Advertisement for orthosearch.org.uk
Results 1 - 20 of 1507
Results per page:
Bone & Joint Open
Vol. 5, Issue 8 | Pages 688 - 696
22 Aug 2024
Hanusrichter Y Gebert C Steinbeck M Dudda M Hardes J Frieler S Jeys LM Wessling M

Aims

Custom-made partial pelvis replacements (PPRs) are increasingly used in the reconstruction of large acetabular defects and have mainly been designed using a triflange approach, requiring extensive soft-tissue dissection. The monoflange design, where primary intramedullary fixation within the ilium combined with a monoflange for rotational stability, was anticipated to overcome this obstacle. The aim of this study was to evaluate the design with regard to functional outcome, complications, and acetabular reconstruction.

Methods

Between 2014 and 2023, 79 patients with a mean follow-up of 33 months (SD 22; 9 to 103) were included. Functional outcome was measured using the Harris Hip Score and EuroQol five-dimension questionnaire (EQ-5D). PPR revisions were defined as an endpoint, and subgroups were analyzed to determine risk factors.


Orthopaedic Proceedings
Vol. 106-B, Issue SUPP_16 | Pages 19 - 19
19 Aug 2024
Macheras G Kostakos T Tzefronis D
Full Access

Total hip arthroplasty (THA) for congenital hip dysplasia (CDH) presents a challenge. In high-grade CDH, key surgical targets include cup placement in the anatomical position and leg length equality. Lengthening of more than 4 cm is associated with sciatic nerve injury, therefore shortening osteotomies are necessary. We present our experience of different shortening osteotomies including advantages and disadvantages of each technique. 89 hips, in 61 pts (28 bilateral cases), for high CDH were performed by a single surgeon from 1997 to 2022. 67 patients were female and 22 were male. Age ranged from 38 to 68 yrs. In all patients 5–8cm of leg length discrepancy (LLD) was present, requiring shortening femoral osteotomy. 12 patients underwent sequential proximal femoral resection with trochanteric osteotomy, 46 subtrochanteric, 6 midshaft, and 25 distal femoral osteotomies with simultaneous valgus correction were performed. All acetabular prostheses were placed in the true anatomical position. We used uncemented high porosity cups. Patients were followed up for a minimum of 12 months. All osteotomies healed uneventfully except 3 non-unions of the greater trochanter in the proximal femur resection group. No femoral shaft fractures in proximally based osteotomies. No significant LLD compared to the unaffected or reconstructed side. 2 patients suffered 3 and 5 degrees malrotation of the femur in the oblique sub-trochanteric group. 3 patients suffered transient sciatic nerve palsies. Shortening femoral osteotomies in the treatment of DDH are necessary to avoid injury to the sciatic nerve. In our series, we found transverse subtrochanteric osteotomies to be the most technically efficient, versatile and predictable in their clinical outcome, due to the ability to correct rotation and preserve the metaphyseal bone integrity, allowing for better initial stem stability. Distal femoral osteotomies allowed for controllable correction of valgus knee deformity


Orthopaedic Proceedings
Vol. 106-B, Issue SUPP_16 | Pages 82 - 82
19 Aug 2024
Courington R Ferreira R Shaath MK Green C Langford J Haidukewych G
Full Access

When treating periprosthetic femur fractures (PPFFs) around total hip arthroplasty (THA)], determining implant fixation status preoperatively is important, since this guides treatment regarding ORIF versus revision. The purpose of this study was to determine the accuracy of preoperative implant fixation status determination utilizing plain films and CT scans.

Twenty-four patients who underwent surgery for Vancouver B type PPFF were included in the study. Two joint surgeons and two traumatologists reviewed plain films alone and made a judgment on fixation status. They then reviewed CT scans and fixation status was reassessed. Concordance and discordance were recorded. Interobserver reliability was assessed using Kendall's W and intraobserver reliability was assessed using Cohen's Kappa. Ultimately, the “correct” response was determined by intraoperative findings, as we routinely test the component intraoperatively.

Fifteen implants were found to be well-fixed (63%) and 9 were loose. Plain radiographs alone predicted correct fixation status in 53% of cases. When adding the CT data, the correct prediction only improved to 55%. Interestingly, concordance between plain radiographs and CT was noted in 82%. In concordant cases, the fixation status was found to be correct in 55% of cases. Of the 18% of cases with discordance, plain films were correct in 43% of cases, and the CT was correct in 57%. Interobserver reliability demonstrated poor agreement on plain films and moderate agreement on CT. Intraobserver reliability demonstrated moderate agreement on both plain films and CT.

