Radiographic Assessment of a Canine Uncemented Porous-Coated Anatomic Total Hip Prosthesis

Thirty uncemented porous-coated anatomic total hip prostheses in 24 dogs were evaluated radiographically at 6 months, 12 months, or 18 months after arthroplasty. All dogs were fully weight bearing and free of lameness at the time of follow-up. Radiographs were assessed for prosthetic component alignment, percentage of femoral canal fill, resorptive and formative bony changes, and changes at the implant bone interface. No evidence of infection, loosening, or failure of component fixation was seen. Settling of the acetabular component and subsidence of the femoral stem was seen early after surgery. Cortical atrophy was seen in all femurs and was the most significant bony remodeling change. Early results of clinical and radiographic evaluation of uncemented total hip arthroplasty in dogs have been encouraging.     

Extraosseous cement granuloma associated with total hip replacement in 6 dogs

OBJECTIVE: To report the clinical, radiographic, and pathologic features of extraosseous cement granuloma (ECG), a low occurrence, long-term complication of total hip replacement (THR). STUDY DESIGN: Retrospective clinical study. Sample Population-Six client-owned dogs. METHODS: Medical records for 6 dogs that had a cemented modular THR and developed ECG were reviewed for the clinical, radiologic, and histopathologic findings of ECG. Morphologic and elemental analyses of retrieved particles were conducted in 2 dogs using scanning electron microscopy (SEM) and energy-dispersive x-ray spectroscopy (EDS). RESULTS: Three golden retrievers and 3 Labrador retrievers had progressive lameness in the operated limb 4 to 7 years after THR and developed ECG. Other clinical features included an acute deterioration to non-weight bearing associated with pathologic fracture of the proximal femur in 2 dogs, and obstipation in 1 dog. In 5 dogs, there was a large caseous mass surrounding the proximal femur. Obstipation in 1 dog was caused by a similar mass adjacent to bone cement used for fixation of the acetabular cup. Radiographic changes included a multilobular soft tissue mass, containing irregular mineral densities, in the gluteal mass surrounding the proximal femur, THR interface deterioration, osteolysis, new bone formation, and implant subsidence were consistent with chronic aseptic loosening. The histopathologic diagnosis was sterile granuloma associated with a massive heterogeneous particulate burden, which was assumed to have resulted from severe wear of all prosthesis components. CONCLUSIONS: ECG is a manifestation of severe aseptic loosening of cemented total hip replacement, with the generation of wear debris probably responsible for granuloma formation. CLINICAL RELEVANCE: Because revision can offer a good prognosis in early aseptic femoral loosening, yearly radiographic monitoring of the recipients is recommended.     

Short-term evaluation of dorsal acetabular augmentation in 10 canine total hip replacements

OBJECTIVE: To report the use of dorsal acetabular augmentation (DAA) in canine total hip arthroplasty (THA) and to evaluate the clinical and radiographic outcome after a minimum of 1 year follow-up. STUDY DESIGN: Clinical study. ANIMALS: Nine dogs that had THA. METHODS: Ten hips requiring acetabular augmentation for optimum acetabular cup implantation for THA were evaluated retrospectively. The excised femoral head and neck was used as the corticocancellous bone graft for index THA surgeries, and the dorsal crest of the ipsilateral ilial wing was used as the bone graft in 1 dog that had a revision surgery. Clinical and radiographic follow-up examinations were performed at approximately 8 weeks and 1 year postoperatively. Long-term radiographic examinations included use of a standing dorsal acetabular rim (DAR) projection for assessment of graft-recipient boundary. Functional clinical outcome was assessed using direct patient evaluation and telephone interview. RESULTS: One dog failed to regain limb function after surgery. Suspected aseptic loosening of the acetabular cup was observed on radiographs approximately 12 weeks postoperatively. Excluding this single explantation, follow-up physical and radiographic examinations indicated good implant and graft positioning and normal function for all dogs. CONCLUSIONS: Acetabular augmentation permitted implantation of an appropriately sized acetabular cup while maintaining complete DAR coverage of the prosthesis. Nine out of 10 hips had a successful outcome with minimal radiographic and no functional abnormalities. Further follow-up is necessary to determine the long-term outcome of DAR augmentation in canine THA. CLINICAL RELEVANCE: Acetabular augmentation with an autogenous corticocancellous bone graft constructed from the femoral head and neck can provide complete DAR coverage to permit use of an appropriately sized acetabular cup in THA.     

