Dow

nloadedfrom

https://journals.lww.com

/jaaosby

BhDMf5ePH

Kav1zEoum1tQ

fN4a+kJLhEZgbsIH

o4XMi0hC

ywCX1AW

nYQp/IlQ

rHD3XG

JiJSDa6kLdjliR

zOOsR

+bI3gZWJ99prw

iH9JO

uTYY=on

08/14/2018

Downloadedfromhttps://journals.lww.com/jaaosbyBhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3XGJiJSDa6kLdjliRzOOsR+bI3gZWJ99prwiH9JOuTYY=on08/14/2018

Research Article

Femoral Derotation Osteotomy inAdults for Version Abnormalities

Abstract

Background: Version abnormalities of the femur can cause pain andhip joint damage due to impingement or instability. A retrospectiveclinical review was conducted on patients undergoing asubtrochanteric derotation osteotomy for either excessiveanteversion or retroversion of the femur.Methods: A total of 55 derotation osteotomies were performed in 43patients: 36 females and 7 males. The average age was 29 years(range, 14 to 59 years). The osteotomies were performed closed withan intramedullary saw. Fixation was performed with a variety ofintramedullary nails. Twenty-nine percent of patients had aretroversion deformity (average, 29� of retroversion; range, 12�to 223�) and 71% had excessive anteversion of the femur(average, 137� of anteversion; range, 122� to 153�). The etiologywas posttraumatic in 5 patients (12%), diplegic cerebral palsy in 2patients (5%), Prader-Willi syndrome in 1 patient (2%), and idiopathicin 35 patients (81%). Forty-nine percent underwent concomitantsurgery with the index femoral derotation osteotomy, including hiparthroscopy in 40%, tibial derotation osteotomy in 13%, and aperiacetabular osteotomy in 5%. Tibial osteotomies were performedto correct a compensatory excessive external tibial torsion that wouldbe exacerbated in the correction of excessive femoral anteversion.Results: No patient was lost to follow-up. Failures occurred in threehips in three patients (5%): two hip arthroplasties and one nonunionthat healed after rerodding. There was one late infection treatedsuccessfully with implant removal and antibiotics with an excellentfinal clinical outcome. At an average follow-up of 6.5 years (range, 2 to19.7 years), themodifiedHarrisHipScore improvedby 29points in theremaining 52 cases (P , 0.001, Wilcoxon signed-rank test). Theresults were rated as excellent in 75%, good in 23%, and fair in 2%.Subsequent surgery was required in 78% of hips, 91% of which wereimplant removals.Conclusions: A closed, subtrochanteric derotation osteotomy of thefemur is a safe and effective procedure to treat either femoralretroversion or excessive anteversion. Excellent or good results wereobtained in 93%, despite the need for subsequent implant removal inmore than two-thirds of the patients.

In the surgical treatment of hipdisorders, a major cause of failure

is either insufficient correction or a

failure to fully recognize the under-lying deformities causing pain andjoint damage.1-3 One type of femoral

Robert L. Buly, MD, MS

Branden R. Sosa, HS

Lazaros A. Poultsides, MD, MSc,PhD

Elaine Caldwell, BS, RN

S. Robert Rozbruch, MD

From the Hospital for Special Surgery,New York, NY.

Correspondence to Dr. Buly:bulyr@hss.edu

J Am Acad Orthop Surg 2018;00:1-10

DOI: 10.5435/JAAOS-D-17-00623

Copyright © 2018 The Author(s).Published by Wolters Kluwer Health,Inc. on behalf of the AmericanAcademy of OrthopaedicSurgeons.This is an open accessarticle distributed under the terms ofthe Creative Commons Attribution-NonCommercial-NoDerivativesLicense 4.0 (CC BY-NC-ND), whichpermits downloading and sharing thework provided it is properly cited. Thework cannot be changed in any way orused commercially without permissionfrom the journal.

Month 2018, Vol 00, No 00 1

deformity that is still frequentlyoverlooked are rotational defor-mities of the femur, that is, excessiveanteversion or femoral retroversion.These rotational deformities mayoccur alone or may coexist withacetabular dysplasia4-6 or varioustypes of hip impingement.2,5,7-9

Rotational deformities may also beassociated with cerebral palsy10 andlabral tears11 and are not unusualafter the fixation of femoral shaftfractures.12-14

Excessive femoral anteversion cancause instability, damage of the artic-ular cartilage and acetabular labrum,and eventually osteoarthritis.15-17

Furthermore, it can cause a decreasein the length of the abductor leverarm,18 posterior extra-articularimpingement,9 and ischiofemoralimpingement.19 Finally, excessive fem-oral anteversion may cause increasedhip and knee adduction moments, anintoeing gait and patellofemoral mal-tracking, with resultant knee pain andarthritis.20-22

