Modified hemipelvectomy techniques in dogs and cats appear well tolerated with good functional outcomes

Maureen A. Griffin Department of Clinical Sciences and Advanced Medicine, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA

Search for other papers by Maureen A. Griffin in
Current site
Google Scholar
PubMed
Close
 DVM, DACVS-SA
,
Johnny Altwal College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO

Search for other papers by Johnny Altwal in
Current site
Google Scholar
PubMed
Close
 BS
,
William T.N. Culp Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California-Davis, Davis, CA

Search for other papers by William T.N. Culp in
Current site
Google Scholar
PubMed
Close
 VMD, DACVS-SA
,
Bernard Seguin Department of Clinical Sciences, Flint Animal Cancer Center, Colorado State University, Fort Collins, CO

Search for other papers by Bernard Seguin in
Current site
Google Scholar
PubMed
Close
 DVM, MS, DACVS
, and
Brandan Wustefeld-Janssens Department of Clinical Sciences, Flint Animal Cancer Center, Colorado State University, Fort Collins, CO

Search for other papers by Brandan Wustefeld-Janssens in
Current site
Google Scholar
PubMed
Close
 BVSc, DECVS

Abstract

OBJECTIVE

To describe the clinical characteristics, procedural techniques, complications, and outcomes of dogs and cats undergoing any of the following modified hemipelvectomy techniques: concurrent partial sacrectomy and/or partial vertebrectomy, osseous excision crossing midline, and reconstruction without the use of local musculature.

ANIMALS

23 client-owned animals (20 dogs and 3 cats) that underwent modified hemipelvectomy techniques. Animals that underwent traditional (nonmodified) hemipelvectomy techniques were excluded.

PROCEDURES

The medical records of 3 academic institutions were reviewed, and data were recorded and analyzed.

RESULTS

Modified hemipelvectomy was performed with partial sacrectomy and/or vertebrectomy in 11 dogs, excision crossing pelvic midline with concurrent limb amputation in 5 dogs and 2 cats, and closure without use of native muscle or mesh in 4 dogs and 1 cat. Surgery was performed for tumor excision in all cases. Excision was reported as complete in 16 of 23, incomplete in 6 of 23, and not recorded in 1 of 23 animals. All animals survived to discharge. Only animals undergoing partial sacrectomy/vertebrectomy (4/11) experienced postoperative mobility concerns. Major intra- or post-operative complications (grades 3 and 4) occurred in 2 dogs that underwent partial sacrectomy/vertebrectomy, and 1 of these animals experienced a complication that resulted in death. The median time to death or last follow-up was 251 days (range, 3 to 1,642).

CLINICAL RELEVANCE

The modified hemipelvectomy techniques reported in this cohort were overall well tolerated with good functional outcomes. These findings support the use of these modified hemipelvectomy techniques in dogs and cats, and previous notions regarding tolerable hemipelvectomy procedures should be reconsidered. However, additional studies with larger numbers of patients undergoing modified hemipelvectomy techniques are needed to gain more information.

Abstract

OBJECTIVE

To describe the clinical characteristics, procedural techniques, complications, and outcomes of dogs and cats undergoing any of the following modified hemipelvectomy techniques: concurrent partial sacrectomy and/or partial vertebrectomy, osseous excision crossing midline, and reconstruction without the use of local musculature.

ANIMALS

23 client-owned animals (20 dogs and 3 cats) that underwent modified hemipelvectomy techniques. Animals that underwent traditional (nonmodified) hemipelvectomy techniques were excluded.

PROCEDURES

The medical records of 3 academic institutions were reviewed, and data were recorded and analyzed.

RESULTS

Modified hemipelvectomy was performed with partial sacrectomy and/or vertebrectomy in 11 dogs, excision crossing pelvic midline with concurrent limb amputation in 5 dogs and 2 cats, and closure without use of native muscle or mesh in 4 dogs and 1 cat. Surgery was performed for tumor excision in all cases. Excision was reported as complete in 16 of 23, incomplete in 6 of 23, and not recorded in 1 of 23 animals. All animals survived to discharge. Only animals undergoing partial sacrectomy/vertebrectomy (4/11) experienced postoperative mobility concerns. Major intra- or post-operative complications (grades 3 and 4) occurred in 2 dogs that underwent partial sacrectomy/vertebrectomy, and 1 of these animals experienced a complication that resulted in death. The median time to death or last follow-up was 251 days (range, 3 to 1,642).

CLINICAL RELEVANCE

The modified hemipelvectomy techniques reported in this cohort were overall well tolerated with good functional outcomes. These findings support the use of these modified hemipelvectomy techniques in dogs and cats, and previous notions regarding tolerable hemipelvectomy procedures should be reconsidered. However, additional studies with larger numbers of patients undergoing modified hemipelvectomy techniques are needed to gain more information.

Introduction

Multiple hemipelvectomy techniques have been described in dogs and cats, with historical classifications being based on total versus partial hemipelvectomy (ie, removal of the entire hemipelvis vs a portion of the hemipelvis, respectively) and external versus internal hemipelvectomy (ie, concurrent amputation vs preservation of the limb, respectively).1,2 Several studies have described not only the traditional techniques used but also outcomes of dogs and cats undergoing hemipelvectomy, with the largest study to date including 84 dogs and 16 cats.36 In these studies, all described techniques included excision of unilateral osseous structures only (without any resections crossing midline), appendicular skeletal excision only (without any concurrent resections of the sacrum and/or vertebrae), and reconstruction of the abdominal wall via primary muscular closure, use of a muscle flap, or synthetic mesh. Although two recent case reports describe a dog and cat that underwent modified hemipelvectomy with excised osseous structures crossing midline (involving both hemipelves) with reportedly good limb function postoperatively, concurrent amputation (external hemipelvectomy) was not performed in either case.7,8 Additional case reports document dogs undergoing nontraditional hemipelvectomy techniques, but no reports to date have documented dogs or cats undergoing modified hemipelvectomy with concurrent sacrectomy, vertebrectomy, osseous excisions crossing midline with concurrent limb amputation (external hemipelvectomies), or closure techniques without local muscular tissues or mesh.912 Also, though reported, very limited data exists on veterinary patients undergoing hemipelvectomy with mesh reconstruction of the abdominal wall or with any contralateral osseous excision.

