This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

Prevention of infection in the treatment of one thousand and twenty-five

open fractures of long bones: retrospective and prospective analyses
RB Gustilo and JT Anderson
J Bone Joint Surg Am. 1976;58:453-458.

This information is current as of March 11, 2011

Reprints and Permissions Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.
Publisher Information The Journal of Bone and Joint Surgery
20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org
 SECONDARILY INFECTED TOTAL JOINT REPLACEMENTS BY HEMATOGENOUS SPREAD 453

11. D'AMBRosiA, R. D.; LIPSCOMB,P. R.; MCCLAIN, E. J.; WISSINGER, H. A.; and MCDOWELL, J. H., III: Prophylactic Anticoagulation in Total
Hip Replacement. Surg. , Gynec. and Obstet. , 140: 523-527, 1975.
12. DEHAVEN, K. E.; EvART5, C. M.; WILDE, A. H.;C0LL1Ns, H. R.; NELSON, CARL; and RAZZANO, C. D.: Early Results of Charnley-Müller
Total Hip Reconstruction. Orthop. Clin. North America, 4: 465-472, 1973.
13. DUPONT, J. A., and CHARNLEY, JOHN: Low-Friction Arthroplasty of the Hip for the Failures of Previous Operations. J. Bone and Joint Surg.,
54-B: 77-87, Feb. 1972.
14. FITZGERALD, R. H., JR.: PETERSON, L. F. A.; WASHINGTON, J. A. II; VAN SCOY, R. E.; and COVENTRY, M. B.: Bacterial Colonization of
Wounds and Sepsis in Total Hip Arthroplasty. J. Bone and Joint Surg.. 55-A: 1242-1250, Sept. 1973.
15. GRISTINA, A. G.; ROvERE, G. D.; and SHoji, HIROMU: Spontaneous Septic Arthritis Complicating Rheumatoid Arthritis. J. Bone and Joint
Surg.. 56-A: 1180-1184, Sept. 1974.
16. JOHNSTON, R. C.: Clinical Follow-Up of Total Hip Replacement. Clin. Orthop., 95: 118-126, 1973.
17. KELLGREN, J. H.; BALL, J.; FAIRBROTHER,R. W.; and BARNES, K. L.: Suppurative Arthritis Complicating Rheumatoid Arthritis. British Med.
J., I: 1193-1200, 1958.
18. LAZANSKY, M. G.: Complications Revisited. The Debit Side of Total Hip Replacement. Clin. Orthop., 95: 96-103, 1973.
19. MCKEE, G. K., and WATSON-FARRAR, J.: Replacement of Arthritic Hips by the McKee-Farrar Prosthesis. J. Bone and Joint Surg., 48-B:
245-259. May 1966.
20. MURRAY. W. R.: Results in Patients with Total Hip Replacement Arthroplasty. Clin. Orthop. , 95: 80-90, 1973.
21 . PATTERSON, F. P., and BROWN, C. S.: The McKee-Fan-ar Total Hip Replacement. Preliminary Results and Complications of 368 Operations
Performed in Five General Hospitals. J. Bone and Joint Surg., 54-A: 257-275, March 1972.
22. WILSON, P. D.; AMSTUTZ, H. C.; CZERNIECKI, A.; SALVATI, E. A.; and MENDES, D. G.: Total Hip Replacement with Fixation by Acrylic
Cement. A Preliminary Study of 100 Consecutive McKee-Farrar Prosthetic Replacements. J. Bone and Joint Surg., 54-A: 207-236, March 1972.