Femoral bone adaptation to unstable long-term cemented total hip arthroplasty in dogs

OBJECTIVE: To evaluate femoral adaptation after unstable long-term cemented total hip arthroplasty (cTHA) in dogs. STUDY DESIGN: Clinical study. ANIMALS: Four dogs. METHODS: Paired femurs were examined from client-owned dogs that were donated to a retrieval program after death from causes unrelated to their cTHA. Mean (+/-SEM) dog age was 10.0+/-1.5 years and implant duration was 6.0+/-1.4 years. Implanted femurs had radiographic changes associated with implant loosening and gross mechanical instability at femur retrieval. Femurs were evaluated at 3 levels relative to implant length. Cortical area and medullary area were measured at each level, and cortical porosity was assessed at each level in 4 quadrants and in 3 regions. Implanted femurs were compared with the contralateral non-implanted femurs. RESULTS: Cortical area and cortical porosity were increased at all levels in femurs with unstable implants. Implanted femurs had increased porosity in all quadrants and regions at the proximal 2 levels, and increased porosity in only the cranial quadrant and mid-cortical area at the distal level, when compared with contralateral non-implanted femurs. Corresponding medullary areas were not different. CONCLUSIONS: Significant histomorphometric changes occur in femurs after unstable cTHA. The patterns of periosteal bone formation and endosteal bone resorption support mechanisms of stress shielding and wear debris-mediated osteolysis as factors that may contribute to femoral adaptation and implant loosening. CLINICAL RELEVANCE: Despite popularity and excellent return to function with cTHA, aseptic loosening remains a serious long-term complication. Substantial net bone loss and unfavorable environment with unstable cTHA may make revision surgery less successful.     

Modular hybrid total hip arthroplasty. Experimental study in dogs

Background

This prospective experimental study evaluated the surgical procedure and results of modular hybrid total hip arthroplasty in dogs.

Methods

Ten skeletally mature healthy mongrel dogs with weights varying between 19 and 27 kg were used. Cemented modular femoral stems and uncemented porous-coated acetabular cups were employed. Clinical and radiographic evaluations were performed before surgery and at 30, 60, 90, 120, 180 and 360 days post-operation.

Results

Excellent weight bearing was noticed in the operated limb in seven dogs. Dislocation followed by loosening of the prosthesis was noticed in two dogs, which were therefore properly treated with a femoral head osteotomy. Femoral fracture occurred in one dog, which was promptly treated with full implant removal and femoral osteosynthesis.

Conclusions

The canine modular hybrid total hip arthroplasty provided excellent functionality of the operated limb.     

Canine uncemented porous-coated anatomic total hip arthroplasty: results of a long-term prospective evaluation of 50 consecutive cases

OBJECTIVE: To evaluate the long-term clinical and radiographic results of a canine uncemented porous-coated anatomic (PCA) total hip arthroplasty (THA). STUDY DESIGN: Prospective study of consecutive clinical patients using survival analysis. ANIMALS: Forty-one dogs that underwent PCA THA; nine had bilateral PCA THA (50 prostheses). METHODS: Gait observation, orthopedic examination, and radiographic assessment were conducted before THA, 6 months after THA, and yearly thereafter. A zonal analysis system was used to document osseous changes in the femur and the acetabulum. Acetabular cup and femoral stem subsidence and migration, femoral canal fill, and implant orientation were measured. Survival analysis of the procedure was conducted. RESULTS: Long-term follow-up was available for 37 dogs (46 prostheses). The median follow-up was 63 months. Limb function was normal for 37 limbs and abnormal for 9 limbs because of dislocation (n = 3), lumbosacral disease (n = 2), degenerative myelopathy (n = 1), autoimmune disease (n = 1), brain tumor (n = 1), or osteosarcoma of the femur (n = 1). All prosthetic stems and cups were fixed by bone ingrowth fixation. Osteolysis was not observed. Bone infarction occurred in five femoral canals (four dogs). The 6-year survival rate for the procedure was 87% (95% confidence interval, 72%-96%). CONCLUSIONS: Long-term fixation of the uncemented PCA acetabular cup and stem is successful in dogs, and long-term clinical function is excellent.     

