RESTORATION  OF   LOSS   OF ^ BONE  (     ,

 INCLUDING AN  ANALYSIS OF THE W^SfeSlS^CASks ^
   OF FRACTURE TREATED BY  BONE GRAFT AT U.  S.
     ARMY GENERAL HOSPITAL NO. 3, COLONIA,  N. J.*

           FRED  H.  ALBEE,  M.D.,  Sc.D.
                     ASSISTED BY
             ELMER  P.  WEIGEL, M.D.
                      NEW  YORK

   In the task of restoring loss  of bone substance and
 function in the wide variety  of traumatisms  resulting
 from the war, the plastic surgeon is  confronted with
 a correspondingly varied array of mechanical problems.
 The recognition of the  underlying biologic and physio-
 logic significance  of  tissue growth and metabolism  is
 a fundamental requirement in the successful treatment
 of these cases.  The surgical  repair of bone, and more
 particularly  the  use of  the  bone graft  in  cases of
 pseudarthrosis with  or without  bone  loss,  is  based
 not  only  on the  ultimate  establishment  of  adequate
 fixation of the bone fragments,  but also on the attain-
 ment of a proper  environment for the nourishment of
 the graft.   This entails the exact  coaptation of parts
 of the graft to respective parts of the host  bone; in
 other words, the  adequate and extensive contact of
 all four corresponding bone layers,  namely, periosteum,
 cortex, endosteum and marrow.

      MECHANICAL STRESS  AND BONE  GROWTH
  Throughout his plastic work, both in civilian practice
 and  army  experience,  the author has been greatly
 impressed  with  the  striking  influence  exerted by
 mechanical  stress on the  growth  and metabolism of
 bone.  In cases of loss of substance of  long duration,
 in the radius,  humerus or any long bone, the bone
 cortex has often become reduced to one-fifth its normal
 thickness, in fact, almost to eggshell consistency, largely
 owing to removal of the stimulus of mechanical stress.
 Such a condition is, of course, in direct sequence to the
general  physiologic law of bone growth ; it is. in  fact.
  * Read before the  Congress of the Italian Orthopedic Association,
Bologna, Oct. 18. 1919.
u 
2
;i magnification of  Wolff's  law.   If bone,  whose  nour-
ishment   and   blood   Mipply  have  not   been  greatly
impaired, should suffer so materially as a result of  loss
of the stimulus of mechanical stress, how much greater
must be  the effect  of the  same inhibitory influences on
any free bone graft  whose  blood  supply and nourish-
ment are not vet established.
of on/err *hp  aorta
  Fig. 1.—Requisite apposition of bone  layers of graft with respective
parts of hn>.t bone.  The graft is of sufficient length to extend well beyond
the  eburnated area of the host fragments, coming into generous contact
with the healthy vascular  marrow  substance. The cross-section at li shows
the  eburnated plug of bone where, as a  rule  in pseudarthrosis, it closes
the  marrow  cavity in  the  fractured ends.   Here  the gutter has been
deepened sufficiently to receive the full amount of marrow of the graft,
which  is demonstrated (lower drawing and cross-section at D) in exten-
sive contact  with  the marrow  of the host fragments.  The arrows on
the  marrow in the host fragments indicate the direction of blood  supply
to the graft  from  its principal normal source, namely, the marrow  sub-
stance of the host.  By the  insertion of the  inlay, with its full amount
of marrow, a continuous  marrow bridge is formed, extending from the
healthy marrow  of both host  fragments through  the  gutter in  the
eburnated ends.   This marrow bridge plays  a most  important  role as
conductor of blood vessels and osteogenic cells from  one host fragment
to the other.  The cross-sections at D and £ show also the cabinet-maker
fit of  the inlay graft with the host fragments, which not  only affords
mechanical fixation of parts,  but also favors the stimulus to  bone growth
from frictional irritation, emphasized by  Roux.