Prevention of Infection in the Treatment of One Thousand and

Twenty-five Open Fractures of Long Bones
RETROSPECTIVE AND PROSPECTIVE ANALYSES

BY RAMON B. GUSTILO, M.D.*, AND JOHN T. ANDERSON, M.D.t, MINNEAPOLIS, MINNESOTA

ABSTRACT: In 673 open fractures of long bones There is universal agreement that open fractures re
(tibia and fibula, femur, radius and ulna, and quire emergency treatment, including adequate
humerus) treated from 1955 to 1968 at Hennepin débridement and irrigation of the wound. Beyond these
County Medical Center, Minneapolis, Minnesota, and two basic tenets, opinions differ as to the following:
analyzed retrospectively, the infection rate was 12 per I . Primary or secondary closure. If the wound is left
cent from 1955 to 1960 and 5 per cent from 1961 to open and secondary direct skin closure is not possible,
1968. In a prospective study from 1969to 1973,352 pa when does one do skin-grafting, or create a cross-leg pedi
tients were managed as follows: débridementand copi cle flap or a rotational flap?
ous irrigation, primary closure for Type I and II frac 2. Use of primary internal fixation. What are the in
tures and secondary closure for Type III fractures, no dications for primary internal fixation and if internal
primary internal fixation except in the presence of as fixation is delayed, when is the proper time for open reduc
sociated vascular injuries, cultures of all wounds, and tion and internal fixation?
oxacillm-ampicillin before surgery and for three days 3. Use of antibiotics. Should they be used routinely?
postoperatively. In 158 of the patients in the prospec What antibiotics should be used and for how long?
tive study the initial wound cultures revealed bacterial In an attempt to answer these questions we carried out
growth in 70.3 per cent and the infection rate was 2.5 retrospective and prospective analyses of a total of I ,025
per cent. Sensitivity studies suggested that cephalospo open fractures of all the long bones treated at Hennepin
rin is currently the prophylactic antibiotic of choice. County Medical Center, Minneapolis, Minnesota, from
For the Type III open fractures (severe soft-tissue in 1955 to 1973.
jury, segmental fracture, or traumatic amputation),
the infection rates were 44 per cent in the retrospective Retrospective Study, 1955 to 1968
study and 9 per cent in the prospective study. As previously reported 14.15, 673 open fractures of
long bones in 602 patients were treated at Hennepin
Prevention of wound sepsis remains the prime objec County Medical Center from 1955 to 1968. The data on
tive in the management of open fractures. The reported in these fractures were collected by review of the charts,
fection rates in these fractures, which range from 3 to 25 follow-up letters, and telephone calls. Of the 673 frac
per cent, are a challenge to every surgeon who treats tures, 583 (86.6 per cent) were followed for at least one
them 6-8,lO.I4,15,2i,22.26 month until soft-tissue healing had occurred. The remain
ing ninety patients failed to return for any follow-up.
* Hennepin County Medical Ccnter. Fifth and Portland South.
Minneapolis, Minnesota 55415.
All patients were cared for by the general surgical or
1@
Universityof Minnesota,Minneapolis.Minnesota. orthopaedic resident assigned to the Trauma Service with a