Femoral Medullary Infarction Prevalence with the Zurich Cementless Canine Total Hip Arthroplasty

Objective— To document the prevalence of femoral medullary infarction associated with the Zurich Cementless Total Hip Replacement (ZCTHR) system in dogs.
Study Design— Case series.
Animals— Dogs (n=35) with 41 ZCTHR implants.
Methods— Medical records (February 1999–December 2002) were reviewed for dogs that had ZCTHR and at least 1 year follow-up with radiographic evaluation. Thirty-five dogs (41 ZCTHR) met the inclusion criteria. Femoral morphologic data, implant to bone relationships, and medullary infarcts were recorded. Data were analyzed for associations between infarct occurrence and morphologic details and dog characteristics.
Results— Eight of 41 femurs had radiographic evidence of infarcts (19.5%). Dogs with infarcts were significantly younger (mean [±SD] age, 18.5±5.2 months) compared with those without infarction (mean age, 44.4±5.6months; P =.027). None of the other variables were significantly different between dogs with and without infarcts. Three femurs with infarcts ultimately developed stem loosening.
Conclusion— A higher prevalence of femoral medullary infarcts was identified with ZCTHR compared with incidence reported for other total hip systems. Younger dogs were more likely to develop infarction.
Clinical Relevance— Femoral infarction appears to be associated with stem loosening. Adjustments in surgical technique or delaying surgery beyond 18 months of age may reduce incidence of infarction but needs further evaluation.     

Revision of Cemented Total Hip Arthroplasty with Cementless Components in Three Dogs

Objective— To report revision of cemented total hip arthroplasty failure with cementless components in 3 dogs.
Study Design— Clinical case reports.
Animals— Dogs with total hip arthroplasty failure (n=3).
Methods— Cementless total hip arthroplasty revision was performed in 3 dogs with previously cemented femoral and acetabular components. All dogs required revision of the femoral component and 1 dog also required revision of the acetabular component.
Results— Revisions resulted in a stable functional prosthesis with successful bone integration.
Conclusions— Use of cementless components may be a viable option for revision of loosened cemented prosthesis after explantation of failed cemented canine total hip replacements.
Clinical Relevance— Failed canine-cemented total hip arthroplasties can be successfully revised with cementless components.     

Limb-sparing surgery in a dog with osteosarcoma of the proximal femur

OBJECTIVE: To report successful limb-sparing surgery in a dog with a proximal femoral osteosarcoma (OSA) using a composite allograft-prosthetic technique. STUDY DESIGN: Case report. ANIMAL: Client-owned dog. METHODS: A stage IIB OSA of the proximal aspect of the femur was resected in accordance with oncologic and limb-sparing principles. The osseous defect was reconstructed with a proximal femoral allograft and cemented, long-stemmed femoral prosthesis. Soft tissue reconstruction was achieved by suturing host tendons to their respective allogeneic tendons on the allograft. Coxofemoral joint function was preserved using standard total hip arthroplasty techniques. RESULTS: Limb-sparing surgery of the proximal aspect of the femur using a composite allograft-prosthetic technique resulted in excellent limb function. Postoperative complications included aseptic loosening of the femoral composite graft and allograft nonunion, which required revision, traumatic implant luxation, and local tumor recurrence. Limb function was excellent after surgical stabilization of the allograft nonunion but deteriorated after implant luxation 270 days postlimb-sparing surgery. Pulmonary and skeletal metastases were diagnosed and local tumor recurrence suspected 596 and 650 days postoperatively, respectively. The dog was euthanatized 688 days after limb-sparing surgery as a result of progressive local and metastatic disease. CONCLUSIONS AND CLINICAL RELEVANCE: Limb-sparing surgery for dogs with primary bone tumors of the proximal aspect of the femur is feasible with good functional results.     