Ultimately, information on the clinical findings, techniques, complications, and short- and long-term outcomes in dogs and cats that undergo these modified hemipelvectomy procedures is lacking. Our primary objective was to describe the preoperative findings, procedural techniques, perioperative complications, and outcomes of dogs and cats undergoing modified hemipelvectomy techniques involving concurrent sacrectomy and/or vertebrectomy, osseous excision crossing midline, and/or reconstruction without local muscular tissues.

Materials and Methods

The medical record databases of 3 academic teaching hospitals were retrospectively searched to identify dogs and cats that underwent hemipelvectomy for any indication. Dogs or cats that had hemipelvectomy procedures with any of the following components (modified techniques) were included: sacrectomy, vertebrectomy, excision crossing pelvic midline, closure technique without muscular tissue (ie, utilizing mesh, omentum, other natural or synthetic materials, or closure with subcutaneous tissue and skin only). Sacrectomy, vertebrectomy, and osseous excisions crossing midline are depicted (Figures 13). Information obtained from the medical records included signalment, history, indication for hemipelvectomy, preoperative pelvic limb function, preoperative staging, preoperative imaging of the pelvis, hemipelvectomy surgical techniques, intra- and post-operative complications, survival to discharge, histopathology results, postoperative limb function and mobility, neoadjuvant and adjuvant treatments, postoperative imaging of the pelvis, and timing and cause of death. Data regarding tumor type, hemipelvectomy modification procedure, complications, mobility, and outcome for each patient are demonstrated (Supplementary Table S1).

Figure 1
Figure 1

Modified hemipelvectomy with partial sacrectomy. In all images, the location of sacral osteotomy is denoted with a dashed red line, and the portion of excised sacrum has been faded. Given the variety of concurrent ipsilateral hemipelvectomy procedures, additional osteotomy locations are not demonstrated in these images. A—Excision of the sacral wing. B—Excision of the lateral third of the sacrum. C—Excision of slightly less than half the sacrum with the osteotomy just ipsilateral to midline. D—Excision of the spinous processes (median sacral crest) of the sacrum.

Citation: Journal of the American Veterinary Medical Association 261, 10; 10.2460/javma.22.12.0592

Figure 2
Figure 2

Modified hemipelvectomy with partial vertebrectomy. In both images, the location of vertebral osteotomy is denoted with a dashed red line, and the portion of excised vertebra has been faded. Given the variety of concurrent ipsilateral hemipelvectomy procedures, additional osteotomy locations are not demonstrated in these images. A—Excision of the transverse processes of L6 (cranial osteotomy) and L7 (caudal osteotomy). B—Excision of the spinous process of L7.

Citation: Journal of the American Veterinary Medical Association 261, 10; 10.2460/javma.22.12.0592

Figure 3
Figure 3

Modified hemipelvectomy with excision crossing midline. The location of contralateral pubic/ischial osteotomy was within the region demonstrated (1 [pink shading]) for cases in which the transection was within the medial third of the contralateral hemipelvis and within the region demonstrated (2 [blue shading]) for cases in which the transection was within the middle third of the contralateral hemipelvis. In all cases, hemipelvectomy was classified as partial and mid-to-caudal and associated with limb amputation (external hemipelvectomy), as demonstrated by the faded portion of pelvis and red dashed line depicting the additional osteotomy location.

Citation: Journal of the American Veterinary Medical Association 261, 10; 10.2460/javma.22.12.0592

Limb use was characterized as functional (weight-bearing with adequate orthopedic and neurologic function) or nonfunctional, and any orthopedic and neurologic deficits noted on examination were described. Preoperative lameness, prior to modified hemipelvectomy with limb amputation procedures, was graded on a scale from 0 to 5 (Supplementary Table S2).13 Postoperatively, mobility/ambulation was described without use of a lameness score (owing to the amputated status of most patients). Neurologic deficits were also described. Complications were listed as grades 1 through 4 in accordance with the CLASSIC (Classification for Intraoperative Complications) criteria for intraoperative complications and the Accordion criteria for postoperative complications (Supplementary Tables S3 and S4).14

Survival time was defined as days from modified hemipelvectomy to death or euthanasia. Follow-up time was defined as days from modified hemipelvectomy to last follow-up in animals that were alive at last follow-up.

Descriptive statistics were calculated for all measured variables. Continuous variables were reported as median (range), and categorical variables were reported as number with or without percentage.

Results

Preoperative findings for all animals

Between 1996 and 2021, 20 dogs and 3 cats from 3 institutions satisfied the inclusion criteria. The median age of dogs was 9.1 years (range, 2 to 13.3 years). Dogs consisted of 10 of 20 (50%) female spayed, 8 of 20 (40%) male castrated, and 2 of 20 (10%) male intact. Dog breeds included Labrador Retriever (4/20 [20%]), mixed breed (4/20 [20%]), Golden Retriever (2/20 [10%]), Rottweiler (2/20 [10%]), and 1 each (5%) of the following breeds: Belgian Tervuren, Great Dane, Cocker Spaniel, Staffordshire Terrier, German Shepherd Dog, Bernese Mountain Dog, Irish Setter, and Flat-Coated Retriever. The median weight of dogs at the time of surgery was 30.4 kg (range, 12.2 to 72 kg). For cats, the median age was 10.3 years (range, 5.6 to 15.3 years). Cats consisted of 2 of 3 (66.7%) female spayed and 1 of 3 (33.3%) male castrated. Cat breeds included domestic shorthair (2/3 [66.7%]) and Maine Coon (1/3 [33.3%]). The median weight of cats at the time of surgery was 5.2 kg (range, 5.1 to 8.2 kg).

