- Email: [email protected]
Comparison of Wound Complications and Deep Infections With Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients
Accepted Manuscript Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients Richard L. Purcell, MD, Nancy L. Parks, MS, John P. Cody, MD, William G. Hamilton, MD PII:
S0883-5403(17)30686-1
DOI:
10.1016/j.arth.2017.07.047
Reference:
YARTH 56027
To appear in:
The Journal of Arthroplasty
Received Date: 28 October 2016 Revised Date:
22 June 2017
Accepted Date: 27 July 2017
Please cite this article as: Purcell RL, Parks NL, Cody JP, Hamilton WG, Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients, The Journal of Arthroplasty (2017), doi: 10.1016/j.arth.2017.07.047. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Comparison of Wound Complications and Deep Infections with
RI PT
Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients
1 Anderson Orthopaedic Research Institute PO Box 7088 Alexandria, VA 22307
TE D
2 Inova Joint Replacement Center Mt. Vernon Hospital 2501 Parker's Lane, Suite 200 Alexandria, VA 22306
M AN U
SC
Richard L. Purcell, MD1,3 Nancy L. Parks, MS1 John P. Cody, MD1,3 William G. Hamilton, MD1,2
EP
3 Walter Reed National Military Medical Center 8901 Wisconsin Ave. Department of Orthopaedics, Building 19, 2nd Floor Bethesda, MD 20889
AC C
Please address all correspondence to: Nancy L. Parks, MS Anderson Orthopaedic Research Institute PO Box 7088 Alexandria, VA 22307 Phone: (703) 619-4423 FAX: (703) 799-5982 Email: [email protected]
ACCEPTED MANUSCRIPT
Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients
RI PT
Introduction: The purpose of this study was to compare the posterior approach (PA) to the direct anterior approach (DAA) among obese and non-obese total hip arthroplasty (THA) patients to determine if obese DAA patients have a higher risk of infection or wound complications
SC
compared to obese PA patients.
Methods: We retrospectively evaluated 4651 primary THA performed via DAA or PA between
M AN U
2009-2015. Patients were divided into four study groups based on approach and BMI: 1)DAA <35kg/m2, 2)DAA ≥35kg/m2, 3)PA <35kg/m2, 4)PA ≥35kg/m2. Infection rates and wound complications were compared.
Results: The rate of deep infection in Groups 1 and 3 (non-obese) was 0.28% and 0.36% (p=0.783); and in Groups 2 and 4 (obese) was 2.35% and 2.7% (p=0.80). The rate of wound
TE D
complications between Groups 1 and 3 (non-obese) was 1.0% and 0.3% (p=0.005). Between Groups 2 and 4 (obese) the rates of complications were 1.7% and 1.4% (p=1.0). There was no
EP
difference in re-operation rates for wounds between Groups 1 and 3 or between Groups 2 and 4 (p=0.217, p=0.449).
AC C
Conclusion: In the largest available series DAA THA experienced higher rates of superficial wound complications compared to PA THA regardless of BMI. However there was no difference in deep infection rates between the two approaches. In the subset of patients with BMI ≥35kg/m2 there was no increased risk of deep infection wound complications in obese DAA patients compared to their obese PA counterparts.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Keywords: Total Hip Arthroplasty, Anterior Approach, Posterior Approach, Obesity, Wound Complications, Infection
RI PT
ACCEPTED MANUSCRIPT
Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior
AC C
EP
TE D
M AN U
SC
Approaches in Obese Hip Arthroplasty Patients
1
ACCEPTED MANUSCRIPT ABSTRACT
2
Background: The purpose of this study was to compare the posterior approach (PA) to the direct
3
anterior approach (DAA) among obese and non-obese total hip arthroplasty (THA) patients to
4
determine if obese DAA patients have a higher risk of infection or wound complications compared to
5
obese PA patients.
6
Methods: We retrospectively evaluated 4651 primary total hip cases performed via anterior or
7
posterior approach between 2009-2015. Patients were divided into four study groups based on
8
approach and BMI: 1) DAA <35kg/m2, 2) DAA ≥35kg/m2, 3) PA <35kg/m2, 4) PA ≥35kg/m2.
9
Infection rates and wound complications were compared.
SC
RI PT
1
Results: The rate of deep infection in Groups 1 and 3 (non-obese anterior vs. posterior) was 0.28% and
11
0.36% (p=0.783); and in Groups 2 and 4 (obese anterior vs. posterior) was 2.35% and 2.7% (p=0.80).
12
The rate of wound complications between Groups 1 and 3 (non-obese) was 1.0% and 0.3% (p=0.005).
13
Between Groups 2 and 4 (obese) the rates of complications were 1.7% and 1.4% (p=1.0). There was no
14
difference in re-operation rates for wounds between Groups 1 and 3 or between Groups 2 and 4
15
(p=0.217, p=0.449).
16
Conclusion: In the largest available series, there was no difference in deep infection rates between the
17
two approaches. In the subset of obese patients with BMI ≥35kg/m2 there was no increased risk of
18
deep infection or wound complications in DAA patients compared to PA patients. However, anterior
19
hip cases experienced higher rates of superficial wound complications compared to posterior cases
20
across all BMIs.
