ORIGINAL ARTICLE

Wound Healing Complications in Closed and Open

Calcaneal Fractures
Stephen K. Benirschke, MD and Patricia A. Kramer, PhD

Key Words: calcaneus, fracture, wound healing, ORIF, open reduc-

Objectives: To determine the rate of serious infection in closed and tion, internal fixation
open calcaneal fractures that were treated with open reduction and
internal fixation (ORIF) via an extensile lateral approach. (J Orthop Trauma 2004;18:1–6)

Design: Retrospective review.

Setting: Level 1 trauma center.
Patients: Two groups of patients with calcaneal fractures treated
with ORIF via an extensile lateral approach by the senior author are
T he orthopaedic community continues to debate whether
operative intervention in calcaneal fractures is appropri-
ate.1,2 One of the principal reasons cited for treating this injury
included. The first group contained 341 closed fractures in patients with closed or limited approaches is the belief that operative
injured during the period 1994–2000. The second group included 39
intervention does not improve the outcome for some patient
open calcaneal fractures in patients injured during the period 1989–
2000.
groups (eg, children3 or patients with simple fractures4). Em-
pirical evidence suggests, however, that restoration of calca-
Main Outcome Measurements: The age, sex, pre-existing medi- neal anatomy is the only way patients can have a normal gait5
cal conditions, compliance history, mechanism of injury, soft tissue or normal joint contact pressures.6 We believe that the ac-
status, presence of serious infection, and treatment of the infection cepted treatment of fractures in other weight-bearing joints—
were recorded for each patient. Data were gathered by review of pa-
anatomic reduction, stable internal fixation, atraumatic surgi-
tient records and by telephone interview when medical records were
incomplete. The rate of serious infection in the closed and open cal technique, and early mobilization—is equally appropriate
samples was determined. A literature review yielded 15 reports that for calcaneal fractures and that the extensile lateral approach is
contained sufficient detail to calculate the rate of serious infection. the best exposure for anatomic reconstruction.
Even when orthopaedic traumatologists acknowledge
Results: Of patients, 1.8% with closed fractures and 7.7% with open that restoration of calcaneal anatomy is important, the possi-
fractures experienced serious infections that required intervention be-
bility of serious infection can discourage them from advocat-
yond oral antibiotics. All of these feet eventually healed their inci-
sions and fractures. The calculations from data obtained from the lit-
ing operative intervention. Reports7–20 in the literature have
erature review indicate rates of serious infection of 0–20% for closed indicated disparity in the rate of infection associated with open
and 19–31% for open calcaneal fractures. reduction and internal fixation (ORIF), even when an extensile
lateral approach is used. The utility of these reports is limited
Conclusions: When done correctly in compliant patients, ORIF for by several factors, including small sample sizes, short follow-
calcaneal fractures via the extensile lateral approach (which allows up times, multiple surgeons, or multiple approaches. To ad-
for restoration of calcaneal anatomy after substantial disruption) does
dress these issues, we report herein on the rate of serious in-
not expose the patient to undue risk of serious infection.
fection in 341 closed and 39 open calcaneal fractures treated by
the senior author with ORIF via an extensile lateral approach
and a two-layer closure.

Accepted for publication June 30, 2003.

From the Department of Orthopaedics, Harborview Medical Center, Univer- MATERIALS AND METHODS
sity of Washington, Seattle, Washington. Our samples consisted of 341 closed displaced calcaneal
The authors received no financial support in connection with this project. fractures in 322 patients injured from 1994 to 2000 and 39
The article submitted does not contain information about medical devices. open fractures in 38 patients injured from 1989 to 2000. The
Reprints: Patricia A. Kramer, PhD, Department of Orthopaedics, Box 359798,
325 Ninth Avenue, Seattle, WA 98104-2499 (e-mail: pakramer@u.
open fracture sample includes a larger time frame to increase
washington.edu). the total sample size. All fractures were treated by the senior
Copyright © 2003 by Lippincott Williams & Wilkins author in a Level 1 trauma facility.

