Journal of Orthopaedic Trauma Publish Ahead of Print

DOI: 10.1097/BOT.0000000000001653

Impact of Prolonged Skeletal Traction in Patients with Acetabular fractures

Adam R Boissonneault, MB BCh BAO1, Mara Schenker, MD1, Jake Wilson, MD1, Andrew

D
Schwartz, MD1, Christopher Staley, BS1, Michael Maceroli, MD1

TE
1
Department of Orthopaedic Surgery, Emory University School of Medicine, Atlanta, Georgia,

USA
EP
Corresponding author contact details:

Adam Boissonneault

Department of Orthopaedic Surgery, Emory University School of Medicine

57 Executive Park South

C

Atlanta, GA 30329

Phone: 404-778-1567
C

Fax: 404-778-8192
A

Conflicts of Interest and Source of Funding:

None declared.

Presented in part at the Annual Meeting of the Orthopaedic Trauma Association, Orlando FL,

October 2018.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Abstract

Objectives. To explore the association between increased time in traction and in-hospital

pulmonary complications in patients with acetabular fractures

Design. Retrospective.

Setting. Level I trauma center.

D
Patients/participants. One-hundred ninety consecutive patients.

Intervention. Application of skeletal traction prior to fixation of acetabular fracture.

TE
Main outcome measurements. The primary outcome measure was pulmonary complication as

defined by pulmonary embolism, pneumonia, and acute respiratory distress syndrome (ARDS).

Secondary outcome measures included length of ICU stay (in days), total length of hospital stay
EP
(in days), deep hardware-associated infection, subsequent conversion to total hip arthroplasty

(THA), urinary tract infection (UTI), and lower extremity deep venous thrombosis (DVT).

Results. The mean time in traction for patients that suffered a pulmonary complication was 210

hours compared to 62 hours for those that did not (p<0.001). After controlling for ISS, chest
C

injury, and concomitant long bone injury requiring intramedullary nailing, the odds of

developing a pulmonary complication for patients that spent longer than 120 hours in traction
C

were over 40 times higher than those treated within 5 days (p<0.001). The mean ICU stay for

patients that spent at least 120 hours in traction was 17 days compared to 5 days for those treated
A

in less than 120 hours (p<0.001).

Conclusion. Early definitive fixation and decreased time in skeletal traction is associated with a

lower rate of complications in patients with acetabular fractures. Our results would suggest that

fixation of acetabular fractures prior to 120 hours (5 days) confers a significant risk reduction

benefit.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Level of Evidence. Level III.

Keywords: acetabular fracture; skeletal traction; pulmonary complication; damage control

Orthopaedics

Introduction

D
Early operative stabilization of femur fractures has been shown to decrease pulmonary

complications and length of intensive care unit (ICU) stay in the multiply injured patient 1-8.

TE
Despite the abundance of literature on early stabilization of long bone injuries, there is limited

evidence on the optimal timing for fixation of displaced acetabular fractures in similar patients 9-
10.
EP
Limitations to the current literature on timing of stabilization of pelvis and acetabular fractures

include a lack of information on pulmonary complications9, arbitrarily subgrouping patients into

only “early” (within first 24 hours) and “late” (after 24 hours) fixation9-10, and including isolated
C

pelvic ring injuries9-10. Unlike femoral shaft and pelvic ring injuries, which may be definitively

fixed via expedient surgeries utilizing small incisions or percutaneous techniques, displaced
C

acetabular fractures often require more extensive soft tissue dissection and longer operative times

to achieve anatomic fracture reduction. Patients with acetabular fractures and associated hip
A

instability are often placed in skeletal traction until definitive fixation. There is currently a

paucity of data on the impact of time in traction on outcomes in such patients.

The primary aim of this study was to review all patients with displaced acetabular fractures that

were placed in skeletal traction and to determine the association between time in traction and

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 pulmonary complications. We hypothesized that increased time in traction and its inherent,

mechanically-induced supine positioning would lead to increased rates of in-hospital pulmonary

complications.