  A  more  suitable environment   for  successful  bone

growth  is  established  by  the  cabinet-maker  fit  of

the properly  inserted  inlay graft   than  by any  other

known technic.   At the  same time the   biologic  laws

that  obtain  in  the  transplantation of  all varieties of

tissues are fulfilled, since  corresponding tissue-layers 
3
are  brought  in  apposition,  thereby furnishing ideal
conditions for the rapid and complete establishment of
the  blood  supply.   Under  such  conditions,  Wolff's
law  of bone  growth is given favorable opportunity to
exert  its influence  on bone proliferation and on  the
adequate adjustment of the  bone architecture.  More-
over, by the  inlay technic, the full influence of Roux's
  Fig. 2 (Case 1).  Wound  in which  an officer, hit  at Ourcq  River
by fragments of a high explosive shell, lost  about  i]/2 inches of  the
upper third of  the humerus,  including  the entire head.  In  this case,
destruction of practically all the musculature  of the shoulder rendered
the patient incapable of shoulder motion.
law of frictional irritation is ideally provided for, since
extensive plane surfaces of the graft are brought into
the closest proximity with equally extensive plane sur-
faces of the host fragments. 
4
                       TEC II NIC
  In  work  on  bone  tissues,  which  easily  dry  on
exposure to  the air.  operative  speed  is  necessary;
moreover, in  the repair of bone that  from  lack of  the
stimulus  afforded  bv  mechanical stress has become
almost  eggshell-like  in consistency,  great delicacy of
technic and operative  speed are fundamental require-
ments in the  difficult work of inserting the necessary
inlay.   It would be impossible to execute such accurate
inlay  technic by the former laborious  methods  with
mallet  and chisel,  or osteotome.  In  work of  this
nature, in which the operating  field is frequently lim-
  Fig.  3  (Case  1).—Diagrammatic  drawing  (anterior view)  to show
the trusswork of tibial grafts in position and fixed with kangaroo tendon.
Graft C is inlaid into the shaft of the  humeral  fragment at li, and
mortised into the acromion process at U.  Graft A is mortised into  the
glenoid body of  the scapula at B, and attached by means of kangaroo
tendon to Graft C at the point where the  latter  graft meets the end
of the humeral fragment.
ited, where fragility of bone may be a constant menace
to success, and  in which an accurate cabinet-maker fit
of parts is indispensable, an electrically  driven rotary
twin-saw seems absolutely essential.  In contradiction
to a recent assertion,  it is emphatically stated that the
motor-saw, when  properly used,  does  not heat  nor
glaze the bone.  During various operations  in the past
few months, the author has made repeated attempts to
determine whether heat was generated by  the motor-
saw when used properly, and if so, to what degree.  In
every instance it has been found that the most delicately 
5
adjusted  thermometers   have  failed  to  register  an
increase  in temperature,  even  to the  fraction  of  a
degree, when placed directly  on  the motor-saw or on
the bone  immediately following  the withdrawal of  the
instrument.
  Fig. 4  (Case 1).—The  plastic  operation in this case  was performed
four and one-half months  after  the  patient  was  wounded  and  two
months after the wound had healed.  The roentgenogram demonstrates
restoration of loss of bone by two tibial grafts, eight weeks after opera-
tion.  Graft 1 was inlaid into the shaft of the humerus and mortised into
the acromion process.   Graft 2 was mortised into the glenoid body of
the scapula  and contacted with Graft  1 at the  point where that graft
met the  humerus.  During the union  of the grafts the arm  was  held
by a plaster-of-Paris shoulder spica in an elevated anterior posture and
in such   relation  to  the  scapula  that  the  powerful thoracic  muscles
controlling this  bone might later move it and in a large measure restore
its loss of function by causing the scapulothoracic  motion to be com-
pensatory for loss of shoulder motion.



   Methods of Internal Fixation.—For the  inlay graft,

accurately  cut  and fitted  by  motor-saw technic,  the

fixation afforded by kangaroo  sutures is adequate and

preferable to that of  all metal agents (such as  plates, 
6
 nails,  screws or wire).   Kangaroo tendon, from the
 standpoint  of mechanical  strength, absorbability  and
 tolerance by  the tissues, surpasses all known  fixation
 agents.   It  remains in situ  sufficiently long  for the
 purpose of fixation, and begins to be absorbed within
 forty days.  Moreover, it  is sufficiently elastic to allow
 the plane surfaces  of an inlay graft to rub in a micro-
 scopic amount on the contiguous plane Mir faces of the
 gutter of the host   fragments, thus favoring the  fric-
 tional  irritation law of  Roux.
   The practice of  inserting  metal plates  (as  recom-
 mended   by  some  surgeons)   is   absolutely  contra-
 indicated in this work.   The use of metal plates as the
 internal  fixation agent not  only   robs  the  graft of
 mechanical  stress,  but  an  influence is thereby added
 that contributes strongly toward the stirring up of old
 infections,  if such  remain in the  tissues.  Whereas,
 in tissues retaining only a  slight amount of an original
 infection, a bone-graft  operation  speedily done with
 minimum trauma may  bring satisfactory results,  the
 introduction of  a foreign body  (such as a metal plate,
 nails or  screws) adds  a second devitalizing element,
 which, in  the battle of tissues, may turn the scales
 unfavorably, with the result that infection again breaks
 out.