Femoral retroversion, on the otherhand, causes damage due to impinge-ment between the femoral neck andacetabulum, which may result in dam-age to the labrum and articular carti-lage, ultimately resulting inosteoarthritis of the hip.16,23 Otherpotential retroversion problems in-clude an increased risk of slippedcapital femoral epiphysis24 and sus-ceptibility to a traumatic posterior hipdislocation.25,26 Residual, untreatedfemoral retroversion may be a reasonwhy hip preserving surgeries may fail,especially after the arthroscopic treat-ment of hip impingement.1-3

Despite the important role of fem-oral rotational deformity in the

pathogenesis of hip disease, there islittle written about the outcomes oftreatment. This article describes thetechnique and outcomes of a closed,derotation osteotomy of the femurto correct either excessive femoralanteversion or retroversion as partof a hip preservation effort. Thequestion to be answered is whetherhip pain as a result of version abnor-malities of the femur can be alleviatedby this type of surgery.

Methods

Patients selected for the index proce-dure had hip pain secondary toincreased femoral anteversion orfemoral retroversion. Surgery wasoffered if the patient had failed allnonsurgical treatment measures andif the range of motion aberrationcorrelated with the version abnor-mality (ie, excessive hip internal rota-tion with excessive anteversion or alack of internal rotation associatedwith femoral retroversion). Patientswith coxa vara (a neck-shaft angleof,125�) or coxa valga (a neck-shaftangle of .140�) were excluded, withthe understanding that a varus orvalgus derotation intertrochantericosteotomy would be more appropri-ate to address the concomitant neck-shaft angulation.27

In addition to plain radiographs, allpatients underwent preoperativeMRI to assess the condition of thearticular cartilage and labrum andthree-dimensional CT scans to accu-rately define the anatomic deform-ities. The measurement of femoralversion was performed by the CTtechnique described by Murphy

et al.28 All readings and measure-ments were performed by board-certified musculoskeletal radiologists.The osteotomies were performed

with the patient in the supine positionunder regional, hypotensive anes-thesia. The operated leg was drapedfree, and traction was not used. Anintramedullary hand saw was usedthat did not require exposure of theosteotomy site.14 A small, longitu-dinal skin incision was made justproximal to the greater trochanter.The isthmus of the femur was over-reamed by 0.5 mm in accordancewith the nail to be used. Thesubtrochanteric region was thenreamed 0.5 mm larger than thediameter of the proposed intra-medullary saw. Rotational controlwas achieved by placing 1/8-inchsmooth Steinmann pins into thefemur proximal and distal to theosteotomy in the desired amountof rotational correction (Figure 1).The location and progress of theosteotomy were controlled byfluoroscopy (Figure 2). The angularcorrection was controlled visually byusing flat, triangular guides from ablade plate instrument set (Figure 3).The osteotomy was performed

in the subtrochanteric region byinserting the hand saw, which wasthen rotated in a stepwise fashionwithprogressive protrusion of the bladefrom the cam.The distal fragmentwasthen rotated to align the two pinsparallel, thus effecting the rotationalcorrection. The goal was to achieveapproximately 15� of femoral ante-version. Fixation was then achievedusing a variety of trochanteric entryintramedullary nails that were lockedproximal and distal to the osteotomy.

Dr. Buly or an immediate family member has stock or stock options held in Blue Belt Technology and serves as a board member, owner,officer, or committee member of the Maurice Mueller Foundation of North America and the International Society for Hip Arthroscopy.Dr. Rozburch or an immediate family member has received royalties from Stryker; is a member of a speakers’ bureau or has made paidpresentations on behalf of NuVasive, Smith & Nephew, and Stryker; serves as a paid consultant to NuVasive, Smith & Nephew, and Stryker;and serves as a board member, owner, officer, or committee member of the Limb Lengthening Reconstruction Society. None of the followingauthors or any immediate family member has received anything of value from or has stock or stock options held in a commercial company orinstitution related directly or indirectly to the subject of this article: Mr. Sosa, Dr. Poultsides, and Ms. Caldwell.

Femoral Derotation Osteotomy in Adults

2 Journal of the American Academy of Orthopaedic Surgeons

The intramedullary devices used were42 TriGen Trochanteric AntegradeNails (Smith & Nephew), five Tro-chanteric Fixation Nails and threeIntramedullary Femoral Nails (DePuySynthes), four GammaNails (Stryker),and one piriformis fossa entryPhoenix Femoral Nail (ZimmerBiomet).Concomitant hip arthroscopy was

performed just prior to the osteotomyif the magnetic resonance image re-vealed labral and/or articular carti-lage lesions or the presence of a camlesion of the femoral neck that wouldimpinge if a retroverting derotationfemoral osteotomy was to be per-formed for excessive anteversion.A concomitant periacetabular

osteotomy was performed at thesame setting if there was coexisting,severe dysplasia that required cor-rection along with the femoral ver-

sion. The periacetabular osteotomywas performed first with the samepreparation and drape setup used forboth procedures.A concomitant tibial/fibular oste-

otomy was performed if the patienthad a compensatory external tibialtorsion coexisting with excessivefemoral anteversion, as described byTönnis and Heinecke.15 This proce-dure was done to prevent an exag-gerated external foot progressionangle that would result from der-otating the excessively antevertedfemur in patients with this rotationaldeformity. The tibia was eitherinternally rotated with gradual cor-rection using an external hexapodframe in the supramalleolar orproximal tibial regions or correctedacutely over an intramedullary nail,depending on the morphology of thetibial deformity.