The ability to determine fixation status for proximal PPFFs around uncemented femoral components remains challenging. The addition of routine CT scanning did not significantly improve accuracy. We recommend careful intraoperative testing of femoral component fixation with surgical dislocation if necessary, and the surgeon should be prepared to revise or fix the fracture based on those findings.


Orthopaedic Proceedings
Vol. 106-B, Issue SUPP_16 | Pages 37 - 37
19 Aug 2024
Rego P Mafra I Viegas R Silva C Ganz R
Full Access

Executing an extended retinacular flap containing the blood supply for the femoral head, reduction osteotomy (FHO) can be performed, increasing the potential of correction of complex hip morphologies. The aim of this study was to analyse the safety of the procedure and report the clinical and radiographic results in skeletally mature patients with a minimum follow up of two years.

Twelve symptomatic patients (12 hips) with a mean age of 17 years underwent FHO using surgical hip dislocation and an extended soft tissue flap. Radiographs and magnetic resonance imaging producing radial cuts (MRI) were obtained before surgery and radiographs after surgery to evaluate articular congruency, cartilage damage and morphologic parameters. Clinical functional evaluation was done using the Non-Arthritic Hip Score (NAHS), the Hip Outcome Score (HOS), and the modified Harris Hip Score (mHHS).

After surgery, at the latest follow-up no symptomatic avascular necrosis was observed and all osteotomies healed without complications. Femoral head size index improved from 120 ± 10% to 100 ± 10% (p<0,05). Femoral head sphericity index improved from 71 ± 10% before surgery to 91 ± 7% after surgery (p<0,05). Femoral head extrusion index improved from 37 ± 17% to 5 ± 6% (p< 0,05). Twenty five percent of patients had an intact Shenton line before surgery. After surgery this percentage was 75% (p<0,05). The NAHS score improved from a mean of 41 ± 18 to 69 ± 9 points after surgery (p< 0,05). The HOS score improve from 56 ± 24 to 83 ± 17 points after surgery (p< 0,05) and the mHHS score improved from 46 ± 15 before surgery to 76 ± 13 points after surgery (p< 0,05).

In this series, femoral head osteotomy could be considered as safe procedure with considerable potential to correct hip deformities and improve patients reported outcome measures (PROMS).

Level of evidence - Level IV, therapeutic study

Keywords - Femoral head osteotomy, Perthes disease, acetabular dysplasia, coxa plana


Securing the osteotomized greater trochanter (GT) during total hip arthroplasty (THA) for dislocated dysplastic hips (DDH) poses a significant challenge. This study evaluates the union rate and effectiveness of a 2-strand transverse wiring technique utilizing the lesser trochanter for wire anchorage and tensioning. A digastric anterior slide trochanteric osteotomy was performed in 106 patients (118 hips) undergoing THA for DDH. Following uncemented stem insertion, the GT was transferred and fixed to the lateral cortex of the proximal femur using monofilament stainless steel wires. In 72 out of 106 patients (80 hips), the GT was fixed with 2 transverse wire cerclages threaded through 2 drill holes in the base of the lesser trochanter, spaced vertically 5–10 millimeters apart. The wires were wrapped transversely over the GT and tightened, avoiding contact with its tendinous attachments. Patients were regularly monitored, and GT union was assessed clinically and radiographically. Patient ages ranged from 20 to 57 years (mean 35.5), with a follow-up period ranging from 1.5 to 12 years (mean 6.2). The mean union time was 3.3 months (range 2–7). Among all hips, two developed stable nonunion and single wire breakage, but no fragment displacement (2.5%). Two hips exhibited delayed union, eventually healing at 6 and 7 months after surgery. Reattachment of the greater trochanter utilizing a 2-strand transverse wire cerclage anchored at the base of the lesser trochanter demonstrated a high rate of union (97.5%) following THA in dislocated DDH cases