                ANALYSIS  OF  CASES
   A careful analysis of the first 100 cases of fracture
 treated by bone  graft at U. S. Army General Hospital
 Xo. 3 during the period from July  15, 1918, to  May 1,
 1919, has yielded, it is believed, valuable data in regard
 to possibilities of treatment.
   Of  the  100  cases,  seventy-nine  involved   bones
 in which injury resulted  from  high  explosive shell,
 machine  gun bullet, or shrapnel; seventeen were simple
 fractures of the long bones; the remaining four cases
 were compression  fractures of  the spinal vertebrae.
 All the simple fractures and the  spine  cases have been
 successfully treated, in  that they have healed, in each
 instance,  without  infection,  and  have  shown bone
growth   by  roentgenographic  examination  within a
 reasonable period after  transplantation  of the  graft.
In every case of fracture  of  the long bones,  function
has been restored,  while in the  spine  cases  there  has
been an inhibition of symptom--. 
7
  Of the seventy-nine cases of fracture by projectiles,
treated by bone graft, sixty-five, or 82 per cent., were
for loss of substance varying in amount from one-half
inch to six inches, and averaging about two inches;
ten cases were for nonunion without  loss of  bone; the
remaining four cases  were for malunion.  These cases
are classified anatomically in Table 1.

 TABLE 1.—SEVENTY-NINE CASES OF FRACTURE TREATED
       BY BONE GRAFT, GROUPED ACCORDING TO
                 THE BONES INJURED
Site of Injury				No.	of Cases 23
Ulna..............					17
Radius and ulna . .					4
					12
					2
					3
	and	upper			1
					62
Tibia ..............					12
					4
Patella ............					1
	extremities :ases consid(				17
Total number of c					79
                 SPECIAL  METHODS
  Some  of  the  various methods  of  plastic  repair
employed in  this series of fractures, more than four
fifths of  which  involved loss of  bone,  are briefly
described in  the following groups of cases:
  Restoration of Loss of Bone at Shoulder.—Owing
to the exposure of the upper  portion of the body in
trench warfare,  shoulder injuries,  and  particularly
those involving the upper part of the humerus, have
been frequent in the recent war.   Through the rather
extensive practice of  certain  surgeons  at  the front,
more especially of the French,  of removing large por-
tions of  bone at or  near the  shoulder-joint in such
injuries as these, there has  resulted a notable group
of cases  in  which  shoulder function is very nearly
negligible, if  not entirely destroyed,  on account of the
loss  of bony  framework over which  the shoulder mus-
cles  might play.   Of all surgical conditions, none pre-
sents a picture of greater helplessness than a dangling
arm  from which the upper portion of the humerus is
missing. 
8
  The high frequency of shoulder and forearm injuries
is  strikingly borne out in  Table 1.  In  this  series of
>eventy-nine  cases  of  fracture by  war  projectiles,
injurv to bones of the upper extremities, as compared
with  those  of the lower,  has occurred  in a  ratio of
nearly 4 to 1.  (>f the total  number of cases of injurv
in the upper extremity, the humerus has been  involved
in fourteen instances, or more than  2? per cent,  t ases
of loss of substance in the humerus with  loss of shoul-
der function have l>een classified in  two groups,  with
respect  to treatment:
  Fig. 5 (Case 1).—Type of brace applied after removal of the plaster
spica.  This brace is so adjusted as to allow the arm to descend to a
lower position than that maintained during the union of the grafts to
scapula and humerus.  This new position  brings  a slight amount of
lateral  stress on  the  grafts thus  stimulating  their  hypertrophy  and
development. It  is  obvious that this new  posture causes the lower
angle of the scapula to separate itself from the thoracic cage.
  Group  1.  Restoration  of  Shoulder  Motion  and
Function:   This  class consists of cases  in  which  the
humerus has  been destroyed, but the musculature  has
been  sufficiently  preserved to  enable the surgeon to
hope  for a  return of shoulder-joint motion  and  func-
tion, provided the bone  be replaced.   The author  has
restored motion and  function in such cases by trans-
planting into the humeral fragment the head and upper 
   Fig. 6  (Case 4).—Diagrammatic drawings 1' to 10' illustrate  the author's technic  in cases in which the end of the bone  fragment (distal
or upper  fragment  in this case) is conical  in  shape and  too  small  in diameter to  receive  the  usual  inlay  graft  which  is  shown in  the
proximal,  or lower,  fragment  in  these drawings.  A  wedge-shaped piece of cortex. H, is  removed from the upper fragment.  A  split at the  end
of the  wedge  cavity may, or may  not, be  made  by means of  a  thin osteotome,  as shown  at /.  If the fragment is small in diameter  and
osteoporotic, as is usual in these cases, the bone may be bent on both sides of  the cavity in the region of  /, and  the wedge cavity may  be
enlarged by driving a  wedge-ended graft  of larger diameter  into  it, as demonstrated in  Step 9'.   In  Step 10', the wedge end  of  the graft
is  (irmly  immobilized in  the distal or  upper fragment by  means  of kangaroo tendon.  The other  end of  the graft is  then  forced  into  the
proximal fragment by means of a  strong  clamp at the same time  that the graft is being levered  endwise by means of a narrow osteotome, the
object  of  this  procedure  being twofold;  first,  to  restore  as  far  as possible the length of the ulna, and  secondly,  to  bring end  stress  on
the graft  as a  stimulus to  bone growth.   These drawings were made with the distal  fragment of the ulna  uppermost because of  the position
of the  arm  during  operation. 
 
