Injury, Int. J.

Care Injured (2006) 37, 194—202

www.elsevier.com/locate/injury

Salvage of failed dynamic hip screw fixation

of intertrochanteric fractures
G.Z. Said, O. Farouk *, A. El-Sayed, H.G. Said

Department of Orthopaedic Surgery and Traumatology, Assiut University Hospitals,

71516 Assiut, Egypt

Accepted 19 September 2005

KEYWORDS Summary Twenty-six patients with failed dynamic hip screw fixation of intertro-
Failed DHS; chanteric fractures were included in this study. The mean age of the patients was 61
Intertrochanteric years (range, 38—84 years). Average limb shortening was 2.4 cm; 18 patients were
fracture; treated with revision internal fixation and eight patients with prosthetic replace-
Trochanteric nonunion ment. The decision depended on the physiological age of the patient, quality of bone,
and condition of the femoral head and the acetabulum. The revision internal fixation
group included DHS reinsertion in eight patients, valgus osteotomy and revision DHS
fixation in six, while four patients were treated by valgus osteotomy and insertion of
single-angled 1308 plate. The prosthetic replacement group included cemented
Thompson endoprothesis in five patients and cemented total hip arthroplasty in
three. The mean follow-up period was 31 months (range, 15—72 months). All patients
of revision internal fixation group achieved fracture healing without bone grafting.
Time to union averaged 17 weeks. Average gain in length was 2 cm Avascular necrosis
of the femoral head occurred in one patient. Six patients of the prosthetic replace-
ment group achieved good functional outcome and pain-free gait. The remaining two
had unsatisfactory result.
# 2005 Elsevier Ltd. All rights reserved.

Introduction lag screw cutout, or excessive lag screw sliding with

medialisation of the distal fracture fragment.
Primary insertion of dynamic hip screw (DHS) for Buciuto et al.2 reported significant technical fail-
trochanteric fractures is not always successful, ures in their series. Mechanical stability after inter-
especially in unstable fractures.3,6 Madsen et al.9 nal fixation is dependent on the quality of bone,
reported significant secondary fracture displace- fracture personality, quality of reduction and choice
ment in 34% of cases, leading to a varus malunion, of the implant. Implant placement in the biomecha-
nically ideal position, however, is probably the most
* Corresponding author. Tel.: +20 12 2443531;
important factor.8 The aim of this study is to report
fax: +20 88 2333327. on surgical salvage for failed DHS fixation of inter-
E-mail address: osama_farouk@yahoo.com (O. Farouk). trochanteric fractures.

0020–1383/$ — see front matter # 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.injury.2005.09.011
Salvage of failed dynamic hip screw fixation 195

Patients and methods ii. Functional assessment was done using subjective
and objective information, based on pain, limb
Between January 1998 and January 2004, twenty-six shortening, and walking ability.
patients with failed DHS fixation of intertrochan-
teric fracture were referred to our institution. All
patients were preoperatively assessed, surgically Results
treated and followed up by the authors.
Details of the patients are shown in Table 2. The
Inclusion criteria patients were aged 38—84 years (mean 61 years)
with a male to female ratio of 9—17. The mean
Early and late DHS failures with un-united intertro- period of follow-up was 31 months (range, 15—72
chanteric fractures in adults and elderly patients months). Eight fractures were stable type A1 and 18
were included. The exclusion criteria were active were unstable (type A2 in 15 and type A3 in three)
infection and united fractures with accepted func- according to AO fracture classification.
tional outcome.
All patients were evaluated both clinically and DHS failure
radiologically. Fractures were classified according
to the AO classification.11 Causes of DHS failure were Time between initial surgery and DHS failure ranged
analysed. between 2 and 52 weeks and averaged 22 weeks.
Causes of DHS failure were inadequate placement of
Treatment groups the DHS in eight patients, fracture instability with
secondary varus displacement in 10 patients and
The patients were sorted under four subgroups severe osteoporosis of the femoral head in eight
according to the method of surgical treatment patients.
applied (Table 1).
In the revision internal fixation group, the Fracture union
implant was inserted to near the subchondral bone
for good purchase. In one patient of group I, in All patients in groups I and II achieved fracture
addition to reinsertion of DHS, a narrow six-hole union. The osteotomy also united in all patients
DCP was laterally mounted like a trochanteric sta- of group II. The mean time to union was 17 weeks
bilizing plate to prevent medialisation of the shaft (range, 14—24 weeks). Bone grafting was not
(Fig. 1). In this case, a 1-cm shorter DHS screw was needed in any case. All had satisfactory radiological
introduced deep into the neck to allow for the result (Figs. 1 and 2) without any implant failure.
sliding mechanism inside the plate barrel. Bone
grafting was not added in any patient. Functional outcome

Assessment 1. Pain: Twenty patients were pain free at the last

follow up. Four patients from group II had occa-
i. Fracture union in groups I and II was considered sional hip pain that did not interfere with their
radiographically if callus formation was seen in daily activities. The remaining two patients were
three of four cortices on anteroposterior and from group III, and had persistent hip and ante-
lateral views. Clinical union was considered rior thigh pain.
when there was painless hip range of movement 2. Limb shortening: Preoperative average limb
and painless full weight bearing. shortening measured 2.4 cm (range, 1—6 cm).

