THE TORN SHOULDER CAPSULE 259

ing the acromion by replacing it after osteotomy is desirable transplanted tendon to depress the humeral head. Inferome-
because of the ease and rapidity of closure, the good cos- dial pressure will tuck the proximal end of the humerus into
metic result, and the biomechanical advantage of preserving the glenoid socket.
the acromion as the origin of the deltoid muscle 2#{149} This The results in this small series of fourteen patients
method also allows excellent exposure of the entire rotator were satisfactory - 78 per cent of the patients had good or
cuff area. In the presence of an inadequate repair of the excellent results. It should be emphasized that the fourteen
rotator cuff, the method is predicated on the idea that the patients had very severe damage to the shoulder cap-
biceps can be utilized functionally as a substitute for the sule, and the results therefore are to be rated with this selec-
rotator cuff. One can readily visualize the effect of the tivity in mind.

References
1 DEPALMA, A. F.:
Surgery ofthe Shoulder. Philadelphia, J. B. Lippincott Co., 1950.
2. HAZLETT, J. W. : Personal communication.
3. MCLAUGHLIN, H. L. : Lesions of the Musculotendinous Cuff of the Shoulder. I. The Exposure and Treatment of Tears with Retraction. J. Bone and
Joint Surg., 26: 31-51, Jan. 1944.
4. NEER, C. S. II: Anterior Acromioplasty for the Chronic Impingement Syndrome in the Shoulder. A Preliminary Report. J. Bone and Joint Surg.,
54-A: 41-50, Jan. 1972.
5. WOLFGANG, G. L. : Surgical Repair ofTears ofthe Rotator Cuffofthe Shoulder. Factors Influencing the Result. J. Bone and Joint Surg. , 56-A: 14-26,
Jan. 1974.

The Development of the Tibiofemoral Angle in Children

BY PENTTI SALENIUS, M.D.*, AND ElLA VANKKA, M.D.*, HELSINKI, FINLAND

From the Orthopaedic Hospital of the Invalid Foundation, Helsinki

The development of the tibiofemoral angle in children comprised 1 ,480 examinations of the tibiofemoral angle
at different ages has been of extensive interest for many done roentgenographically and clinically. In the examina-
years As far as we know there are no published
-#{176}. series in tion, the extremity was positioned with the patellae straight
which the development itself was followed longitudinally ahead. If there was torsion or bowing of the tibia, a Ion-
during growth. In most reports only examples of different gitudinal axis was estimated between the patelba and the
tibiofemoral angles have been reported, and in some mid-point of the ankle joint. If there was a divergence be-
cases 8,9 the angle was expressed as the distance between tween the angles in both legs, their mean was taken as the
knees or medial malleoli. In some reports 2 the proportional representative figure. The results were processed by com-
number of varus or valgus knees at different ages has been puter. As the results of the clinical examinations correlated
expressed in per cent. Very few roentgenographic directly with those of the roentgenographic examinations,
investigations 8.9 have been published. As it is an important only the latter were used for analysis. The material was
question in clinical orthopaedic surgery whether to correct divided into groups according to age, each group containing
extreme varus or valgus knees by osteotomy, we collected an average of forty patients, with six-month intervals from
the present series in order to determine the tibiofemoral birth to sixteen years. The largest group was that between
angle in children of various ages. the ages of zero and six months, containing ninety-four pa-
tients. The oldest group was the smallest and contained only
Material and Methods
nine patients. The tibiofemoral angle was measured on the
This series includes 979 patients from the Paediatric roentgenogram by drawing a longitudinal axis midway be-
Clinic of the University of Helsinki. These patients were in tween the femoral and tibial diaphyseal cortices. The angle
the hospital for a variety of reasons and the knees and legs between these two longitudinal lines was measured in de-
were roentgenographically examined for reasons unrelated grees. In the case of newborn infants drawing the lines was
to their illness. The series also includes 300 patients from difficult because their femora are not straight; therefore, the
the Orthopaedic Hospital of the Invalid Foundation, Hel- femoral line drawn represented our best estimate of the lon-
sinki. Of these patients, fifty-nine were examined roent- gitudinal axis of the femur. In the case of the older children
genographically twice and fifty-two were examined three or the drawing of the longitudinal lines presented no difficulty.
more times, at intervals of six months. The entire series thus The results were processed by computer and the output gave
* Orthopaedic Hospital of the Invalid Foundation, Tenholantie 10,
the mean of the angle in every age group and the correlation
Helsinki 28, Finland. between the age of the patients and the tibiofemoral angle.

VOL. 57-A, NO. 2, MARCH 1975

260 PENTTI SALENIUS AND ElLA VANKKA

DEVELOPMENT OF THE TIBIO-FEMORAL

ANGLE DURING GROWTH
VARUS +
+1 VALGUS

U)

- 50
U)

-10o

tO
0
0)

0I 0 0 0
Ca .1 .$1 -I +1

FIG. I
The development of the tibiofemoral angle in children during growth. The results are based on I ,480 measurements of the tibiofemoral angle of chil-
dren at different ages. The mean of the measurements is in the middle and on both sides of this is the error of the mean, which was an average of 4.4
degrees. The standard deviation was 8 degrees.