9.
end of the fibula to replace the upper  portion of the
humerus that has been  destroyed.  He has  resorted
to this operation many times in his civil practice, as
well as in military work, and has invariably found that
the patient can  functionate satisfactorily without the
upper end of the fibula.  In these cases, whenever pos-
sible, the principal muscles of the shoulder, such as the
pectoralis major, the supraspinatus and the subscapu-
laris,  are firmly affixed subperiosteally to the  trans-
planted head and neck of the fibula.
  Group 2.   Shoulder Function Restored by  Compen-
satory Scapulothoracic  Motion (Figs. 2, 3, 4 and 5) :
This group includes cases in which, in addition to loss
of  bone, the musculature of the shoulder  has been
destroyed or severely injured to  such a degree that one
cannot  hope  to secure  a return of  shoulder-joint
motion.  In these cases, the loss of bone is restored by
ankylosing the humerus to the scapula by a truss-work
of  tibial grafts, usually two in number.  During the
union of the grafts, the arm is immobilized in an ele-
vated anterior posture,  and is held in such relation to
the scapula that the powerful scapulothoracic muscles
which control this bone may later move it.   Thus, by
causing the scapulothoracic motion to  compensate for
loss of shoulder motion of the arm, the lost motion of
the arm and  the shoulder is restored to a surprising
degree.
  In restoring loss of bone near joints, this technic has
been employed :  By means of an osteotome or chisel, a
wedge-shaped mortise is made  in the  joint  fragment
and is  extended under the  capsule,  and sometimes
completely through to the joint cartilage, without dam-
aging the joint.   One end of the  graft is driven into
this wedge-shaped mortise. The other end of the graft
is then inlaid into the long bone  fragment by  the usual
inlay  technic and  is  fixed  with kangaroo  tendon
(Fig. 3).
  Restoration of Long  Bones.—Restoration  of the
shaft of the humerus, femur, tibia or any long,  large
bone has been accomplished by means of a graft inlaid
by  the  author's usual inlay  technic,  as illustrated in
Figure 6, in the larger  fragment  at A, B,  C and E.
On bones  of the forearm, or  on any  bone  of small
diameter, or in cases in which a bone of large diameter
has become  conical-ended, as  frequently occurs, the 
10
mode of repair which has been  resorted to is  desig-
nated, for purpose of  description, as the  "fishpole
technic  (Fig.  f> H, I and /). since a similar method is
employed bv the artisan in mending a fishpole.
  The influence   of stress  on  the hypertrophy  and
metabolism of bone has been more strikingly illustrated
in bone graft restoration of the  shafts of long bones
than in anv similar class of work.  However small the
diameter of  the graft, provided  it be protected  from
fracture  by external support and at the same time be
allowed to withstand stress,  it will eventually  restore
the  lost  bone in  almost  every  anatomic  particular,
namelv, in diameter, strength and external contour, as
well as in respect to the internal architecture.
Immobilization  in  Extremity  U'ork.— tty  making
  Fig. 7 (Case 2). — Extreme laxity of left arm due to loss of substance
in shaft of humerus as result of wound by machine-gun bullet at Can-
tigny.  The plastic operation  for  restoration of  bone and  function  in
this case was performed seven months after injury and two months after
wound  had healed.
the most  of  all known  mechanical joints and by the
insertion of kangaroo tendon in such ways as to afford
the best internal fixation, in conjunction with the  most
perfect external fixation by plaster-of-Paris dressings,
the extremity being placed in various "positions of neu-
tral muscle-pull,"  immobilization of the involved  frag-
ments  has been found possible.  Too great emphasis
cannot be laid on  the importance of putting absorbable
ligatures in the skin, so that carefully applied  plaster-
of-Paris dressings need not be disturbed for a period
of at least eight weeks after implantation of the graft.
Plain catgut Xo. 0 or Xo. 1 with suture-holes  puddled 
11
with tincture of iodin, serves admirably for this pur-
pose.   Chromic catgut, No. 0 or Xo.  1, is also suitable.
The buried sutures in the soft parts should always be
small in diameter and as limited in number as possible.
The  only  adequate  postoperative  dressing in these
cases is  the plaster-of-Paris splint,  applied with the
utmost care and molded to the bony contours of the
extremity.  It should always include at least one joint
above and one joint below the bone involved, with due
  Fig. 8 (Ca-c 2).—Loss  of about  iy2 inches of shaft of humerus.