Postoperatively, epidural patient-controlled anesthesia was used if atibial osteotomy was not performed.Intravenous patient-controlled anes-thesia was used instead with tibialosteotomies to allow monitoring ofthe lower leg and vigilance regardinga possible compartment syndrome.No braces or casts were used aftersurgery. There were no range ofmotion restrictions. Weight bearingas tolerated was permitted withcrutches unless a concomitant peri-acetabular or tibial osteotomy wasperformed, in which case the weightbearing was restricted to 20% for6 weeks. Follow-up examinationswith AP and lateral radiographswereperformed at 6 weeks, 3 months,6 months, and 1 year after surgery.The modified Harris Hip Score(mHHS) was used, and scores weredocumented before surgery and at

Figure 1

Schematic illustration demonstrating the osteotomy technique.

Robert L. Buly, MD, MS, et al

Month 2018, Vol 00, No 00 3

the latest follow-up. The minimumfollow-up time was 24 months.

Results

Starting in 1997, a total of 81 femoralosteotomies have been performed in67 patients. Forty-three patientshaving undergone 55 derotationosteotomies had a minimum follow-up of 2 years. All bilateral cases weredone staged. There were 36 females

and 7males; the average agewas 29.0years (range, 14 to 59 years).The deformity etiology was post-traumatic in 5 patients (12%), diple-gic cerebral palsy in 2 patients (5%),Prader-Willi syndrome in 1 patient(2%), and idiopathic in 35 patients(81%) (Table 1). All hips had aTönnis arthritis grade of zero (noevidence of arthritis).15 Twenty-ninepercent of the hips (16 hips in 14patients) had a retroversion defor-mity (average, 29� of retroversion;

range, 12� anteversion to 224�retroversion). Seventy-one percent(39 hips in 29 patients) had excessiveanteversion of the femur (average, 137� of anteversion; range, 122�to 153� anteversion). The averagerotational correction was 24� for theretroverted hips (range, 18� to 35�)and 23� (range, 15� to 40�) for an-teverted hips (Table 2). All 16 ret-roverted hips were considered tohave “severely diminished ante-version” by the criteria of Tönnis,

Figure 2

A, Photograph showing the Winquist intramedullary saw. B, Photograph showing the saw blade extended from the cammechanism. C–E, Intraoperative fluoroscopic images showing gradual transection of the lateral and medial cortices,followed by osteotomy completion.

Femoral Derotation Osteotomy in Adults

4 Journal of the American Academy of Orthopaedic Surgeons

whereas 37 of 38 excessively ante-verted hips (97%) were consideredto be “severely increased” (.25� ofanteversion), with 1 (3%) being“moderately increased” (21� to 25�of anteversion).15

The clinical hip range of motionassessment for all patients includedmeasuring internal and external ro-tations, with the hip flexed to 90�.Retroverted hips typically lacked orhad diminished internal rotation.Conversely, hips with excessiveanteversion had more internal ro-tation than external rotation. Forthe cases with excessive anteversion,the preoperative average of internalrotation at 90� of hip flexionwas173� (range,145� to 110�) andexternal rotation 122� (range, 25�to 160�). After osteotomy, theinternal rotation diminished to 126�(range, 15� to 45�), whereas theexternal rotation improved to 148�(range, 115� to 70�). This wassignificant at P , 0.01 (Wilcoxonsigned-rank test). For the cases ofretroversion, the preoperative averageof internal rotation at 90� of hipflexion was 21� (range, 220� to110�) and external rotation 181�(range, 150� to 190�). After oste-otomy, the internal rotation improvedto 123� (range, 110� to 135�),whereas the external rotation dimin-ished to 142� (range, 130� to 50�).This was significant at P , 0.001(Wilcoxon signed-rank test).