AC C
EP
TE D
M AN U
10
Keywords: Total Hip Arthroplasty, Anterior Approach, Posterior Approach, Obesity, Wound Complications, Infection
2
ACCEPTED MANUSCRIPT Introduction: It is well-documented in the orthopaedic literature that obesity is a major risk factor for
22
complications in all patients that undergo orthopaedic care. With the projected rise of both obesity and
23
the need for total hip arthroplasty (THA), obesity-related complications can only be expected to follow
24
a similar trend. In the setting of such surgical hardships, many have begun to focus on expediting
25
patient discharge and improving surgical outcomes. Multiple reports on the favorable outcomes
26
following direct anterior approach (DAA) THA have been published.[1-3] Despite some theoretical
27
advantages, various authors have brought into question some risks associated with DAA THA. Recent
28
reports of increased wound complications in obese DAA patients seem to reflect the systemic effects
29
of increased body mass rather than the soft-tissue anatomy associated with the anterior approach.[4-7]
30
Purcell et al. were the first to report increased rates of deep infection requiring revision arthroplasty in
31
obese DAA patients. They found obese DAA hip patients were at seven-times greater risk than their
32
non-obese counterparts.[8] The purpose of the current study was to determine whether the anterior
33
approach alone is responsible for this complication or if obesity plays the dominant role.
SC
M AN U
TE D
34
RI PT
21
This study is the largest available cohort to directly compare rates of infection and wound complications in two of the most commonly used THA approaches. The purpose was to compare the
36
rates of postoperative infection and wound complications in patients who underwent primary THA via
37
the DAA and posterior approach (PA). Secondly, by stratifying patients based on BMI, we aimed to
38
determine if obese DAA patients are at higher risk of infection compared to obese PA patients. We
39
hypothesized that although significant differences in complication rates may exist between non-obese
40
and obese patients, there would be no difference in infection rates between obese DAA and obese PA
41
patients.
AC C
EP
35
3
ACCEPTED MANUSCRIPT 42
Materials and Methods:
43
This study is a single-center, multi-surgeon, retrospective cohort analysis. Institutional Review Board (IRB) approval was given for this study. We queried our institution’s total joint arthroplasty
45
database to identify all primary total hip arthroplasties performed through either a direct anterior
46
approach or posterior approach between 2009 and 2015. Reasons for THA included either
47
osteoarthritis or avascular necrosis. There was a minimum of one-year follow up, unless a primary or
48
secondary endpoint of infection or wound dehiscence was reached in less than one year. We identified
49
4651 consecutive primary THAs that were available for data analysis. 2424 cases were performed via
50
the DAA, and 2227 cases were performed via the PA. All patients routinely received antibiotics prior
51
to skin incision and for 24 hours postoperatively. The wound closure technique during this six-year
52
period was 2-0 Vicryl subcutaneously and running sutures or staples with adhesive for the skin closure.
53
DVT prophylaxis was the same for all patients at our institute, which is 325 mg aspirin with
54
mechanical pumps for low-risk patients and coumadin for high-risk patients.
M AN U
SC
RI PT
44
55
TE D
Patients were separated into four study groups based on preoperative body mass index (BMI in kg/m2):
57
Group 1: DAA with BMI less than 35
58
Group 2: DAA with BMI greater than or equal to 35
59
Group 3: PA with BMI less than 35
60
Group 4: PA with BMI greater than or equal to 35
61
BMI was calculated using the traditional formula of weight in kilograms divided by height in meters
62
squared. The primary outcome measure was postoperative deep infection requiring a revision hip
63
arthroplasty. Periprosthetic infections were defined by the Musculoskeletal Infection Society criteria, a
64
standard used by arthroplasty surgeons for defining deep infection.[9] Revision consisted of either
65
head-liner exchange with debridement or two-stage full revision. Superficial wound dehiscence
66
(SWD), a secondary outcome, was defined as any wound separation that required additional
AC C
EP
56
4
ACCEPTED MANUSCRIPT conservative treatment such as oral antibiotics, topical wound healing agents, or prolonged period of
68
dressing changes, or a return to the operating room for wound revision and closure. All outcome measures
69
were analyzed via a Fisher’s exact test, with significance set at 0.05.
70
Results:
71
Demographics:
72
RI PT
67
In total, 4651 hip arthroplasty cases met the inclusion criteria. We identified 2127 patients in Group 1, 297 in Group 2, 1933 in Group 3, and 294 in Group 4. Table 1 indicates the number, age,
74
gender, and BMI of the cases in each study group. Group 1 consisted of 888 males, 1239 females, a
75
mean age of 63 (range 19-95) years, and mean BMI of 26.7 kg/m2. Group 2 consisted of 118 males,
76
179 females, a mean age of 60 (range 29-84) years, and a mean BMI of 39.0 kg/m2. Group 3 consisted
77
of 934 males, 999 females, a mean age of 63 (range 20-96) years, and a mean BMI of 27.4 kg/m2.
78
Group 4 consisted of 112 males, 182 females, a mean age of 61 (range 28-85), and a mean BMI of 39.7
79
kg/m2 (Table 1).