attention given to  position, which  is of the greatest
importance.  By way of illustration, in cases of injury
of the upper portion of the ulna, the arm should always
be put up  straight,  never with the elbow flexed.   In
cases of  injury at or near the lesser trochanter of the
femur, the extremity should always be immobilized in
a plaster spica with the thigh abducted and flexed.
  Synthetic Grafting  of Tissues  in  Construction  of
Neil' Fingers.—In  two  cases  of loss of four fingers 
                          12

with the adjoining metacarpal bones as a result of high
explosive  shell and  shrapnel  wounds, the hands  were
completely helpless,  so  far as  grasping and  holding
were  concerned.  In these cases,  function has  been
restored to a great extent by the synthetic construction
of new digits.   One of these cases  is  illustrated in
Figures 12, 13 and 14.   By providing an apposing sur-
face for the  thumb, the usefulness  of the member has
in each instance been restored.
 Fig. 9 (Case 2i.  Method of immobilization of fractures of humeru-.
In this case the dressing was allowed to remain on for eight weeks.
  In plastic  work of  this nature, which  involves  the
transplantation of more than one kind of tissue, a two-
step or multiple-step  procedure  is  the only  method
whereby a successful sequence may be expected.  It is
essential,  for example, in handling soft  parts and bone
(as in the construction of new fingers, in plastic repair
of the jaw, etc.)  that skin and subcutaneous tissues be
firmly united with the host-tissue and that circulation
therewith be  well established before  implantation of 
13
bone.  Fase of technic and the possibility of obtaining
ample soft tissues and bone have also led the author to
recommend strongly this type of operative procedure.
  Repair of  the  Mandible.—Probably no branch  of
surgical repair presents greater difficulty of mechanical
fitting and adjusting than in injury to the lower jaw,
involving extensive loss of bone.  Owing to the irregu-
  Fig. 10 (Case 2).—Tibial  bone graft firmly united in position, ten
weeks after plastic operation.
larity of contour of the jaw  fragments, hardness of
the bone and lack of anvil stability, such work demands
an accuracy and precision of technic that can be secured
only by the use of delicately adjusted motor tools, such
as the author's  tiny circular saws, burrs, drills, end-
mills and the like.
  Of primary  importance  is the cosmetic  result.  In
many cases this  depends entirely  on  the construction
of a suitable graft  framework over which to-restore 
14
the contours of  the  face.   Such a  framework must
secure the adequate fixation of the jaw fragments, as
well as restoring, in many instances,  loss of bone sub-
stance.   In his plastic work on  the  jaw,  the  surgeon
will do well to cooperate, so far as possible, with the
prosthetic dentist in the application of the most efficient
intradermal splints.  Use  of  such splints is, however, in
many  instances,  impossible,  owing to  the  exten-
sive loss of teeth and of bone.  In the latter cases,
the sole means  of  fixation must be  provided by  the
graft, which should be molded and firmly inlaid into
each fragment, adequate in any emergency (such  as
removal of the  dental splint on account  of pressure
necrosis, etc.) to supply  the requisite  fixation.
  In other work, such as the restoration of the shafts
of long bones, neck of femur,  etc., the tibia has been
found  a satisfactory source  from which to obtain  the
graft.   In  work on the  jaw involving extensive bone
loss, however,  the  tibial dimensions are  not always
sufficient to supply  a graft of the necessary curve and
size.  In such an event, the  side of the ilium, adjacent
to the  anterior superior  spine, is the only bone, with
the possible exception of the outer table of  the skull,
which  is of adequate dimension to allow the proper
modeling of the graft,  which is done  by means of
motor-driven tools in a manner resembling the scroll-
work of the cabinet-maker.
  Relief of Compression Fractures of  the  Vertebrae.
—The use of the bone graft as a means for relief of
compression  fractures of  the  vertebral  bodies  has
afforded highly  satisfactory results.   On  account of
meager bone growth and inadequate bone repair around
crushed vertebral bodies,  and because of constant inter-
ference with bony union  by  the respiratory and volun-
tary motion, nature, unassisted, does not bring about
the proper repair in these cases, even though efficient
external means  of immobilization be  employed for
months, and sometimes even for years.  To supply the
deficiency resulting from lack of bone repair, the bone
graft offers a sure means  of relief, and its indications in
such cases  are as definite as in  any pathologic or trau-
matic conditions encountered.  It is  inserted into the
spinous processes by precisely the same technic as has
been devised by the author for the treatment of Pott's
disease of the spine (Fig. 15). 
15
  Preoperative Treatment of Persistent Infection.—In
our first series of  100 fracture  cases treated by bone
graft,  we  found that postoperative wound infections
were more frequent  in the last twenty-five cases than
previously.  However, these cases did  not include in
all instances  those  wounded at a later date, but rather
those in which the  primary wound took  longer to heal;
and the later infections, following the plastic operations,
are attributed to a more severe  primary infection and
to a less  complete immunization  on the part of the
patient.
  Fig. 11 (Case  2).—Stability of arm from union of graft, ten weeks
after operation.  Such a posture was impossible before operation.
  At the time of the plastic operation, positive cultures
were occasionally found in the scar tissue around the
bone ends, and small sequestrums were removed from
the bone fragments.  Not all of these cases, however,
showed infection following the operation.   In those
cases which did, the infection was usually of a mild,
low-grade  type,  apparently  owing to  the  attenuation
of  the  infecting  organism  or to a  partial previous
immunization  of the  patient against that organism. 
16
   The scar, on account of its  low-grade tissue and its
deficient blood supply,  furnishes a  most  unfavorable
environment for the reception and nourishment of the
graft.   To obviate the possibility of subsequent infec-
tion in  cases  which,  from  their past  history,  seem
unfavorable, and in order  to furnish  healthy tissues
in which to  implant the bone graft later, it  has been
the author's practice to excise the scar at a preliminary
operation, laying bare  the  bone ends, and  replacing
the scar by plastic flaps  of  healthy  skin, subcutaneous
tissue, muscle,  fat, etc.  If, as has usually been the case,
  Fig.  1J (Case J).—Left hand of an American soldier who lost com-
pletely  the four fingers and  adjoining metacarpal surface  following a
high explosive shell wound  at Chateau-Thierry.  One may note the
absence of any apposing surface when the thumb is flexed, as a result
of which nothing can be grasped or held.

