What Is the Evidence?

Samuel P. Franklin Comparative Orthopaedic Laboratory College of Veterinary Medicine, University of Missouri, Columbia, MO 65211.

Search for other papers by Samuel P. Franklin in
Current site
Google Scholar
James L. Cook Comparative Orthopaedic Laboratory College of Veterinary Medicine, University of Missouri, Columbia, MO 65211.

Search for other papers by James L. Cook in
Current site
Google Scholar

Click on author name to view affiliation information


A 6-year-old spayed female Newfoundland weighing 57.5 kg (126.5 lb) was evaluated at the University of Missouri Veterinary Medical Teaching Hospital for a 5-month history of right hind limb lameness manifested as a decrease in physical activity difficulty rising from a recumbent position, and disinterest in swimming, one of the dog's preferred activities before the lameness developed. The dog was a household pet that lived indoors and was allowed to roam freely on the owners' farm. The owners initially sought advice from their local veterinarian who had sequentially prescribed 2 types of NSAIDs; the owners reported minimal improvement with administration of either NSAID and so were referred to the teaching hospital for further evaluation of their dog.

General, orthopedic, and neurologic examinations were performed. The only pertinent abnormalities were detected during the orthopedic examination and included signs of pain in the dog during manipulation of the right elbow joint, extension of the right stifle joint, and extension of both hip joints.

The dog was sedated with nalbuphine hydrochloride (0.5 mg/kg [0.23 mg/lb]) and dexmedetomidine hydrochloride (5 μg/kg [2.3 μg/lb]) for radiographic examination. Mediolateral and craniocaudal radiographic views of the right elbow joint revealed changes consistent with a fragmented medial coronoid process and secondary osteoarthritis. Mediolateral and craniocaudal radiographic views of the right stifle joint showed no evidence of lesions. Lateral and ventrodorsal hip-extended radiographic views of the pelvis and hind limbs revealed changes indicative of severe osteoarthritis secondary to hip dysplasia, with more severe subluxation, remodeling, and osteophytosis noticed in the right versus left joint. The relative severity of radiographic changes was consistent with the owner's observation that the patient was more frequently lame on the right hind limb.

The diagnostic findings, options for treatment, associated complications, and prognoses were discussed with the owner. The complexity of the case given the multiple orthopedic abnormalities noted was discussed thoroughly, and the supervising clinician and the owner agreed that the first priority for surgical treatment would be the right hip joint as it seemed most clinically debilitating. The owner's objective was to pursue treatment that would alleviate the dog's discomfort and improve hind limb function defined as renewed interest in and ability to play and swim without reluctance or difficulty.

Formulation of the Clinical Question

The problems identified were pain and impaired ability to ambulate attributable to osteoarthritis secondary to hip dysplasia with concurrent osteoarthritis of the right elbow joint. The treatment options offered included continued nonsurgical management, right femoral head and neck excision arthroplasty (FHNE), or right total hip arthroplasty (THA). Because nonsurgical management had been attempted without success in meeting the owners' goals for their dog, surgical treatment of the right hip joint was believed to be indicated at this point.

Clinical Question

Is the function of large dogs with hip joint osteoarthritis treated by means of cemented or cementless THA superior to that of large dogs treated by means of FHNE, and is there a difference in owner satisfaction or complication rates between the 2 procedures?

Evidentiary Search Strategy

A targeted literature search was performed for the most recent and pertinent articles regarding FHNE and all commercially available prostheses for THA, including cemented and cementless systems. Sources identified through this search mechanism were supplemented with articles known to the authors or referenced in other manuscripts.