Material and methods

D
This retrospective observational study reports on 190 consecutive patients that presented to a

single American College of Surgeons – verified Level I trauma center for surgical fixation of a

TE
displaced acetabular fracture between January 1, 2013 and December 31, 2016. Inclusion criteria

included all patients that were placed in skeletal traction at time of evaluation in the emergency

room. All patients were placed in skeletal traction within the emergency room and not as a
EP
separate procedure in the operating room. At our institution, indications for application of

skeletal traction include hip dislocation with inability to maintain concentric hip reduction,

displaced fractures without concentric reduction of the femoral head, fractures with incarcerated

intra-articular bony fragments, and all both column acetabular fractures. Patients with
C

concomitant pelvic ring injuries were included.

C

After institutional review board approval was obtained, medical records were queried

retrospectively. Time in traction was recorded in total hours and defined as the time from
A

placement of skeletal traction, as indicated by procedure/nursing notes, to the time of arrival in

the operating room for definitive fixation. The Injury Severity Score (ISS) and Abbreviated

Injury Score (AIS) were extracted from a prospectively collected institutional trauma registry.

Chest injury was defined as Chest AIS ≥ 2. Acetabular fractures were classified as described by

Judet et al. and also by the Orthopaedic Trauma Association (OTA) system after review of

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 anterior-posterior and Judet radiographs and preoperative computed tomography (CT) imaging
11-12
. Pelvic ring injuries were classified as described by Burgess et al. and also the OTA system

after review of anterior-posterior and inlet/outlet radiographs and preoperative CT imaging 12-13.

The primary outcome measure was pulmonary complication as defined by pulmonary embolism,

D
pneumonia, and acute respiratory distress syndrome (ARDS). Pulmonary embolism (PE) was

diagnosed by CT PE protocol scans. Patients empirically treated for suspected PE but without a

TE
diagnostic CT PE protocol scan were not included as a positive PE. Pneumonia was defined as

documented clinical (purulent sputum, leukocytosis) and radiological (pulmonary infiltrate on

chest radiograph) evidence of infection with positive respiratory cultures (≥ 10,000 colony-
EP
forming units (CFU) / milliliter (mL) for bronchoscopy sample and ≥ 100,000 CFU/mL for

sputum sample) that required treatment with intravenous (IV) antibiotics. ARDS was defined in

accordance with criteria outlined by the most recent ARDS Definition Task Force 14.
C

Secondary outcome measures included length of ICU stay (in days), total length of hospital stay

(in days), post-operative blood transfusion, deep hardware-associated infection, subsequent

C

conversion to total hip arthroplasty (THA), urinary tract infection (UTI), and lower extremity

deep venous thrombosis (DVT). Deep hardware-associated infection was defined as patients that
A

required return to the operating room for irrigation and debridement and subsequent treatment

with IV antibiotics. DVT was diagnosed by venous duplex scan.

The distribution of continuous numerical data including demographic, temporal, and ISS data

were examined in descriptive histograms and box plots, and a Kolmogorov Smirnov test was

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 used to confirm a normal distribution. Time in traction (reported in hours) was examined as both

a continuous and categorical variable. A Pearson bivariate correlation analysis was performed to

evaluate the correlation between time in traction and length of ICU stay, length of hospital stay,

and also ISS. The product-moment correlation coefficient (r) was reported with associated P

value. Correlation was characterized as poor (0.00-0.20), fair (0.21-0.40), moderate (0.41-0.60),

D
good (0.61-0.80), or excellent (0.81-1.00) in accordance with previously accepted guidelines 15.