VOL. 58.A, NO. 4. JUNE l976

454 R. B. GUSTILO AND J. 1. ANDERSON

member of the attending staff consulting. The manage random, a problem in all retrospective studies. However,
ment of these fractures was essentially the same through we believe that the fractures treated with primary internal
out the period of this study, except for the antibiotics used. fixation could have been treated by plaster alone or by pins
Treatment during this period included: inserted above and below the fracture site and incorporated
1. Adequate débridement and copious irrigation of in a plaster cast. Fifty-two open fractures (thirty-four
the wound. tibiae and eighteen femora) treated by primary nailing or
2. Primary closure whenever possible, regardless of plating had a 19 per cent infection rate, and 238 open frac
the severity of the fracture and the soft-tissue injury. tures (194 tibiae and forty-four femora) were treated with
3. Internal fixation used at the discretion of the out internal fixation and had an infection rate of 5 per cent.
surgeon and not in accordance with any rigid criteria.
Some fractures treated by internal fixation could have been Primary Versus Secondary Wound Closure
treated in a plaster cast alone or with pins above and below In a retrospective study it is not possible to make
the fracture incorporated in a plaster cast. statistically valid comparisons. Obviously some of the
4. Routine antibiotics. During the first five years wounds that were left open were the most contaminated or
(1955 to 1960), penicillin, 40,000 to 60,000 units, was had the most tissue loss. Of forty-four open fractures
given intravenously every four to six hours in combination treated by initial débridement and secondary closure, 20
with streptomycin, 0.5 to 1.0 gram twice daily, starting per cent became infected, while only 6 per cent of the 495
postoperatively and continuing for seven to ten days. From fractures treated by primary closure became infected. Be
1961 to 1966, penicillin, 600,000 to 1.0 million units, and cause of these excellent results with primary closure, in
Chloromycetin (chloramphenicol), 0.5 to 1.0 gram, were 1966 we advocated primary closure for all open fractures
given intravenously every six hours starting one to three except when the surgeon thought that the débridementand
hours preoperatively and continuing for seven to ten days. irrigation had not been entirely satisfactory. From 1955 to
From 1967 to 1968, penicillin, 1.0 million units every six 1960, 18 per cent of the open fractures were left open;
hours, and oxacillin, 1.0 gram every four hours intra from 1961 to 1966,4 percent;and from 1967 to 1968,
venously, were started one to three hours before operation only 1 percent.
and continued for seven to ten days. Occasionally, when These over-all results might suggest that primary
the wound was severely contaminated, kanamycin, 0.5 closure is indicated for all types of open fractures. How
gram twice a day, was also given. ever, we observed that in the majority of the thirty-eight
infected cases the injuries were segmental fractures, ex
Results
tensive soft-tissue lacerations, or traumatic amputations
Infection Rate treated by primary closure. Furthermore, of the twenty
The infections were usually evident during the first one patients with open segmental tibial fractures and as
month after surgery, the majority being recognized during sociated extensive soft-tissue injuries treated from 1961 to
the first seven days. Drainage developed in thirty-eight 1968, the sixteen treated by primary closure (three of them
(6.5 per cent) of the wounds, of which thirty-one had posi by primary internal fixation as well) had an infection rate
tive and five had negative cultures, and two were not cul that was a staggering 44 per cent, while only one of the
tured. All thirty-eight wounds were considered infected fivetreated by delayedclosurehad an infection.
because of the purulent drainage and characteristic ap The conclusions from this retrospective study, there
pearance of the wounds. Sensitivity studies were done on fore, were that primary internal fixation increases the mci
all pathogenic bacteria recovered and the antibiotic pro dence of infection; that primary closure after segmental
grams were changed as indicated. The common patho open fracture, open fracture with extensivelaceration
genic organisms isolated were Staphylococcus aureus and avulsion of the soft tissues, or traumatic amputation
coagulase positive in twenty-six patients, Escherichia coli results in a high infection rate; and that prophylactic an
in three, Proteus in two, and Pseudomonasand a gram tibiotic therapy is essential in the treatment of open frac
negative rod in one each. tures.
The over-all incidence of wound infections from 1955
to 1960 was 11.85 per cent and from 1961 to 1968, 5.24 Prospective Study of352 Open Fractures, 1969 to 1973
per cent. These rates were considered low and appeared to From 1969 to 1973, all open fractures were classified
justify the methods of treatment, considering the high in into three categories:
cidence of infection reported in previous series of open Type I —¿An open fracture with a wound less than
fractures 6-8.10.14,15,20,21,26 one centimeter long and clean.
Type II —¿An open fracture with a laceration more
Primary Internal Fixation than one centimeter long without extensive soft-tissue
The incidence of infection in open fractures of the damage, flaps, or avulsions.
femur and tibia treated by primary internal fixation was Type III —¿Either an open segmental fracture, an
compared with that in fractures treated without internal open fracture with extensive soft-tissue damage, or a
fixation. The fractures were not matched or selected at traumatic amputation. Special categories in Type III were

THE JOURNAL OF BONE AND JOINT SURGERY

 PREVENTION OF INFECTION IN THE TREATMENT OF OPEN FRACTURES OF LONG BONES 455

gunshot injuries, any open fracture caused by a farm in the delayed infections, was possible for 326 of these frac
jury, and any open fracture with accompanying vascular tures, of which seventy-eight were Type I; 18 1, Type II;
injury requiring repair. and sixty-seven, Type III. Therefore, there was no
Guidelines forthetreatment of thesefractures were follow-up at all on twenty-six of the 352 fractures.
established prospectively and followed with only minor
Results
alterations. These guidelines were as follows:
1. All open fractures were treated as emergencies. Bacterial Flora and Sensitivities
2. Cultures were obtained routinely on admission Bacteriological studies were complete in the 158 con
and before wound closure or application of the postopera secutive patients seen after 1971 . Of these, eighty (50.7
tive dressing after 1971. per cent) showed a positive wound culture on admission,
3. Oxacillin, 1.0 gram every four hours, and ampicil and thirty-one (20 per cent) had a negative culture mi
lin, 1.0 gram every six hours, were given intravenously tially, but a positive culture at wound closure following ir
before, during, and for three days after surgery. Occasion rigation and débridement. Therefore, a total of I I 1 pa
ally, when the wounds were severely contaminated, gen tients (70.3 per cent) had a contaminated wound as shown
tamicin, three to five milligrams per kilogram of body by a positive culture either on admission or at wound do
weight, was given instead of ampicillin to “¿cover― sure.
gram-negative bacteria. If the wound was left open, an As shown in Table I, approximately fifteen organisms
tibiotics were continued for more than three days in ac were isolated on 143 occasions. Eighty-six of the isolates