Reported comorbidities included osteoarthritis (n = 2), allergies (2), ocular disease (2), squamous cell carcinoma of the mandibular lip excised concurrently at the time of modified hemipelvectomy (1), prior mast cell tumor excision (1), prior perianal adenoma excision (1), hypothyroidism (1), proteinuria (1), hypoadrenocorticism (1), mitral valve disease (1), prior tail amputation for trauma (1), liver enzyme elevation (1), chronic large bowel diarrhea (1), and historical seizures (1).

All animals that underwent modified hemipelvectomy had surgery performed for tumor excision. Three animals underwent modified hemipelvectomy for recurrent neoplastic disease. One cat received modified hemipelvectomy for a recurrent injection site sarcoma that was previously excised and treated with adjuvant radiation therapy approximately 1 year prior to presentation. One dog had a soft tissue sarcoma (peripheral nerve sheath tumor) excised in the hip/ischial region by its primary veterinarian 4 weeks prior to presentation with rapid recurrence. Another dog previously underwent traditional external hemipelvectomy for suspected chondroblastic osteosarcoma with subsequent local recurrence, prompting revision modified hemipelvectomy with partial sacrectomy; this same dog developed a pulmonary mass (osteosarcoma) that was excised 6 months prior to the modified hemipelvectomy and also received carboplatin chemotherapy prior to modified hemipelvectomy. One additional dog received 2 doses of neoadjuvant chemotherapy (administered by the primary veterinarian; agent not specified) for chondrosarcoma prior to surgery. No other animals received neoadjuvant therapy prior to modified hemipelvectomy.

On preoperative staging, 21 of 23 (91.3%) animals had no evidence of metastatic disease, 1 of 23 (4.3%) animals had multiple small miliary soft tissue opacities throughout the lung fields on thoracic radiographs with differentials including osteomas versus early pulmonary metastatic disease, and 1 of 23 (4.3%) animals (dog with recurrent osteosarcoma with prior traditional hemipelvectomy and limb amputation) had previously undergone lung lobectomy for excision of suspected metastatic osteosarcoma.

Sacrectomy and vertebrectomy

Modified hemipelvectomy was performed with partial sacrectomy and/or partial vertebrectomy in 11 dogs: partial sacrectomy in 8 dogs, partial vertebrectomy in 1 dog, and both partial sacrectomy and partial vertebrectomy in 2 dogs. No cats underwent modified hemipelvectomy with partial sacrectomy or vertebrectomy.

Preoperatively, 1 of 11 dogs had undergone previous limb amputation. Of the other 10 dogs, 2 had no apparent lameness, 1 had grade 1 lameness, 3 had grade 2 lameness, 3 had grade 4 lameness, and 1 had grade 5 lameness. No dogs were noted to have neurologic deficits preoperatively.

Preoperatively, all 11 dogs had advanced imaging of the pelvis performed, with CT in 10 of 11 dogs and MRI in 1 of 11 dogs (this dog underwent partial sacrectomy). Abnormal pelvic and axial osseous structures on imaging involved the ilium (9/11), sacrum (8/11), L6 and/or L7 vertebrae (3/11), acetabulum (3/11), pubis (1/11), and coccygeal vertebrae (1/11), with multiple dogs having multiple osseous structures affected. All dogs had unilateral osseous abnormalities with no abnormalities crossing midline.

For the 10 dogs that underwent modified hemipelvectomy with partial sacrectomy, the extent of sacrum excised was undefined in 3 of 10, lateral third in 2 of 10, slightly less than half the sacrum with the osteotomy just ipsilateral to midline in 2 of 10, sacral wing in 1 of 10, lateral 1 cm in 1 of 10, and spinous processes (median sacral crest) in 1 of 10. For the 3 dogs that underwent modified hemipelvectomy with partial vertebrectomy, the extent of vertebrae excised involved the spinous process of L7 in 1 of 3, transverse processes of L6 and L7 in 1 of 3, and transverse process of L7 in 1 of 3. All partial sacrectomy and partial vertebrectomy procedures were ipsilateral to the hemipelvectomy procedures or on midline in association with the spinous processes. Described osteotomy locations for dogs of this modified hemipelvectomy group are depicted (Figures 1 and 2). In addition, examples of osteotomy locations on CT scans of patients in this group are provided (Supplementary Figure S1).

Modified hemipelvectomy was performed with associated limb amputation (external hemipelvectomy) in 9 of 11 dogs; 1 of 11 dogs underwent modified hemipelvectomy with preservation of the limb (internal hemipelvectomy), and 1 of 11 dogs had prior limb amputation. The dog that had limb preservation underwent excision of the cranial portion of the ilial wing (transected at the level of the sacroiliac joint) as well as the spinous processes of the sacrum and L7 vertebra; amputation for wide excision was recommended and declined by the owner preoperatively. Modified hemipelvectomy was classified as partial in 8 of 11 dogs and total in 3 of 11 dogs. For the dogs that underwent modified partial hemipelvectomy, excised portions of the pelvis were mid-to-cranial in 5 of 8 dogs and cranial in 3 of 8 dogs. None of these dogs had hemipelvectomy excisions that crossed midline. One dog that underwent partial sacrectomy had mesh placed for reconstruction of the abdominal wall; all other dogs had routine closure of the body wall utilizing residual muscular tissues. Intraoperative complications were reported in 2 of 11 dogs: both complications were classified as grade 1 and associated with tumor rupture intraoperatively.

All 11 dogs survived to discharge with a median of 2 days (range, 1 to 6 days) of hospitalization. Postoperative complications occurred in 4 of 11 (36.4%) dogs: 2 dogs had grade 1 complications (inappetence, weight loss), 1 dog had both grade 2 (blood transfusion for anemia) and grade 3 (wound dehiscence that required surgical management and antimicrobials) complications within 30 days postoperatively, and 1 dog had both grade 1 (seroma, suspect partial necrosis of skin incision) and grade 4 (L7 vertebral fracture) complications within 30 days postoperatively. The dog with the grades 2 and 3 complications had undergone partial sacrectomy and vertebrectomy with routine closure, and histopathology diagnosed hemangiosarcoma of the bone and muscle with metastasis in the subcutaneous tissues. The dog with the grades 1 and 4 complications had undergone modified total hemipelvectomy with nearly half of the sacrum excised for osteosarcoma (reported as completely excised).

