Original ArticleArticle original

AN ORTHOPEDIC SURGEON SURVEY ON THE TREATMENTOF DISPLACED FEMORAL NECK FRACTURE: OPPOSING VIEWS

David Chua, MD;*‡¶ Susan B. Jaglal, PhD;†‡§ Joseph Schatzker, MD*‡§

From the *Department of Surgery and †Department of Preventive Medicine and Biostatistics, University of Toronto, Toronto, Ont.

‡M.E. Müller Program in Research, Sunnybrook Health Science Centre, University of Toronto

§Musculoskeletal Health Status Working Group, Toronto

¶Clinical fellow, supported by the Hospital Manpower Development Program for Advanced Orthopaedic Trauma, Singapore

Accepted for publication Dec. 12, 1996

Correspondence to: Dr. Susan Jaglal, M.E. Muller Program in Research, Ste. D-514, Sunnybrook Health Science Centre, 2075 Bayview Ave., Toronto ON M4N 3M5; tel 416 480-6100 x2641; fax 416 480-5886

© 1997 Canadian Medical Association (text and abstract/résumé)

OBJECTIVE: To examine the reasons for practice variation in the treatment of displaced femoral neck fractures.DESIGN: A survey, asking surgeons to choose either hemiarthroplasty or internal fixation for 2 different fe-male patients with a displaced femoral neck fracture.SETTING: The Canadian Orthopaedic Association Meeting, Halifax, May 1995.PATIENTS: The scenario in the first patient was of an independent 70-year-old woman with no pre-existing med-ical conditions. The scenario in the second patient was of a housebound 84-year-old woman with co-morbidity.MAIN OUTCOME MEASURES: Proportion of surgeons choosing either hemiarthroplasty or internal fixationfor each case scenario. Distribution of reasons to explain the treatment decision.RESULTS: Ninety-nine surgeons responded. For the case of the 70-year-old woman, 47% chose hemiarthro-plasty and 53% chose internal fixation (p = 0.60), and for the 84-year-old woman, 96% chose hemiarthro-plasty. These findings were consistent within the subgroups of teaching surgeons and community practicesurgeons. Surgeons with 10 years or less of practice tended to favour hemiarthroplasty whereas those withmore than 15 years’ practice favoured internal fixation. Important reasons for treatment choice were avoid-ance of reoperation in the hemiarthroplasty group (85%) and better hip function in the fixation group(83%), durability (83%) and ease of revision (77%).CONCLUSION: The surgeon’s interpretation of the importance of reoperation and function underlies thedifferences in treatment decision regarding the management of femoral neck fractures in elderlypatients.

OBJECTIF : Analyser les raisons pour lesquelles la méthode de réduction des fractures du col du fémur avecdéplacement varie dans la pratique.CONCEPTION : Sondage dans le cadre duquel on a demandé aux chirurgiens de choisir l’hémiarthroplastie oula fixation interne chez deux patientes différentes qui ont d’une fracture du col du fémur avec déplacement.CONTEXTE : Réunion de l’Association canadienne d’orthopédie, Halifax, mai 1995.PATIENTES : Dans le premier cas, le scénario était celui d’une femme autonome de 70 ans qui n’avait aucunproblème médical antérieur. Dans le deuxième, le scénario était celui d’une femme de 84 ans confinée à lamaison et qui avait une comorbidité.PRINCIPALES MESURES DES RÉSULTATS : Proportion des chirurgiens qui ont choisi soit l’hémiarthroplastie,soit la fixation interne, dans chaque cas. Répartition des raisons justifiant la décision.RÉSULTATS : Quatre-vingt dix-neuf chirurgiens ont répondu. Dans le cas de la femme de 70 ans, 47 % ontchoisi l’hémiarthroplastie et 53 %, la fixation interne (p = 0,60). Dans celui de la femme de 84 ans, 96 %ont choisi l’hémiarthroplastie. Ces résultats étaient uniformes dans les sous-groupes de chirurgiens qui en-seignaient et ceux qui pratiquaient dans la communauté. Les chirurgiens comptant 10 ans ou moins de

14846 August/97 CJS /Page 271

CJS, Vol. 40, No. 4, August 1997 271

It is generally accepted that dis-placed fractures of the femoralneck in young adults are best

managed by primary internal fixa-tion.1,2 Because of the high rate ofnonunion in patients older than 85years, this fracture is most often treatedby hemiarthroplasty.3,4 But how tomanage the majority of patients be-tween the ages of 65 and 85 years iscontroversial.5,6 Since 1975, over 100studies have examined the treatmentof femoral neck fractures.7 Of these,only a handful have been randomizedtrials or comparative studies and nonehave provided conclusive evidence onwhich method of treatment is appro-priate for individual cases.8–15