TABLE I

of
OrganismNo.
IsolatesSensitivities*CephalothinPenicillinOxacillinCleocinAmpicillinGentamicinCarbenicillinChloramphenicolGram

positiveStaphylococcus
pos.1210020100100————Staphylococcus
coag.
neg.41100578595————Diphtheroids14————————AlphaStreptococcus,groupD1100100——————Bacillus
coag.

1————————Alpha
subtilis1
Streptococcus7505000————Gram
negativeEscherichia
co/i486———50100—100Klebsiella
enterobacter2083———6100—100Klebsiella
speciesI0———100100—100Enterobacter
00Pseudomonasspecies60———I 71 00—I
species170———0100100100Proteus
species3100———100100—100Herellea30———66100—100Achromobacter

species2———————100Other
gram-neg. rods1100————100—100

* The sensitivity studies were done by the Kirby-Bauer method. A dash indicates that no studies were done.

cordance with the finding by wound culture or the appear were gram positive and fifty-seven were gram-negative
ance of the wound. Usually, antibiotics were continued for organisms. Thirty-two of the 158 wounds yielded mixed
three days after secondary wound closure in the Type III bacterial growth.
fractures. Based on the sensitivity studies of these organisms,
4. Thorough débridement and copious irrigation the cephalosporins (cephalothin and cefalozin) appeared to
were emphasized, and for the more recently treated Type be the most effective antibiotics for prophylaxis after open
III injuriesjet lavage was used i3.16@ fractures. These agents are effective against all gram
5. No primary internal fixation was employed except positive organisms including Staphylococcus coagulase
in rare cases in which vascular injury required repair. Ex positive and negative, and the majority of Escherichia
ternal skeletal fixation or traction was the preferred coli, Klebsiella, and Proteus, excluding Pseudomonas.
method of immobilization. We did not find that anaerobes were particularly
6. Primary closure was performed in Type I and II significant in these studies, presumably because the
fractures and delayed closure, in Type III lesions. anaerobic culture technique at our institution was not satis
From 1969 to 1973, 352 open fractures were treated factory prior to 1974.
at Hennepin County Medical Center according to the pro
tocol described. Ofthese fractures, 81 per cent were Types Wound Closure
I and II and 19 per cent, Type III. Follow-upfor at least Of the sixty-seven Type III fractures. 19 per cent
six weeks, sufficient to determine the incidence of all but were treated by delayed secondary closure, either direct