An independent samples t-test was used to compare mean time in traction between patients who

TE
developed a pulmonary complication or not. An analysis of variance was used to compare

differences in mean time in traction by fracture type; a post-hoc Bonferroni analysis was

performed to explore pair-wise differences. Multi-variate logistic regression was performed to

EP
evaluate the odds of suffering a pulmonary complication based on various time in traction cut-

offs. Co-variates included in the regression model included ISS, chest injury, head injury, and

concomitant long bone injury requiring intramedullary nailing. All statistical analyses were

performed with Stata statistical software (StataCorp. 2015. Stata Statistical Software: Release 14.
C

College Station, TX: StataCorp LP.).

C

Results

The mean (standard deviation (SD)) age of the cohort was 38 (SD 15) years old. Of the 190
A

acetabular fractures, 27 (14%) were associated with a concomitant pelvic ring injury. The most

common acetabular fracture type was a transverse and posterior wall fracture (31%), followed by

a posterior wall fracture (28%) (Table 1). Acetabular fractures were associated with a hip

dislocation in 121 (64%) cases and a concomitant long bone injury in 36 cases (19%). Of the 36

long bone injuries, 22 (61%) underwent intramedullary nailing.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 The overall mean (SD) time in traction was 72 (SD 62) hours. Time in traction had a poor

correlation with ISS (r=0.18, p=0.014) (Figure 1), but a moderate correlation with both length of

ICU stay (r=0.54, p<0.001) and total length of hospital stay (r=0.53, p<0.001). An analysis of

variance demonstrated that time in traction was not related to fracture morphology (p=0.356)

D
(Table 1).

TE
The overall pulmonary complication rate for the entire cohort was 6.3% (12 of 190 patients).

Including multiple pulmonary complications within a single patient, there were 8 (4%) total

cases of ARDS, 8 (4%) cases of pneumonia, and 1 (0.5%) PE. The mean time in traction for
EP
patients that suffered a pulmonary complication was 210 hours compared to 62 hours for those

that did not (p<0.001). The mean ISS for patients that suffered a pulmonary complications was

significantly higher than those that did not, 20 (SD 11) versus 14 (SD10) (p=0.043). Patients that

had a pulmonary complication also spent a significantly longer time in ICU than those that did
C

not, 29 (SD 14) compared to 5 (SD 3) days (p<0.001). Pulmonary complications were also

associated with long bone injury requiring intramedullary nailing (p=0.001), chest injury
C

(p<0.001), and head injury (p<0.001).

A

Of the 115 (60%) fractures treated within 72 hours of initial application of skeletal traction, there

were 2 (2%) patients that suffered a pulmonary complication (Table 2). The 2% pulmonary

complication rate was significantly less than the 13% pulmonary complication rate for patients

that spent longer than 72 hours in traction (p=0.001). After controlling for ISS, chest injury, head

injury, and concomitant long bone injury requiring intramedullary nailing, patients that spent less

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 than 72 hours in traction were almost 7 times less likely to develop a pulmonary complication

(odds ratio (OR) 0.15, 95% confidence interval (CI) 0.03 to 0.75; p=0.022) (Table 3). In contrast,

patients that spent longer than 120 hours in traction had a significantly higher pulmonary

complication rate of 45% (p<0.001) (Table 2). After controlling for ISS, chest injury, head

injury, and concomitant long bone injury requiring intramedullary nailing, the odds of

D
developing a pulmonary complication for patients that spent longer than 120 hours in traction

were over 33 times higher than those treated within 5 days (OR 33.9, 95% CI 6.8-168.9;

TE
p<0.001) (Table 3). For patients that spent 135 hours or longer in traction – greater than one

standard deviation above the overall cohort 72-hour mean time in traction – the odds of

developing a pulmonary complication were almost 70 times higher than for patients treated in
EP
less than 135 hours (OR 68.8, 95% CI 13 to 368; p<0.001). Patients that spent an intermediate

73-120 hours in traction had a 2% pulmonary complication rate (p=0.104) (Table 2). There were