80
Primary Outcome: Periprosthetic joint infections
M AN U
TE D
81
SC
73
The overall rate of deep infection requiring a revision in the entire study cohort was 0.60% (28/4651). The average time to revision for infection was 3.1 months after primary surgery. The rate of
83
deep infection for DAA cases was 0.54% (13/2424) compared to 0.67% (15/2227) for all PA cases
84
(p=0.574). Among the non-obese groups, the rate of infection was not different: 0.28% (6/2127) for
85
Group 1 and 0.36% in Group 3 (7/1933) (p=0.783). Likewise in the two obese patient groups, the
86
infection rate was not different: 2.35% (7/297) in Group 2 and 2.70% (8/294) in Group 4 (p=0.80).
87
Comparing the effect of obesity on infection rates within each approach group, the rate of infection in
88
the DAA groups (Groups 1 and 2) were 0.28% and 2.35%, respectively (p=0.0003) This difference
89
reflects a relative risk (RR) of 8.4 for patients with BMI above 35 with a power analysis showing 91%
90
power. Comparing the PA cohort, the rates of infection were 0.36% and 2.70% for Groups 3 and 4,
91
respectively (p=0.0002, RR=7.5, Power 92%). Comparing all non-obese patients (Groups 1 and 3
AC C
EP
82
5
ACCEPTED MANUSCRIPT 92
combined) to all obese patients (Groups 2 and 4 combined), the rates of infection were 0.32% and
93
2.54%, respectively (p<0.0001, RR=7.9).
94
Secondary Outcome: Superficial wound dehiscence (SWD)
95
The overall rate of SWD in the entire study cohort was 0.75% (35/4651). All of the wound complications occurred within the first 8 weeks of surgery, representing a failure of initial wound
97
healing, and the average timeframe was 1.1 months after surgery. Table 2 presents the wound
98
complication rates in each study group. The rate of SWD for all DAA patients was 1.07% (26/2424)
99
compared to 0.40% (9/2227) for all PA patients (p=0.01) This represents a RR of 2.7. Among the non-
SC
RI PT
96
obese groups, the rate of SWD was 1.0% (21/2127) for Group 1, compared to 0.26% (5/1933) in
101
Group 3 (p=0.005) This relative risk is up to 3.8 and a post hoc power analysis shows 84% power
102
comparing these groups. In the obese patient groups, the rate of SWD was 1.68% (5/297) in Group 2,
103
nearly identical to 1.40% (4/294) in Group 4 (p=1.0). Again, comparing the effect of obesity within
104
each approach group, the rate of SWD in DAA groups (Groups 1 and 2) were 1.0% and 1.68%,
105
respectively (p=0.238, RR=1.7). In the PA cohort, the rates of SWD were 0.26% and 1.40% for
106
Groups 3 and 4, respectively (p=0.021, RR=5.4). Comparing all non-obese patients (Groups 1 and 3
107
combined) to all obese patients (Groups 2 and 4 combined) the rates of SWD were 0.64% and 1.52%,
108
respectively (p=0.035, RR=2.4). For patients that required a return to the operating room for
109
debridement or closure for their SWD, there was no difference comparing the entire DAA cohort
110
(13/2424) to the entire PA cohort (9/2227) (p=0.53), and no difference between Groups 1 and 3
111
(p=0.217) or between Groups 2 and 4 (p=0.448).
112
Total complications combined:
TE D
EP
AC C
113
M AN U
100
The total rate of all studied complications for the entire cohort was 1.35% (63/4651). The DAA
114
cases had a total complication rate of 1.61%, which was not significantly different than 1.13% for the
115
PA cases (p=0.128). In the two non-obese cohorts, the total rates of complications were 1.27% for
116
Group 1 and 0.62% for Group 3 (p=0.036, RR=2.0). In the obese groups, the total rates of 6
ACCEPTED MANUSCRIPT 117
complications were virtually identical at 4.04% for Group 2 and 4.08% for Group 4 (p=1.0).
118
Comparing all non-obese patients (Groups 1 and 3 combined) to all obese patients (Groups 2 and 4
119
combined) the total rates of all complications was 0.96% compared to 4.06% (p=0.0001, RR=4.2).
120
122
Discussion:
RI PT
121
The purpose of the current study was to compare rates of infection and superficial wound dehiscence among patients undergoing THA via either the direct anterior approach or posterior
124
approach. The main objectives were to determine if DAA patients were at higher risk of infection
125
requiring revision arthroplasty compared to their PA counterparts, and more specifically, if obese DAA
126
patients had a higher risk than obese PA patients.
M AN U
SC
123
Our results demonstrate that for the studied complications, as a whole, DAA patients are not at
128
significantly higher risk of deep infection compared to their PA counterparts. Furthermore, there were
129
no significant differences in deep infection rates when comparing the two approaches in both non-
130
obese (Groups 1 and 3, p=0.783), and obese patient groups (Groups 2 and 4, p=0.80). In keeping with
131
the reported literature, we found highly significant differences in rates of infection when comparing
132
non-obese and obese patients that underwent DAA (Groups 1 and 2, p=0.0003) and those who
133
underwent PA (Groups 3 and 4, p=0.0002). Obesity was a more substantial risk factor for infection and
134
wound complications than surgical approach.
