3. small Anirn. Pmct. (1975) 16, 649-659.

Fractures of the distal femur in the dog and cat

P. J . A L C A N T A R A * AND A. C . S T E A D
Department of Veterinary Surgery,
Royal (Dick) School of Veterinary Studies,
University of Edinburgh, Edinburgh

ABSTRACT
A series of fractures of the distal femur in the dog and cat are described and
classified anatomically from radiographs. Treatment is discussed and the
long-term results of surgery in twenty-four cases are described.

INTRODUCTION
Fractures of the distal femur i.e., fractures close to, or involving the condyles, are
common in dogs and cats. A series of eighty-two such fractures represented 32 %
of the total femoral fractures treated in this clinic over a 6 year period.
Various methods have been recorded for the treatment of the supracondylar
types of fracture. Knight (1956) described the use of an obliquely inserted lag screw
from the anterior diaphpis down into the condyles. Lawson (1958) favoured two
Rush pins, Singleton (1 966) a retrograde intramedullary pin, Walker & Hickman
(1963) a 3-holed plate and screws and Sumner-Smith & Dingwall (1973) recom-
mended crossed Kirschner wires inserted from the diaphysis down into the con-
dyles. Hinko (1974) used a variation of the lag screw technique described by
Knight, by inserting two screws, one from either condyle into the diaphysis.
While the techniques of fixation for this type of fracture are well documented,
there seems little information on the efficacy of treatment and its long-term results,
although Lacote (1971) stated that functional recovery was rarely complete.
Hinko, recorded femoral shortening of 9mm in an 8f-month-old dog, 2f months
after surgical treatment for a supracondylar fracture.
A review was made of the cases treated in this clinic to establish the types of
fracture in the distal femur and to record the results of surgical treatment.
* Present address : Department of Veterinary Surgery, Faculty of Veterinary Medicine, Univer-
sity of Lourenco Marques, Lourenco Marques, Mozambique.
649
650 P . J . ALCANTARA A N D A . C. STEAD

METHODS
Fractures were divided into three classes, as judged by their radiographic appear-
ance. The divisions were as follows:
(1) Distal Diaphyseal (DD)
Fractures of the distal shaft above the metaphysis (Fig. 1).

FIG. 1 Distal diaphyseal fracture (dog).

(2) Metaphyseal (M)
Fractures between the distal shaft and the epiphysis. This group included some
in which the fracture line extended to the growth plate for a small part of its
length (Fig. 2).

FIG.2. Metaphyseal fracture (dog).

 DISTAL FEMUR FRACTURES 65 1

(3) Epiphyseal-Metaphyseal (EM)

Fractures along the growth plate, but including in the distal fragment portions
of metaphysis attached postero-medially, postero-laterally, or both (Fig. 3).

FIG.3. Epiphyseal-metaphyseal fracture (dog).

(4) Epiphyseal Separations (ES)

Fracture wholly along the growth plate in juveniles or along the line of the
closed growth plate in adults (Fig. 4).

FIG.4.Epiphyseal separation (cat).

(5) Intercondylar (IC)

Fractures separating the 2 femoral condyles or producing chip fractures of the
condyles (Fig. 5).
Fractures in four dogs and one cat were not classified due to poor or lost radio-
graphs.
Twenty-four cases were re-examined clinically and radiographically 1-6 years
post-operatively. Lameness, stifle movement, changes in femoral and tibia1 lengths
(measured from lateral X-rays, using the normal leg as a control) and radiographic
652 P. J. ALCANTARA AND A . C . STEAD

FIG. 5. Intercondylar fracture (cat) with separation of condyles from shaft.

changes in the stifle joint were recorded. The presence of osteophytes within the
stiflejoint was taken as evidence ofosteoarthritis.

RESULTS
There were eighty-two fractures of the distal femur of which sixty-one were in dogs
(29% of the total canine femoral fractures) and twenty-one in cats (45.7% of the
total feline femoral fractures).
Thirty dogs and eight cats had right limb involvement and thirty-one dogs and
thirteen cats left. One dog and one cat had bilateral fractures.
Thirty-six dogs and ten cats were male and twenty-four dogs and ten cats were
female.
The initiating trauma, where known, was a road accident in forty-five cases,
falling from a height in ten, blows in three and dog bites in two. I n twenty-two
instances, the cause ofinjury was unknown.
Fig. 6 shows the distribution of the fractures according to age. Sixty-seven
fractures happened between 2 and 10 months of age.
Table 1 illustrates the frequency of the fractures related to the state of the growth
plate, whether open or closed. These judgements were made from radiographic
examinations. The growth plate was considered open if the radiolucent line, repre-
senting cartilage, was present between the epiphysis and metaphysis. I t was con-
sidered closed if that line was absent, or showed as a radiopaque line. I n epi-
physeal separations, the status of the growth plate was assessed from radiographs
 1. Distribution of fractures related to growth plate status as judged radiographically for dogs and cats
TABLE