74 patients (40%) with an ISS > 15. For patients with ISS > 15, there was a 50% pulmonary

complication rate for those that spent longer than 120 hours in traction compared to a 2%
C

pulmonary complication rate for those that were treated within 5 days (p<0.001) (Figure 1).
C

Patients that were treated within 72 hours had a significantly shorter mean ICU stay (p=0.002)

and mean total hospital stay (p<0.001) compared to patients that spent longer than 72 hours in
A

traction (Table 2). The mean ICU stay for patients that spent at least 120 hours in traction was 17

days compared to 5 days for those treated in less than 120 hours (p<0.001). The mean total

length of hospital stay was 27 days for patients that spent over 120 hours in traction compared to

a mean of 12 days for those treated before 120 hours (p<0.001). Patients that developed a DVT

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 or deep hardware-associated infection also had a significantly higher mean time in traction

(Table 4).

Discussion

The results of the present study suggest that prolonged skeletal traction and delayed definitive

D
fixation of acetabular fractures are associated with increased in-hospital complications. While

early definitive management (<72 hours) was protective, delayed fixation (>120 hours) was

TE
associated with a 40 times increased risk of pulmonary complication. Additionally, our results

confirm previous studies that demonstrate an association between increased time to surgery and

increased length of ICU and total hospital stay9-10.

EP
Good clinical results have been shown when acetabular fractures are surgically managed up to 3

weeks after injury16. Many surgeons routinely delay treatment of acetabular fractures,

particularly those that require an anterior approach, several days or longer after injury17. In a
C

study by Furey et al., almost 75% of posterior wall acetabular fractures were fixed after 24

hours18. And similarly, in a study by Dailey et al., the mean time to surgery for posterior wall
C

fractures was 3 days and the mean time to surgery for fractures that required an anterior approach

was 4 days19. The overall mean time in traction for the current study was 72 hours, which is
A

consistent with these prior studies and recommendations16-19.

Early total care in the adequately resuscitated trauma patient with a long bone injury has been

shown to improve outcomes1-9. The current literature on the impact of time to treatment of

acetabular fractures use a similar 24-hour treatment threshold to define early versus late

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 fixation9-10. However, given that the majority of acetabular fractures are fixed between 2 and 5

days16,18-20, these findings are not generalizable or clinically applicable. It is not reasonable, or

safe, to suggest that all acetabular fractures should be fixed within 24 hours to improve

outcomes. In contrast, the current study offers an improvement to the literature in its examination

of time to fixation of acetabular fractures in more clinically relevant time periods. Our data

D
would suggest a more practical treatment threshold of 5 days before further surgical delay

becomes deleterious to outcomes.

TE
There are several limitations to the current investigation and the results must be interpreted in

this context. First, this is a retrospective investigation and therefore causality is unable to be
EP
definitively determined. Second, while the model used in this study controlled for many possible

confounding variables (ISS, chest injury, and concomitant long bone injury requiring

intramedullary nailing), it is difficult to completely isolate time in traction as an independent

variable as there are a multitude of factors that contribute to delays to fixation in these clinically
C

complicated patients. Similarly, while we controlled for concomitant chest injury, ISS and long

bone injury, many patients with acetabular fractures are multiply injured and the role that this
C

plays in outcomes is difficult to fully control for. Finally, we chose to focus on time in traction as

our surrogate for time to surgery as we feel it is this group of patients, in a mechanically
A

restricted supine position, that are most vulnerable to recumbent-related complications. The

results may not be applicable to patients not placed in skeletal traction and further propensity

controlled matching studies are underway to further explore any specific role that the actual

placement of skeletal traction may have.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Despite these limitations, this study addresses many of the shortcomings of prior studies. The

study excludes isolated pelvic ring injuries which are amenable to expedient percutaneous

techniques and could inappropriately skew the results. Additionally, this investigation is a

consecutive series of patients that presented to an urban, level-one trauma center. Given that this

is the setting of most acetabular fixation, the results should be generalizable to other such

D
centers. Our results are also enhanced by our methodology as we included only culture proven

infections (UTI and PNA) and utilized a recent, standardized definition of ARDS, thereby

TE
avoiding artificially inflating the incidence of such serious complications.