EP
Many recent studies have identified obesity as a significant risk factor for wound complication
AC C
135
TE D
127
136
and infection following DAA THA.[4-8] In the first available study of its kind, it was demonstrated
137
that obese patients have a 7-fold increased risk for deep infection following DAA THA compared to
138
non-obese individuals.[8] Reports regarding infectious complications in other THA surgical
139
approaches are limited; however, Dowsey et al demonstrated the same effect in obese patients
140
following PA THA.[10] The results of the current study demonstrate that the anterior approach alone is
7
ACCEPTED MANUSCRIPT 141
not associated with significantly higher rates of infection compared to the posterior approach, in both
142
non-obese and obese patients.
143
In addition to evaluating rates of deep infection, we compared the rates of superficial wound dehiscence between the study groups. As previously mentioned, recent literature has demonstrated
145
increased rates of wound healing complications in obese patients following DAA THA. In keeping
146
with these previous studies, we identified significantly increased rates of SWD in the entire cohort of
147
DAA patients compared to the entire cohort of PA patients (p=0.01, RR=2.7). In comparing the two
148
approaches, we also identified higher rates of SWD in the non-obese DAA patients compared to non-
149
obese PA patients (Groups 1 and 3, p=0.0048); however there was no significant difference in the
150
obese cohorts (Groups 2 and 4, p=1.0). Previous authors have proposed explanations for the increased
151
rate of wound complications in obese patients following DAA THA, such as the overhanging
152
abdominal pannus and propinquity to the genitalia, raising concerns about hygiene and contamination
153
[4, 11]. We identified a RR of 2.7 for developing SWD following DAA compared to PA, regardless of
154
BMI, and a similar RR of 2.4 for developing SWD in obese compared to non-obese patients, regardless
155
of approach. The overlying abdominal pannus may play a role in the increased rate of wound
156
complications following DAA.[12] Our data suggests that the presence of obesity is likely a main
157
contributing factor, resulting in systemic immune dysfunction and subsequent wound healing
158
complications, regardless of surgical approach used, as explained by the high rate of complications in
159
obese PA patients.
SC
M AN U
TE D
EP
AC C
160
RI PT
144
Similar to the greater incidence of wound complications in the entire DAA cohort in the
161
current study, previous studies have also reported on increased rates of reoperations for wound healing
162
following DAA THA due to the thinner skin and higher shear forces. [4, 5, 7] Christensen et al.
163
reported a 1.4% reoperation rate for wound healing issues following DAA, compared to 0.2%
164
reoperation rate in PA patients, however this was a review of only 500 DAA patients and 1200 PA
165
patients [5]. In contrast, we identified a 0.5% reoperation rate in the entire DAA cohort, compared to 8
ACCEPTED MANUSCRIPT 166
0.4% in the entire PA group. Although ‘wound complications’ may require a wide variety of
167
conservative treatments, a broader consensus occurs on patients requiring a return to the operating
168
room for revision of wound closure. These reoperations gave a clearer representation of ‘wound
169
complications’ given the increased risk and morbidity imparted on the patient. Weaknesses inherent to any retrospective review hold true for this study. Patient-specific
RI PT
170
medical co-morbidities are not routinely recorded in our institutional database and therefore,
172
differences in co-morbidities between the groups cannot be gauged. Secondly, as the THAs in this
173
study were performed by multiple surgeons, the threshold for wound treatments may differ and may
174
explain differences in the rates of complications. As with many arthroplasty studies evaluating rates of
175
infection, the current study may be subject to Type II error, potentially contributing to our inability to
176
find differences in infection and SWD between the obese patient cohorts. However in groups that were
177
significantly different the power was over 80%. Lastly, BMI is a continuous variable, affecting
178
arthroplasty outcomes while as low as 28 kg/m.[13] We chose a cutoff value of 35 kg/m2 in
179
distinguishing outcomes in non-obese and obese patients based on a recent study by Lubbeke et al.
180
which found that patients with a BMI below 35 kg/m2 had similar infection rates and cited a threshold
181
of greater than 35 kg/m2 for being a significant risk factor for the development of infection.[14]
M AN U
TE D
In conclusion, this is the largest study to directly compare the rates of postoperative infection
EP
182
SC
171
and wound complications among obese and non-obese hip patients with two of the most common
184
surgical approaches. In accordance with the literature, this study also found higher rates of superficial
185
wound complications following DAA hip arthroplasty regardless of BMI. However, no significant
186
difference in deep infection rates between the two approaches was shown. In the subset of patients
187
with BMI above 35 kg/m2, there was no increased risk of deep infection or SWD in DAA patients
188
compared to their PA counterparts. Regardless of surgical approach used, obese patients had higher
189
rates of deep infection and SWD compared to non-obese patients. Obesity was a greater risk factor for
190
deep infection and wound complication than surgical approach. Obese DAA patients are at no higher
AC C
183
9
ACCEPTED MANUSCRIPT 191
risk of reoperation for infection or SWD compared to obese PA patients. Prospective studies are
192
needed to examine the rates of infection and reoperation in obese patients following DAA and PA hip
193
replacement.
AC C
EP
TE D
M AN U
SC
RI PT
194
10
ACCEPTED MANUSCRIPT REFERENCES 195
1.
Bergin, P.F., et al., Comparison of minimally invasive direct anterior versus posterior total
196
hip arthroplasty based on inflammation and muscle damage markers. J Bone Joint Surg Am,
197
2011. 93(15): p. 1392-8. 2.