Zurich Cementless Total Hip Replacement: Retrospective Evaluation of 2nd Generation Implants in 60 Dogs

Objective— To evaluate the results of application, and identify complications, of the 2nd generation of Zurich Cementless Total Hip Replacement (ZCTHR).
Study Design— Case series.
Animals— Client-owned dogs (n=60) that had ZCTHR (n=65).
Methods— Dogs with ZCTHR (2001–2003) with a minimum follow-up ≥6 months were evaluated. Data included signalment, cup position, longest follow-up, complications, management of complications and outcome.
Results— Mean follow-up was 22.68 months. Eleven cases (17%) had postoperative complications: femoral fracture (n=1; 1.5%), prosthesis luxation (7; 11%), cup loosening (2; 3%), and implant failure (1; 1.5%); 9 cases were successfully revised. Explantation of implants was performed in 1 case because of infection, and 1 dog was euthanatized after reluxation.
Conclusions— ZCTHR can restore function in dogs affected by disabling diseases of the coxofemoral joint. The press-fit fixation of the cup allowed for corrections in cases of incorrect positioning. Cases with aseptic loosening were revised successfully by impacting larger cups. Newer stems of this generation are shot peening treated to increase their resistance to breakage. In our cases, infection is a disastrous event, leading to implant removal. After resolution of complications, a successful final outcome was achieved in 97% of THR.
Clinical Relevance— ZCTHR offers a reliable alternative for treating dogs with disabling diseases of the hip joints.     

Radiographic, Densitometric, and Biomechanical Effects of Recombinant Canine Somatotropin in an Unstable Ostectomy Gap Model of Bone Healing in Dogs

Objective —To determine the effect of recombinant canine somatotropin (STH) on radiographic, densitometric, and biomechanical aspects of bone healing using an unstable ostectomy gap model.
Study Design —After an ostectomy of the midshaft radius, bone healing was evaluated over an 8–week period in control dogs (n = 4) and dogs receiving recombinant canine STH (n = 4).
Animals or Sample Population—Eight sexually intact female Beagle dogs, 4 to 5 years old. Methods—Bone healing was evaluated by qualitative and quantitative evaluation of serial radiographs every 2 weeks. Terminal dual-energy x-ray absorptiometry and three-point bending biomechanical testing were also performed.
Results —Dogs receiving STH had more advanced radiographic healing of ostectomy sites. Bone area, bone mineral content, and bone density were two to five times greater at the ostectomy sites of treated dogs. Ultimate load at failure and stiffness were three and five times greater in dogs receiving STH.
Conclusions —Using the ostectomy gap model, recombinant canine STH enhanced the radiographic, densitometric, and biomechanical aspects of bone healing in dogs. Clinical Relevance—Dogs at risk for delayed healing of fractures may benefit from treatment with recombinant canine STH.     

Cemented Total Knee Replacement in 24 Dogs: Surgical Technique, Clinical Results, and Complications

Objective— To characterize the performance of cemented total knee replacement (TKR) in dogs.
Study Design— Preclinical research study.
Animals— Skeletally mature, male Hounds (25–30 kg; n=24) with no preexisting joint pathology.
Methods— Dogs had unilateral cemented TKR and were evaluated at 6, 12, 26, or 52 weeks (6 dogs/time point) by radiography, bone density analysis, visual gait assessment, and direct measurement of thigh circumference and stifle joint range of motion as indicators of functional recovery. At study end, the stability of the cemented tibial component was determined by destructive mechanical testing.
Results— Joint stability was excellent in 16 dogs (67%) and good in 8 dogs. None of the tibial components had evidence of migration or periprosthetic osteolysis whereas 1 femoral component was loose at 52 weeks. There was an early and significant decrease in tibial bone density, likely because of disuse of the operated limb. Dogs returned to full activity by 12 weeks. The tibial cement–bone interface maintained its strength over 52 weeks.
Conclusions— Cement provides stable fixation of the tibial component in canine TKR.
Clinical Relevance— Cemented TKR yields adequate clinical function and stifle joint excursion in the dog. Clinical studies are needed to determine the long-term fate of cemented TKR implants, to assess the influence of implant design on implant fixation and wear, and to obtain objective functional data.     