Overall, 7 of 11 (63.6%) dogs had no reported mobility concerns postoperatively. One of the 2 dogs that had partial sacrectomy with nearly half of the sacrum excised was noted to be ambulatory at the time of discharge, though it had difficulty rising and ambulating for long periods of time. This dog’s postoperative mobility and ambulation acutely declined to a nonambulatory status within 8 days postoperatively, and pelvic radiographs performed 14 days postoperatively revealed a complete, displaced, oblique fracture of the body of L7; this dog was subsequently euthanized. Of note, the other dog that had partial sacrectomy with sacral excision near midline had improved mobility over the course of hospitalization and was ambulatory with some support at the time of discharge, though this dog was ultimately lost to follow-up 5 days postoperatively. A second dog with postoperative mobility concerns was reported to have hind end weakness both prior to and following surgery (modified cranial internal hemipelvectomy with excision of the spinous processes of both the sacrum and L7 vertebra) and definitive radiation therapy. A third dog that had received modified hemipelvectomy with partial sacrectomy (with excision of the lateral third of the sacrum) was considered to be weakly ambulatory until death 13 days postoperatively when the dog experienced hypovolemic shock from a hemorrhagic event associated with residual hemangiosarcoma. A fourth dog that had received modified hemipelvectomy with partial vertebrectomy (with excision of the transverse process of L7) for osteosarcoma was reported to be ambulating well postoperatively until developing acute and progressive apparent pain and inability to walk 47 days postoperatively; CT of the pelvis and thorax 52 days postoperatively revealed disease recurrence with extension into the spinal canal as well as pulmonary nodules, and the dog was subsequently euthanized. This dog had undergone modified mid-to-cranial hemipelvectomy with partial vertebrectomy for osteosarcoma. No neurologic deficits were noted in any dogs postoperatively, aside from fecal incontinence in the dog that also received definitive adjuvant radiation therapy (incontinence suspected as a late radiation effect).

Postoperative pelvic imaging was performed in 4 of 11 dogs (radiographs in 2, CT in 2). In addition to the dog with pelvic radiographs 14 days postoperatively revealing an L7 vertebral fracture, pelvic radiographs were performed 29 days postoperatively in a dog that underwent modified mid-to-cranial hemipelvectomy with partial sacrectomy for osteosarcoma and showed evidence of tumor recurrence at the site of partial sacrectomy and transverse process of L7. In addition to the dog with CT showing disease recurrence 52 days postoperatively, pelvic CT was performed for radiation planning 20 days postoperatively in a dog that underwent modified cranial internal hemipelvectomy with partial sacrectomy and vertebrectomy for chondrosarcoma and showed reactive changes in the region of the residual sacrum.

Hemipelvectomy excision crossing midline

Modified hemipelvectomy involving partial excision of the contralateral pelvis, or bilateral hemipelves, was performed in 5 dogs and 2 cats.

Preoperatively, 1 of 5 dogs and 2 of 2 cats had no apparent lameness, 1 of 5 dogs had grade 1 lameness, and 3 of 5 dogs had grade 5 lameness. Also, none of the dogs or cats undergoing modified hemipelvectomy excisions crossing midline were noted to have neurologic deficits preoperatively.

Preoperatively, all 7 animals had advanced imaging of the pelvis performed, with CT in 6 of 7 and MRI in 1 of 7. Abnormal pelvic structures on imaging involved the ischium (7/7), pubis (6/7), and acetabulum (3/7), with multiple animals having multiple osseous structures affected. Osseous abnormalities in all cases were unilateral, though the mass extended near or past midline or compressed pelvic canal structures on midline in all dogs but neither cat. No dogs or cats had osseous abnormalities of the sacrum or vertebrae.

For these 7 animals, a portion of the contralateral (relative to tumor gross disease extent and side of amputation) pubis and ischium were excised with the ipsilateral hemipelvis in 6 of 7 cases (including both cats), and a portion of the contralateral ischium without contralateral pubis was excised with the ipsilateral hemipelvis in 1 of 7 case. The transected portions of ischium/pubis were within the medial third of the contralateral hemipelvis in 5 of 7 cases (including both cats) and within the middle third of the contralateral hemipelvis in 2 of 7 cases. No concurrent partial sacrectomy or vertebrectomy was performed in any case. Modified hemipelvectomy was performed with associated limb amputation (external hemipelvectomy) in all 7 cases. Modified hemipelvectomy was classified as partial and mid-to-caudal in all 7 cases. Described osteotomy locations for animals of this modified hemipelvectomy group are depicted (Figure 3). In addition, an example of osteotomy locations on CT scan for a patient of this group is provided (Supplementary Figure S1). All 7 animals had routine closure of the body wall utilizing residual muscular tissues. No intraoperative complications were reported in any dog or cat.

All 7 animals survived to discharge with a median of 2 days (range, 1 to 6 days) of hospitalization. Postoperative complications occurred in 2 of 7 (28.6%) animals following hospital discharge: 1 dog and 1 cat developed grade 2 complications (surgical site infections requiring antimicrobial treatment).

No animals had reported mobility issues or neurologic deficits postoperatively. Postoperative pelvic imaging was performed in 1 cat and no dogs: pelvic radiographs 3 days postoperatively in a cat with pelvic osteosarcoma showed postsurgical changes with no concerning lesions.

Reconstruction without native muscular tissues

Modified hemipelvectomy involving closure without use of native muscular tissues (primary apposition or flap) was performed with mesh in 1 dog and without mesh in 5 animals, including 4 dogs and 1 cat; the dog with mesh used for closure had concurrent partial sacrectomy, and data for this case has been included in the previous section.