The extent of treatment variationfor femoral neck fractures is exempli-fied in a number of studies. In the ME-DOS study,16 a prospective multicentrestudy of the incidence of hip fracturesin 6 Mediterranean countries, therewas wide variation between centres inthe use of arthroplasty, ranging from9.5% in rural Turkey to 83% inToulouse, France. In a recent studycomparing treatment in The Nether-lands and Sweden, 94% of displacedfemoral neck fractures were treatedwith hemiarthroplasty in The Nether-lands, whereas practically all weretreated with internal fixation in Swe-den.17 In Ontario there was a 9-foldvariation (9% to 83%) in the use ofhemiarthroplasty among countieswhere at least 1 femoral neck fracturewas treated per month.18

The main purpose of this studywas to examine the reasons con-tributing to practice variation in thetreatment of displaced femoral neck

fractures. We also examined patientand clinical factors that may be im-portant in decision-making.

METHODS

Sample

The target population was Cana-dian orthopedic surgeons. To obtain alarge enough sample, a questionnairewas included in the registration pack-age of those who attended the 1995Canadian Orthopaedic Association an-nual meeting in Halifax. In total, 320packages were distributed at the con-ference. To maintain anonymity andimpartiality in responses, we purposelyomitted any tracking mechanisms inthe survey form. Therefore follow-upwas not possible, and participation wasentirely voluntary.

Survey

The questionnaire polled orthope-dic surgeons on their preferred treat-ment: either hemiarthroplasty or in-ternal fixation for 2 distinctly differentpatients with a fractured hip. The firstpatient was a 70-year-old woman withno pre-existing medical conditionsand a Garden grade 3 fracture of thefemoral neck. She was independentbefore her fracture and had no symp-toms of joint pain or stiffness. The sec-ond patient was an 84-year-oldwoman who had mild hypertensionand diabetes and a Garden grade 3(Fig. 1) fracture. She was houseboundand a bit confused.Surgeons were also asked to select

reasons for their treatment choice and

state whether they would change theirtreatment decision if the fracture wasGarden grade 4.We developed the reasons for treat-

ment choice from the literature andfrom the clinical impressions of seniorsurgeons. We also asked surgeons toidentify general patient and local fac-tors that they considered were impor-tant in deciding on either treatment.Demographic information regard-

ing frequency of performing hip frac-ture surgery, numbers of years in prac-tice and size and type of institutionwas also collected.

Analysis

Frequency distributions were cal-culated for each clinical scenario anddemographic information. To exam-ine reasons for choosing hemiarthro-plasty, only surgeons who indicatedthis as their treatment choice wereincluded in the denominator. Thiswas then repeated for surgeonschoosing internal fixation. Resultswere also stratified by the number ofyears in practice and whether the in-stitution of practice was teaching orcommunity.The entire sample was used to

analyse characteristics identified bysurgeons as being important in mak-ing a treatment decision for hemi-arthroplasty or internal fixation. Allanalyses were performed using the sta-tistical package STATA, version 4.0(Stata Corp., College Station, Tex.).Differences in the results were

analysed by the Student’s t-test. Aprobability value of less than 0.05 wasconsidered significant.

CHUA, JAGLAL, SCHATZKER

14846 August/97 CJS /Page 272

272 JCC, Vol. 40, No 4, août 1997

pratique ont eu tendance à privilégier l’hémiarthroplastie, tandis que ceux qui avaient plus de 15 ans depratique ont privilégié la fixation interne. Les justifications importantes du traitement choisi ont été la pos-sibilité d’éviter une nouvelle intervention chez les sujets qui ont subi une hémiarthroplastie (85 %) et unmeilleur fonctionnement de la hanche chez tous ceux qui ont subi une fixation (83 %), la durabilité (83 %)et la facilité de révision (77 %).CONCLUSION : L’interprétation que le chirurgien a fait de l’importance d’une nouvelle intervention et dufonctionnement sous-tend les différences dans le choix de la méthode de réduction des fractures du col dufémur chez les patients âgés.

RESULTS

Response rate

Overall, 99 surgeons responded tothe survey (Table I), for a responserate of 31%. Ninety-four percent ofthe respondents treated more than 10hip fractures per year, and the mediannumber of years in practice was 11.Approximately 61% of the respon-dents were associated with a teachinginstitution.