Previous surgery had been per-formed in 26 hips (47%) (Table 1).Twenty-seven hips (49%) underwentconcomitant surgery with the indexfemoral derotation osteotomy, in-cluding hip arthroscopy with labraldébridement and chondroplasty in16 (29%), 6 hip arthroscopies withan additional femoral osteochon-droplasty (11%), a tibial derotationosteotomy in 7 (13%), and an ipsi-lateral periacetabular osteotomy with3 of the femoral osteotomies (5%)(Table 1). Three of the ipsilateraltibial osteotomies were performed inthe supramalleolar region and two inthe proximal tibia, with external framefixation. Two of the tibial osteotomieswere performed at midshaft withimmediate rotational correction andfixation with an intramedullary nail.No patient was lost to follow-up.

One patient died of cancer 12.7 yearsafter surgerywithahip scoreof85.Theaverage time to femoral osteotomyunion was 3.3 months (range, 2 to16 months). All the tibial and pelvicosteotomies healed uneventfully.At an average follow-up of 6.5 years

(range, 2 to 19.7 years), the mHHSimproved by 27 points (P , 0.001,Wilcoxon signed-rank test) from 66to 93 points. When taken separately,there was a statistically significantimprovement in both the groups witheither retroversion or excessive ante-version (Table 2). The results wererated as excellent in 70%, good in

23%, and fair in 7%, including therevision femoral osteotomy and twototal hip replacements.Failure was defined as conversion to

total hip arthroplasty, refixation of theindexosteotomy, or anmHHSof,70.Failures occurred in three hips in threepatients (5%): two hip arthroplastiesand one rerodding for a femoralnonunion. A total hip arthroplastywasperformed 46 months after osteotomyin a 46-year-old woman with Ehlers-Danlos syndrome with only minimalosteoarthritic change seen on MRIand a normal joint space on plainradiographs (Tönnis stage zero). Thepatient continues to do well with thecontralateral osteotomy and has anmHHS of 74 points. Another hipreplacement was performed in an 18-year-old man with Prader-Willi syn-drome 15 months after the indexprocedure because of the failure ofthe concomitant periacetabular oste-otomy. The third failure was in a 26-year-old woman with Ehlers-Danlossyndrome with a nonunion thatwas rerodded successfully. All threepresented initially with excessiveanteversion.Subsequent surgery was required in

78% of hips, 39 of 43 (91%) wereimplant removals. The implant wasremoved in patientswith radiographicevidence of bone union and only ifthere was notable pain refractory tononsurgical treatment, usually irrita-tion from the screw heads or a thigh

Figure 3

A, Triangles used to set the degree of rotation correction.B, A 20� triangle was used to set the correction between the proximaland distal pins in a case of excessive anteversion. C, The femoral nail is inserted while maintaining rotational correction.

Robert L. Buly, MD, MS, et al

Month 2018, Vol 00, No 00 5

ache that occurred with loading orunloading the femur (Figure 4). Arevision femoral osteotomy was per-formed in one hip (2%) by anothersurgeon to add additional rotational

stability while performing a surgicalhip dislocation in a 39-year-oldwoman with anterior capsular defi-ciency and instability after four pre-vious surgeries. Currently, 9 months

after the revision femoral osteotomy,the mHHS was 85 points.A late infection occurred in one

femur of a 14-year-old girl 8monthsafter surgery on the second femur.The organism was a minimally re-sistant Staphylococcus aureus andwas treated successfully with im-plant removal and antibiotics. Atfollow-up, the patient’s mHHS was100 points.

Discussion

It has been reported that osteo-arthritis may occur with eitherfemoral retroversion or increasedanteversion.15-17,23,29,30 Retrover-sion of the femur, either alone or incombination with other defor-mities, can cause hip damage sec-ondary to impingement.16,23,29,30

The damage with excessive ante-version occurs at the periphery of theacetabulum secondary to high com-pressive and shear forces on thearticular cartilage and labrum, caus-ing hip pain and arthritis.15-17

The onset of pain in patients withdysplasia occurs earlier if thereis coexistent excessive combinedanteversion.4 Increased femoral an-teversion has also been associatedwith posterior greater trochantericimpingement,9,31 decreased abductor

Table 1

Summary of Results

Total patients 43 (36 females and 7males)

Average age 29.0 y (range, 14 to 59 y)

Total osteotomies 55

Excessively anteverted 39 (71%)

Retroverted 16 (29%)

Etiology (patients)

Idiopathic 35 (81%)

Posttraumatic 5 (12%)

Cerebral palsy 2 (5%)

Prader-Willi syndrome 1 (2%)

Previous surgery (hips)

Total 26 (47%)

Hip arthroscopy 1 femoralosteochondroplasty

13 (24%)

Open reduction 1 internal fixation 5 (9%)

Previous femoral osteotomy 3 (5%)

Hip arthroscopy 1 labral débridement 3 (5%)

Femoral lengthening 1 (2%)

Slipped capital femoral epiphysis pinning 1 (2%)

Concomitant surgery (hips)

Total: all ipsilateral 27 (49%)

Hip arthroscopy 1 labrum 1chondroplasty

16 (29%)