Barrett, W.P., S.E. Turner, and J.P. Leopold, Prospective randomized study of direct anterior
RI PT
198 199
vs postero-lateral approach for total hip arthroplasty. J Arthroplasty, 2013. 28(9): p. 1634-
200
8. 3.
Zawadsky, M.W., et al., Early outcome comparison between the direct anterior approach
SC
201
and the mini-incision posterior approach for primary total hip arthroplasty: 150 consecutive
203
cases. J Arthroplasty, 2014. 29(6): p. 1256-60.
204
4.
Watts, C.D., et al., High Risk of Wound Complications Following Direct Anterior Total Hip Arthroplasty in Obese Patients. J Arthroplasty, 2015. 30(12): p. 2296-8.
205 206
M AN U
202
5.
Christensen, C.P., T. Karthikeyan, and C.A. Jacobs, Greater prevalence of wound complications requiring reoperation with direct anterior approach total hip arthroplasty. J
208
Arthroplasty, 2014. 29(9): p. 1839-41. 6.
7.
8.
9.
218
Parvizi, J., et al., New definition for periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res, 2011. 469(11): p. 2992-4.
216 217
Purcell, R.L., et al., Severely Obese Patients Have a Higher Risk of Infection After Direct Anterior Approach Total Hip Arthroplasty. J Arthroplasty, 2016.
214 215
Jewett, B.A. and D.K. Collis, High complication rate with anterior total hip arthroplasties on a fracture table. Clin Orthop Relat Res, 2011. 469(2): p. 503-7.
212 213
EP
Total Hip Arthroplasty. J Arthroplasty, 2015. 30(8): p. 1384-7.
210 211
Russo, M.W., et al., Increased Complications in Obese Patients Undergoing Direct Anterior
AC C
209
TE D
207
10.
Dowsey, M.M. and P.F. Choong, Obesity is a major risk factor for prosthetic infection after primary hip arthroplasty. Clin Orthop Relat Res, 2008. 466(1): p. 153-8. 11
ACCEPTED MANUSCRIPT 11.
Arthroplasty. J Arthroplasty, 2016.
220 12.
222 223
p. 294-9. 13.
226
Wagner, E.R., et al., Effect of Body Mass Index on Complications and Reoperations After Total Hip Arthroplasty. J Bone Joint Surg Am, 2016. 98(3): p. 169-79.
224 225
Nieman, D.C., et al., Influence of obesity on immune function. J Am Diet Assoc, 1999. 99(3):
RI PT
221
Jahng, K.H., et al., Risk Factors for Wound Complications After Direct Anterior Approach Hip
14.
Lubbeke, A., et al., Body mass and weight thresholds for increased prosthetic joint infection rates after primary total joint arthroplasty. Acta Orthop, 2016. 87(2): p. 132-8.
SC
219
AC C
EP
TE D
M AN U
227
12
ACCEPTED MANUSCRIPT
Acknowledgements:
AC C
EP
TE D
M AN U
SC
RI PT
The authors would like to acknowledge Inova Health System for institutional research support.
ACCEPTED MANUSCRIPT
Table 1: Demographics of the four study groups
Average AGE (years) age range N male percent male
63 20-96
Group 4: PA BMI >35 294 (13% of total) 61 28-85
63 19-95
62.9 19-96
888 42%
118 40%
934 48%
112 38%
2052 44%
26.7 15.6-34.9
39 35-53.3
27.4 16.3-34.9
AC C
EP
TE D
M AN U
Average BMI (kg/m2) BMI range
Group 3: PA BMI<35 1933
Total 4651
RI PT
N
Group 2: DAA BMI>35 297 (12% of total) 60 29-84
SC
Group 1: DAA BMI<35 2127
Study Group
39.7 35-60
28.6 15.2-60
ACCEPTED MANUSCRIPT
Table 2: Rates of infection and wound complications in each study group
Group 4: PA BMI >35 294 (13% of total)
all DAA 2424
Revised for Infection
N=28
13
6
7
15
7
8
% Wound complications (Re-OP in parentheses)
0.60%
0.54%
0.28%
2.35%
0.67%
N=35 (22)
26 (13)
21 (11)
%
0.75%
1.07%
1%
N= 63
39
27
1.35%
1.60%
1.27%
0.36%
p-values Group 2 vs. 4
0.78
0.8
0.0048
1
0.037
1
2.70%
5 (2)
9 (9)
5 (5)
4 (4)
1.68%
0.40%
0.26%
1.40%
12
24
12
12
4.04%
1.07%
0.62%
4.08%
TE D
p-values Group 1 vs. 3
RI PT
all PA 2227
Group 3: PA BMI<35 1933
SC
Total 4651
Either of above
Group 2: DAA BMI>35 297 (12% of total)
M AN U
Study Group N
Group 1: DAA BMI<35 2127
AC C
EP
Bolded p-values are <0.05 and indicate statistically significant differences between the groups.