Case scenarios

There was almost universal agree-ment (95%) that the 84-year-oldwoman should be treated by hemi-arthroplasty.The results for the 70-year-old

woman were more interesting: 46%chose a hemiarthroplasty and 54%chose internal fixation. When strati-fied by hospital affiliation, no signifi-cant difference (p = 0.60) in treatmentchoice was observed among surgeonsfrom teaching and community hospi-tals (Table II). In addition, the 24%increase in use of hemiarthroplasty fora similar patient with a Garden grade4 fracture is entirely due to surgeonswho initially chose internal fixation forthe Garden grade 3 fracture andchanged their treatment choice tohemiarthroplasty.

Treatment choice by years inpractice

When treatment choice by years inpractice was examined for the 70-year-old woman, a clear pattern emerged(Fig. 2). Those in practice less than 10years tended to favour hemiarthro-plasty, whereas those in practice morethan 15 years favoured internal fixa-tion. However, the choice was evenlysplit among those in practice for 11 to15 years.

Reasons for treatment choice

The most frequently quoted reasonfor choosing a hemiarthroplasty for the70-year-old patient was “it is less likelyto result in a second operation” (85%).Early mobilization (66%) and earlyfunction and pain relief (64%) also fig-ured highly in this decision (Table III).Two reasons for choosing internal fixa-tion for the 70-year-old woman ratedhighly: “hip function after successful fix-ation is better” (83%) and durability ofa successful result (83%). Ease of revi-sion to a hemiarthroplasty or total hipreplacement was also important (77%).Although 95% chose a hemiarthro-

plasty for the 84-year-old patient, thereasons for choosing this procedurediffered, depending on whether thesurgeon chose hemiarthroplasty or in-ternal fixation for the 70-year-old pa-

DISPLACED FEMORAL NECK FRACTURE

14846 August/97 CJS /Page 273

CJS, Vol. 40, No. 4, August 1997 273

FIG. 1. Garden classification, which is based on the relationship of the medial trabeculae (compressiontrabeculae) in the femoral head and pelvis. I = grade 1, incomplete fracture; II = grade 2, completefracture not displaced; III = grade 3, complete fracture displaced less than 50%; IV = grade 4, completefracture displaced more than 50%. (Reprinted from Schatzker J, Tile M, editors. The Rationale of Opera-tive Fracture Care, 2nd ed, Fig. 14.6, page 330, 1996, by permission of Springer-Verlag, New York.)

Table I

1–5 26 (27)

6–10 19 (20)

11–15 12

Characteristics of Responding Surgeons

(12)

16–20 11 (11)

> 20

Characteristic

29 (30)

No. of hip fractures treatedper year*

*94 respondents †97 respondents

< 10

> 10

Hospital affiliation*

Teaching 57

88

6

No. (%)

(61)

(94)

(6)

Large community 22 (23)

Small community 15 (16)

Years in practice†

tient. Of the surgeons who chosehemiarthoplasty for the 70-year-oldpatient, 71% indicated that the reasonthis was also indicated for the 84-year-old patient was that “early functionand pain relief are better” with hemi-arthroplasty. On the other hand, ofthose who chose internal fixation forthe 70-year-old patient and hemi-arthroplasty for the 84-year-old pa-tient, only 43% indicated that “earlyfunction and pain relief are better” was

an important reason for choosing thisprocedure in the 84-year-old patient.

Patient factors

Irrespective of the treatment choicefor the 70-year-old patient, surgeonstended to agree on the importance ofgood pre-fracture walking ability(73%) and independence (71%), pres-ence of comorbidity (77%) and de-mentia (74%) as important prognostic

indicators. The other factors that werealso rated highly were osteoporosis(63%) and comminution of the frac-ture (63%).