Short-term aseptic loosening of the femoral component in canine total hip replacement: effects of cementing technique on cement mantle grade

OBJECTIVES: To evaluate the effects of different cementing techniques on radiographic cement mantle grade and short-term aseptic loosening of the femoral component in canine total hip replacement (THR). STUDY DESIGN: Retrospective study. SAMPLE POPULATION: Radiographs of 284 dogs that had THR. METHODS: Immediate postoperative radiographs of 284 dogs that had cemented THR were reviewed by 4 surgeons and 1 radiologist and assigned a cement mantle grade using a 4-grade classification system. Dog age and weight at surgery, cementing technique (1st, 2nd, 3rd generation), complications (type and timing), and follow-up time were retrieved and analyzed. RESULTS: Mean cement mantle score for 2nd generation technique was significantly higher than 3rd generation technique and both were significantly higher than 1st generation technique (P<.001). Aseptic loosening was the 2nd most common complication observed with an incidence of 2.1%. Mean time from surgery to last follow-up examination was 122 days. No statistically significant difference in incidence of aseptic loosening was identified among different cement mantle grades or cementing techniques. CONCLUSIONS: Advanced cementing techniques resulted in better cement mantles based on grading of immediate postoperative radiographs, however grading did not predict short-term aseptic loosening. Cementing technique seemingly did not affect the incidence of short-term aseptic loosening of the femoral component for dogs in this study. CLINICAL RELEVANCE: Our study suggests that advances in cementing technique may result in improvements in the radiographic grade of cement mantles. With respect to aseptic loosening of the femoral component, our data only suggest that short-term (3 months-3 years) loosening cannot be predicted by immediate postoperative radiographic evaluation of cement mantle.     

Implantation of an Uncemented Total Hip Prosthesis Technique and Initial Results of 100 Arthroplasties

A porous-coated modular total hip system was developed for uncemented implantation in dogs. The operative technique was developed in cadaver bones and live animals. One hundred uncemented total hip arthroplasties were performed in 92 dogs. Results were successful in 98% of the joints during follow-ups of 3 months or more. Complications included three luxations of the prosthetic joint, two fissure fractures of the femoral cortex, and one complete displacement of the acetabular component from its bony bed. Four complications were resolved successfully.     

Measurement of the elastic properties of the cancellous bone in the femoral head of the dog

The degenerative wear and pathologic damage of the joints are reasons for total endoprotheses in man as well as in dogs. The main problem is the aseptic loosening of the protheses. By usig the finite-element-method, the total endoprothesis is designed with new features, with the purpose of preventing loosening and being better adapted to load transmission. In order to simulate the femur of the dog for the numerical analysis, a material law is developed. By taking into account the anisotropy and the local density of the cancellous bone in the femoral head, the young's modules are experimentally determined. The measurements are performed by ultrasonic methods on femoral heads of euthanised dogs. The results show planar isotropic cancellous bone.     

Hinged Ilizarov External Fixation for Correction of Antebrachial Deformities

Objective —To evaluate hinged circular external fixation for correction of antebrachial deformities in dogs.
Study Design —Uncontrolled clinical trial.
Animal Population —Seven client-owned dogs.
Methods —Six dogs had one radius corrected and one dog had both radii corrected. Preoperative planning included measurement of the craniocaudal and mediolateral angular deformities, rotational deformity, length deficit, origin of deformity, graphical or mathematical determination of the amplitude and direction of the actual limb deformity, and frame assembly.
Results —Preoperatively, function and cosmesis were assessed to be fair to poor in all dogs. Deformity correction started 48 to 60 hours postoperatively and ranged from 0.46 mm to 1.36 mm twice daily. Hospitalization time ranged from 4 to 6 days. Corrections were mostly made by the owners, at home. Lengthening and angular correction ranged from 3 to 38 mm and 18° to 48°. Mean residual deformities were 2.7% of radial length and 2.7°. The time duration with the circular external fixators in place ranged from 29 to 71 days. Two additional surgeries were necessary in one dog because of wire breakage. Mean follow-up was 40 months. Long-term function and cosmesis were good to excellent in all dogs.
Conclusion—Although complications were present in six of seven dogs, the outcome of hinged Ilizarov external fixation was successful in all dogs treated for deformities of the antebrachium.
Clinical Relevance —Despite complex preoperative planning, the placement of hinged circular external fixators is straightforward, and allows precise correction of complex antebrachial deformities with minimal tissue trauma.     