S0883-5403(17)30686-1
DOI:
10.1016/j.arth.2017.07.047
Reference:
YARTH 56027
To appear in:
The Journal of Arthroplasty
Received Date: 28 October 2016 Revised Date:
22 June 2017
Accepted Date: 27 July 2017
Please cite this article as: Purcell RL, Parks NL, Cody JP, Hamilton WG, Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients, The Journal of Arthroplasty (2017), doi: 10.1016/j.arth.2017.07.047. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Comparison of Wound Complications and Deep Infections with
RI PT
Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients
1 Anderson Orthopaedic Research Institute PO Box 7088 Alexandria, VA 22307
TE D
2 Inova Joint Replacement Center Mt. Vernon Hospital 2501 Parker's Lane, Suite 200 Alexandria, VA 22306
M AN U
SC
Richard L. Purcell, MD1,3 Nancy L. Parks, MS1 John P. Cody, MD1,3 William G. Hamilton, MD1,2
EP
3 Walter Reed National Military Medical Center 8901 Wisconsin Ave. Department of Orthopaedics, Building 19, 2nd Floor Bethesda, MD 20889
AC C
Please address all correspondence to: Nancy L. Parks, MS Anderson Orthopaedic Research Institute PO Box 7088 Alexandria, VA 22307 Phone: (703) 619-4423 FAX: (703) 799-5982 Email: [email protected]
ACCEPTED MANUSCRIPT
Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients
RI PT
Introduction: The purpose of this study was to compare the posterior approach (PA) to the direct anterior approach (DAA) among obese and non-obese total hip arthroplasty (THA) patients to determine if obese DAA patients have a higher risk of infection or wound complications
SC
compared to obese PA patients.
Methods: We retrospectively evaluated 4651 primary THA performed via DAA or PA between
M AN U
2009-2015. Patients were divided into four study groups based on approach and BMI: 1)DAA <35kg/m2, 2)DAA ≥35kg/m2, 3)PA <35kg/m2, 4)PA ≥35kg/m2. Infection rates and wound complications were compared.
Results: The rate of deep infection in Groups 1 and 3 (non-obese) was 0.28% and 0.36% (p=0.783); and in Groups 2 and 4 (obese) was 2.35% and 2.7% (p=0.80). The rate of wound
TE D
complications between Groups 1 and 3 (non-obese) was 1.0% and 0.3% (p=0.005). Between Groups 2 and 4 (obese) the rates of complications were 1.7% and 1.4% (p=1.0). There was no
EP
difference in re-operation rates for wounds between Groups 1 and 3 or between Groups 2 and 4 (p=0.217, p=0.449).
AC C
Conclusion: In the largest available series DAA THA experienced higher rates of superficial wound complications compared to PA THA regardless of BMI. However there was no difference in deep infection rates between the two approaches. In the subset of patients with BMI ≥35kg/m2 there was no increased risk of deep infection wound complications in obese DAA patients compared to their obese PA counterparts.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Keywords: Total Hip Arthroplasty, Anterior Approach, Posterior Approach, Obesity, Wound Complications, Infection
RI PT
ACCEPTED MANUSCRIPT
Comparison of Wound Complications and Deep Infections with Direct Anterior and Posterior
AC C
EP
TE D
M AN U
SC
Approaches in Obese Hip Arthroplasty Patients
1
ACCEPTED MANUSCRIPT ABSTRACT
2
Background: The purpose of this study was to compare the posterior approach (PA) to the direct
3
anterior approach (DAA) among obese and non-obese total hip arthroplasty (THA) patients to
4
determine if obese DAA patients have a higher risk of infection or wound complications compared to
5
obese PA patients.
6
Methods: We retrospectively evaluated 4651 primary total hip cases performed via anterior or
7
posterior approach between 2009-2015. Patients were divided into four study groups based on
8
approach and BMI: 1) DAA <35kg/m2, 2) DAA ≥35kg/m2, 3) PA <35kg/m2, 4) PA ≥35kg/m2.
9
Infection rates and wound complications were compared.
SC
RI PT
1
Results: The rate of deep infection in Groups 1 and 3 (non-obese anterior vs. posterior) was 0.28% and
11
0.36% (p=0.783); and in Groups 2 and 4 (obese anterior vs. posterior) was 2.35% and 2.7% (p=0.80).
12
The rate of wound complications between Groups 1 and 3 (non-obese) was 1.0% and 0.3% (p=0.005).
13
Between Groups 2 and 4 (obese) the rates of complications were 1.7% and 1.4% (p=1.0). There was no
14
difference in re-operation rates for wounds between Groups 1 and 3 or between Groups 2 and 4
15
(p=0.217, p=0.449).
16
Conclusion: In the largest available series, there was no difference in deep infection rates between the
17
two approaches. In the subset of obese patients with BMI ≥35kg/m2 there was no increased risk of
18
deep infection or wound complications in DAA patients compared to PA patients. However, anterior
19
hip cases experienced higher rates of superficial wound complications compared to posterior cases
20
across all BMIs.
AC C
EP
TE D
M AN U
10
Keywords: Total Hip Arthroplasty, Anterior Approach, Posterior Approach, Obesity, Wound Complications, Infection
2
ACCEPTED MANUSCRIPT Introduction: It is well-documented in the orthopaedic literature that obesity is a major risk factor for
22
complications in all patients that undergo orthopaedic care. With the projected rise of both obesity and
23
the need for total hip arthroplasty (THA), obesity-related complications can only be expected to follow
24
a similar trend. In the setting of such surgical hardships, many have begun to focus on expediting
25
patient discharge and improving surgical outcomes. Multiple reports on the favorable outcomes
26
following direct anterior approach (DAA) THA have been published.[1-3] Despite some theoretical
27
advantages, various authors have brought into question some risks associated with DAA THA. Recent
28
reports of increased wound complications in obese DAA patients seem to reflect the systemic effects
29
of increased body mass rather than the soft-tissue anatomy associated with the anterior approach.[4-7]
30
Purcell et al. were the first to report increased rates of deep infection requiring revision arthroplasty in
31
obese DAA patients. They found obese DAA hip patients were at seven-times greater risk than their
32
non-obese counterparts.[8] The purpose of the current study was to determine whether the anterior
33
approach alone is responsible for this complication or if obesity plays the dominant role.