Preoperatively, the 1 cat had no apparent lameness, 3 of 4 dogs had grade 4 lameness, and 1 of 4 dogs had grade 5 lameness. Also, 2 of 5 animals (all dogs) were noted to have neurologic deficits preoperatively: 1 dog was weight bearing with mild neurologic dysfunction and severe apparent pain in the lumbar/sacral region, and the other dog was non–weight bearing with absence of proprioception and deep pain sensation.

Preoperatively, 4 of 5 animals (including 3 dogs and 1 cat) had advanced imaging of the pelvis performed with CT; the other dog had nuclear scintigraphy scan without CT or MRI. Abnormal pelvic structures on imaging involved the acetabulum (4/5), ischium (3/5), ilium (3/5), and pubis (2/5), with multiple animals having multiple osseous structures affected. Osseous abnormalities in all cases were unilateral. No animals had osseous abnormalities of the sacrum or vertebrae.

Modified hemipelvectomy was performed with associated limb amputation in all 5 cases. Modified hemipelvectomy was classified as total in 4 of 5 (including the cat) and partial (mid-to-caudal) in 1 of 5. No hemipelvectomy excisions crossed midline. No concurrent partial sacrectomy or vertebrectomy was performed in any case. For these 5 animals, the following closure techniques were utilized: closure of subcutaneous tissues and skin only (4/5, including the cat) and mobilization of omentum from the exposed abdomen with apposition of the omentum to surrounding body wall and musculature followed by closure of subcutaneous tissues and skin (1/5). No prophylactic procedures (such as cystopexy or colopexy) were performed in any case to prevent herniation of organs. The only reported intraoperative complication (1/5 [20.0%]) was a grade 1 complication in a dog undergoing modified total hemipelvectomy, in which a small portion of the sacral wing was inadvertently excised during disarticulation of the sacroiliac joint.

All 5 animals survived to discharge with a median of 2 days (range, 1 to 4 days) of hospitalization. Postoperative complications occurred in 2 of 5 (40.0%) animals: 1 dog prior to discharge (grade 1 characterized by hypovolemia) and 1 dog within 30 days of discharge (grade 1 characterized by minor incisional dehiscence).

No animals had reported mobility issues or neurologic deficits postoperatively. Postoperative pelvic imaging was not performed in any case.

Histopathology results and long-term outcomes for all animals

Modified hemipelvectomy histopathology results in dogs were consistent with osteosarcoma in 10 of 20, chondrosarcoma in 3 of 20, soft tissue sarcoma in 3 of 20, hemangiosarcoma in 2 of 20, osteochondrosarcoma in 1 of 20, and synovial cell sarcoma in 1 of 20. The soft tissue sarcomas were reported as grade 2 in 2 dogs and grade 3 in 1 dog, and the synovial cell sarcoma was reported as grade 1. Modified hemipelvectomy histopathology results in cats were consistent with osteosarcoma in 1 of 3, soft tissue sarcoma (grade not reported) in 1 of 3, and injection site sarcoma (grade not reported) in 1 of 3. Excision was reportedly complete in 16 of 23, incomplete in 6 of 23, and not recorded in 1 of 23 cases.

Adjuvant chemotherapy with doxorobucin, carboplatin, and/or toceranib phosphate was administered in 5 of 20 dogs. No cats received adjuvant chemotherapy. Adjuvant definitive radiation therapy was administered in 1 of 20 dog and no cats.

At the time of study completion, 18 of 23 animals were known to be dead and 5 of 23 were lost to follow-up (Supplementary Table S1). The median time to survival or last follow-up was 251 days (range, 3 to 1,642 days). Of the animals that died, 13 of 18 were euthanized, 3 of 18 died of natural causes, and the etiology of death was unknown for 2 of 18. Death was associated with suspected metastatic disease in 7 of 18 animals, primary disease progression or recurrence in 4 of 18 animals, disease not related to the neoplastic indication for modified hemipelvectomy in 2 of 18 animals, both metastatic disease and primary disease progression in 1 of 18 animals, mobility compromise associated with postoperative vertebral fracture in 1 of 18 animals, and unknown in 3 of 18 animals. Therefore, overall, cause of death was tumor-associated in 13 of 18 animals (including the dog with postoperative vertebral fracture) and either was not tumor associated or was unknown in 5 of 18 animals.

Discussion

This report documents several modified hemipelvectomy techniques not previously described, including concurrent partial sacrectomy and/or vertebrectomy, hemipelvectomy excisions crossing midline with concurrent amputation, and hemipelvectomy abdominal closure without native local muscular tissue or mesh, with overall low incidence of major intra- and post-operative complications and good functional outcomes in the majority of animals. The outcomes of these dogs and cats challenge several widely held notions, including tolerable proportion of sacrum that can be excised, potential for functional compromise with disruption of the contralateral pelvic structures when concurrent limb amputation is performed, and the requirement for reconstruction of the resulting hemipelvectomy abdominal defect with local muscular tissues or mesh.