The PubMed electronic database of medical literature published between 1948 and January 2010 was searched with the following search term: “(femoral head ostectomy OR femoral head osteotomy OR femoral head and neck excision OR femoral head and neck resection OR femoral head excision) AND (total hip arthroplasty OR total hip replacement) AND dog.” This initial search identified only 1 report that included discussion of both procedures; it was a review of treatment options for hip dysplasia in dogs published in 1994. Two separate searches were then performed for each treatment type (eg, femoral head and neck excision and synonyms for 1 search and total hip arthroplasty or replacement for the second search). The search for reports on FHNE yielded 37 articles, and that for reports on THA yielded 90 articles. The articles were individually assessed for usefulness in addressing the clinical question on the basis of the title, a review of the abstract, or both.

After initial assessments, 5 manuscripts were identified as having potential relevance to the case scenario with regard to FHNE; 2 involved prospective controlled studies with objective outcome data, 1 involved a retrospective case series, 1 was a review article, and 1 was a case report. An additional prospective study was included that was not retrieved with the PubMed search because it was familiar to the authors and included objective force plate data. Of the reports pertaining to THA, 51 were potentially applicable to clinical veterinary patients and most involved single cases or case series detailing risk factors for intraoperative and postoperative complications. Importantly, 1 report involved an experimental prospective study in which kinetic data were used to objectively quantify outcome in dogs treated by means of THA. An additional prospective study that made use of a pressure-sensitive walkway to obtain objective outcome data was included because the authors were aware of its recent publication. Finally, an additional publication identified during the literature review was included because, although not identified with the PubMed search, it was the only report available of a cohort study in which FHNE and THA were compared.

A more exhaustive search using broader terms, different databases, and pursuit of all reports cited in the aforementioned publications would assuredly have yielded additional manuscripts pertinent to the clinical question. However, the search strategy used was believed the most likely type of search to be performed in a clinical situation to retrieve articles that would provide the current best evidence for practical and efficient clinical decision making.

Quality of the Evidence

The evidentiary value of retrieved articles was evaluated by use of a hierarchy pyramid based on type of study design. With regard to FHNE, 3 reports were deemed of low evidentiary value (the literature review, case report, and retrospective case series) for broad application to the clinical scenario. The 3 randomized controlled studies in which objective outcome measures were used for dogs with naturally occurring hip joint osteoarthritis and orthopedically normal dogs with experimental surgery were considered of higher evidentiary value. However, these studies were also suboptimal with respect to direct applicability to the clinical problem because they did not compare FHNE with THA.

Similarly, the literature search for reports on THA did not identify any studies that also included a cohort of dogs treated with FHNE. The reports considered to provide the greatest evidentiary value included 3 case series in which objective data were obtained via force plate or pressure-sensitive walkway from dogs with naturally occurring hip joint osteoarthritis or orthopedically normal dogs with experimental surgery, with the contralateral limb used as a control limb. Most reports on THA were of low evidentiary value because they consisted of literature reviews, case reports, and cases series without a comparative cohort or control subjects. The only cohort study in which FHNE and THA were compared was identified through cited literature.

Review of the Evidence

The available evidence that most directly addressed the clinical question consisted of 1 cohort study1 that involved force plate analysis to evaluate outcomes after unilateral THA, bilateral THA, and unilateral THA and contralateral FHNE in both orthopedically normal dogs and dogs with naturally occurring hip dysplasia.1 This study yielded variable results, with load bearing in THA limbs greater, equivalent to, or less than that in the contralateral unaffected limbs in various animals. Weight bearing was equivalent or greater in FHNE-treated limbs versus THA-treated limbs in dogs with both procedures performed, and there was equivalent load bearing between limbs with bilateral THA. The ideal aspect of this study was that it included the cohorts we sought to compare and involved an objective primary outcome measure. Unfortunately, the direct applicability of this evidence to our patient was limited in that it did not account for advances in THA over the last 33 years, had small sample sizes for each cohort (n = 4 or 5 dogs) without associated statistical comparison, lacked randomization, and lacked a comparison between dogs that underwent unilateral FHNE and unilateral THA.