Tibial derotation osteotomy 7 (13%)

Hip arthroscopy 1 osteochondroplasty 6 (11%)

Periacetabular osteotomy 3 (5%)

Table 2

Comparison of the Anteverted Versus Retroverted Cases

ConditionHips(%)

PeriacetabularOsteotomy

(%)

TibialOsteotomy

(%)AverageDeformity

AverageCorrection

PreoperativemHHS

PostoperativemHHS

Excessivefemoralanteversion

39 (71) 3 (8) 7 (18) 137� (122�653�) 23� (15�–40�) 64 94 (P , 0.01)a

Femoralretroversion

16 (29) 0 (0) 0 (0) 29� (12� to 224�) 24� (18�–35�) 70 96 (P , 0.05)a

Total(patients)

55 (43) 3 (5) 7 (13) — — 65 94 (P , 0.001)a

mHHS = modified Harris Hip Scorea Wilcoxon signed-rank test.

Femoral Derotation Osteotomy in Adults

6 Journal of the American Academy of Orthopaedic Surgeons

power by ,28% due to diminishedfemoral offset,18 and hip instability.32

Psoas irritation may be due to ante-rior instability; the tendon may actas a dynamic stabilizer and releasingit may exacerbate the problem.2,33

Patients with symptomatic ischiofe-moral impingement with diminishedclearance between these two structuresare more likely to have excessivefemoral anteversion compared toasymptomatic patients.34 Other prob-lems associated with excessive femoralanteversion include increased hip andknee adduction moments, an intoeinggait and patellofemoral maltracking,pain, and arthritis.20-22

The arthroscopic treatment of hipimpingement may fail if bonydébridement is inadequate. In addi-tion, these procedures may also fail ifcoexisting femoral retroversion isnot detected or treated.2,35 Six of thepatients in this study with retrover-sion of the femur had previouslyundergone an arthroscopic débride-ment with initial symptomatic relief,but eventually had a relapse ofpainful impingement and a lack ofinternal rotation.There is no uniform agreement as to

the value of normal femoral and ace-tabular version. The reported valuesfor the acetabulum range from 13� to20� of anteversion in three studies,averaging 17�.30,36,37 For the femur,the range of anteversion is 10� to 20�in three studies, averaging 15�.15,37,38The McKibbin index is the sum ofacetabular version and femoral ver-sion; “normal” is approximately 30�,and values.60� are considered to behighly unstable.15 The goal of surgi-cal correction was to approach 15� offemoral version. Three-dimensionalCT scans were used to measure ver-sion as precisely as possible and havelong been considered the benchmarktechnique to measure version.39

Version abnormalities of the femurcan also be treated with an inter-trochanteric osteotomy. These tech-niques allow the correction of an

abnormal neck-shaft angle (ie, coxavara or coxa valga) and rotationaldeformities by rotating the distalfragment by the desired amount ofcorrection before applying the bladeplate.27 This technique may not benecessary if the neck-shaft angle isnormal, and the only femoral defor-mity is purely rotational. There is

also no shortening of the abductormuscle fibers as occurs with a varusproducing intertrochanteric oste-otomy. Because of the much lowerprofile compared to a blade plate,there can be considerably less peri-trochanteric bursitis and pain. Asubtrochanteric derotation femoralosteotomy may also be performed by

Figure 4

AP (A) and lateral (B) radiographs of a 17 year-old female who presented withexcessive anteversion, showing a healed femoral osteotomy at 12 monthspostoperatively. The intramedullary nail was subsequently removed.

Robert L. Buly, MD, MS, et al

Month 2018, Vol 00, No 00 7

plating, but it requires a much moreinvasive approach.40 The advantageof the described technique is that itallows for a much less invasiveapproach, lessening surgical mor-bidity and theoretically a lowerchance of infection. In addition, thevastus lateralis is not dissected fromthe femur, maintaining more of theperiosteal blood supply to enhancebone union. In all the cases, it wasnot necessary to expose the oste-otomy site because the transectionwas performed with an intra-medullary bone saw. Other advan-tages include the ability to allowweight bearing as tolerated immedi-ately because the fixation is providedwith a locked intramedullary nail.Placing the distal interlocking screwin the dynamic mode allows com-pression at the osteotomy site withweight bearing. In contrast, patientstreated with an intertrochantericosteotomy and plating are main-tained at 20% weight bearing for atleast 6 weeks after surgery. Inaddition, a pure derotation oste-otomy performed in the subtro-chanteric region does not deform theproximal femur. Should a total hiparthroplasty be required in thefuture, it does not hamper steminsertion as can occur after a previ-ous intertrochanteric osteotomy.The disadvantage of the described