Bone & Joint 360
Vol. 13, Issue 4 | Pages 26 - 29
2 Aug 2024

The August 2024 Shoulder & Elbow Roundup360 looks at: Comparing augmented and nonaugmented locking-plate fixation for proximal humeral fractures in the elderly; Elevated five-year mortality following shoulder arthroplasty for fracture; Total intravenous anaesthesia with propofol reduces discharge times compared with inhaled general anaesthesia in shoulder arthroscopy: a randomized controlled trial; The influence of obesity on outcomes following arthroscopic rotator cuff repair; Humeral component version has no effect on outcomes following reverse total shoulder arthroplasty: a prospective, double-blinded, randomized controlled trial; What is a meaningful improvement after total shoulder arthroplasty by implant type, preoperative diagnosis, and sex?; The safety of corticosteroid injection prior to shoulder arthroplasty: a systematic review; Mortality and subsequent fractures of patients with olecranon fractures compared to other upper limb osteoporotic fractures.


Bone & Joint Open
Vol. 5, Issue 7 | Pages 592 - 600
18 Jul 2024
Faschingbauer M Hambrecht J Schwer J Martin JR Reichel H Seitz A

Aims

Patient dissatisfaction is not uncommon following primary total knee arthroplasty. One proposed method to alleviate this is by improving knee kinematics. Therefore, we aimed to answer the following research question: are there significant differences in knee kinematics based on the design of the tibial insert (cruciate-retaining (CR), ultra-congruent (UC), or medial congruent (MC))?

Methods

Overall, 15 cadaveric knee joints were examined with a CR implant with three different tibial inserts (CR, UC, and MC) using an established knee joint simulator. The effects on coronal alignment, medial and lateral femoral roll back, femorotibial rotation, bony rotations (femur, tibia, and patella), and patellofemoral length ratios were determined.


Bone & Joint Open
Vol. 5, Issue 6 | Pages 524 - 531
24 Jun 2024
Woldeyesus TA Gjertsen J Dalen I Meling T Behzadi M Harboe K Djuv A

Aims

To investigate if preoperative CT improves detection of unstable trochanteric hip fractures.

Methods

A single-centre prospective study was conducted. Patients aged 65 years or older with trochanteric hip fractures admitted to Stavanger University Hospital (Stavanger, Norway) were consecutively included from September 2020 to January 2022. Radiographs and CT images of the fractures were obtained, and surgeons made individual assessments of the fractures based on these. The assessment was conducted according to a systematic protocol including three classification systems (AO/Orthopaedic Trauma Association (OTA), Evans Jensen (EVJ), and Nakano) and questions addressing specific fracture patterns. An expert group provided a gold-standard assessment based on the CT images. Sensitivities and specificities of surgeons’ assessments were estimated and compared in regression models with correlations for the same patients. Intra- and inter-rater reliability were presented as Cohen’s kappa and Gwet’s agreement coefficient (AC1).


Bone & Joint Research
Vol. 13, Issue 6 | Pages 294 - 305
17 Jun 2024
Yang P He W Yang W Jiang L Lin T Sun W Zhang Q Bai X Sun W Guo D

Aims

In this study, we aimed to visualize the spatial distribution characteristics of femoral head necrosis using a novel measurement method.

Methods

We retrospectively collected CT imaging data of 108 hips with non-traumatic osteonecrosis of the femoral head from 76 consecutive patients (mean age 34.3 years (SD 8.1), 56.58% male (n = 43)) in two clinical centres. The femoral head was divided into 288 standard units (based on the orientation of units within the femoral head, designated as N[Superior], S[Inferior], E[Anterior], and W[Posterior]) using a new measurement system called the longitude and latitude division system (LLDS). A computer-aided design (CAD) measurement tool was also developed to visualize the measurement of the spatial location of necrotic lesions in CT images. Two orthopaedic surgeons independently performed measurements, and the results were used to draw 2D and 3D heat maps of spatial distribution of necrotic lesions in the femoral head, and for statistical analysis.


Bone & Joint 360
Vol. 13, Issue 3 | Pages 42 - 45
3 Jun 2024

The June 2024 Children’s orthopaedics Roundup360 looks at: Proximal femoral unicameral bone cysts: is ESIN the answer?; Hybrid-mesh casts in the conservative management of paediatric supracondylar humeral fractures: a randomized controlled trial; Rate and risk factors for contralateral slippage in adolescents treated for slipped capital femoral epiphysis; CRP predicts the need to escalate care after initial debridement for musculoskeletal infection; Genu valgum in paediatric patients presenting with patellofemoral instability; Nusinersen therapy changed the natural course of spinal muscular atrophy type 1: what about spine and hip?; The necessity of ulnar nerve exploration and translocation in open reduction of medial humeral epicondyle fractures in children.