VOL. 58.A, NO. 4, JUNE 1976

456 R. B. GUSTILO AND J. T. ANDERSON

skin closure, skin-grafting, or allowing the wound to heal and provides better coverage of the bone.
by granulation tissue. Early (at five days to three weeks) Bacterial culture and sensitivities of the eight infected
pedicle flaps were used in five cases. open fractures from 1969 through I 973 revealed that three
had infections with gram-positive and gram-negative bac
Primary Internal Fixation teria. The predominant organisms were Staphylococcus
In the last five years, only two open fractures were coagulase positive , Klebsiella enterobacter, Escherichia
treated by primary internal fixation. Both were commi co/i, Proteus, and Staphylococcus coagulase negative.
nuted supracondylar fractures, complicated by injuries to
Fracture Healing
the popliteal artery requiring repair. Both wounds became
infected. In seventeen femoral-shaft fractures (thirteen Of the 352 fractures in the prospective study, 197
Type I and II and four Type III) treated by intramedullary were followed with roentgenographic and clinical evalua
nailing ten days or more after fracture, there were no infec tions for at least six months after injury. The other 155
tions. had not been followed for six months, had been trans
ferred to private care, or were lost to follow-up at the
Infection Rate time of writing. Of the 197 fractures which were followed,
Eight (2.4 per cent) of the 326 fractures in the pros 170 (86.4 per cent) united uneventfully and twenty-seven
pective study became infected compared with twenty-four (13.6 per cent) failed to unite and required bone-grafting.
(5.2 per cent) of the 458 fracturesin the seriesfrom 1961
to 1968 and fourteen ( 1 1.8 per cent) of 135 fractures in the
Discussion
series from 1955 to 1960 (p < 0.02). Of the eight infec
tions in the prospective series, six involved fractures of the Adequate Débridement and Copious Irrigation
tibial shaft and two, supracondylar fractures of the femur. Scully and co-workers described criteria for muscle
Six were Type III and two, Type II. Three of the eight viability called the four C's: ( 1) color, (2) capacity to
fractures were associated with vascular injuries, two of the bleed, (3) contractility, and (4) consistency. In our experi
popliteal (in one patient) and one of the posterior tibial ar ence color was a reliable guide and the other three C's cor
tery. Five of these eight infections appeared to be secon related well with muscle viability. However, adequate
dary, developing when the wound cultures became posi débridementremains a difficult technical problem. If there
tive after the wounds had been open for two weeks or more is the slightest doubt in the surgeon's mind as to whether
because we had failed to close them secondarily by re there has been adequate débridetnent of the wound after
peated skin grafts. For the Type III fractures the infection an open fracture, the wound should not be closed regard
rate in this series was 9.9 per cent in contrast to the 44 per less of the type of open fracture. For the surgeon who
cent rate in the group studied retrospectively (p < 0.01). manages on/v an occasional open fracture, the safe rule is
The three open fractures associated with vascular in not to close the wound.
jury and extensive soft-tissue damage were treated by Copious irrigation implies that the wound is irrigated
débridement(considered adequate at the time), copious ir with normal saline solution in large amounts in conjunc
rigation, primary internal fixation, vascular repair, and tion with débridement in order to remove all dirt and
packing the wound open. In the patient with bilateral su foreign material as well as all devitalized tissue. In our in
pracondylar open fractures and bilateral lacerated popliteal stitution, the residents use an average of ten to fourteen
arteries, the operative time was eight hours, while in the liters of normal saline solution for open fractures of long
other patient with a comminuted open fracture in the prox bones. Recently use ofjet lavage, particularly in Type III
imal part of the tibia and laceration of the posterior tibial fractures, accomplished this task more effectively in a
artery, the time was three and one-half hours. To our dis shorter time and with about half as much irrigating solu
may, forty-eight to seventy-two hours later we found more tion.
soft-tissue necrosis in all three wounds, which required
Rationale for Routine Use of Antibiotics
further débridement. Infection was evident several days
later in each instance. In our institution, every open fracture is now consid
In the uninfected Type III wounds, delayed primary ered a severely contaminated wound and, therefore, more
closure by either direct suture or split-thickness skin likely to become infected than a wound made in a clean,
grafting was done five to seven days after the primary pro elective surgical procedure. Studies by Patzakis and as
cedure in all but five cases in which rotational or cross-leg sociates and our own study (Table I) showed that open
pedicle flaps were created. Split-thickness skin-grafting fractures are contaminated. In our series, 70.3 per cent of
was successful over exposed bone when the periosteum the open wounds yielded positive cultures. Therefore, we
was intact; otherwise, we resorted to some type of flap consider the routine use of antibiotics in open fractures as
coverage. To date we have tried delayed (five days to three therapeutic rather than prophylactic, and we think that
weeks after injury) primary cross-leg pedicle grafts to subsequent changes in antibiotic coverage should be
cover large exposed areas of bone five times, all success guided by the sensitivity of the organisms isolated from
ful. This procedure appears to prevent delayed infection the infected open fractures. Use of Chloromycetin was dis
THE JOURNAL OF BONE AND JOINT SURGERY
 PREVENTION OF INFECTION IN THE TREATMENT OF OPEN FRACTURES OF LONG BONES 457