The remaining evidence gathered suggested that limbs treated by means of FHNE in dogs weighing between 12 and 41 kg (26.4 and 90.2 lb) consistently have decreased hip joint range of motion, decreased muscle mass, lameness, and decreased load bearing, compared with observations in contralateral unaffected limbs or control dog limbs. Another study2 revealed that load bearing in FHNE-treated limbs was either not different than or significantly worse than preoperative values in the same limb, depending upon whether an oral NSAID was administered in the postoperative period. The purported benefits of FHNE that are supported by the available evidence include an owner-perceived relief of pain in most affected dogs and a low relative incidence and severity of complications. The value of these potential benefits is supported by reported owner satisfaction rates of 93% to 100%.

For unilateral THA, results of a prospective case series3 suggest that load bearing in the THA-treated limb improves after surgery to achieve near-healthy or healthy load bearing by 6 months after surgery in successfully treated dogs. Success rates, as determined by resolution or lack of complications and satisfactory outcome in the owner's estimation, range from 84% to 97%. However, several complications of THA have been reported, including unacceptable limb function, luxation, neoplasia, implant failure, infection, aseptic loosening of prostheses, pulmonary embolism, granuloma formation, femoral infarction, sciatic neuropraxia, and femur fracture. Although complication rates for dogs undergoing cemented THA have been reported to be as low as 3.8%, more recent retrospective case series4,5 have revealed complication rates as high as 12% to 17%.

Given the aforementioned evidence, what decision would you make?

Clinical Decision and Outcome

A summary of the aforementioned findings supplemented with success and complication rates for FHNE and THA at our hospital as well as the associated financial costs for each procedure were presented to the owners. The corresponding recommendation from the supervising clinician was to pursue THA if the owners were not limited by the financial costs, maximal postoperative load bearing function was a priority, and an approximately 10% to 20% chance for complications requiring additional surgery was acceptable. Femoral head and neck excision arthroplasty was recommended as the preferred treatment if minimizing risk of surgical complications and financial costs was a high priority and if it was understood that full limb function would not be restored. The owners subsequently elected for the dog to undergo THA.

After a preanesthetic CBC, serum biochemical analysis, and urinalysis failed to reveal any clinically relevant abnormalities, the dog was anesthetized and THA was performed through a craniolateral approach to the hip joint. A commercially available, cementless THA systema was used to implant a 28-mm acetabular prosthesis and No. 9 femoral endoprosthesis with a 17-mm femoral head (+6 mm offset). Intraoperative implant positioning, joint stability, and range of motion were judged as satisfactory. Postoperative radiographic views of the pelvis and hind limbs revealed appropriate implant location, orientation, and reduction.

The dog was monitored and treated with analgesics administered IV in the intensive care unit for approximately 16 hours after surgery and then transferred to the surgical ward the day after surgery. Two days after surgery, the dog was discharged from the hospital with owner instructions detailing incision monitoring and activity restriction. The owners were advised to confine the dog when unobserved during the first 8 weeks after surgery and otherwise restrict activity to short leash walks for urination and defecation purposes only. The dog bore weight on the THA-treated limb with no signs of discomfort or surgery-related problems at the time of discharge from the hospital.

Three days after discharge from the hospital, the dog escaped from the house and ran down the steps and into the yard and became acutely lame in its right hind limb. The dog was reexamined at the teaching hospital, and radiographs of the right hind limb were obtained, revealing craniodorsal luxation of the femoral prosthesis. Revision surgery was performed the following day. The degree of ventroversion of the acetabular cup was slightly increased by use of the cup positioner handle with the offset cup positioner head; the femoral prosthesis was then reduced. Two antirotational sutures comprised of 80-lb test nylon monofilament were then placed to prevent excessive external rotation. One suture was anchored through the ilium cranioventral to the acetabulum by drilling a bone tunnel and passing a 3.2-mm toggle to the medial side of the ilium. The second antirotational suture was anchored craniodorsal to the acetabulum with a bone anchor. Both were secured to the greater trochanter through a bone tunnel. The dog recovered without complications and was discharged from the hospital 2 days later with the same instructions for postoperative care as previously given.