satisfactory  healing of  the  wound takes  place,  the
bone  graft  operation  follows after a period  of  from
ten days to two weeks.  By  means  of such a  two-step
operative procedure, healthy  tissue  is provided for the
subsequent  implantation  of  bone,  and a  successful
sequence is more  reasonably  certain.
  Source of Graft Material.—Ow ing to  its accessibility
and  favorable contour,  the  tibia, perhaps  of all the
bones of the body, as a source of graft material affords 
17
the greatest ease of technic.  Moreover, bone from the
tibia is nearly always preferable for grafting purposes,
not only on  account of the dimensions and the plane
surface of this bone, which permit of a wide choice in
the selection of material, but also because of its char-
acteristic  strength  and  osteogenic  activity.   In  the
treatment by bone graft of the foregoing series of 100
cases of fracture, the tibia with  few exceptions fur-
nished  the graft material.  The exceptions were in
cases of sliding  grafts to restore loss of substance in
the humerus, femur or tibia,  and in one instance of
synthetic transplantation  of  tissues to  form  a new
finger, in which case bone was transplanted  from the
clavicle.
   Avoidance of Operative  Trauma.—There  are trau-
matic influences which have an important bearing on
the success of a bone graft operation.  It is necessary
to  avoid certain faults in  technic, such as  excessive