Most dogs that received modified hemipelvectomy with partial sacrectomy and/or vertebrectomy had good short-term outcomes, though 4 of 11 dogs had reported mobility concerns postoperatively. Anecdotally, up to one-third of the sacrum in width (with osteotomy through the sacral foramina) can be safely excised without functional complication, though no prior reports of concurrent partial sacrectomy with hemipelvectomy have been published in dogs or cats.15 In the present report, 2 dogs had partial sacrectomy excisions that were slightly off midline, with just less than one-half of the sacrum excised in that region. Both dogs had adequate mobility at the time of discharge several days postoperatively. However, 1 dog was lost to follow-up shortly after discharge, and the other dog experienced an acute decline in mobility associated with an L7 body fracture and was euthanized 2 weeks postoperatively. The authors postulate that the L7 fracture may have been associated with altered biomechanics and weight bearing associated with excision of nearly half the sacrum in addition to hemipelvectomy. Therefore, although 2 dogs in this report had partial sacrectomies presumed to be greater than one-third of the sacral width, based on these cases there is no evidence to support the tolerance, overall safety, or long-term functional outcome for modified hemipelvectomy with this more extensive partial sacrectomy procedure. In fact, this data supports the potential for major postoperative complication in the form of vertebral fracture following partial sacrectomy of nearly half the sacrum in conjunction with hemipelvectomy. Additional data is needed, and the authors recommend proceeding with caution in more extensive partial sacrectomies. However, of the dogs that underwent partial sacrectomy with excision of one-third or less of the sacral width, overall good functional outcomes were seen in all dogs. Only 1 dog with partial vertebrectomy developed significant mobility concerns, and this dog had tumor recurrence within the vertebral canal such that the mobility compromise was associated with primary disease recurrence rather than the vertebrectomy/hemipelvectomy procedure itself. Therefore, although no significant mobility concerns following modified hemipelvectomy with partial sacrectomy/vertebrectomy were definitively associated with the procedure itself, one case was attributed to disease recurrence and another patient experienced an acute postoperative complication in the form of vertebral fracture, which may have been associated with altered biomechanical forces on the axial skeleton following extensive partial sacrectomy/hemipelvectomy. Furthermore, several cases had relatively short follow-up, and additional data is needed.

The outcomes of the dogs and cats that underwent modified external hemipelvectomy crossing midline allow for several important conclusions. Although there is a theoretical functional concern involved with disrupting the contralateral pelvic osseous structures and muscle attachments when concurrent limb amputation is performed (ie, anatomical disruption of the only remaining pelvic limb support structures), all dogs and cats with modified hemipelvectomy excisions crossing midline had concurrent amputation, and none had postoperative function or mobility compromise reported. The extent of contralateral excision varied, and all animals had excision of pelvic components within the mid or medial third of the contralateral pelvis such that no conclusions can be drawn regarding more extensive contralateral hemipelvectomy excisions with amputation. In addition, with osseous excisions nearing or crossing ventral midline, iatrogenic trauma of the urethra and rectum is a possible complication.2,3,15 However, no dogs or cats in the present study experienced any urethral or rectal complications. It remains important to protect these structures intraoperatively due to their proximity to osteotomies of the pubis and ischium; placement of urethral catheters, rectal syringe cases or tampons, and surgical retractors deep to the site of osteotomy transection can be utilized for these purposes to limit these complications.

Finally, regarding the subset of animals that underwent modified hemipelvectomy (the majority being total) without primary closure of muscular tissues or use of a local muscle flap for closure of the abdomen, these animals all experienced routine recoveries without any major complications or mobility concerns reported. No incidence of abdominal or perineal herniation, septic peritonitis, or major wound healing complications occurred. These findings support that in select cases, closure of subcutaneous tissue and skin alone or in conjunction with native omentum or mesh for a deep closure layer can be well tolerated without complication, and that use of local muscular tissues is not required for abdominal closure in every hemipelvectomy case.

The largest study on traditional hemipelvectomy in dogs and cats reported intraoperative complications in 8 of 100 (8.0%) animals and postoperative complications in 12 of 96 (12.5%) animals.3 If the CLASSIC and Accordion complication schemes are extrapolated to that data, intraoperative complications were grade 1 in 5 dogs, grade 2 in 2 dogs, and grade 3 in 1 dog, and postoperative complications were grade 1 in 10 dogs, grade 2 in 1 dog, and grade 3 in 1 dog. In the present modified hemipelvectomy cohort, intraoperative complications were reported in 3 of 23 (13.0%) cases and were all grade 1, and postoperative complications were reported in 8 of 23 (34.7%) animals, though only 2 of 23 (8.7%) animals had grade 3 or 4 postoperative complications. It is difficult to make direct comparisons between the complication incidence reported in the current cohort of animals undergoing modified hemipelvectomy procedures relative to that in the Bray et al study of animals undergoing traditional hemipelvectomy procedures for several reasons.3 First, the method of intra- and postoperative complication recording and grading was different between the 2 data sets; though we attempted to extrapolate the available data in the Bray et al study to the complication-grading schemes used in the present cohort for comparison, there is potential for error.3 Second, the present modified hemipelvectomy cohort may represent a different population compared with that in the Bray et al study due to the more extensive nature of local disease, resulting in the indication for these modified and more extensive hemipelvectomy techniques.3 If the complication comparisons are accurate and not associated with incomplete information in different retrospective studies, it would appear that the postoperative complication rate in the present cohort of animals undergoing modified hemipelvectomy techniques may be greater than that previously reported for traditional hemipelvectomy techniques. However, it is important to consider the potentially more extensive nature of local disease that necessitated modified hemipelvectomy techniques in the present modified hemipelvectomy cohort compared with the Bray et al traditional hemipelvectomy cohort.3 Ultimately, though, the reported incidence of major perioperative complications (grade 3 or 4) was relatively low (8.7%) in this modified hemipelvectomy cohort, though larger sample sizes and prospective data with standardized follow-up are needed to definitively determine the risk of perioperative complications relative to each of these modified hemipelvectomy techniques.

This study had several limitations. First, due to the retrospective nature, complete clinical information was lacking for some patients. Also, given the small sample size of animals in each subgroup of modified hemipelvectomy techniques, it was not possible to perform statistical analyses with regard to risk factors for complications or outcomes due to the risk for error. Five patients were lost to follow-up, and postoperative pelvic imaging was rarely available in these animals, such that the exact extent of excisions could not be determined in many cases. In addition, the population varied widely relative to neoplastic disease, management, and follow-up. Subsequently, prognostic information regarding survival times relative to surgical procedure cannot be ascertained, and specific outcome data may be attributed to a multitude of differences associated with patient and disease variables rather than modified hemipelvectomy technique. Instead, survival and follow-up data have been provided (Supplementary Table S1) to give information on duration of follow-up for these patients relative to their functional outcomes and complications. Finally, selection bias may have occurred because all cases were contributed by referral academic institutions, and patient and client factors likely influenced the types of treatments administered and follow-up data available.