DISCUSSION

There was almost unanimousagreement that the 84-year-old pa-tient should be treated with a hemi-arthroplasty. However, there was aclear difference in opinion for thetreatment of the 70-year-old patient,with a virtual 50–50 split. Surgeonswith 10 years or less of practice tendedto favour hemiarthroplasty whereasthose with more than 15 years of prac-tice favoured internal fixation.Approximately 24% of surgeons

who chose internal fixation for theGarden grade 3 fracture would changeto a hemiarthroplasty if faced with aGarden grade 4 fracture. This impliesa belief that there is a significant dif-ference in the rate of failure betweenthe two grades of fracture after inter-nal fixation. Yet the published 2-yearrates of failure in women for Gardengrades 3 and 4 femoral neck fracturesare 29% and 35% respectively.3

The fact that there is a significantreoperation rate (30%) associated withinternal fixation, is well supported inthe literature.3,7 For the surgeons pre-ferring internal fixation, the choicewould seem to be for the hope of su-perior hip function and durability,trading this for an inevitably high re-operation rate within the first 2 years.Conversely, the important considera-tion for surgeons preferring hemi-arthroplasty seems to be the avoidanceof early reoperation within the first 2years. The trade-off is the need to re-vise a proportion of these proceduresto total joint arthroplasty some 5 to10 years later because of pain or loos-ening of the prosthesis.19–21 Given that30% of hip fracture patients die withinthe first year after fracture, this signifi-

CHUA, JAGLAL, SCHATZKER

14846 August/97 CJS /Page 274

274 JCC, Vol. 40, No 4, août 1997

Table II

Garden grade

3 95 5 0.212

Treatment Choice for Each Case Scenario by Garden Grade and Hospital Affiliation

4 99 1

Hospital affiliation

Case/Garden grade/affiliation

70-year-old woman

Teaching

Garden grade

96 4

3

0.738

Community

4

94 6

Hospital affiliation

70

46

Hemiarthroplasty,% of respondents

30

54

Internal fixation,% of respondents

0.001

p value

Teaching 44 56 0.571

Community 49 51

84-year-old woman

0–5 5–10 11–15 16–20 >20

0

10

20

30

40

50

60

70

%

Years

FIG. 2. Treatment choice of surgeons for the 70-year-old patient by number of years in practice.White bars = hemiarthroplasty, black bars = internal fixation.

cantly reduces the number requiring alate revision procedure.Almost all surgeons agreed on the

general patient and pathological fac-tors that should be considered in man-aging femoral neck fractures. Yet, ourfindings imply that 2 surgeons seeingthe same patient, having the sameknowledge base and considering thesame factors may make entirely differ-ent treatment choices. Although thereis agreement on these factors, sur-geons differ in their interpretation ofhow these factors influence outcomein a clinical setting.The limitations of this study need

to be addressed. First, the generaliz-ability of our findings is limited to theopinions of orthopedic surgeons andnot general surgeons, who may carefor some of these patients. Second, wenote that 61% of respondents werefrom teaching institutions. Since therewas no correlation between practicesetting and treatment choice, our re-sults should be generalizable to all or-thopedic surgeons treating hip frac-ture patients. Third, a relatively lowresponse rate of 35% was achieved.However, 93% of respondents treatedat least 10 femoral neck fractures per

year, so they are surgeons who are wellacquainted with this injury. A re-sponse rate of 30% is standard for asurvey with no “follow up” of respon-dents.22 We intentionally decided toforgo follow-up because absoluteanonymity in respondents was impor-tant in preserving data integrity.In conclusion, a major factor dri-

ving surgeon preference seems to bethe view of the impact of reoperationon these patients. Those preferringhemiarthroplasty for the 70-year-oldpatient saw avoidance of reopera -tion as an important consideration,whereas those preferring internal fix-ation did not view reoperation withsuch negativity. The phys ical functionand overall clinical and radiographicresults of these revised cases alsoneed to be addressed in anotherstudy. This study underscores theneed for a randomized trial to deter-mine whether internal fixation orhemiarthroplasty is the optimal treat-ment for displaced femoral neck frac-tures in patients between the ages of65 and 79 years.

The authors thank the Canadian OrthopaedicAssociation for assistance with this survey.

References

1. Swiontkowski MF, Winquist RA,Hansen ST Jr. Fractures of thefemoral neck in patients between theages of twelve and forty-nine years. JBone Joint Surg [Am] 1984;66(6):837-46.

2. Robinson CM, Court-Brown CM,McQueen MM, Christie J. Hip frac-tures in adults younger than 50 yearsof age. Epidemiology and results.Clin Orthop 1995;312:238-46.

3. Barnes R, Brown JT, Garden RS,Nicoll EA. Subcapital fractures of thefemur. A prospective review. J BoneJoint Surg [Br] 1976;58:2-24.

4. Bentley G. Treatment of nondis-placed fractures of the femoral neck.Clin Orthop 1980;152:93-7.

5. Robinson CM, Saran D, Annan IH. In-tracapsular hip fractures. Results ofmanagement adopting a treatment pro-tocol. Clin Orthop 1994;302:83-91.