Two months afterward, the dog was returned for a reevaluation. Physical and radiographic examinations confirmed functional use of the THA-treated limb with no apparent lameness or gait alteration and appropriate healing and positioning of the implants. Instructions were given to progressively return the dog to full, unrestricted function over the subsequent 6 weeks. The dog was reevaluated at the teaching hospital 14 months after surgery for reasons unrelated to orthopedic disease. At that time, it had returned to full-intended function and radiographic examination confirmed adequate positioning of the implants without migration or evidence of loosening. The owners reported complete satisfaction with the outcome after THA in this dog.


Although the current best evidence for clinical decision making for surgical treatment of osteoarthritis secondary to hip dysplasia in the dog of this report included several reports of relatively high evidentiary value, none directly addressed the clinical question by head-to-head comparison of well-delineated cohorts for the 2 procedures being considered (FHNE and THA). In situations such as the one described here, the clinician is restricted to use of indirect comparisons among studies or to historical data, extrapolation from available evidence, and personal experiences and opinions. When this occurs, it is ethical to use the evidence that the clinician considers most relevant to make decisions in the best interest of patient and owner as opposed to simply doing nothing because of a lack of high-quality evidence. However, communications with and recommendations made to owners must be honest and accurate. When evidence is lacking, then statements suggesting superiority of one procedure over another should not be made. The basis for the recommendation, whether derived from review of multiple highest-level-of-evidence, peer-reviewed articles or the clinician's clinical experiences with a few cases, should be fully disclosed so that appropriate information is communicated to owners.

In the scenario reported here, the available evidence suggested that THA may be associated with a higher degree of function in treated limbs and FHNE may be associated with fewer and less severe complications but also that owner satisfaction rates are similar between these 2 surgical approaches. The similar owner satisfaction rates may be the result of the methods of data collection, owner and practitioner or surgeon biases, a true lack of a difference, or different expectations from owners in each group. Because reports of head-to-head cohort studies with predefined criteria for determining owner satisfaction were not available for the present scenario, the validity of comparing owner satisfaction rates from different studies cannot be ascertained. Unfortunately, until these types of studies are completed, owners cannot truly perform a valid cost-benefit analysis and make a fully informed decision regarding their pet's surgery. The recommendation of the supervising clinician, hopefully based on the peer-reviewed literature and personal experience, will continue to substantially influence owner decisions. As such, detailed information regarding hospital-specific complication rates and types as well as long-term patient outcomes should be determined, recorded, and provided to owners.


BFX, Biomedtrix, Boonton, NJ.


  • 1.

    Dueland RBartel DLAntonson E. Force-plate technique for canine gait analysis of total hip and excision arthroplasty. J Am Anim Hosp Assoc 1977; 13: 547552.

    • Search Google Scholar
    • Export Citation
  • 2.

    Charette BDupuis JMoreau M, et al. Assessing the efficacy of long-term administration of tolfenamic acid in dogs undergoing femoral head and neck excision. Vet Comp Orthop Traumatol 2003; 16: 232237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3.

    Budsberg SCChambers JNLue SL, et al. Prospective evaluation of ground reaction forces in dogs undergoing unilateral total hip replacement. Am J Vet Res 1996; 57: 17811785.

    • Search Google Scholar
    • Export Citation
  • 4.

    Bergh MSGilley RSShofer FS, et al. Complications and radio-graphic findings following cemented total hip replacement: a retrospective evaluation of 97 dogs. Vet Comp Orthop Traumatol 2006; 19: 172179.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5.

    Guerrero TGMontavon PM. Zurich cementless total hip replacement: retrospective evaluation of 2nd generation implants in 60 dogs. Vet Surg 2009; 38: 7080.

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 79 0 0
Full Text Views 741 703 74
PDF Downloads 92 57 6