SC
M AN U
TE D
34
RI PT
21
This study is the largest available cohort to directly compare rates of infection and wound complications in two of the most commonly used THA approaches. The purpose was to compare the
36
rates of postoperative infection and wound complications in patients who underwent primary THA via
37
the DAA and posterior approach (PA). Secondly, by stratifying patients based on BMI, we aimed to
38
determine if obese DAA patients are at higher risk of infection compared to obese PA patients. We
39
hypothesized that although significant differences in complication rates may exist between non-obese
40
and obese patients, there would be no difference in infection rates between obese DAA and obese PA
41
patients.
AC C
EP
35
3
ACCEPTED MANUSCRIPT 42
Materials and Methods:
43
This study is a single-center, multi-surgeon, retrospective cohort analysis. Institutional Review Board (IRB) approval was given for this study. We queried our institution’s total joint arthroplasty
45
database to identify all primary total hip arthroplasties performed through either a direct anterior
46
approach or posterior approach between 2009 and 2015. Reasons for THA included either
47
osteoarthritis or avascular necrosis. There was a minimum of one-year follow up, unless a primary or
48
secondary endpoint of infection or wound dehiscence was reached in less than one year. We identified
49
4651 consecutive primary THAs that were available for data analysis. 2424 cases were performed via
50
the DAA, and 2227 cases were performed via the PA. All patients routinely received antibiotics prior
51
to skin incision and for 24 hours postoperatively. The wound closure technique during this six-year
52
period was 2-0 Vicryl subcutaneously and running sutures or staples with adhesive for the skin closure.
53
DVT prophylaxis was the same for all patients at our institute, which is 325 mg aspirin with
54
mechanical pumps for low-risk patients and coumadin for high-risk patients.
M AN U
SC
RI PT
44
55
TE D
Patients were separated into four study groups based on preoperative body mass index (BMI in kg/m2):
57
Group 1: DAA with BMI less than 35
58
Group 2: DAA with BMI greater than or equal to 35
59
Group 3: PA with BMI less than 35
60
Group 4: PA with BMI greater than or equal to 35
61
BMI was calculated using the traditional formula of weight in kilograms divided by height in meters
62
squared. The primary outcome measure was postoperative deep infection requiring a revision hip
63
arthroplasty. Periprosthetic infections were defined by the Musculoskeletal Infection Society criteria, a
64
standard used by arthroplasty surgeons for defining deep infection.[9] Revision consisted of either
65
head-liner exchange with debridement or two-stage full revision. Superficial wound dehiscence
66
(SWD), a secondary outcome, was defined as any wound separation that required additional
AC C
EP
56
4
ACCEPTED MANUSCRIPT conservative treatment such as oral antibiotics, topical wound healing agents, or prolonged period of
68
dressing changes, or a return to the operating room for wound revision and closure. All outcome measures
69
were analyzed via a Fisher’s exact test, with significance set at 0.05.
70
Results:
71
Demographics:
72
RI PT
67
In total, 4651 hip arthroplasty cases met the inclusion criteria. We identified 2127 patients in Group 1, 297 in Group 2, 1933 in Group 3, and 294 in Group 4. Table 1 indicates the number, age,
74
gender, and BMI of the cases in each study group. Group 1 consisted of 888 males, 1239 females, a
75
mean age of 63 (range 19-95) years, and mean BMI of 26.7 kg/m2. Group 2 consisted of 118 males,
76
179 females, a mean age of 60 (range 29-84) years, and a mean BMI of 39.0 kg/m2. Group 3 consisted
77
of 934 males, 999 females, a mean age of 63 (range 20-96) years, and a mean BMI of 27.4 kg/m2.
78
Group 4 consisted of 112 males, 182 females, a mean age of 61 (range 28-85), and a mean BMI of 39.7
79
kg/m2 (Table 1).
80
Primary Outcome: Periprosthetic joint infections
M AN U
TE D
81
SC
73
The overall rate of deep infection requiring a revision in the entire study cohort was 0.60% (28/4651). The average time to revision for infection was 3.1 months after primary surgery. The rate of
83
deep infection for DAA cases was 0.54% (13/2424) compared to 0.67% (15/2227) for all PA cases
84
(p=0.574). Among the non-obese groups, the rate of infection was not different: 0.28% (6/2127) for
85
Group 1 and 0.36% in Group 3 (7/1933) (p=0.783). Likewise in the two obese patient groups, the
86
infection rate was not different: 2.35% (7/297) in Group 2 and 2.70% (8/294) in Group 4 (p=0.80).