Table 1 The four groups of patients according to the type of surgery

Surgical group Number Indications
I. Repeat DHS 8 Short screw, good bone stock
II. Subtrochanteric valgus osteotomy 10 Varus neck
DHS re-fixation 6 Good bone stock of femoral head
Single angled 1308 plate 4 Good bone stock in infero-medial quadrant
III. Haemiarthroplasty 5 Excavated femoral head
IV. Total hip replacement 3 Osteoporosis, acetabular damage
196 G.Z. Said et al.

Figure 1 (a) A 38-year-old man (case no. 1) presented 8 months after DHS operation. Note translucency around the
screw and barrel, excessive backing of the screw and medialisation. (b) Lateral view confirmed poor fracture reduction
and non-union. (c and d) The fracture united after reduction of the fracture fragments and reinsertion of DHS to near the
subchondral bone. A narrow DCP was put as a trochanter-stabilizing plate.
 Salvage of failed dynamic hip screw fixation
Table 2 Summary of patients’ data, pre-operative clinical evaluation, surgical procedure and outcome of all patients
Patients Initial DHS fixation Pre-operative clinical evaluation Fracture Surgical Complications Functional outcome
type (AO procedure
No. Age Sex Time to Cause of Limb Femoral Acetabular classification) Pain Walking
failure failure shortening head state
(weeks) (cm)
1 38 Male 32 Mal-placed DHS 1.5 Good Good A2 DHS reinsertion Free Without
support
2 55 Female 22 Unstable fracture 3 Good Good A2 DHS reinsertion + Free Without
valgus osteotomy support
3 55 Female 16 Unstable fracture 2.5 Good Good A2 DHS reinsertion + Occasional Without
valgus osteotomy support
4 40 Female 12 Mal-placed DHS 1.5 Good Good A3 DHS reinsertion Free Without
support
5 58 Male 18 Unstable fracture 6 Good Good A2 DHS reinsertion + Free Without
valgus osteotomy support
6 42 Male 2 Mal-placed DHS 1 Good Good A2 DHS reinsertion Free Without
support
7 70 Female 24 Femoral head 1.5 Excavated Good A1 Thompson Free Without
osteoporosis prosthesis support
8 60 Female 20 Unstable fracture 3.5 Good Good A2 DHS reinsertion + Free Without
valgus osteotomy support
9 75 Female 20 Femoral head 2 Excavated Good A1 Thompson Free One-arm
osteoporosis prosthesis support
10 64 Male 16 Unstable fracture 3 Good Good A2 1308 Plate Free Without
inferomedial + valgus support
quadrant osteotomy
11 48 Male 12 Mal-placed DHS 1.5 Good Good A3 DHS reinsertion Free Without
support
12 80 Female 32 Femoral head 2 Excavated Abraded A1 Total hip Free One-arm
osteoporosis replacement support
13 50 Female 12 Mal-placed DHS 1.5 Good Good DHS reinsertion Free Without
support
14 84 Female 40 Femoral head 2 Excavated Severe A1 Total hip Free One-arm
osteoporosis osteoporosis replacement support
15 52 Female 16 Mal-placed DHS 1.5 Good Good A2 DHS reinsertion Free Without
support
16 62 Male 24 Unstable fracture 3 Good Good A2 1308 Plate + Free Without
infero-medial valgus support
quadrant osteotomy
17 43 Female 14 Mal-placed DHS 2 Good Good A3 DHS reinsertion Free Without

197
support
 198
Table 2 (Continued )
Patients Initial DHS fixation Pre-operative clinical evaluation Fracture Surgical Complications Functional outcome
type (AO procedure
No. Age Sex Time to Cause of Limb Femoral Acetabular classification) Pain Walking
failure failure shortening head state
(weeks) (cm)
18 72 Male 36 Femoral head 1.5 Excavated Osteoporosis A1 Thompson Persistent Indoor +
osteoporosis prosthesis walker
19 55 Female 16 Mal-placed DHS 2 Good Good A2 DHS reinsertion Free Without
support
20 75 Female 30 Femoral head 2 Excavated Osteoporosis A1 Thompson Free One-arm
osteoporosis prosthesis support
21 68 Female 28 Femoral head 3 Excavated Good A1 Thompson Periprosthetic Persistent Indoor +
osteoporosis prosthesis fracture walker
22 66 Male 20 Unstable 4 Good Good A2 1308 Plate Avascular Occasional Without
fracture infero-medial + valgus necrosis support
quadrant osteotomy of femoral
head
23 82 Female 52 Femoral head 3 Excavated Severe A1 Total hip DVT Free One-arm
osteoporosis osteoporosis replacement support

G.Z. Said et al.