DOG CAT
Growth plate Growth plate
Fracture Total Total
Open Closed Open Closed

DD 8 (14%) 9 .(15.8%) 17 (29.8%) 2 (10%) 2 (10%) 4 (20%)

M 11 (19.3%) 3 (5.3%) 14 (24.6%) 2 (10%) 0 (0%) 2 (10%)
EM 18 (31.6%) 4 (7%) 22 (38.6%) 2 (10%) 0 (0%) 2 (10%)
ES 1 (1.8%) 0 (0%) 1 (1.8%) 8 (40%) 3 (15%) 11 (55%)
IC 0 (0%) 3 (5.3%) 3 (5.3%) 1 (5%) 0 (0%) 1 (5%)
Total 38 (66.7%) 19 (33.4%) 57 (100.1%) 15 (75%) 5 (25%) 20 (100%)
654 P . J . A L C A N T A R A A N D A . C . STEAD

o < 2 c 4 C 8 c 10
Age in months
< 12 c 24 -
FIG. 6. Total fractures of distal femur related to age for dogs and cats.

of the opposite femur. I n the dog, growth plate closure was considered to occur
between 6 and 8 months of age (Sumner-Smith, 1966) and in the cat, between 8
and 9 months (Schaeffer, 1932).
Most fractures occurred in animals with open epiphyses. Epiphyseal separation
was rare in the dog and common in the cat. Presumably, where it occurred in adult
cats, there was a line ofweakness still present.
Treatments adopted were Rush pins in fifty-five instances, oblique lag screw in
thirteen, crossed Kirschner wires inserted from the condyles in five, retrograde in-
tramedullary pin in one and no treatment in five. One case of intercondylar frac-
ture was treated by transverse lag screws. Two, with intercondylar fracture plus
separation of the condyles from the shaft were treated with transverse lag screws
and Rush pins. One case, with a large chip from the bearing surface of the medial
condyle, was fixed by three countersunk Kirschner wires inserted from the arti-
cular surface.
After surgery, routine re-examination of forty-four cases was made between 6
and 16 weeks post-operatively. Critical examinations were not made but the
records show that fourteen animals still had some degree of lameness.
Nineteen dogs and five cats were re-examined 1-6 years after surgical treatment.
Seven of these dogs showed no clinical or radiographic abnormality (Table 2). T h e
remaining twelve dogs and five cats showed changes tabulated in Table 3.
 DISTAL F E M U R F R A C T U R E S 655
TABLE
2. Animals re-examined in which no abnormality was found

Time since
Case Growth Age at time Fracture treatment
number Species plate of fracture type Treatment (years)

18 Open 10 w M None 3k
19 Open 7m M RP 34
20 Open 6m M RP 31
21 Closed 10 m DD RP 6
22 Closed 8m M 0s 3
23 Open 5m DD RP 2
24 Closed 10 m DD 0s 2
~~ ~

R P = Rush pins; 0s = Oblique lag screws.

In addition, Nos. 1, 11 and 15 showed lateral bowing of the distal femur. I n dog
15, this was possibly associated with the development of patella luxation.
Nos. 3, 4 and 15 required surgical treatment for medial patella luxation follow-
ing the fracture treatment. At the time of this long-term examination, the patella
in these three dogs was stable within the trochlea groove.
No. 2 showed medial subluxation of its patella after treatment of its fracture.
This was untreated and was still present 4 years later. It is not known whether this
patella subluxation was directly associated with the fracture.
No. 1 was originally treated with an oblique lag screw, but following disintegra-
tion of the bone and breakdown of this fixation, the fracture was stabilized with
Rush pins.
There was lengthening of the ipsilateral tibia of 9mm in No. 2,3mm in No. 5 and
20mm in No. 17. This was in apparent compensation for the femoral shortening.
The two lag screws in No. 14 did not cross the growth plate.