Transcutaneous tibial implants: a surgical procedure for restoring ambulation after amputation of the distal aspect of the tibia in a dog

Objective— To report surgical planning, technique, and long-term outcome of custom transcutaneous tibial implants used to restore ambulation after bilateral pelvic limb amputation in a dog.
Study Design— Case report.
Animals— A 4-year-old, 25.5 kg, female spayed, Siberian Husky.
Methods— Computed tomographic scans of the pelvic limbs were used to build sterolithographic models of the remaining tibia after bilateral amputation of the distal aspect of the tibiae. The sterolithographic models facilitated fabrication of implants that would replace the missing distal segments of the tibiae. Custom implants were surgically placed in both limbs.
Results— Assisted ambulation was restored immediately postoperatively and unassisted locomotion occurred at 7 days. At 14 months, aseptic loosening of the right implant occurred, it was removed and a 2nd transcutaneous implant was fabricated and implanted. At 26 months after initial surgery and 17 months after revision of the right implant, function is restored at a walk, trot, and run.
Conclusion— Based on over a 2-year follow-up, transcutaneous tibial implants allowed for restoration of locomotion.
Clinical Relevance— Transcutaneous tibial implants offer a potentially viable treatment option for restoring ambulation after amputation of the distal aspect of the tibiae in the dog.     

Femoral medullary infarction secondary to canine total hip arthroplasty

OBJECTIVE: To evaluate the prevalence of femoral intramedullary infarction after total hip arthroplasty (THA) and to determine whether any specific femoral morphology predisposes to bone infarction. STUDY DESIGN: Retrospective clinical study. SAMPLE POPULATION: All dogs from our hospital population undergoing THA between 1984 and 1997 with radiographic follow-up available at 1 year or more postoperatively. METHODS: A case control study was conducted within the THA group to determine risk factors predisposing to femoral infarction after THA. Medical records and radiographs were reviewed. Data were collected on clinical parameters, femoral morphology, prosthesis, and bone changes. Radiographic diagnosis was confirmed using histopathology in 11 femora. Radiographs of 50 age-matched control dogs weighing more than 20 kg with coxofemoral degenerative joint disease were randomly chosen to determine the prevalence of bone infarction in nonoperated dogs. RESULTS: Ninety-one dogs with 110 THA were included in the study. Fifteen of the 110 femora with THA had radiographic evidence of infarction (14%). Infarction was not present in any femora in the control group. There was no significant difference in the prevalence of infarction between dogs that received cemented or uncemented prostheses. Clinical signs were not reported in any patient that developed femoral infarction. Young age (P = .03) and a distance between the greater trochanter and nutrient foramen greater than 79 mm (P = .008) predisposed dogs to femoral infarction. Over time, three infarcts decreased in size radiographically, five remained unchanged, and three expanded. An osteosarcoma developed at the site of a bone infarct in one dog. CONCLUSION: Femoral intramedullary infarction occurred in 15 of 110 THA. Young age at the time of THA and a greater distance between the greater trochanter and the nutrient foramen predisposed to infarction. CLINICAL RELEVANCE: Intramedullary infarction occurs after canine THA. These bone infarcts do not appear to cause clinical signs; however, they may present a diagnostic challenge. Malignant transformation could potentially result from medullary infarction.     

Femoral bone adaptation to stable long-term cemented total hip arthroplasty in dogs

OBJECTIVE: To quantify long-term bone adaptation after stable cemented total hip arthroplasty (cTHA) in dogs. STUDY DESIGN: Clinical study. ANIMALS: Fourteen dogs. METHODS: Femoral specimens were collected from client-owned dogs that were donated after death because of causes unrelated to their cTHA. Mean (+/-SEM) dog age was 11.4+/-0.7 years and implant duration was 5.3+/-0.7 years. Implant stability was established from radiographic signs and gross mechanical stability. Femurs were evaluated at 3 levels based on implant length: proximal stem (PS), mid-stem (MS), and distal to stem (DS). Cortical area, medullary area, and porosity were measured at each level. Implanted femurs were compared to contralateral nonimplanted femurs. RESULTS: Cortical area and cortical porosity were significantly increased in implanted femurs compared to nonimplanted femurs. Cortical area was increased at the MS and DS levels, and porosity was increased at the PS and MS levels in implanted femurs. Porosity was greatest in the endosteal region at the PS and MS levels in implanted femurs. CONCLUSIONS: Significant differences in femoral geometry and cortical porosity were detected after long-term stable cTHA. Net bone loss proximally and increased bone mass distally support stress shielding as a important mechanical factor associated with bone adaptation. Distribution of porosity shifts to endosteal regions after long-term cTHA. CLINICAL RELEVANCE: Significant site-specific femoral adaptation occurs in response to stable cTHA and may precede implant loosening.