continued in the early phase of the study because of the solution and also shortening the operating time. At the end
rare complication of agranulocytopenia. ofthe procedure the wound is packed open. and five to six
Since our culture and sensitivity studies of the bacte days later split-thickness skin grafts are applied, followed
rial flora in open fractures and of the bacteria isolated from if need be by repeated skin grafts as a good base of granu
the infected wounds suggested that the best single anti lation tissue develops.
biotic for prophylaxis is a cephalosporin (cephalothin or During the long intervals that such wounds are open,
cefalozin), in 1974 we began using cephalosporins exclu secondary infections, usually with gram-negative or
sively as the antibiotics for open fractures. ganisms, may be a problem since these organisms are usu
The dosage and duration of antibiotic administration ally difficult to control by antibiotics alone. Prolonged
remain unsettled questions. Antibiotics were given most aminoglycoside therapy, particularly for Pseudomonas, is
commonly for seven to ten days in all previous studies, in dangerous and of questionable value. Daily soaks of the
cluding our own. We could find no evidence to support open wound with solutions such as 0.25 per cent silver ni
this practice. The rationale appears to be that the wound trate, hydrogen peroxide, and acetic acid, or with plain
heals in seven to ten days. On the other hand, recent clini normal saline, help to control these infections and enhance
cal reports 2.22 have shown that prophylactic antibiotics the formation of good granulation tissue. A cross-leg pedi
given for a shorter time, such as three to five days, are cle flap or rotational flap to cover any exposed bone or
probably just as effective in preventing wound infection large soft-tissue defect, done early, may prevent secon
and have the advantage that if infection develops, it will be dary contamination and subsequent infection. We treated
manifest while the patient is still in the hospital. In our five patients with cross-leg pedicle flaps to cover extensive
present regimen prophylactic antibiotics for open fractures soft-tissue defects in the leg five days to three weeks after
are given one to three hours before surgery, during injury, while extraskeletal fixation was used to immobilize
surgery, and for three days thereafter. They are then the unstable open fractures. The results were satisfactory
stopped whether the patient is febrile or not. If the wound in all five patients.
becomes infected, gram stain and culture are obtained, and Another major problem is the comminuted open frac
surgical débridementand irrigation are done immediately. ture with vascular injury and extensive soft-tissue damage.
The appropriate antibiotic is administered as soon as the This injury usually requires extensive débridenient and
sensitivities of the bacteria are known. primary repair of the damaged blood vessels. In the past it
If the wound is left open initially, antibiotics are con has been our policy to fix all fractures associated with ar
tinued depending on the appearance of the wound and the terial injury internally, in order to provide stability for the
results of its culture. It is difficult to keep a large open repaired vessel. To do this, however, increases the operat
wound sterile for a long period of time, and it should be ing time as well as the amount of dissection around the
closed secondarily with either a skin graft or a cross-leg fracture site, with consequent additional surgical trauma
pedicle flap as soon as possible. If a wound is not closed, which predisposes to wound sepsis.
secondary contaminants such as Pseudomonas and other The danger of vascular disruption from unstable frac
gram-negative bacteria, for which the antibiotics used are ture fragments prompted many authors to advise internal
not effective, appear and become pathogenic. fixation of the fracture prior to arterial repair in patients
The aminoglycosides (kanamycin and gentamicin) with combined bone and arterial injuries. However, ex
may be used occasionally in severely contaminated open perience with injuries treated during the Vietnam war
fractures, primarily for gram-negative bacteria, but the pa raised questions about the necessity for internal fixation in
tient must be watched cat.fully for nephrotoxic effects. severely traumatized limbs 11.18.24 A study of civilian
The serum creatinine and blood urea nitrogen values femoral and tibial fractures by Connolly and co-workers
should be determined daily. In patients who have had mul also did not substantiate the need for internal fixation,
tiple trauma with severe open fractures, the amino since 4.5 to 6.8 kilograms of skeletal traction was shown
glycosides predispose to acute renal shutdown. These to immobilize these fractures safely and effectively after
drugs should be used only if the anticipated beneficial ef arterial repair, especially when there was a severely
fects are deemed essential after careful weighing of the p0- contaminated wound or when the blood flow had to he
tential benefits and dangers. restored as soon as possible. The method of managing
the fracture, therefore, probably should not take preced
Unresolved Problems ence over treatment of the related serious soft-tissue
During the last five years we have focused our atten injury 5.
tion on Type III fractures characterized by extensive soft In open fractures with arterial injury it is difficult to
tissue loss, instability, and large areas ofexposed bone re determine the extent of avascular and devitalized tissue
quiring soft-tissue coverage. In these cases thorough initially. In our experience. despite prompt arterial repair,
débridementof the devitalized tissue is absolutely essen non-viable and necrotic tissues not recognized at the initial
tial, and during the process large areas of bone are ex procedure were found two to three days later. This over
posed. Use of the jet lavage apparatus certainly helps in sight was obviously the result of inadequate circulation not
cleaning these wounds, reducing the amount of irrigating evident initially.
VOL. 58-A,NO. 4.JUNE 1976
458 R. B. GUSTILO AND J. T. ANDERSON