DISCUSSION
Fractures of the distal femur are basically a problem in young animals. Beyond 2
years of age, distal diaphyseal fractures occurred almost exclusively. Intercondylar
fractures were unusual at any age.
The rarity of epiphyseal separation in the dog, as compared with the cat was a
surprise. Epiphyseal-metaphyseal fractures seemed to be the equivalent. Examina-
tion of boiled-out dog femurs revealed that the canine metaphysis has four pro-
tuberances, two anterior and two posterior, fitting into corresponding depressions
in the epiphysis. The posterior pair are very much smaller and weaker than the
anterior but are seated in deep fossae (Fig. 7).
In fractures of the distal femur, with the exception of the intercondylar type, it
is usual for the proximal fragment to displace anteriorly to the distal fragment
(Figs 1-4). These fractures are most likely with the lower limb fixed. I n a dog with
 TABLE
3. Findings in animals re-examined 1-6 years post-operatively

Femoral
Growth plate shortening
state in mm and
and age at Time since as a % of Restricted
Case time of Fracture treatment normal femur Osteo- joint
number Species fracture type Treatment (years) length Lameness arthritis movement
9
r
1 44m Open DD RP 2Y 35mm 20.6% ++ n
2 3m Open M 9
RP 5Y 20mm 8.5% + 2:
3 4+m Open M RP 1Y lOmm 6.4% + +3
- - 9
4 9rn Closed M RP 5Y ;d
5 4m Open EM RP 1Y 20mm 13.3% - >
6 7rn Open EM RP lY - - + 9
7 2.
12m Closed ES RP 4tY 5mm 4% U
8 5hm Open ES RP lY 8mm 6.8% 9
3 4m Open ES RP 1Y 5mm 4.8% -
10 9rn Open ES - n
RP 6Y 3mm 2.6%
11 6m Open EM 0s 12mm 8.5% v1
4Y ++ 4
12 7rn Open EM 0s 4tY 4mm 2.1% + M
13 7m Closed DD 0s - - 9
4Y +' 0
14 6m Open DD 0s 1t Y +
15 18m Closed IC Screws - -
4Y +-
16 4m Open M KW lY 6mm 5.3%
17 4m Open ? RP 2Y 30mm 16%

R P = Rush Pins; 0 s = Oblique lag screw; KW = Kirschner Wire; - = Absent; + = Present (slight); + + = Present (gross).
 DISTAL F E M U R FRACTURES 65 7

FIG. 7. Lateral (on left) and posterior (on right) views of separated metaphysis and
epiphysis of immature dog femur showing the weak posterior metaphyseal protuberances
and the deep fossae in which they sit.

an open growth plate, an anterior, or antero-laterally directed force would tend to

displace the metaphysis first upwards and then forwards from the epiphysis, at the
same time shearing one, or both posterior metaphyseal protuberances, leaving
them in situ in their epiphyseal depressions.
The cat also has four metaphyseal protuberances but the posterior pair are

FIG. 8. Lateral (on left) and posterior (on right) views of separated metaphrjis and
epiphysis of immature cat femur showing the small posterior metaphyseal protuberances
with their stronger attachment and the shallow fossae in which they sit.
658 P . J . ALCANTARA AND A . C . STEAD