technique is that bone healing isslower, averaging 3 to 4 months, andin some cases even longer, whichmaybe due to the diminished healingpotential of cortical bone versuscancellous bone. In addition, there ismuch less surface area at the site ofthe transverse subtrochanteric oste-otomy than with an intertrochantericor supracondylar type. Another dis-advantage is potential damage to thehip abductors because of the reamingnecessary to insert a nail. Care wastaken to enter the greater trochanterthrough the posterosuperior “barearea” if possible to leave theabductors minimally disrupted. At

follow-up, no patient had notablehip abductor weakness or Trende-lenburg limp or sign.It was necessary to perform a con-

comitant tibialosteotomy insevencases(13%) with excessive femoral ante-version and a compensatory externaltibial torsion instead of the usualintoeing gait associated with excessiveanteversion, dubbed as “miserablemalalignment syndrome.”21 Surgicalcorrection of increased femoral ante-version requires externally rotating thedistal fragment. In these patients, therewould have been a greatly exaggeratedexternal foot progression angle.A concomitant periacetabular oste-

otomy was performed in patients withsevere acetabular dysplasia and coex-isting femoralmalrotationwhere it wasfelt that correction of only one or theother would leave a notable deformitythat is often an indication for surgerywhen occurring alone. This procedurewas performed in three patients (5%).Concomitant hip arthroscopy wasperformed for two reasons: to addressintra-articular pathology (ie, tornacetabular labrum and articular car-tilage damage) that would ordinarilynot be accessed during the osteotomyand to remove a sizable cam lesionthat would impinge after a femoralretroverting osteotomy.The failures all occurred in patients

with excessive anteversion and con-nective tissue disorders: Ehlers-Danlos or Prader-Willi syndrome.Interestingly, failure did not occur inthe contralateral osteotomy of thetwo Ehlers-Danlos patients. Collagenabnormalities associated with theseconditions may have contributed tothe problems of instability and poorbone healing.Although subsequent surgery was

required in 78% of hips, 93% of thesewere implant removals. Overall, 70%of patients underwent removal of theimplant. Although generally bettertolerated than a blade plate afterintertrochanteric osteotomy, most pa-tients had either irritation from the in-

terlocking screws or a thigh ache thatresolved in most cases after implantremoval. Other than the hip arthro-plasties or osteotomy revisions, theremainder of cases were hip arthro-scopic débridements in two patients.Winquist14 reported the ability to

perform a closed osteotomy and in-tramedullary nailing to correct sim-ple rotational deformities.Chapman et al13 reported closed

osteotomy nailing performed in 31patients for leg-length inequality and6 with rotational deformities. Preop-erative rotational deformities aver-aged 58� and all were corrected towithin 5� of normal.Stahl et al12 treated 14 patients with

posttraumatic rotational deformitiesof the femur that ranged from 26� to63� with a closed technique over anintramedullary nail. Postoperative CTscans revealed excellent correction ofthe deformity within 4� in all cases.Kamath et al40 reported 28 rota-

tional femoral osteotomies in 26patients, 93% for excessive femoralanteversion. Clinical outcomes werenot reported. After two initial fail-ures for nonunion, all subsequentlywent on to union with refixation.Pailhe et al41 reported nine der-

otation osteotomies in six adolescents(average age, 13.6 years) for excessiveanteversion. The technique was donewith a distal supracondylar oste-otomy and fixationwith an antegradeintramedullary nail. The averagecorrection was 19�. Patient-reportedoutcome scores were not recorded.All patients were satisfied or verysatisfied and had better foot pro-gression angles and less internalrotation on range of motion testing.Putz et al10 performed 96 der-

otation femoral osteotomies (proxi-mal or distal) in 63 adult cerebralpalsy patients with excessive ante-version. Although patient-reportedoutcome scores were not recorded,the group experienced statisticalimprovement in foot progressionangle and passive and stance range of

Femoral Derotation Osteotomy in Adults

8 Journal of the American Academy of Orthopaedic Surgeons

motion. Tibial rotation osteotomywas required in 16.7% of cases tocompensate for excessive externaltibial torsion.10 In the present study, asimilar need for concomitant tibialderotation osteotomy was noted.The limitation of this study is that it

is a retrospective case series without acontrol group. However, it is a single-surgeon series with a consistenttechnique over a 20-year period. Thepresent study seems to be the onlyseries in which a patient-recorded out-come score was used. In addition, nopatients were lost to follow-up.

Conclusion

In conclusion, hip pain and deterio-ration can be caused by a variety ofdeformities, acting either alone or incombination. There can be consider-able overlap with acetabular dyspla-sia, hip impingement, and neck-shaftabnormalities, while femoral versionmay be diminished, normal, or exces-sive. It is important to identify all thedeformities present to ensure the bestchance of success after hip preserva-tion surgery. Version abnormalities ofthe hip, often overlooked, must beassessed because of the damagecaused by these deformities. A closed,subtrochanteric derotation osteotomyof the femur is a safe and effectiveprocedure to treat either femoral ret-roversion or excessive anteversion.Excellent or good resultswereobtainedin 93%, with a statistically significantimprovement in themHHS, despite theneed for subsequent implant removal inmore than two-thirds of the patients.Failures may occur in patients withgenetic defects associated with abnor-mal collagen or bone density.