Salvage of failed dynamic hip screw fixation 199

Figure 2 (a) A 58-year-old man (case no. 5) had DHS fixation for trochanteric fracture. The DHS lag screw cutout. The
DHS was removed. (b) The patient presented to our institution 6 weeks later with coxa vara, full external rotation and
6 cm shortening. (c) Reinsertion of DHS and valgus osteotomy was done after correction of external rotation and
shortening. (d) Follow-up 10 months postoperatively with complete union of the fracture.

Average gain in length was 2 cm (range, 1—4 cm) Complications

postoperatively. Residual shortening of 2 cm
remained in two patients. Two patients had post-operative complications. One
3. Walking ability: All patients of groups I and II had DVT and the other had a periprosthetic fracture
could walk without support at the latest follow in the early postoperative period, which was treated
up. Two patients from group III were walking with successfully with plate fixation.
one arm support and one patient was able to walk Avascular necrosis of the femoral head occurred
without support. The remaining two could walk in one patient who had valgus osteotomy and 1308
indoors using walker with difficulty, and were plate fixation. The blade penetrated the femoral
listed for conversion to total joint replacement. head gradually and became intra-articular. Hard-
The three patients of group IV could walk with ware removal was done after complete union of
one arm support. fracture and osteotomy site (Fig. 3).
200 G.Z. Said et al.

Figure 3 (a) A 66-year-old man (case no. 22) presented with non-union of basi-trochanteric fracture and cutout of DHS
lag screw. (b) DHS removal, valgus osteotomy and insertion of 1308 blade-plate were done. (c) Follow-up 8 months
postoperatively; the blade penetrated the femoral head with radiological picture of avascular necrosis. (d) Good healing
of the non-union and osteotomy sites after metal removal.
Salvage of failed dynamic hip screw fixation 201

Re-operations will lead to non-union and loss of fracture fixation.

The lag screw should be inserted in a central posi-
Two re-operations were reported. The first was tion inside the femoral head and should be
plate fixation of a periprosthetic fracture in the advanced to near the subchondral bone for best
early postoperative period. The second was hard- purchase.7,11 Improper placement of the lag screw
ware removal after fracture union because of intra- may lead to loss of fracture fixation and DHS failure.
articular penetration of the implant.
Patient selection

Discussion The type of surgical treatment after failed DHS

fixation depended on the physiological age of the
The results of salvage procedures after failed DHS patient, quality of bone, and condition of the
fixation are few in the orthopaedic literature. Wu femoral head and the acetabulum. The policy fol-
et al.14 reported on 14 intertrochanteric fractures lowed was that whenever the patient was young and
with failed DHS. All were treated by reinsertion of fit, and there was still good bone stock in the
a lag screw inferiorly in the femoral head, cement femoral head, revision internal fixation was done.
augmentation and subtrochanteric valgus osteot- If the patient was fragile and the femoral head was
omy. All healed at a mean of 5 months. Haiduke- found excavated from the previous internal fixation,
wych and Berry4—6 reported the largest series for replacement arthroplasty was decided.
salvage of trochanteric fractures after failed
initial fixation using different types of implants Implant selection
in two retrospective studies over 20 years. They
reported successful treatment of 20 patients with The implant used for revision internal fixation was
revision internal fixation and 60 patients with selected according to the quality and location of
prosthetic replacement. remaining bone stock in the femoral head. It was
In the current study we report on salvage proce- possible to reinsert a DHS when there was still good
dures for failed DHS fixation of intertrochanteric bone stock in the femoral head to hold the threads
fractures. The following points are to be discussed. of another screw. A single-angled 1308 plate was
inserted when there was good bone stock only in the
DHS failure infero-medial quadrant of the femoral head. Hae-
miarthroplasty was done, if the femoral head was
The result of DHS fixation depends on patient and excavated, but with healthy acetabulum. Total hip
surgeon factors. The patient related factors replacement was decided when the acetabulum
include: (1) fracture stability, (2) bone quality, was abraded by a protruding screw and in severe
and (3) femoral neck shaft angle. Unstable frac- osteoporosis.
tures, with loss of medial calcar continuity, tend to
fall into varus displacement when stabilised. Repositioning osteotomy
Severe osteoporosis allows cutting of the lag screw
through the hollow femoral head and loss of frac- A subtrochanteric valgus osteotomy was added to
ture reduction. The neck shaft angle decreases facilitate bringing the plate of the implant to the
gradually with age. The average neck-shaft angle shaft of the femur. The osteotomy also helped in
was found to be 1258 in a population with an stabilization of the fracture and correction of the
average age of 69.9 years.12 To insert DHS of 1358 shortening.1,10,13,14 Müller10 and Bartonicek et al.1
angle in that hip, the lag screw will take a supero- used double-angled plate to fix the subtrochanteric
lateral position inside the femoral head facilitating valgus osteotomy. Single-angled 1308 plate was
its cutout. In these cases, it is preferable to insert used in this series. Fixation of the subtrochanteric
the lag screw in the long axis of neck and head. The osteotomy by a single-angled plate allowed later-
plate will stick out about 108 away from the shaft alization and normal orientation of the femoral
and valgus osteotomy of 108 below the hole of screw shaft, to counteract the medialisation and the ver-
insertion should be done. The surgeon related fac- tical orientation of the femoral shaft produced by
tors to prevent DHS failure include: (1) adequate the osteotomy.13
fracture reduction with good contact of bone frag-
ments, (2) correction of varus displacement, and Bone grafting
(3) proper placement of the lag screw. Poor frac-
ture reduction with lack of good bone contact Haidukewych and Berry5 reported on open reduction
across the fracture site or persistent varus position and bone grafting in all their patients with revision
202 G.Z. Said et al.