Therefore, we now recommend in such cases that a cated is believed to have helped in the prevention of this
complete débridementbe carried out, as is done in all open complication.
fractures, the damaged blood vessels repaired, and the
Conclusions
fracture immobilized in traction, leaving the wound open
to be closed secondarily by direct closure, split-thickness 1. Open fractures require emergency treatment, in
skin-grafting, or a cross-leg pedicle flap. cluding adequate débridementand copious irrigation.
The use of large metallic fixation devices such as 2. Primary closure is indicated for Type I and II open
plates and intramedullary nails may be indicated rarely to fractures, but delayed primary closure, including split
provide stability and to facilitate wound care so that the thickness skin grafts or appropriate flaps, should be used
soft tissues around the fracture site may be preserved. In for Type III open fractures.
only one patient, treated in 1974, was a Lottes nail used 3. Internal fixation by plates or intramedullary nails
for a Type III fracture of the tibia associated with an exten should not be used. External skeletal fixation by skeletal
sive, almost circumferential degloving soft-tissue wound traction, pins above and below the fracture site incorpo
of the thigh. By stabilizing the tibia, care of the thigh and rated in a plaster cast, or such devices as the Roger Ander
leg wounds, which otherwise would have been very son or Muller apparatus is recommended.
difficult, was made much easier. We do not close wounds 4. Open fractures associated with arterial injury re
primarily when large metallic devices are used in the mi quiring repair should be treated in skeletal traction
tial care of fractures, and we advocate strongly that ex whenever possible instead of by primary internal fixation.
traskeletal fixation devices, such as pins incorporated in 5. Antibiotics should be administered before and
plaster, a Roger Anderson device, or a Muller apparatus, during surgery, the antibiotics of choice currently being
be used whenever stability is needed to facilitate care of an the cephalosporins in therapeutic doses. If the wound is
open wound. closed primarily, the antibiotics are stopped on the third
It is worthy of note that there have been no cases of postoperative day. If the wound is closed secondarily, the
gas gangrene in over 1,000 patients treated at Hennepin antibiotics are continued for another three days after this
County Medical Center. The method of treatment advo procedure.
References
1. ALTEMEIER, W. A.; CULBERTSON, W. R.; SHERMAN, ROGER; COLE, WILLIAM; ELSTUN, WESLEY; and FULTZ, C. T.: Critical Reevaluation of
Antibiotic Therapy in Surgery. J. Am. Med. Assn. , 157: 305-309, 1955.
2. BOYD, R. J.; BURKE, J. R.; and COLTON, THEODORE:A Double-Blind Clinical Trial of Prophylactic Antibiotics in Hip Fractures. J. Bone and
Joint Surg., 55-A: 1251-1258, Sept. 1973.
3. BRAy, E. A.: The Management of Open Fractures of the Extremities. In Instructional Course Lectures, The American Academy of Orthopaedic
Surgeons. Vol. 13, pp. 227-233. Ann Arbor, J. W. Edwards, 1956.
4. CONNOLLY, J. F.; WHITTAKER, DAVID; and WILLIAMS, EUGEN: Femoral and Tibial Fractures Combined with Injuries to the Femoral or Popliteal
Artery. A Review of the Literature and Analysis of Fourteen Cases. J. Bone and Joint Surg.. 53-A: 56-68. Jan. 1971.