In conclusion, this report represents the first documentation of dogs and cats undergoing modified hemipelvectomy with concurrent partial sacrectomy, partial vertebrectomy, external hemipelvectomy excisions crossing midline, and reconstruction techniques not utilizing muscular tissues or mesh for body wall closure. The outcomes of these animals lend support to use of these techniques in dogs and cats when indicated, and based on these cases, dogma regarding tolerable hemipelvectomy procedures and constraints should be reconsidered. Overall, these modified hemipelvectomy techniques appear to be well tolerated, with a low incidence of major complications, and can result in adequate functional outcomes. Additional studies with larger numbers of dogs and cats undergoing these modifications are needed to gain more information and to determine the tolerable extent of partial sacrectomy/vertebrectomy and contralateral osseous excision, as well as to explore scenarios in which closure with subcutaneous tissue and skin alone or with omentum may not be well tolerated.

Supplementary Materials

Supplementary materials are posted online at the journal website: avmajournals.avma.org

Acknowledgments

The authors received no grant funding in association with the cases described in this report. The authors declare that there were no conflicts of interest.

The authors thank Chrisoula Toupadakis Skouritakis for assistance with composing Figures 1 through 3 for this manuscript.

References

  • 1.

    Kramer A, Walsh PJ, Seguin B. Hemipelvectomy in dogs and cats: technique overview, variations, and description. Vet Surg. 2008;37(5):413-419. doi:10.1111/j.1532-950X.2008.00405.x

    • Search Google Scholar
    • Export Citation
  • 2.

    Bray JP. Hemipelvectomy: modified surgical technique and clinical experiences from a retrospective study. Vet Surg. 2014;43(1):19-26. doi:10.1111/j.1532-950X.2013.12085.x

    • Search Google Scholar
    • Export Citation
  • 3.

    Bray JP, Worley DR, Henderson RA, et al. Hemipelvectomy: outcome in 84 dogs and 16 cats. A veterinary society of surgical oncology retrospective study. Vet Surg. 2014;43(1):27-37. doi:10.1111/j.1532-950X.2013.12080.x

    • Search Google Scholar
    • Export Citation
  • 4.

    Barbur LA, Coleman KD, Schmiedt CW, Radlinsky MG. Description of the anatomy, surgical technique, and outcome of hemipelvectomy in 4 dogs and 5 cats. Vet Surg. 2015;44(5):613-626. doi:10.1111/vsu.12324

    • Search Google Scholar
    • Export Citation
  • 5.

    Ferreira PI, Cassanego GR, Basso PC, Müller DCM. Partial hemipelvectomy: an alternative for improving dogs’ life quality. Cienc Rural. 2021;52(5):e20200525. doi:10.1590/0103-8478cr20200525

    • Search Google Scholar
    • Export Citation
  • 6.

    Alexander J, Carb A. Subtotal hemipelvectomy in the dog. J Vet Orthop. 1979;1:9-14.

  • 7.

    Sharma S, Boston SE, Mosley C, Boylan M. Internal hemipelvectomy with ischiectomy, partial acetabulectomy, and femoral head and neck excision in a cat with a pelvic osteochondroma. J Am Vet Med Assoc. 2021;259(4):401-405. doi:10.2460/javma.259.4.401

    • Search Google Scholar
    • Export Citation
  • 8.

    Gordon C, Nakahara N, Thomson C, Mitchell R. Novel radical pelvectomy technique to treat chondrosarcoma in a large-breed dog. Aust Vet J. 2021;99(12):513-516. doi:10.1111/avj.13118

    • Search Google Scholar
    • Export Citation
  • 9.

    Endo Y, Sakai T, Fukui S, et al. Mid-to-caudal partial hemipelvectomy with limb preservation for ischial tumor in a dog. J Vet Med Sci. 2022;84(2):218-222. doi:10.1292/jvms.21-0414

    • Search Google Scholar
    • Export Citation
  • 10.

    Downey AC, Mathews KG, Borst L. Cranial internal hemipelvectomy (iliectomy) with limb sparing for a dog with ilial chondrosarcoma: a case report. Clin Case Rep. 2022;10(1):e05262. doi:10.1002/ccr3.5262

    • Search Google Scholar
    • Export Citation
  • 11.

    Gilman O, Doran I, Matiasovic M. Limb function-preserving ischiectomy for canine osteosarcoma. J Small Anim Pract. 2020;61(10):653. doi:10.1111/jsap.13201

    • Search Google Scholar
    • Export Citation
  • 12.

    Oramas A, Boston SE, Skinner OT. Iliectomy with limb preservation for a dog with ilial osteosarcoma: surgical description and case report. Vet Surg. 2020;49(3):607-613. doi:10.1111/vsu.13329

    • Search Google Scholar
    • Export Citation
  • 13.

    Sumner-Smith G. Gait analysis and orthopedic examination. In: Slatter D, ed. Textbook of Small Animal Surgery. 2nd ed. WB Saunders; 1993;1577-1586.

    • Search Google Scholar
    • Export Citation
  • 14.

    Follette CM, Giuffrida MA, Balsa IM, et al. A systematic review of criteria used to report complications in soft tissue and oncologic surgical clinical research studies in dogs and cats. Vet Surg. 2020;49(1):61-69. doi:10.1111/vsu.13279

    • Search Google Scholar
    • Export Citation
  • 15.