 TABLE 2.—SUMMARY OF TIME REQUIRED IN BONE GRAFT
      OPERATIONS, ACCORDING TO SITE OF INJURY
                          i----Time of Operation---.,
  Bone                      Average      Shortest
Humerus..........................  1 hour      45 minutes
Radius ...........................  47 minutes   25 minutes
Ulna  .............................  33 minutes   19 minutes
Femur............................  55 minutes   21 minutes
Tibia  ............................  23 minutes   14 minutes
length of operating time, with the resultant drying of
the graft or host  tissues from  contact  with the  air,
rough use of  the  retractors, poor  mechanical fit of
graft, absence  of coaptation of similar bone layers, or
direct trauma  from wedging or crushing by the chisel
and mallet.
  Table  2 permits a comparison of the average operat-
ing time  with the shortest in certain cases, and has been
inserted  because it is believed that  in  no other class
of surgery does length of operating  time so materially
influence results.  In this work, in which we are deal-
ing with rigid tissues, every effort  must be  made to
maintain to the fullest possible extent the viability of
the transplanted bone, a condition  largely dependent
on the early  union  of graft tissue with host tissue and
on the adequate establishment of nourishment.  Drying
of the tissues from contact with the  air should be  con- 
IS
   Fig.  13  (Case  3).   To  restore  lo-t function of  hand, the  synthetic
 transplantation  of »t,ft  tissues and bone was undertaken in a two-step
 operative procedure.  Skin and  soft parts were first  turned up  from the
 chest wall  to form a boneless finger.  Through  a pedicle  left attached
 to  the  chest wall, supply of blood was furnished until  circulation  with
 the hand was  thoroughly established.  The  hand and arm  were immo-
 bilized  in  plaster for  four weeks.  In  the  second operative  step, the
 boneless  linger  was first cut loose from  the  chest wall.  A tibial graft,
 inserted  through  the  soft  parts,  was then mortised firmly  into  the os
 magnum.   A sliver graft, indicated by the arrow,  was affixed  alongside
 for increased osteogenesis.   The  roentgenogram  was  taken four weeks
 after the implantation  of the  grafts, which now  have become firmly
united to  the bones of the hand. 
19
stantly guarded against, not only by the use  of  saline
solution, but by completing the operation in the shortest
possible time consistent with good work and with min-
imum trauma.
  Final Results  in Forty-Eight Cases.—Too short  a
time has elapsed to pass judgment, at the present date,
on the entire series of  100  cases of fracture.   How-
ever,  in forty-eight  cases that were  treated before
March 1, 1919, and have, therefore, afforded  opportu-
nity for observation  over a period  of at  least  ten
months, we feel  justified in reporting  definite conclu-
sions.
  Of these  forty-eight  cases, the results  in six  are
questionable; the grafts are  still in situ and the  roent-
gen ray reveals bone growth, but the wounds  were pri-
marily  infected  and there yet  remain one  or more
sinuses.  Three of these questionable  cases,  however,
show favorable indications  of  ultimate good results;
the other three cases will probably be failures.
  Four of the forty-eight cases are definite failures.  Of
these, one case, a sliding inlay from a tibia to a  femur
to stabilize a resected knee,  was complicated  by pneu-
monia five weeks after the operation ; sinuses broke out
on both sides of the knee,  with a resulting infection of
the entire scar tissue of the knee and a failure of part
of the graft to "take."   Of two radius  cases that were
failures, one showed a positive Wassermann  after,
but not before, the operation; and the  patient himself
removed the fixation dressing on three different occa-
sions.   The other radius case showed  a bad  infection
of  dense  scar tissue.  The fourth, and last case of
failure, was  a humerus  case with loss  of three inches
of substance, and with much scar tissue.  The wound
broke down,  and examination by roentgen ray revealed
that the lower end of the graft was not attached  to the
distal end of the humerus.
  Subtracting this  group of ten cases,  of which four
are definite failures and six questionable cases, we have
thirty-eight  cases remaining, all of which have shown
perfect  results  in  respect to  postoperative  primary
healing of the wound,  proliferation of new  bone as
demonstrated by the roentgen ray, and restoration of
function.   This yields  a rate of  79 per cent, perfect
results,  with a possibility of an ultimate 85 per cent.,
should three of the questionable cases prove successful. 
                         2D

  A complete report of all the plastic operations per-
formed at U.  S. Army (leneral Hospital Xo.  3  will be
published as soon  as sufficient time has elapsed  to per-
mit of trustworthy deductions.

                      SUMMARY
  The following conclusions are based not only on a
study of cases and results at L\ S. Army (leneral Hos-
pital' Xo. 3. at Colonia. X. J., but also on the author's
previous experience with over 1,800 cases  treated  by
bone graft in  civilian  practice, as well as in extensive
  Fig. 14 (Case 3).—New finger six weeks after last operation.  The
patient is now able to  grasp and hold objects with his  thumb and
grafted finger.

animal experimentation.  It is believed that the careful
observance of these points is essential to success in this
class of work.
  1. Early  Observation of Wound.—A  careful  study
of the wound should be made before it  has healed, if
possible.  The type of infecting organism (Streptococ-
cus hemolyticus, gas bacillus, etc.),  the  nature of  the
clean-up operation and the manner  of healing of  the
wound should be noted.
  2. Time to  Operate.—In a few cases it is permissible
to operate after the wound has been  completely healed
for a period of two months, while in  others, on account 
21
of possible latent infection, it may be advisable to delay
the  final  plastic work for  at least six  months.   In
some of the unfavorable latter cases, a two-step opera-
tive method  may be  followed,  consisting of a prelim-
inary  excision  of  scar tissue  with replacement by a
  Fig  15  (Cas( 4).--Case of  compression  fracture of  third, fourth
and fifth lumbar vertebrae caused by fall in  dugout.  Panel of verte-
brae (from elav model  of ease), showing relief of condition by  a tibia
graft inlaid  into spinous  processes  from first  lumbar  to  hrst sacral
vertebrae.


healthy skin flap,  muscle, fat, etc..  followed  after  a

period'of from ten days to two weeks by the final bone

plastic operation.