References

References printed in bold type arethose published within the past 5years.

1. Ross JR, Larson CM, Adeoye O, Kelly BT,Bedi A: Residual deformity is the mostcommon reason for revision hiparthroscopy: A three-dimensional CTstudy. Clin Orthop Relat Res 2015;473:1388-1395.

2. Fabricant PD, Fields KG, Taylor SA,Magennis E, Bedi A, Kelly BT: The effect offemoral and acetabular version on clinicaloutcomes after arthroscopicfemoroacetabular impingement surgery. JBone Joint Surg Am 2015;97:537-543.

3. Clohisy JC, Nepple JJ, Larson CM, Zaltz I,Millis M: Persistent structural disease is themost common cause of repeat hippreservation surgery. Clin Orthop RelatRes 2013;471:3788-3794.

4. Kohno Y, Nakashima Y, Akiyama M, FujiiM, Iwamoto Y: Does native combinedanteversion influence pain onset in patientswith dysplastic hips?Clin Orthop Relat Res2015;473:3716-3722.

5. Tibor LM, Liebert G, Sutter R,Impellizzeri FM, Leunig M: Two or moreimpingement and/or instability deformitiesare often present in patients with hip pain.Clin Orthop Relat Res 2013;471:3762-3773.

6. Thawrani DP, Feldman DS, Sala DA: Notall hip dysplasias are the same: PreoperativeCT version study and the need for reversebernese periacetabular osteotomy. J PediatrOrthop 2017;37:47-52.

7. Bedi A, Dolan M, Magennis E, Lipman J,Buly R, Kelly BT: Computer-assistedmodeling of osseous impingement andresection in femoroacetabularimpingement. Arthroscopy 2012;28:204-210.

8. Fabricant PD, Bedi A, De La Torre K, KellyBT: Clinical outcomes after arthroscopicpsoas lengthening: The effect of femoralversion. Arthroscopy 2012;28:965-971.

9. Siebenrock KA, Steppacher SD, Haefeli PC,Schwab JM, Tannast M: Valgus hip withhigh antetorsion causes pain throughposterior extraarticular FAI. Clin OrthopRelat Res 2013;471:3774-3780.

10. Putz C, Wolf SI, Geisbusch A, Niklasch M,Doderlein L, Dreher T: Femoral derotationosteotomy in adults with cerebral palsy.Gait Posture 2016;49:290-296.

11. DolanMM, Heyworth BE, Bedi A, Duke G,Kelly BT: CT reveals a high incidence ofosseous abnormalities in hips with labraltears. Clin Orthop Relat Res 2011;469:831-838.

12. Stahl JP, Alt V, Kraus R, Hoerbelt R,ItomanM, Schnettler R: Derotation of post-traumatic femoral deformities by closedintramedullary sawing. Injury 2006;37:145-151.

13. Chapman ME, Duwelius PJ, Bray TJ,Gordon JE: Closed intramedullary femoralosteotomy: Shortening and derotation

procedures. Clin Orthop Relat Res 1993:245-251.

14. Winquist RA: Closed intramedullaryosteotomies of the femur. Clin OrthopRelat Res.1986:155-164.

15. Tonnis D, Heinecke A: Acetabular andfemoral anteversion: Relationship withosteoarthritis of the hip. J Bone Joint SurgAm 1999;81:1747-1770.

16. Eckhoff DG: Effect of limb malrotation onmalalignment and osteoarthritis. OrthopClin North Am 1994;25:405-414.

17. Terjesen T, Benum P, Anda S, SvenningsenS: Increased femoral anteversion andosteoarthritis of the hip joint. Acta OrthopScand 1982;53:571-575.

18. Scheys L, Spaepen A, Suetens P, Jonkers I:Calculated moment-arm and muscle-tendon lengths during gait differsubstantially using MR based versusrescaled generic lower-limbmusculoskeletalmodels. Gait Posture 2008;28:640-648.

19. Gomez-Hoyos J, Schroder R, Reddy M,Palmer IJ, Martin HD: Femoral neckanteversion and lesser trochantericretroversion in patients with ischiofemoralimpingement: A case-control magneticresonance imaging study. Arthroscopy2016;32:13-18.

20. Eckhoff DG, Montgomery WK, KilcoyneRF, Stamm ER: Femoral morphometry andanterior knee pain. Clin Orthop Relat Res1994:64-68.