In conclusion, early definitive fixation and decreased time in skeletal traction (along with its
EP
necessitated recumbency) is associated with a lower rate of complications in patients with

acetabular fractures. While the principles of damage control orthopedics must continue to be

observed, our results would suggest that fixation of acetabular fractures prior to 120 hours (5

days) confers a significant risk reduction benefit. Future work should focus on prospective
C

validation of these results.

C
A

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Table 1. Fracture classification

Pulmonary

complication (n)

Total, n (% with

D
pulmonary Traction time

Classification No Yes complication) (mean (SD))

TE
Posterior wall 52 2 54 (4) 56 (41)

Posterior column 1 1 2 (50) 49 (1)

Anterior wall 0 0 0 (0) -

EP
Anterior column 3 0 3 (0) 121 (11)

Transverse 6 1 7 (14) 92 (129)

Transverse and posterior

wall 54 4 58 (7) 72 (65)

C

Posterior column and wall 17 1 18 (6) 67 (92)

T-type 17 2 19 (11) 79 (48)

C

Associated both column 17 1 18 (6) 92 (60)

Anterior column and PHT 11 0 11 (0) 82 (38)

A

PHT = posterior column hemitransverse

Traction time reported in

hours

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Table 2. Association between time in traction and complications

Hardware- Length of Total

Pulmonary associated ICU stay hospital
Time in complicatio DVT, UTI, n infection, n (mean stay (mean

D
traction No. n, n (%) n (%) (%) (%) days) days)
0 to 72 hrs 115 2 (2%) 5 (4%) 4 (3%) 0 (0%) 4.2 10.2
73 to 120 6

TE
hrs 55 1 (2%) 4 (7%) (11%) 2 (4%) 6.2 15.9
2 2
> 120 hrs 20 9 (45%) (10%) (10%) 1 (10%) 17.1 27.2
EP
Table 3. Odds ratios for pulmonary complication by time in traction

Pulmonary
Time in complication (OR(95%
C

traction No. CI)) p-value

0 to 72 hours 115 0.15 (0.03 to 0.75) p=0.022
C

73 to 120 hours 55 0.19 (0.02 to 1.74) p=0.142

> 120 hours 20 33.9 (6.8 to 168.9) p<0.001
OR = odds ratio; CI = confidence interval
A

*Multivariate logistic regression after controlling for ISS, chest

injury, head injury, and intramedullary nailing

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Table 4. Secondary outcomes

Event (time in traction mean(SD))

No. (%) No Yes p-value

Blood transfusion 58 (31%) 70 (54) 76 (77) p=0.490

DVT 11 (6%) 68 (53) 138 (134) p=0.002

D
UTI 12 (6%) 69 (61) 105 (65) p=0.054

Infection 3 (2%) 70 (60) 173 (117) p=0.004

TE
Subsequent THA 10 (5%) 71 (63) 73 (32) p=0.933

DVT = deep venous thrombosis; UTI = urinary tract infection; THA = total hip

arthroplasty
EP
References
C

1. Bone LB, Johnson KD, Weigelt J, et al. Early versus delayed stabilization of femoral

fractures. A prospective randomized study. J Bone Joint Surg Am. 1989 Mar;71(3):336-
C

40.

2. Behrman SW, Fabian TC, Kudsk KA, et al. Improved outcome with femur fractures:
A

early vs. delayed fixation. J Trauma. 1990 Jul;30(7):792-7.

3. Charash WE, Fabian TC, Croce MA. Delayed surgical fixation of femur fractures is a

risk factor for pulmonary failure independent of thoracic trauma. J Trauma. 1994

Oct;37(4):667-72.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 4. Bone LB, Anders MJ, Rohrbacher BJ. Treatment of femoral fractures in the multiply

injured patient with thoracic injury. Clin Orthop Relat Res. 1998 Feb;(347):57-61.