    Liptak J, Dernell WS, Farese J, Bray JP. Musculoskeletal tumors. In: Kudnig S, Seguin B, eds. Veterinary Surgical Oncology. 2nd ed. John Wiley & Sons; 2022:720-806. doi:10.1002/9781119089124.ch16

    • Search Google Scholar
    • Export Citation
  • Figure 1

    Modified hemipelvectomy with partial sacrectomy. In all images, the location of sacral osteotomy is denoted with a dashed red line, and the portion of excised sacrum has been faded. Given the variety of concurrent ipsilateral hemipelvectomy procedures, additional osteotomy locations are not demonstrated in these images. A—Excision of the sacral wing. B—Excision of the lateral third of the sacrum. C—Excision of slightly less than half the sacrum with the osteotomy just ipsilateral to midline. D—Excision of the spinous processes (median sacral crest) of the sacrum.

  • Figure 2

    Modified hemipelvectomy with partial vertebrectomy. In both images, the location of vertebral osteotomy is denoted with a dashed red line, and the portion of excised vertebra has been faded. Given the variety of concurrent ipsilateral hemipelvectomy procedures, additional osteotomy locations are not demonstrated in these images. A—Excision of the transverse processes of L6 (cranial osteotomy) and L7 (caudal osteotomy). B—Excision of the spinous process of L7.

  • Figure 3

    Modified hemipelvectomy with excision crossing midline. The location of contralateral pubic/ischial osteotomy was within the region demonstrated (1 [pink shading]) for cases in which the transection was within the medial third of the contralateral hemipelvis and within the region demonstrated (2 [blue shading]) for cases in which the transection was within the middle third of the contralateral hemipelvis. In all cases, hemipelvectomy was classified as partial and mid-to-caudal and associated with limb amputation (external hemipelvectomy), as demonstrated by the faded portion of pelvis and red dashed line depicting the additional osteotomy location.

  • 1.

    Kramer A, Walsh PJ, Seguin B. Hemipelvectomy in dogs and cats: technique overview, variations, and description. Vet Surg. 2008;37(5):413-419. doi:10.1111/j.1532-950X.2008.00405.x

    • Search Google Scholar
    • Export Citation
  • 2.

    Bray JP. Hemipelvectomy: modified surgical technique and clinical experiences from a retrospective study. Vet Surg. 2014;43(1):19-26. doi:10.1111/j.1532-950X.2013.12085.x

    • Search Google Scholar
    • Export Citation
  • 3.

    Bray JP, Worley DR, Henderson RA, et al. Hemipelvectomy: outcome in 84 dogs and 16 cats. A veterinary society of surgical oncology retrospective study. Vet Surg. 2014;43(1):27-37. doi:10.1111/j.1532-950X.2013.12080.x

    • Search Google Scholar
    • Export Citation
  • 4.

    Barbur LA, Coleman KD, Schmiedt CW, Radlinsky MG. Description of the anatomy, surgical technique, and outcome of hemipelvectomy in 4 dogs and 5 cats. Vet Surg. 2015;44(5):613-626. doi:10.1111/vsu.12324

    • Search Google Scholar
    • Export Citation
  • 5.

    Ferreira PI, Cassanego GR, Basso PC, Müller DCM. Partial hemipelvectomy: an alternative for improving dogs’ life quality. Cienc Rural. 2021;52(5):e20200525. doi:10.1590/0103-8478cr20200525

    • Search Google Scholar
    • Export Citation
  • 6.

    Alexander J, Carb A. Subtotal hemipelvectomy in the dog. J Vet Orthop. 1979;1:9-14.

  • 7.

    Sharma S, Boston SE, Mosley C, Boylan M. Internal hemipelvectomy with ischiectomy, partial acetabulectomy, and femoral head and neck excision in a cat with a pelvic osteochondroma. J Am Vet Med Assoc. 2021;259(4):401-405. doi:10.2460/javma.259.4.401

    • Search Google Scholar
    • Export Citation
  • 8.

    Gordon C, Nakahara N, Thomson C, Mitchell R. Novel radical pelvectomy technique to treat chondrosarcoma in a large-breed dog. Aust Vet J. 2021;99(12):513-516. doi:10.1111/avj.13118

    • Search Google Scholar
    • Export Citation
  • 9.

    Endo Y, Sakai T, Fukui S, et al. Mid-to-caudal partial hemipelvectomy with limb preservation for ischial tumor in a dog. J Vet Med Sci. 2022;84(2):218-222. doi:10.1292/jvms.21-0414

    • Search Google Scholar
    • Export Citation
  • 10.

    Downey AC, Mathews KG, Borst L. Cranial internal hemipelvectomy (iliectomy) with limb sparing for a dog with ilial chondrosarcoma: a case report. Clin Case Rep. 2022;10(1):e05262. doi:10.1002/ccr3.5262

    • Search Google Scholar
    • Export Citation
  • 11.

    Gilman O, Doran I, Matiasovic M. Limb function-preserving ischiectomy for canine osteosarcoma. J Small Anim Pract. 2020;61(10):653. doi:10.1111/jsap.13201

    • Search Google Scholar
    • Export Citation
  • 12.

    Oramas A, Boston SE, Skinner OT. Iliectomy with limb preservation for a dog with ilial osteosarcoma: surgical description and case report. Vet Surg. 2020;49(3):607-613. doi:10.1111/vsu.13329

    • Search Google Scholar
    • Export Citation
  • 13.

    Sumner-Smith G. Gait analysis and orthopedic examination. In: Slatter D, ed. Textbook of Small Animal Surgery. 2nd ed. WB Saunders; 1993;1577-1586.

    • Search Google Scholar
    • Export Citation
  • 14.

    Follette CM, Giuffrida MA, Balsa IM, et al. A systematic review of criteria used to report complications in soft tissue and oncologic surgical clinical research studies in dogs and cats. Vet Surg. 2020;49(1):61-69. doi:10.1111/vsu.13279

    • Search Google Scholar
    • Export Citation
  • 15.

    Liptak J, Dernell WS, Farese J, Bray JP. Musculoskeletal tumors. In: Kudnig S, Seguin B, eds. Veterinary Surgical Oncology. 2nd ed. John Wiley & Sons; 2022:720-806. doi:10.1002/9781119089124.ch16

    • Search Google Scholar
    • Export Citation

Advertisement