J Orthop Trauma • Volume 18, Number 1, January 2004 1

Benirschke and Kramer J Orthop Trauma • Volume 18, Number 1, January 2004

Sample Selection tions can appear spontaneously months after primary healing is
All patients with calcaneal fractures that were treated by complete. The descriptive statistics of the sample are given in
the senior author from January 1, 1994, to December 31, 2000, Table 1, including age, sex, pre-existing medical conditions,
were retrieved from our institution’s orthopaedic trauma data- compliance history, mechanism of injury, soft tissue status,
base. A total of 371 fractures initially were obtained; 30 of presence of serious infection, and treatment of that infection.
these were excluded because they were treated with a treat-
ment approach other than ORIF via the extensile lateral ap- Compliance
proach due to the nature of the injury or patient. No serious Noncompliance was noted when a patient failed to con-
infections were seen in these eliminated 30 fractures. The same form to physician orders. Typical noncompliance included
trauma database was queried for all open calcaneal fractures smoking before surgery or during the first 3 postoperative
from January 1, 1989, to December 31, 2000. A total of 46 months, failure to take prescribed antibiotics, and weight bear-
fractures were obtained, and 7 were eliminated because of a ing before cleared to do so. In all cases of noncompliance, the
treatment approach other than ORIF through an extensile lat- events were noted in the medical record (eg, “broken splint due
eral incision. All necessary amputations were performed emer- to weight bearing”).
gently. The open sample included 3 wounds classified as O1,
Outcome
14 as O2, 19 as O3A, and 3 as O3B.21
Patients requiring treatment beyond oral antibiotics
were classified as having serious infections. Symptoms in-
Treatment cluded persistent serous discharge, erythema, diffuse cellu-
All selected fractures were managed with ORIF via an lites, and other evidence of infection. Treatments for patients
extensile lateral approach. The proximate goal for each reduc- with serious infections included courses of intravenous antibi-
tion was the restoration of (1) calcaneal height, width, and otics appropriate to the infectious organism, irrigation and dé-
length; (2) the articular surface of the anterior, middle, bridement procedures, and hardware removal (Table 2). Hard-
and posterior facets; (3) the anatomic relationship of all three ware that was removed due to prominence was not counted as
subtalar joint surfaces; and (4) the correct orientation of the indicative of serious infection. All of our patients with serious
tuberosity. Re-establishing the height, width, length, and tu- infections went on to heal their incisions and their calcaneal
berosity orientation involved gross anatomic reduction, fractures. No patient required amputation.
whereas restoring the relationship of the articular surfaces re-
quired microreduction of the anterior fragments to the body Literature Review
and reconstruction of the critical angle of Gissane. Anatomic We reviewed the (1998 and later) literature on calcaneal
reduction was achieved in all fractures. All incisions were fractures treated with ORIF for reports with details that in-
closed with a two-layer approach, using inverted mattress su- cluded number of patients and infections, the characteristics of
tures in the deep layer and a horizontal modification of the the infection, and the treatment protocols for serious infec-
Allgöwer-Donati suture in the superficial layer. A bulky dress- tions. Table 3 lists the studies that met our criteria. We calcu-
ing was applied, and the limb was placed in a well-padded lated a rate of serious infection for each published study,
splint with the ankle in a neutral position and slight forefoot counting as serious any infection that required hospitalization,
pronation to lock the subtalar joint. Intravenous antibiotics surgery, or intravenous antibiotics after the initial fixation. We
were administered during surgery, and oral antibiotics (typi- avoided using superficial and deep as descriptors of infection
cally trimethoprim/sulfamethoxazole [Bactrim DS]) were pre- because the usage of these terms is not consistent in the litera-
scribed for patients whose incisions were not sealed at the time ture. We assumed that when delivery method was not given,
of discharge (usually 3 days after surgery) or who subse- oral antibiotics were administered; that when soft tissue status
quently presented with any symptom diagnostic of infection. was not explicitly stated, the fractures were closed; and that
Nasal oxygen was used until the patient’s oxygen saturation ultimate treatments were listed (ie, a patient with an amputa-
reached 100% on room air. Patients were non–weight bearing tion may have had courses of intravenous antibiotics and irri-
for a minimum of 3 months and were seen by a physical thera- gation and débridements, but was listed only under the ampu-
pist while in the hospital and during the transition from non– to tation category).
full–weight bearing.22,23
Statistical Analysis
We present no statistical analyses due to the small
Data sample size of patients with serious infections.
Data were gathered from the patient’s medical records
and, when the follow-up time was less than 1 year, from a tele- RESULTS
phone interview between the senior author and the patient. A In this study, 335 of 341 closed fractures and 36 of 39
minimum follow-up of 1 year was necessary because infec- open fractures healed without intervention other than oral an-