internal fixation. Bone grafting was not found neces- 4. Haidukewych GJ, Berry DJ. Hip arthroplasty for salvage of
sary in any of our patients. We believe that the failed treatment of intertrochanteric hip fractures. J Bone
Joint Surg 2003;8(A):899—904.
problem is mechanical rather than biological, as the 5. Haidukewych GJ, Berry DJ. Salvage of failed internal fixation
intertrochanteric region of the femur is well vascu- of intertrochanteric hip fractures. Clin Orthop 2003;412:
larised and has excellent surrounding soft tissue 184—8.
coverage. 6. Haidukewych GJ, Berry DJ. Salvage of failed treatment of hip
The limitations of the current study include the fractures. J Am Acad Orthop Surg 2005;13:101—9.
7. Jacobs RR, McClain O, Armstrong HJ. Internal fixation of
retrospective design and the possible selection bias. intertrochanteric hip fractures: a clinical and biomechanical
However, the limitations do not undermine the study. Clin Orthop 1980;146:62—70.
conclusion that for salvage of failed DHS we can 8. Kaufer H. Mechanics of the treatment of hip injuries. Clin
achieve union with revision internal fixation for Orthop 1980;146:53—61.
9. Madsen JE, Naess L, Aune AK, et al. Dynamic hip screw with
physiologically younger patients with good remain-
trochanteric stabilizing plate in the treatment of unstable
ing bone stock, while older patients with low- proximal femoral fractures: a comparative study with the
demand activities and poor bone quality, or a gamma nail and compression hip screw. J Orthop Trauma
damaged hip articular surface are treated with 1998;12(4):241—8.
hip arthroplasty. 10. Müller ME. Intertrochanteric osteotomy: indication, preo-
perative planning, technique. In: Schatzker J, editor. The
intertrochanteric osteotomy. Berlin: Springer-Verlag; 1984.
11. Müller ME, Allgöwer M, Schneider R, Willenegger H. Manual of
References internal fixation: techniques recommended by the AO-ASIF
group. Berlin: Springer-Verlag; 1991.
1. Bartonicek J, Skala-Rosenbaum J, Dousa P. Valgus osteotomy 12. Noble PC, Alexander JW, Lindhal LJ. The anatomical basis
for malunion and nonunion of trochanteric fractures. J of femoral component design. Clin Orthop 1988;235:148—
Orthop Trauma 2003;17(9):606—12. 65.
2. Buciuto R, Uhlin B, Hammerby S, Hammer R. RAB-plate versus 13. Said GZ, Gaballa MA, Said HZ, Elkady H. Subtrochanteric
Richards CHS plate for unstable trochanteric hip fractures: a valgus osteotomy fixation by single or double angled plate.
randomised study of 233 patients with 1-year follow-up. Acta Pan Arab J Orthop Trauma 1999;3(2):103—7.
Orthop Scand 1998;69:25—8. 14. Wu CC, Shih CH, Chen WJ, Tai CL. Treatment of cutout
3. Ecker ML, Joyce III JJ, Kohl EJ. The treatment of trochanteric of a lag screw of a dynamic hip screw in an intertro-
hip fractures using a compression screw. J Bone Joint Surg chanteric fracture. Arch Orthop Trauma Surg 1998;117:
1975;57(A):23—7. 193—6.