6. Zuckerman JD. Hip fracture [review;see comments]. N Engl J Med 1996;334(23):1519-25. Comments in: N Engl J Med 1996;334(26): 1994-6.

7. Lu-Yao GL, Keller RB, Littenberg B,Wennberg JE. Outcomes after dis-placed fractures of the femoral neck. JBone Joint Surg [Am] 1994;76:15-25.

8. Söreide O, Molster A, Raugstad TS.Internal fixation versus primary pros-thetic replacement in acute femoralneck fractures: a prospective, ran-domized clinical study. Br J Surg1979; 66:56-60.

9. Sikorski JM, Barrington R. Internalfixation versus hemiarthroplasty forthe displaced subcapital fracture ofthe femur. A prospective randomisedstudy. J Bone Joint Surg [Br] 1981;63(3):357-61.

10. Broos PL, Stappaerts KH, Luiten EJ,Gruwez JA. Endoprosthesis. The bestway to treat unstable intracapsularhip fractures in elderly patients. Un-fallchirurg 1987;90:347-50.

11. Rodriguez J, Herrara A, Canales V,Serrano S. Epidemiologic factors,morbidity and mortality after femoralneck fractures in the elderly. A com-parative study: internal fixation vs.hemiarthroplasty. Acta Orthop Belg1987;53:472-9.

DISPLACED FEMORAL NECK FRACTURE

14846 August/97 CJS /Page 275

CJS, Vol. 40, No. 4, August 1997 275

Table III

Distribution of Reasons for Treatment Decision for the Independent 70-Year-Old Patient WithNo Comorbidity

Reasons for choosinghemiarthroplasty

Less likely to result in asecond operation

Early mobilization easier

Earlier function andbetter pain relief

Operation morepredictable

Personal results arebetter

38

60

64

66

85

% ofrespondents

Lower morbidity andmortality

Less physiological insultto the patient

Ease of revision if needed

More lasting result

Better hip function

Reasons for choosinginternal fixation

38

54

77

83

83

% of respondents

Not convinced of adefinite advantage of oneprocedure over the other

20 Not convinced of adefinite advantage of oneprocedure over the other

27

CHUA, JAGLAL, SCHATZKER

14846 August/97 CJS /Page 276

276 JCC, Vol. 40, No 4, août 1997

12. Skinner P, Riley D, Ellery J, Beau-mont A, Coumine R, Shafighian B.Displaced subcapital fractures of thefemur: a prospective randomizedcomparison of internal fixation, hemi-arthroplasty and total hip replace-ment. Injury 1989;20(5):291-3.

13. Gebhard JS, Amstutz HC, Zinar DM,Dorey FJ. A comparison of total hiparthroplasty and hemiarthroplasty oftreatment of acute fracture of thefemoral neck. Clin Orthop 1992;282:123-31.

14. Parker MJ. Internal fixation orarthroplasty for displaced subcapitalfractures in the elderly? Injury 1992;23:521-4.

15. van Vugt AB, Oosterwijk WM, GorisRJ. Osteosynthesis versus endopros-thesis in the treatment of unstable in-tracapsular hip fractures in the elderly.Arch Orthop Trauma Surg 1994;113:39-45.

16. Lyritis GP, Johnell O. Orthopaedicmanagement of hip fracture. Bone1993;14:S11-S17.

17. Berglund-Rödén M, Swierstra BA,Wingstrand H, Thorngren K.Prospective comparison of hip frac-ture treatment. 856 cases followed for4 months in The Netherlands andSweden. Acta Orthop Scand 1994;65(3): 287-94.

18. Jaglal SB, Chua D, Schatzker J. Tem-poral trends and geographic varia-tions in surgical fractures. J Trauma.In press.

19. D’Arcy J, Devas M. Treatment offractures of the femoral neck by re-placement with the Thompson pros-thesis. J Bone Joint Surg [Br] 1976;58: 279-86.

20. Koefed H, Kofod J. Moore prosthesisin the treatment of fresh femoral neckfractures: a critical review with specialattention to secondary acetabular de-generation. Injury 1983;14: 531-4.

21. Phillips TW. The Bateman bipolarfemoral head replacement: a fluoro-scopic study of movement over afour-year period. J Bone Joint Surg[Br] 1989;69:761-4.

22. Choi BC, Pak AW, Purdham JT. Ef-fects of mailing strategies on responserate, response time, and cost in aquestionnaire study among nurses.Epidemiology 1990;1:72-4.