2 © 2003 Lippincott Williams & Wilkins

J Orthop Trauma • Volume 18, Number 1, January 2004 Serious Infections in Calcaneal Fractures

TABLE 1. Patient Characteristics

Closed Open
Patient statistics
No. displaced calcaneal fractures 341 39
No. patients 304 38
No. males 245 24
No. females 59 15
Average age in years (range) 39 (3–74) 38 (5–77)
No. feet of smokers 107 6
No. feet of diabetics 10 0
No. feet of patients with questionable compliance 32 10
No. feet of patients involved in L&I claims 84 7
Injury statistics
No. left calcanei 165 14
No. right calcanei 176 25
No. bilateral fractures 37 0
No. tongue-type fractures 134 14
No. joint depression fractures 192 23
No. other fracture types 36 2
No. with fracture blisters 32 2
No. injuries caused by falls 276 19
No. injuries caused by motor vehicle accidents 61 18
No. injuries caused by other accidents 4 2
Treatment statistics
Average time to surgery in days (range) 14 (0–222) 15 (1–29)
No. feet with bone grafts (allograft) 226 24
No. patients who received oral antibiotics at discharge 160* 16†
Clinical outcome
Average follow-up time in years 1.8 (1–10) 3.1 (1–10)
No. feet with hardware removal 132 19
Average time to hardware removal in years (range) 1.4 (0.7–5.5) 1.6 (0.9–2.7)
No. feet whose incisions healed without intervention 335 36
No. feet requiring intervention to heal 6 3
*Of 333 patients for whom data are available.
†Of 35 patients for whom data are available.

tibiotics. Six closed (1.8%) and three open (7.7%) fractures patient (B), who was nondiabetic, nonsmoking, and compliant,
required additional interventions, although all eventually experienced persistent drainage from the apex of his lateral
healed. The care of the patients who required intervention is incision, developed erythema at 4 months, and required hospi-
given in comprehensive detail in this section and summarized talization to administer intravenous antibiotics (cefazolin
in Table 2. [Kefzol]). Cultures showed polymicrobial growth. The drain-
age resolved after the intravenous antibiotic course, and the
Infections in the Closed Sample incision was healed and dry by 7 months after the initial sur-
The first patient (A) required an irrigation and débride- gery. The next patient (C) with a closed fracture had significant
ment of a hematoma 14 days after the initial ORIF. This rela- compliance problems, breaking five splints by weight bearing
tively aggressive treatment was indicated because the patient in the first 2 months after surgery. This homeless individual
had concurrent bilateral tibial fractures, was positive for hepa- was hospitalized for cellulitis and received intravenous antibi-
titis B and C, and had a recent history of intravenous drug otics (vancomycin and gentamicin). He was referred to a nurs-
abuse and needle sharing. After this procedure was completed, ing facility until he was able to bear weight and healed without
the patient exhibited a typical progression of healing. Another further incident.

© 2003 Lippincott Williams & Wilkins 3

Benirschke and Kramer J Orthop Trauma • Volume 18, Number 1, January 2004

TABLE 2. Patients requiring interventions beyond oral antibiotics

Patient Complication Treatment Comments

Closed A Hematoma I&D
B Erythema IV antibiotics
C Cellulitis IV antibiotics Noncompliance
D Deep infection I&D, IV antibiotics, hardware removal Posterior sore
E Cellulitis IV antibiotics, hardware removal Care elsewhere
F Cellulitis IV antibiotics, hardware removal Noncompliance
Open G Deep infection Hardware removal
H Cellulitis IV antibiotics Noncompliance
I Cellulitis Sinus tract excision, antibiotic beads

