigh IMLs in which only mild seromas were noted postoperatively.⁴ The authors of that study recommended passive drain placement to decrease the risk of seroma formation because seromas were not seen in any of the cases treated with a passive drain. Four of the six dogs in that study that did not receive a drain developed a seroma.⁴ Closed-suction drains were preferred by surgeons in the current study. Although drains were placed in the majority of dogs, the low incidence of seromas (overall) precluded the authors from performing a meaningful analysis. In the authors’ opinion, a closed-suction drain should be considered in large surgical wounds to help reduce the risk of seroma formation, although a passive drain may also be effective.⁴

Recurrence was not seen in any dogs in the follow-up period. Although the duration of the follow-up period (median, 12 mo) may have limited the identification of recurrence in some dogs, a previous report of IMLs demonstrated no recurrence during a median follow-up period of 17 mo.⁴

Conclusion

IMLs occur with similar frequency in the thoracic and pelvic limbs of dogs. Although a predilection for the caudal thigh seems to exist for pelvic limb IMLs, there was not a predilection site associated with thoracic limb IMLs. Regardless of the anatomic location, the majority of IMLs can be easily evaluated with fine-needle aspiration and ultrasound prior to surgery. As with IMLs of the caudal thigh, IMLs of the axilla are associated with an excellent prognosis and recurrence is considered very unlikely.