5. CONNOLLY, JOHN; WILLIAMS, EUGEN; and WHITTAKER, DAVID: The Influence of Fracture Stabilization on the Outcome of Arterial Repair in
Conibined Fracture-Arterial Injuries. Surg. Forum, 20: 450-452, 1969.
6. COPELAND, C. X., JR., and ENNEKING, W. F.: Incidence of Osteomyelitis in Compound Fractures. Am. Surg. , 31: 156-158, 1965.
7. DALAND, E. M.: The Management of Two Hundred and Thirty-six Compound Fractures Treated at the Massachusetts General Hospital. New
England J. Med. , 210: 983-995, 1934.
8. DAvis, A. G.: Primary Closure of Compound-Fracture Wounds, With Immediate Internal Fixation, Immediate Skin Graft, and Compression
Dressings, J. Bone and Joint Surg. , 30-A: 405-415, April 1948.
9. EMERY, M. A., and MURAKAMI, H.: The Features of Fracture Healing in Cats after Immediate and Delayed Open Reduction. J. Bone and Joint
Surg. , 49-B: 571-579, Aug. 1967.
10. Epps, C. H., JR., and ADAMS, J. P.: Wound Management in Open Fractures. Am. Surg., 27: 766-769, 1961.
1 1. FISHER, G. W.: Acute Arterial Injuries Treated by the United States Army Medical Service in Vietnam, 1965-1966. J. Trauma, 7: 844-855,
1967.
12. FOGELBERG,E. V.; ZITZMAN, E. K.; and STINCHFIELD,F. E.: Prophylactic Penicillin in Orthopaedic Surgery. J. Bone and Joint Surg., 52-A:
95-98, Jan. 1970.
13. GROSS, ARTHUR; CUTRIGHT, D. E.; and BHASKAR, S. N.: Effectiveness of Pulsating Water Jet Lavage in Treatment of Contaminated Crushed
Wounds. Am. J. Surg. , 124: 373-377, 1972.
14. GUSTILO, R. B.: Management of Open Fractures: An Analysis of 673 Cases. Minnesota Med., 54: 185-189, 1971.
15. GUSTILO, R. B.; SIMpsoN, L.; NIXON, R.;RUIZ, A.; and INDECK, WALTER: Analysis ofSll Open Fractures. Clin. Orthop., 66: 148-154, 1969.
16. HAMER, M. L., and STUCKER, P. J.: Scientific Exhibit, Annual Meeting of The American Academy of Orthopaedic Surgeons. San Francisco,
California, February 1971.
I 7. LAM, S. J.: The Place of Delayed Internal Fixation in the Treatment of Fractures of the Long Bones. J. Bone and Joint Surg. , 46-B: 393-397,
Aug. 1964.
18. LEVITSKY, S.; JAMES, P. M.; ANDERSON, R. W.; and HARDAWA,R.: VascularTrauma in Vietnam Battle Casualties. Ann. Surg., 168: 831-836,
1968.
19. MCNAMARA,J. J.; BRIEF, D. K.: STREMPLE,J. F.; and WRIGHT,J. K.: Management of Fractures with Associated Arterial Injury in Combat
Casualties. J. Trauma, 13: 17-19, 1973.
20. ORR, H. W.: The Treatment of Fractures. J. Orthop. Surg. . 3: 23-27, 1921.
21. ORR, H. W.: Compound Fractures. With Special Reference to the Lower Extremity. Am. J. Surg., 46: 733-737, 1939.
22. PATZAKIS, M. J.; HARVEY, J. P., JR.: and IVLER, DANIEL: The Role of Antibiotics in the Management of Open Fractures. J. Bone and Joint
Surg., 56-A: 532-541, April 1974.
23. PAVEL, ALAN; SMITH, R. L. ; BALLARD, ANTHONY; and LARSEN, I. J.: Prophylactic Antibiotics in Clean Orthopaedic Surgery. J. Bone and
Joint Surg. , 56-A: 777-782, June 1974.
24. RICH, M. N., and HUGHES, C. W.: Vietnam Vascular Registry: A Preliminary Report. Surgery. 65: 218-226, 1969.
25. SCULLY, R. E.; ARTZ, C. P.; and SAKO, Y0SHI0: An Evaluation of the Surgeon's Criteria for Determining Viability of Muscle During Debride
ment.Arch.Surg.,73: 1031-1035,1956.
26. VELISKAKIS, K. P.: Primary Internal Fixation in Open Fractures of the Tibial Shaft. The Problem of Wound Healing. J. Bone and Joint Surg.,
41-B: 342-354, May 1959.

THE JOURNAL OF BONE AND JOINT SURGERY