Bone & Joint Open
Vol. 5, Issue 6 | Pages 457 - 463
2 Jun 2024
Coviello M Abate A Maccagnano G Ippolito F Nappi V Abbaticchio AM Caiaffa E Caiaffa V

Aims

Proximal femur fractures treatment can involve anterograde nailing with a single or double cephalic screw. An undesirable failure for this fixation is screw cut-out. In a single-screw nail, a tip-apex distance (TAD) greater than 25 mm has been associated with an increased risk of cut-out. The aim of the study was to examine the role of TAD as a risk factor in a cephalic double-screw nail.

Methods

A retrospective study was conducted on 112 patients treated for intertrochanteric femur fracture with a double proximal screw nail (Endovis BA2; EBA2) from January to September 2021. The analyzed variables were age, sex, BMI, comorbidities, fracture type, side, time of surgery, quality of reduction, pre-existing therapy with bisphosphonate for osteoporosis, screw placement in two different views, and TAD. The last follow-up was at 12 months. Logistic regression was used to study the potential factors of screw cut-out, and receiver operating characteristic curve to identify the threshold value.


The Bone & Joint Journal
Vol. 106-B, Issue 6 | Pages 548 - 554
1 Jun 2024
Ohyama Y Minoda Y Masuda S Sugama R Ohta Y Nakamura H

Aims

The aim of this study was to compare the pattern of initial fixation and changes in periprosthetic bone mineral density (BMD) between patients who underwent total hip arthroplasty (THA) using a traditional fully hydroxyapatite (HA)-coated stem (T-HA group) and those with a newly introduced fully HA-coated stem (N-HA group).

Methods

The study included 36 patients with T-HA stems and 30 with N-HA stems. Dual-energy X-ray absorptiometry was used to measure the change in periprosthetic BMD, one and two years postoperatively. The 3D contact between the stem and femoral cortical bone was evaluated using a density-mapping system, and clinical assessment, including patient-reported outcome measurements, was recorded.


Bone & Joint Open
Vol. 5, Issue 6 | Pages 452 - 456
1 Jun 2024
Kennedy JW Rooney EJ Ryan PJ Siva S Kennedy MJ Wheelwright B Young D Meek RMD

Aims

Femoral periprosthetic fractures are rising in incidence. Their management is complex and carries a high associated mortality. Unlike native hip fractures, there are no guidelines advising on time to theatre in this group. We aim to determine whether delaying surgical intervention influences morbidity or mortality in femoral periprosthetic fractures.

Methods

We identified all periprosthetic fractures around a hip or knee arthroplasty from our prospectively collated database between 2012 and 2021. Patients were categorized into early or delayed intervention based on time from admission to surgery (early = ≤ 36 hours, delayed > 36 hours). Patient demographics, existing implants, Unified Classification System fracture subtype, acute medical issues on admission, preoperative haemoglobin, blood transfusion requirement, and length of hospital stay were identified for all patients. Complication and mortality rates were compared between groups.


Orthopaedic Proceedings
Vol. 106-B, Issue SUPP_6 | Pages 1 - 1
2 May 2024
Mayne A Saad A Botchu R Politis A Wall P McBryde C
Full Access

Radiological investigations are essential in the work-up of patients presenting with non-arthritic hip pain, to allow close review of the complex anatomy around the hip and proximal femur. The aim of this study is to quantify the radiation exposure associated with common radiological investigations performed in assessing young adult patients presenting with non-arthritic hip pain. A retrospective review of our UK tertiary hip preservation centre institutional imaging database was performed. Data was obtained for antero-posterior, cross-table lateral and frog-lateral radiographs, along with data for the low dose CT hip protocol and the Mako CT Hip protocol. The radiation dose of each imaging technique was measured in terms of dose-area product (DAP) with units of mGycm2, and the effective doses (ED, mSv) calculated. The mean effective radiation dose for hip radiographs was in the range 0.03 to 0.83mSv (mean DLP 126.7–156.2 mGycm2). The mean effective dose associated with the low-dose CT hip protocol was 3.04mSv (416.8 mGycm2) and for the Stryker Mako CT Hip protocol was 8.4mSv (1061 mGycm2). The radiation dose associated with use of CT imaging was significantly greater than plain radiographs (p<0.005). Investigation of non-arthritic hip pain can lead to significant ionising radiation exposure for patients. In our institution, the routine protocol is to obtain an anteroposterior radiograph and then a specific hip sequence 3 Tesla MRI including anteversion views. This provides the necessary information in the majority of cases, with CT scanning reserved for more complex cases where we feel there is a specific indication. We would encourage the hip preservation community to carefully consider and review the use of ionising radiation investigations