   3. Immediate Preoperative  Observations.—For  the

purpose of  determining  the  existence of latent  infec- 
T?
 tion. splints should be removed, and deep massage and
 rough manipulation should be practiced for a period of
 from one to two weeks prior to the operation.  During
 this time the temperature should be observed,  and the
 parts should be carefully  examined  for  local tender-
 ness or anv evidence of a recrudescence of infection.
 The field of operation should have a  forty-eight  hour
 preparation, iodin  technic being preferred.
  4. Plan  of  Operation and  Choice  of Incision.—By
 means of roentgenographies and physical examinations.
 the proposed plan  of operation,  especially in respect to
 location of graft, should be determined before  incision
 is made.  If possible, the  skin  incision should not lie
 directly over the  proposed bed of the graft, and the
 operation should be so planned that the graft  may be
 covered without undue tension of skin and, if possible,
 so placed that  it comes in contact with healthy tissue
 instead of scar tissue.  In several cases in which this
 has been accomplished, the graft has  healed in by pri-
 mary union, whereas the scar, even at a conslterable
 distance  from the  graft, has  broken down completely.
 In cases  of extensive loss of  bone, the scar tissue may
 be pushed to one  side, in order that the graft  may lie
 in healthy tissue.  Drainage wicks of any  kind should
 never be inserted at the time of operation.
  5. Duration  of  Operation.—It is  believed that the
 shortest  possible operating time consistent with  good
 work  and with a minimum amount of trauma is requi-
 site to successful results in these cases.
  6. Use  of Motor-Driven Instruments.—These are
 essential: (a) on  account  of the necessity for rapid
 work  in  order  that drying and  traumatization of the
 graft  tissues and host tissues may be avoided; (b) in
 order that a cabinet-maker fit may provide for mechan-
 ical fixation of parts and for the operation of  Roux's
 law of frictional stimulus to bone growth with a  view
 to an early and adequate establishment of  nourishment
to the graft; and  (c) on account of the necessity for
 the fulfilment  of  the  law of  anociassociation.   The
 motor outfit with its various tools to produce automatic
 fits seems indispensable.  The motor-saw, when used by
 the proper technic, does not heat nor glaze the bone.
  7. Adequate  Length  of Graft.—The  graft  should
always, when possible, be of  the inlay type, and  suffi- 
23
ciently long to extend into each fragment for a distance
of at least 2 inches, and always beyond the sclerosed
area.  The gutter should extend well into the healthy
marrow of the host-bone,  with which  the marrow of
the graft should be amply contacted.  It is exceedingly
important that the bridge of marrow  from the marrow
canal of one fragment to that of the other  should be
restored for the transmission of blood vessels, bone
cells, etc.
  8. Type of Graft.—The graft should, if possible, be
autogenous, consisting of all  the bone  layers, namely,
periosteum, complete  thickness of cortex,  endosteum
and marrow  substance; and it should be so inlaid that
the fit is perfect, with exact apposition of  correspond-
ing  layers of graft tissue to those of the host frag-
ments.   Such  a contacting  fit  not  only favors  the
mechanical fixation of the fractured bone and the graft,
but  also the very  potent  frictional  stimulus to bone
growth, emphasized by Roux.
  9. Stujpplemcntal Grafts for the Purpose  of  Osteo-
genesis.—Small  silver  grafts  placed  alongside  the
principal fixation graft are most efficacious in supply-
ing additional foci for bone growth.
  10.  The Graft as Main  Fixative  Agent.—Fixation
should always be secured by the graft itself, and not by
metal  plates  or  other  foreign material, since it is  the
stimulation  from the stress carried by  the graft itself
that is largely responsible  for the  healthy metabolism
of the graft  and for bone  growth.
  11. Suture Material.—The graft should  be  held in
place by a minimum amount of absorbable suture, pref-
erably kangaroo tendon, which is the  ideal material  for
this purpose, in that it is  tolerated  by tissue, readily
absorbable, very strong and reliable.   Fine absorbable
suture material should be used for  the skin and under-
lying soft parts.
  12. Postoperative Fixation.—The  limb  should   be
firmly immobilized by a plaster-of-Paris cast for a
period of  from eight to ten  weeks following operation,
and  as long thereafter as the  roentgen ray shows it to
be necessary.  Emphasis should again be placed on  the
importance of using absorbable skin suture material in
order that the plaster dressing need  not be disturbed
until time for the removal of  the splint. 
Reprinted }rotn The Journal of the  American  Medical  A^^ia
               Feb.  >S,  1920,  Vol.  74. pp.  589-597
                      Copyright, 1920
American Medical Association,  535 N. Dearborn St..  Chicago