Three patients required hardware removal before their the initial surgery, and his foot healed without further compli-
surgical incisions healed, although the hardware was retained cations.
in each until consolidation of the fracture was complete. The next patient (E), a diabetic with bilateral fractures,
The first of these patients (D) presented for initial treat- was referred from an out-of-town clinic and consequently re-
ment with a posterior sore on his heel of uncertain etiology ceived most of his follow-up care in a setting other than our
and required ORIF for a second metatarsal fracture in addi- facility. At the time of discharge from the initial ORIF, his
tion to the calcaneal fracture. His surgical incision healed incisions appeared to be in excellent condition. At his 6-week
well, but he ultimately became infected with Staphylococ- and 4-month follow-up appointments with us, however, the
cus aureus through the posterior sore. He developed a deep apex of the surgical incision on the right foot was not healed
infection and required two episodes of irrigation and débride- completely. He was treated with intravenous antibiotics (cefa-
ment with intravenous antibiotics (vancomycin). His hard- zolin and gentamicin) for Staphylococcus and experienced
ware was removed at slightly less than 10 months from drainage and intermittent cellulitic symptoms until his hard-

TABLE 3. Reports of Infection Rates for Lateral Extensile ORIF (1998–2002)

Soft Tissue Sample Size: No. Fractures Serious Infection Average Time of
Status Reference (No. Patients) Rates* % (No.) Follow-up in Years
Closed Stiegelmar et al,18 2001 13 (13) 0 5.8
Huang et al,14 2002 32 (30) 0 2.9
Geel and Flemister,12 2000 33 (29) 3% (1) 1.7
Raymakers et al,16 1998 33 (31) 0 4.5
Assous and Bharma,9 2000 40 (40) 20% (8) 2.3
Tennent et al,20 2001 51 (47) 7.8% (4) 3.7
Strømsøe et al,19 1998 52 (46) 5.8% (3) 1.8
Shuler et al,17 2001 63 (62) 7.9% (5) >0.5
Naovaratanophas and Thepcharti,15 2001 114 (98) 2.6% (3) 6.8
Combined Al-Mudhaffar et al,8 2000 33 (30) 9% (2)
Aktuglu and Aydogan,7 2002 35 (28) 0 3.2
Brenner et al,10 2001 178 (178) 4.5% (8)
Folk et al,11 1999 190 (179) 21% (40)
Harvey et al,13 2001 218 (181) 1.8% (4) >0.4
Open Brenner et al,10 2001 26 (26) 19% (5) >0.8
Heier et al,24 1999 35 (?) 31% (11) 4.6
*Serious infections are those that require a more extensive intervention than the administration of oral antibiotics. These infections are listed variously using
terms such as hospitalization required, surgical procedure, and IV antibiotics. Also included under this category are infections that required hardware removal
before the infection would heal. In general, superficial infections are those for which oral antibiotics are sufficient. These are not included in this table.

4 © 2003 Lippincott Williams & Wilkins

J Orthop Trauma • Volume 18, Number 1, January 2004 Serious Infections in Calcaneal Fractures