21. Bruce WD, Stevens PM: Surgical correctionof miserable malalignment syndrome. JPediatr Orthop 2004;24:392-396.

22. MacWilliams BA, McMulkin ML, DavisRB, Westberry DE, Baird GO, Stevens PM:Biomechanical changes associated withfemoral derotational osteotomy. GaitPosture 2016;49:202-206.

23. Moya LE, Buly RL, Henn RF, Kelly BT, MaY, Molisani D: Femoral retroversion inpatients with femoroacetabularimpingement: A cofactor in thedevelopment of hip osteoarthritis. J BoneJoint Surg Br 2010;92(suppl IV):526.

24. Gelberman RH, Cohen MS, Shaw BA,Kasser JR, Griffin PP, Wilkinson RH: Theassociation of femoral retroversion withslipped capital femoral epiphysis. J BoneJoint Surg Am 1986;68:1000-1007.

25. Steppacher SD, Albers CE, Siebenrock KA,Tannast M, Ganz R: Femoroacetabularimpingement predisposes to traumaticposterior hip dislocation. Clin OrthopRelat Res 2013;471:1937-1943.

26. Canham CD, Yen YM, Giordano BD: Doesfemoroacetabular impingement cause hipInstability? A systematic review.Arthroscopy 2016;32:203-208.

27. Buly R: Femoral deformities and hiposteotomies, in Anil Ranawat M, BryanKelly M, eds: Musculoskeletal

Robert L. Buly, MD, MS, et al

Month 2018, Vol 00, No 00 9

Examination of the Hip and Knee.Thorofare, NJ, SLACK Incorporated,2010, pp 252–273.

28. Murphy SB, Simon SR, Kijewski PK,Wilkinson RH, Griscom NT: Femoralanteversion. J Bone Joint Surg Am 1987;69:1169-1176.

29. Ganz R, Parvizi J, Beck M, Leunig M,Notzli H, Siebenrock KA:Femoroacetabular impingement: A causefor osteoarthritis of the hip. Clin OrthopRelat Res 2003:112-120.

30. Jamali AA, Mladenov K, Meyer DC, et al:Anteroposterior pelvic radiographs toassess acetabular retroversion: Highvalidity of the “cross-over-sign”. J OrthopRes 2007;25:758-765.

31. Steppacher SD, Zurmuhle CA, Puls M,et al: Periacetabular osteotomy restoresthe typically excessive range of motion indysplastic hips with a spherical head.Clin Orthop Relat Res 2015;473:1404-1416.

32. Kraeutler MJ, Garabekyan T, Pascual-Garrido C, Mei-Dan O: Hip instability: A

review of hip dysplasia and othercontributing factors. Muscles LigamentsTendons J 2016;6:343-353.

33. Ejnisman L, Philippon MJ, Lertwanich P,et al: Relationship between femoralanteversion and findings in hips withfemoroacetabular impingement.Orthopedics 2013;36:e293-e300.

34. Gomez-Hoyos J, Schroder R, Reddy M,Palmer IJ, Khoury A, Martin HD: Isthere a relationship between psoasimpingement and increased trochantericretroversion? J Hip Preserv Surg 2015;2:164-169.

35. Kain MS, Novais EN, Vallim C, Millis MB,Kim YJ: Periacetabular osteotomy afterfailed hip arthroscopy for labral tears inpatients with acetabular dysplasia. J BoneJoint Surg Am 2011;93(suppl 2):57-61.

36. Reynolds D, Lucas J, Klaue K: Retroversionof the acetabulum: A cause of hip pain. JBone Joint Surg Br 1999;81:281-288.

37. Maruyama M, Feinberg JR, Capello WN,D’Antonio JA: The Frank StinchfieldAward: Morphologic features of the

acetabulum and femur: Anteversion angleand implant positioning. Clin Orthop RelatRes 2001;52-65.

38. Hartel MJ, Petersik A, Schmidt A, et al:Determination of femoral neck angle andtorsion angle utilizing a novel three-dimensional modeling and analyticalTechnology based on CT datasets. PLoSOne 2016;11:e0149480.

39. Sugano N, Noble PC, Kamaric E: Acomparison of alternative methods ofmeasuring femoral anteversion. J ComputAssist Tomogr 1998;22:610-614.

40. Kamath AF, Ganz R, Zhang H, GrappioloG, Leunig M: Subtrochanteric osteotomyfor femoral mal-torsion through a surgicaldislocation approach. J Hip Preserv Surg2015;2:65-79.

41. Pailhe R, Bedes L, Sales de Gauzy J, Tran R,Cavaignac E, Accadbled F: Derotationalfemoral osteotomy technique with lockingnail fixation for adolescent femoralantetorsion: Surgical technique andpreliminary study. J Pediatr Orthop B2014;23:523-528.

Femoral Derotation Osteotomy in Adults

10 Journal of the American Academy of Orthopaedic Surgeons