flatter than in the dog and sit rather less deeply in the epiphysis (Fig. 8). Assum-
ing that the fracture is produced as for the dog, the metaphysis would tend to dis-
place upwards and forwards without shearing the posterior protuberances.
If this hypothesis is correct, it would explain the different incidences of epiphy-
seal-metaphyseal fractures and epiphyseal separations in the dog and cat.
Femoral shortening was seen, predictably in those animals with an open growth
plate at the time of fracture. Case No. 7 was an exception in showing shortening of
5mm after growth plate closure. Although the shortening ranged to a maximum of
35mm and 20.6% of the normal femoral length, if this were related to total leg
length, then the expressed percentage would be reduced by approximately two-
thirds (allowing for limb length from proximal tibia to distal phalanges).
It was considered that the femoral shortening was unlikely to contribute to lame-
ness, except possibly in Case No. 1. Indeed three dogs and four cats with recorded
femoral shortening showed no lameness.
Vaughan (1 963) recommended early removal of implants placed across an open
growth plate. I n this series, most implants were left in situ. I n no case where an
implant was removed was a growth plate seen to be open at the time of such re-
moval. Clinical limb shortening and radiographically evident premature growth
plate closure were seen in Case Nos 5 and 16, 4 and 6 weeks respectively after
surgery. Both animals were aged 4 months at the time of fracture.
Of eleven animals treated with Rush pins, femoral shortening was present in
seven and absent in four. As bone growth can occur in some instances despite the
presence of Rush pins across the growth plate, this raises the question of whether
the femoral shortening was due to the implant or to damage at the time of injury.
However, as the animals (dogs) in which no shortening occurred were aged 5,6, 7
and 7 months, possibly there was little growth left in the growth plate.
Femoral shortening occurred where screws, or crossed Kirschner wires were
placed across an open growth plate and measurements were taken (Nos 1 1, 12, 16
and another treated with Kirschner wires).
The concept of tibia1 lengthening to compensate for femoral shortening was
interesting. The three dogs in which this was seen were aged 4 months, or less, at
the time offracture.
Two dogs were permanently lame (Nos 1 and 11). Both had stifle ankylosis,
associated in No. 1 with osteoarthritis and lateral bowing of the distal femur and in
No. 1I'with the presence of a loose screw which had been there for four years.
Lameness was present in six other dogs and one cat. This lameness was very
slight and occasional in nature and probably associated with their osteoarthritis.
Reduction in the normal range ofjoint movement was also assumed to be due to
osteoarthritis.
The precise cause of osteoarthritis in these seven cases was not determined ex-
cept in No. 2 which had a subluxating patella. However, it was noted at the time
of surgery that rupture of the anterior cruciate ligament was a frequent complica-
tion to distal femoral fracture in the dog and cat. Whether this could be associated
 DISTAL FEMUR FRACTURES 659

with later osteoarthritis is not known. Certainly the ‘drawer forward’ sign indicat-
ing anterior cruciate ligament rupture, was never seen.
Of the twenty-four animals re-examined, fifteen-eleven dogs and four
cats (62.5% of those re-examined)-were not lame. Seven, six dogs and one
cat (29.2 %) had virtually normal function, but occasional, slight lameness which
was assumed to be due to osteoarthritis and two dogs (8.3%) had poor function
and permanent lameness.
T h e important problem in the treatment of these fractures is accurate joint re-
alignment and stabilization where the fracture is intra-articular. I n young animals,
growth plate function should be preserved if possible, to allow normal growth in
femoral length. Although there is little evidence that shortening of the femur
causes lameness, it would nevertheless seem wise to select a n implant which would
least disturb an open growth plate. Other difficulties arise from the crumbly nature
of the bone in the distal femur of immature dogs and the small size of the distal
fragment in cats, which is frequently only a sliver of bone and cartilage.
Although our series of screwed cases is small, its incidence of lameness (four of
six re-examined) was greater than for Rush pins (four of fifteen). Oblique lag
screws, if used, should be confined to adult dogs, as immature bone can provide an
inferior grip for screws and interference with bone growth is likely.
Our series of Kirschner wire-treated fractures was too small to draw any con-
clusions, except that in two instances (No. 16 and another) premature growth
plate closure occurred.
It was concluded from our own experience that the use of Rush pins as described
by Lawson offers the best chance of a good functional end result, with less risk of
growth plate interference. Osteoarthritis is likely in some cases but will not neces-
sarily detract from the end result.

ACKNOWLEDGMENTS

We are grateful to M r J. A. Fraser for his advice, to Mrs B. Robbie, Miss G.

McHarrie and Miss C. Somner for radiographic assistance, Mrs I. Stenhouse for
typing and Mrs A. McNeill of the Medical Illustration Unit of Clinical Surgery,
Edinburgh University for the illustrations.

REFERENCES
P.J. (1974) 3. Am. Anim. Hosp. Ass. 10,61.
HINKO,
KNIGHT,
G.C. (1956) Vet. Rec. 68,415.
LACOTE,
E.F.L. (1971) Doctorate Thesis Alfort.
LAWSON,D.D. (1958) Vet. Rec. 70,760.
H. ( 1932) Inaugural Dissertation Hanover.
SCHAEFFER,
W.B. ( 1 966) 3.small Anim. Pract. 7 , 166.
SINGLETON,
SUMNER-SMITH,G. (1966) 3,small Anim. Pract. 7,303.
SUMNER-SMITH,G. & DINGWALL, J.S. (1973) 3. Am. Anim. Hosp. Ass. 9, 171.
VAUGHAN,L.C. (1963) Vet. Rec. 75,292.
WALKER,R.G. & HICKMAN, J. (1963) Vet. Rec. 75,400.