ware was removed 9 months after surgery. After hardware re- DISCUSSION
moval, he healed uneventfully.
Literature Review
Another patient (F) experienced persistent drainage
The rate of serious infection after ORIF of closed calca-
from his incision that required intravenous antibiotics (genta-
neal fractures via the extensile lateral approach was calculated
micin and cefazolin) on two occasions. This patient had sig-
from the literature and ranged from 0–20% in closed fractures
nificant compliance problems, including refusal to take oral
(Table 3). The largest study15 reported three deep infections in
antibiotics consistently, to quit smoking, or to return for fol-
114 fractures for a rate of 2.6%, which is comparable to our
low-up appointments. This patient also had hepatitis C for
rate. The rates from the other, smaller studies also were similar
which he refused treatment, and his wound cultures consis-
to ours except for that of Assous and Bharma,9 in which 8 of 40
tently grew Staphylococcus epidermidis. He reappeared at 1
(20%) fractures required hardware removal before healing. It
year after his initial surgery and was noted to be cellulitic over
is unclear why these investigators obtained a relatively high
the lateral aspect of the incision with a draining wound. His
infection rate, but smoking is implicated as a significant risk
hardware was removed, and he healed without further inter-
factor for infection.
vention.
Several calcaneal series contained open and closed frac-
tures, with rates of serious infection ranging from 0–21%
Infections in the Open Sample (Table 3). The largest series13 contained some crossover with
Three patients with open fractures required treatment be- our series. It differed from our study in that their follow-up
yond oral antibiotics, including the case of a nonsmoking, time was relatively short (ⱖ5 months), and multiple surgeons
compliant patient (G) with bilateral, medially open (O2) frac- were involved. The next two largest studies reported 8 infec-
tures. This patient healed his surgical incisions and open tions in 178 fractures (4.5%)10 and 40 in 190 fractures (21%).11
wounds by primary intention but spontaneously developed It is unclear why Folk et al11 experienced relatively high infec-
wound drainage in the vertical limb of the lateral ORIF inci- tion rates, although they mentioned diabetes, smoking, and
sion 10 months after surgery. He ultimately developed a deep open wounds as risk factors for infection.
infection. Cultures showed coagulase-negative S. epidermidis Two reports on infection in open calcaneal fractures are
and Enterobacter cloacae. The hardware in his left foot was available. The study of Brenner et al,10 mentioned previously,
removed 1 year after initial surgery, and the patient went on to focused on complex foot trauma but included data specific to
heal his incision uneventfully. The right foot did not develop their calcaneal fractures. They had a 19% infection rate in open
any wound-healing problems. Hardware on the right foot was fractures treated with delayed soft tissue closure. The report
removed 15 months after the initial surgery, and the wound from Heier et al,24 in which a 31% infection rate was reported,
healed without intervention. was published only in abstract form, and consequently many
The next patient (H) was treated for a right open (O3B) details are not available.
calcaneal fracture and left open (O3A) tibial-fibular fracture. Although we were unable to use statistics because our
Her incisions and her medially open wound healed. She pro- sample sizes for serious infections were too small, we can
gressed well initially but developed a habit of resting the lateral comment anecdotally on what situations are problematic in
aspect of her foot against her mattress. Eventually, she devel- this large group of patients. The single biggest issue precluding
oped a pressure sore over the hardware, which became in- wound healing without infection is patient noncompliance.
fected. She could not comply with her oral antibiotic regimen. One third of our infections occurred in patients with significant
She developed cellulitis in the foot that required two courses of noncompliance, including inconsistent use of oral antibiotics.
intravenous antibiotics (cefazolin and ciprofloxacin) for S. au- Although noncompliance is a contraindication to surgical in-
reus before she healed. Her hardware was removed approxi- tervention, other frequently cited contraindications were man-
mately 2 years after her injury due to prominence, and deep aged successfully in this series. For instance, diabetes has been
cultures from the lateral wall of the calcaneus were negative mentioned as a risk factor for infection,11,20 but none of our 10
for microbial growth. The incision required to remove her patients with diabetes experienced serious infections. Another
hardware healed without incident. potential problem is exposure to infections from sources in the
The final patient (I) with an open fracture (O3A) and body distant from the calcaneus. Intravenous drug use is a
serious infection experienced intermittent apical drainage that known exposure risk, but other seemingly benign behaviors,
persisted even after hardware removal. Given the option of a such as extensive dental work, can pose risks. The patient with
free flap for wound coverage, he opted for intravenous antibi- the bilaterally open case that we mention may have seeded his
otics (cefazolin and gentamicin), sinus tract excision, and an- calcaneal plate from dental work.
tibiotic (tobramycin) bead insertion. The last-mentioned treat- Our infection rates of 1.8% for closed and 7.7% for open
ment resolved the drainage, and his wound has remained fractures are comparable to the low end of the distribution of
healed for 7 years. the series reported in the orthopaedic literature. Our experi-