Orthopaedic Proceedings
Vol. 106-B, Issue SUPP_6 | Pages 7 - 7
2 May 2024
McCabe P Baxter J O'Connor M McKenna P Murphy T Cleary M Rowan F
Full Access

The burden of metastatic disease presenting with axial skeleton lesions is exponentially rising predominantly due to advances in oncological therapies. A large proportion is these lesions are located in the proximal femora, which given its unique biomechanical architecture is problematic.

These patients are frequently comorbid and require prompt and concise decision making regarding their orthopaedic care in line with recent British Orthopaedic Association guidelines.

We present data detailing the outcomes for patients with proximal femoral metastatic disease referred and treated over a three year period in an Regional Cancer Centre.

We retrospectively reviewed a prospectively maintained database of all patients referred for discussion at MDT with axial skeletal metastatic disease. From this we isolated patients with femoral disease. Demographic data along with primary tumour and metastatic disease site were assessed. Treatment regimens were analysed and compared. Finally predicted and actual mortality data was collated.

331 patients were referred over the analysed time period, of which 99 had femoral disease. 66% of patients were managed conservatively with serial monitoring while 34% underwent operative treatment. 65% of those received an intramedullary fixation while 35% had arthroplasty performed. There was a 51:49 split male to female with Prostate, Lung and Breast being the predominant primary tumours. Concurrent spinal metastatic disease was noted in 62% of patients while visceral mets were seen in 37%. Mortality rate was 65% with an average prognosis of 388 days (1.06years) while average mortality was noted within 291 days (0.8 years).

Proximal femoral metastatic disease accounts for a large volume of the overall mets burden. There is an overall tendency towards conservative management and of those requiring surgery IM nailing was the treatment of choice. The data would indicate that outcomes for these patients are guarded and on average worse than those predicted.


The Bone & Joint Journal
Vol. 106-B, Issue 5 Supple B | Pages 40 - 46
1 May 2024
Massè A Giachino M Audisio A Donis A Giai Via R Secco DC Limone B Turchetto L Aprato A

Aims

Ganz’s studies made it possible to address joint deformities on both the femoral and acetabular side brought about by Perthes’ disease. Femoral head reduction osteotomy (FHRO) was developed to improve joint congruency, along with periacetabular osteotomy (PAO), which may enhance coverage and containment. The purpose of this study is to show the clinical and morphological outcomes of the technique and the use of an implemented planning approach.

Methods

From September 2015 to December 2021, 13 FHROs were performed on 11 patients for Perthes’ disease in two centres. Of these, 11 hips had an associated PAO. A specific CT- and MRI-based protocol for virtual simulation of the corrections was developed. Outcomes were assessed with radiological parameters (sphericity index, extrusion index, integrity of the Shenton’s line, lateral centre-edge angle (LCEA), Tönnis angle), and clinical parameters (range of motion, visual analogue scale (VAS) for pain, Merle d'Aubigné-Postel score, modified Harris Hip Score (mHHS), and EuroQol five-dimension five-level health questionnaire (EQ-5D-5L)). Early and late complications were reported.


The Bone & Joint Journal
Vol. 106-B, Issue 5 | Pages 435 - 441
1 May 2024
Angelomenos V Mohaddes M Kärrholm J Malchau H Shareghi B Itayem R

Aims

Refobacin Bone Cement R and Palacos R + G bone cement were introduced to replace the original cement Refobacin Palacos R in 2005. Both cements were assumed to behave in a biomechanically similar fashion to the original cement. The primary aim of this study was to compare the migration of a polished triple-tapered femoral stem fixed with either Refobacin Bone Cement R or Palacos R + G bone cement. Repeated radiostereometric analysis was used to measure migration of the femoral head centre. The secondary aims were evaluation of cement mantle, stem positioning, and patient-reported outcome measures.