5. Brundage SI, McGhan R, Jurkovich GJ, et al. Timing of femur fracture fixation: effect on

outcome in patients with thoracic and head injuries. J Trauma. 2002 Feb;52(2):299-307.

6. O'Toole RV, O'Brien M, Scalea TM, et al. Resuscitation before stabilization of femoral

D
fractures limits acute respiratory distress syndrome in patients with multiple traumatic

injuries despite low use of damage control orthopedics. J Trauma. 2009 Nov;67(5):1013-

TE
21.

7. Nahm NJ, Vallier HA. Timing of definitive treatment of femoral shaft fractures in

patients with multiple injuries: a systematic review of randomized and nonrandomized

EP
trials. J Trauma Acute Care Surg. 2012 Nov;73(5):1046-63.

8. Harvin JA, Harvin WH, Camp E, et al. Early femur fracture fixation is associated with a

reduction in pulmonary complications and hospital charges: a decade of experience with

1,376 diaphyseal femur fractures. J Trauma Acute Care Surg. 2012 Dec;73(6):1442-8.
C

9. Plaisier BR, Meldon SW, Super DM, et al. Improved outcome after early fixation of

acetabular fractures. Injury. 2000 Mar;31(2):81-4.

C

10. Vallier HA, Cureton BA, Ekstein C, et al. Early definitive stabilization of unstable pelvis

and acetabulum fractures reduces morbidity. J Trauma. 2010 Sep;69(3):677-84.

A

11. Judet R, Judet J, Letournel E. Fractures of the acetabulum: classification and surgical

approach for open reduction. Preliminary report. J Bone Joint Surg Am. 1964

Dec;46:1615-46.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 12. Marsh JL, Slongo TF, Agel J, et al. Fracture and Dislocation Classification Compendium

- 2007: Orthopaedic Trauma Association Classification, Database and Outcomes

Committee. J Orthop Trauma. 2007;21 Supplement 10 pp: S1-S163.

13. Burgess AR, Eastridge BJ, Young JW, et al. Pelvic ring disruptions: effective

classification system and treatment protocols. J Trauma. 1990 Jul;30(7):848-56.

D
14. ARDS Definition Task Force, Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND,

Caldwell E, Fan E, Camporota L, Slutsky AS. Acute respiratory distress syndrome: the

TE
Berlin Definition. JAMA. 2012 Jun 20;307(23):2526-33.

15. Altman DG. Practical Statistics for Medical Research. London: Chapman and Hall; 1991.

16. Matta JM. Fractures of the acetabulum: accuracy of reduction and clinical results in
EP
patients managed operatively within three weeks after the injury. J Bone Joint Surg Am.

1996 Nov;78(11):1632-45.

17. Letournel E, Judet R. Fractures of the Acetabulum. 2nd ed. New York, NY: Springer-

Verlag; 1993.
C

18. Furey AJ, Karp J, OʼToole RV. Does early fixation of posterior wall acetabular fractures

lead to increased blood loss? J Orthop Trauma. 2013 Jan;27(1):2-5.

C

19. Dailey SK, Archdeacon MT. Open reduction and internal fixation of acetabulum

fractures: does timing of surgery affect blood loss and OR time? J Orthop Trauma. 2014
A

Sep;28(9):497-501.

20. Wright R, Barrett K, Christie MJ, et al. Acetabular fractures: long-term follow-up of open

reduction and internal fixation. J Orthop Trauma. 1994 Oct;8(5):397-403.

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 D
X

TE
X

X X
X

EP
X
X X
X
X X

C X
C
A

Figure 1. Correlation between ISS and time in traction

Red “X” denotes patient with pulmonary complication

Copyright Ó 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.