© 2003 Lippincott Williams & Wilkins 5

Benirschke and Kramer J Orthop Trauma • Volume 18, Number 1, January 2004

ences combined with those of other institutions suggest that it treatment of calcaneus fractures: analysis of 190 fractures. J Orthop
Trauma. 1999;13:369–372.
is possible to minimize the chance of serious infection after 12. Geel CW, Flemister AS Jr. Standardized treatment of intra-articular cal-
ORIF of calcaneal fractures. We conclude that the possibility caneal fractures using an oblique lateral incision and no bone graft. J
of serious infection should not contraindicate treating calca- Trauma. 2001;50:1083–1089.
13. Harvey EJ, Grujic L, Early JS, et al. Morbidity associated with ORIF of
neal fractures similar to any other fracture in a weight-bearing intra-articular calcaneus fractures using a lateral approach. Foot Ankle Int.
joint—with anatomic reduction, stable internal fixation, atrau- 2001;22:868–873.
matic surgical technique, and early mobilization. 14. Huang PJ, Huang HT, Chen TB, et al. Open reduction and internal fixation
of displaced intra-articular fractures of the calcaneus. J Trauma. 2002;52:
946–950.
REFERENCES 15. Naovaratanophas P, Thepchatri A. The long term results of internal fixa-
tion of displaced intra-articular calcaneal fractures. J Med Assoc Thai.
1. Barei DP, Bellabarba C, Sangeorzan BJ, et al. Fractures of the calcaneus. 2001;84:36–44.
Orthop Clin North Am. 2002;33:263–285. 16. Raymakers JT, Dekkers GH, Brink PR. Results after operative treatment
2. Sangeorzan BJ, Benirschke SK, Sanders R, et al. The literature on calca- of intra-articular calcaneal fractures with a minimum follow-up of 2 years.
neal fractures is highly controversial. Foot Ankle Int. 2001;22:844–845. Injury. 1998;29:593–599.
3. Brunet JA. Calcaneal fractures in children: long-term results of treatment. 17. Shuler FD, Conti SF, Gruen GS, et al. Wound-healing risk factors after
J Bone Joint Surg Br. 2000;82:211–216. open reduction and internal fixation of calcaneal fractures: does correc-
4. Omoto H, Nakamura K. Method for manual reduction of displaced intra- tion of Bohler’s angle alter outcomes? Orthop Clin North Am. 2001;32:
articular fracture of the calcaneus: technique, indications and limitations. 187–192.
Foot Ankle Int. 2001;22:874–879. 18. Stiegelmar R, McKee MD, Waddell JP, et al. Outcome of foot injuries in
5. Catani F, Benedetti MG, Simoncini L, et al. Analysis of function after multiply injured patients. Orthop Clin North Am. 2001;32:193–204.
intra-articular fracture of the os calcis. Foot Ankle Int. 1999;20:417–421. 19. Stromsoe K, Mork E, Hem ES. Open reduction and internal fixation in 46
displaced intraarticular calcaneal fractures. Injury. 1998;29:313–316.
6. Mulcahy DM, McCormack DM, Stephens MM. Intra-articular calcaneal
20. Tennent TD, Calder PR, Salisbury RD, et al. The operative management
fractures: effect of open reduction and internal fixation on the contact
of displaced intra-articular fractures of the calcaneum: a two-centre study
characteristics of the subtalar joint. Foot Ankle Int. 1998;19:842–848.
using a defined protocol. Injury. 2001;32:491–496.
7. Aktuglu K, Aydogan U. The functional outcome of displaced intra- 21. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of
articular calcaneal fractures: a comparison between isolated cases and type III (severe) open fractures: a new classification of type III open frac-
polytrauma patients. Foot Ankle Int. 2002;23:314–318. tures. J Trauma. 1984;24:742–746.
8. Al-Mudhaffar M, Prasad CV, Mofidi A. Wound complications following 22. Benirschke SK, Kramer PA. Calcaneal fractures. Orthopaedic Knowl-
operative fixation of calcaneal fractures. Injury. 2000;31:461–464. edge Online (http://www5.aaos.org/oko/login.cfm), 2003.
9. Assous M, Bharma MS. Should os calcis fractures in smokers be fixed? A 23. DiGiovanni CW, Benirschke SK, Hansen ST. Foot injuries. In Trafton P,
review of 40 patients. Injury. 2001;32:631–632. ed. Skeletal Trauma. Philadelphia, PA: WB Saunders, 2003.
10. Brenner P, Rammelt S, Gavlik JM, et al. Early soft tissue coverage after 24. Heier KA, Infante AF, Walling AK, et al. The natural history and treat-
complex foot trauma. World J Surg. 2001;25:603–609. ment of open calcaneal fractures. In Orthop Trauma Association 15th An-
11. Folk JW, Starr AJ, Early JS. Early wound complications of operative nual Meeting, 1999.

6 © 2003 Lippincott Williams & Wilkins