Results was 8 degrees and the standard deviations were greater

The main results obtained are shown graphically in for boys and for younger children. In the series of boys the
Figure 1 The tibiofemoral
. angle in the newborn infants and standard deviation was 10 degrees and in girls, 7 de-
in children less than one year old was in pronounced varus grees. The error of the mean in the entire series was 4.4
which then decreased with growth. At the age of about one degrees, 4.8 degrees in boys and 4 degrees in girls. The de-
and one-half years the knees tended to straighten. During vebopment of the tibiofemoral angle in a representative pa-
the second and third years the angle changed to a marked tient is shown in Figures 4-A through 4-D.
valgus position. The valgus position corrected itself in the
Discussion
following years. The development of the tibiofemoral angle
was similar in boys and girls (Figs. 2 and 3). On the graphs In clinical practice it has sometimes been noted that the
three lines are shown (mean error of the mean) sup- development of the tibiofemoral angle in children follows a
plemented by a smoothed curve of the tibiofemoral angles certain pattern, in which there is first a pronounced varus in
drawn by hand. The standard deviation of the entire series newborn infants and infants and subsequently an extreme

On

Ca

Age

U)

0.,-

>

FIG. 2 FIG. 3
The development of the tibiofemoral angle in boys. The development of the tibiofemoral angle in girls.

ThE JOURNAL OF BONE AND JOINT SURGERY

 DEVELOPMENT OF THE TIBIOFEMORAL ANGLE 261

FIG. 4-A FIG. 4-B FIG. 4-C FIG. 4-D

Fig. 4-A: The tibiofemoral angle in a child one year and two months old. The child has not yet learned to walk. The angle is in 2 1 degrees of varus on
the right and 28 degrees of varus on the left.
Fig. 4-B: The tiblofemoral angle ofthe child six months later. The child has been walking for a few months. The angle is in 13 degrees of varus on both
sides.
Fig. 4-C: The chIld has been walking for more than a year. The tibiofemoral angle is in I 2 degrees of valgus on the right and I 3 degrees on the left. The
child is three years old.
Fig. 4-D: The child is now five years old. In the right knee the valgus angle is I 1 degrees and in the left, 12 degrees.

vabgus position. Prior to the present study this had not Children, when they learn to walk, tend to hold their feet
been proved statistically. Our material includes 1 ,480 ex- wide apart to increase stability. This may be why a pressure
aminations of 1 ,279 patients whose tibiofemoral angles on the outer side of the knees exists and the medial part of
were measured and expressed in degrees. It was shown that the epiphyseal plate grows faster, resulting in the valgus
before the age of one year there is pronounced varus posi- position of the tibia. Whatever the reason for the develop-
tion, which changes into valgus when a child is between ment of the normal tibiofemoral angle in children during
eighteen months and three years old. The valgus can some- growth might be, it is obvious from this investigation that
times be extreme. It corrects spontaneously to about 5 to 6 an operative procedure to correct the angle in normal chil-
degrees, where it remains until the age of six to seven years. dren is seldom indicated.

References
1. BLOUNT, W. P.: Bow Leg. Wisconsin Med. J.. 40: 484-487. 1941.
2. BRAGARD, K.: Das genu valgum. Zwei Teile. Zeitschr. f. orthop. chir. (Beilagchefte), 57: 1: 125, 1932.
3. CHRISTIE, AMOS. and STEMPFEL, R. S.: Nonrachitic or Physiologic Bowing. Postgrad. Med.. 17: 306-312. 1955.
4. HOLT, J. F.: LATOURETTE, H. B.; and WATSON, E. H.: Physiological Bowing of the Legs in Young Children. J. Am. Med. Assn. . 154: 390-394,
1954.
5. HUTTER, C. G., JR., and SCOTT, WALTER: Tibial Torsion. J. Bone and Joint Surg., 31-A: 511-518, July 1949.
6. KITE, J. H.: Torsion of the Lower Extremities in Small Children. J. Bone and Joint Surg., 36-A: 51 1-520, June 1954.
7. KNIGHT, R. A.: Developmental Deformities of the Lower Extremities. J. Bone and Joint Surg., 36-A: 521-527, June 1954.
8. MCEVAN, D. W. , and DUNBAR, J. S.: Radiologic Study of Physiologic Knock Knee in Childhood. J. Canadian Assn. Radiol. , 10: 59-62, 1958.
9. MORLEY, A. J. M. : Knock-Knee in Children. British Med. J. , 2: 976-979, 1957.
10. SHERMAN, MARY: Physiologic Bowing of the Legs. Southern Med. J. , 53: 830-836, 1960.

VOL. 57-A, NO. 2, MARCH 1975