Methods

Overall, 75 patients were included in the study and 71 were available at two years postoperatively. Prior to surgery, they were randomized to one of the three combinations studied: Palacos cement with use of the Optivac mixing system, Refobacin with use of the Optivac system, and Refobacin with use of the Optipac system. Cemented MS30 stems and cemented Exceed acetabular components were used in all hips. Postoperative radiographs were used to assess the quality of the cement mantle according to Barrack et al, and the position and migration of the femoral stem. Harris Hip Score, Oxford Hip Score, Forgotten Joint Score, and University of California, Los Angeles Activity Scale were collected.


The Bone & Joint Journal
Vol. 106-B, Issue 5 | Pages 508 - 514
1 May 2024
Maximen J Jeantet R Violas P

Aims. The aim of this study is to evaluate the surgical treatment with the best healing rate for patients with proximal femoral unicameral bone cysts (UBCs) after initial surgery, and to determine which procedure has the lowest adverse event burden during follow-up. Methods. This multicentre retrospective study was conducted in 20 tertiary paediatric hospitals in France, Belgium, and Switzerland, and included patients aged < 16 years admitted for UBC treatment in the proximal femur from January 1995 to December 2017. UBCs were divided into seven groups based on the index treatment, which included elastic stable intramedullary nail (ESIN) insertion with or without percutaneous injection or grafting, percutaneous injection alone, curettage and grafting alone, and insertion of other orthopaedic hardware with or without curettage. Results. A total of 201 patients were included in the study. The mean age at diagnosis was 8.7 years (SD 3.9); 77% (n = 156) were male. The mean follow-up was 9.4 years (SD 3.9). ESIN insertion without complementary procedure had a 67% UBC healing rate after the first operation (vs 30% with percutaneous injection alone (p = 0.027), 43% with curettage and grafting (p = 0.064), and 21% with insertion of other hardware combined with curettage (p < 0.001) or 36% alone (p = 0.014)). ESIN insertion with percutaneous injection presented a 79% healing rate, higher than percutaneous injection alone (p = 0.017), curettage and grafting (p = 0.028), and insertion of other hardware combined with curettage (p < 0.001) or alone (p = 0.014). Patients who underwent ESIN insertion with curettage had a 53% healing rate, higher than insertion of other hardware combined with curettage (p = 0.009). The overall rate of postoperative complications was 25% and did not differ between groups (p = 0.228). A total of 32 limb length discrepancies were identified. Conclusion. ESIN insertion, either alone or combined with percutaneous injection or curettage and grafting, may offer higher healing rates than other operative procedures. Limb length discrepancy remains a major concern, and might be partly explained by the cyst’s location and the consequence of surgery. Therefore, providing information about this risk is crucial. Cite this article: Bone Joint J 2024;106-B(5):508–514


Bone & Joint Open
Vol. 5, Issue 4 | Pages 317 - 323
18 Apr 2024
Zhu X Hu J Lin J Song G Xu H Lu J Tang Q Wang J

Aims

The aim of this study was to investigate the safety and efficacy of 3D-printed modular prostheses in patients who underwent joint-sparing limb salvage surgery (JSLSS) for malignant femoral diaphyseal bone tumours.

Methods

We retrospectively reviewed 17 patients (13 males and four females) with femoral diaphyseal tumours who underwent JSLSS in our hospital.


Bone & Joint Research
Vol. 13, Issue 4 | Pages 184 - 192
18 Apr 2024
Morita A Iida Y Inaba Y Tezuka T Kobayashi N Choe H Ike H Kawakami E

Aims

This study was designed to develop a model for predicting bone mineral density (BMD) loss of the femur after total hip arthroplasty (THA) using artificial intelligence (AI), and to identify factors that influence the prediction. Additionally, we virtually examined the efficacy of administration of bisphosphonate for cases with severe BMD loss based on the predictive model.

Methods

The study included 538 joints that underwent primary THA. The patients were divided into groups using unsupervised time series clustering for five-year BMD loss of Gruen zone 7 postoperatively, and a machine-learning model to predict the BMD loss was developed. Additionally, the predictor for BMD loss was extracted using SHapley Additive exPlanations (SHAP). The patient-specific efficacy of bisphosphonate, which is the most important categorical predictor for BMD loss, was examined by calculating the change in predictive probability when hypothetically switching between the inclusion and exclusion of bisphosphonate.