87
Comparing the effect of obesity on infection rates within each approach group, the rate of infection in
88
the DAA groups (Groups 1 and 2) were 0.28% and 2.35%, respectively (p=0.0003) This difference
89
reflects a relative risk (RR) of 8.4 for patients with BMI above 35 with a power analysis showing 91%
90
power. Comparing the PA cohort, the rates of infection were 0.36% and 2.70% for Groups 3 and 4,
91
respectively (p=0.0002, RR=7.5, Power 92%). Comparing all non-obese patients (Groups 1 and 3
AC C
EP
82
5
ACCEPTED MANUSCRIPT 92
combined) to all obese patients (Groups 2 and 4 combined), the rates of infection were 0.32% and
93
2.54%, respectively (p<0.0001, RR=7.9).
94
Secondary Outcome: Superficial wound dehiscence (SWD)
95
The overall rate of SWD in the entire study cohort was 0.75% (35/4651). All of the wound complications occurred within the first 8 weeks of surgery, representing a failure of initial wound
97
healing, and the average timeframe was 1.1 months after surgery. Table 2 presents the wound
98
complication rates in each study group. The rate of SWD for all DAA patients was 1.07% (26/2424)
99
compared to 0.40% (9/2227) for all PA patients (p=0.01) This represents a RR of 2.7. Among the non-
SC
RI PT
96
obese groups, the rate of SWD was 1.0% (21/2127) for Group 1, compared to 0.26% (5/1933) in
101
Group 3 (p=0.005) This relative risk is up to 3.8 and a post hoc power analysis shows 84% power
102
comparing these groups. In the obese patient groups, the rate of SWD was 1.68% (5/297) in Group 2,
103
nearly identical to 1.40% (4/294) in Group 4 (p=1.0). Again, comparing the effect of obesity within
104
each approach group, the rate of SWD in DAA groups (Groups 1 and 2) were 1.0% and 1.68%,
105
respectively (p=0.238, RR=1.7). In the PA cohort, the rates of SWD were 0.26% and 1.40% for
106
Groups 3 and 4, respectively (p=0.021, RR=5.4). Comparing all non-obese patients (Groups 1 and 3
107
combined) to all obese patients (Groups 2 and 4 combined) the rates of SWD were 0.64% and 1.52%,
108
respectively (p=0.035, RR=2.4). For patients that required a return to the operating room for
109
debridement or closure for their SWD, there was no difference comparing the entire DAA cohort
110
(13/2424) to the entire PA cohort (9/2227) (p=0.53), and no difference between Groups 1 and 3
111
(p=0.217) or between Groups 2 and 4 (p=0.448).
112
Total complications combined:
TE D
EP
AC C
113
M AN U
100
The total rate of all studied complications for the entire cohort was 1.35% (63/4651). The DAA
114
cases had a total complication rate of 1.61%, which was not significantly different than 1.13% for the
115
PA cases (p=0.128). In the two non-obese cohorts, the total rates of complications were 1.27% for
116
Group 1 and 0.62% for Group 3 (p=0.036, RR=2.0). In the obese groups, the total rates of 6
ACCEPTED MANUSCRIPT 117
complications were virtually identical at 4.04% for Group 2 and 4.08% for Group 4 (p=1.0).
118
Comparing all non-obese patients (Groups 1 and 3 combined) to all obese patients (Groups 2 and 4
119
combined) the total rates of all complications was 0.96% compared to 4.06% (p=0.0001, RR=4.2).
120
122
Discussion:
RI PT
121
The purpose of the current study was to compare rates of infection and superficial wound dehiscence among patients undergoing THA via either the direct anterior approach or posterior
124
approach. The main objectives were to determine if DAA patients were at higher risk of infection
125
requiring revision arthroplasty compared to their PA counterparts, and more specifically, if obese DAA
126
patients had a higher risk than obese PA patients.
M AN U
SC
123
Our results demonstrate that for the studied complications, as a whole, DAA patients are not at
128
significantly higher risk of deep infection compared to their PA counterparts. Furthermore, there were
129
no significant differences in deep infection rates when comparing the two approaches in both non-
130
obese (Groups 1 and 3, p=0.783), and obese patient groups (Groups 2 and 4, p=0.80). In keeping with
131
the reported literature, we found highly significant differences in rates of infection when comparing
132
non-obese and obese patients that underwent DAA (Groups 1 and 2, p=0.0003) and those who
133
underwent PA (Groups 3 and 4, p=0.0002). Obesity was a more substantial risk factor for infection and
134
wound complications than surgical approach.
EP
Many recent studies have identified obesity as a significant risk factor for wound complication
AC C
135
TE D
127
136
and infection following DAA THA.[4-8] In the first available study of its kind, it was demonstrated
137
that obese patients have a 7-fold increased risk for deep infection following DAA THA compared to
138
non-obese individuals.[8] Reports regarding infectious complications in other THA surgical
139
approaches are limited; however, Dowsey et al demonstrated the same effect in obese patients
140
following PA THA.[10] The results of the current study demonstrate that the anterior approach alone is
7
ACCEPTED MANUSCRIPT 141
not associated with significantly higher rates of infection compared to the posterior approach, in both
142
non-obese and obese patients.
143
In addition to evaluating rates of deep infection, we compared the rates of superficial wound dehiscence between the study groups. As previously mentioned, recent literature has demonstrated
145
increased rates of wound healing complications in obese patients following DAA THA. In keeping
146
with these previous studies, we identified significantly increased rates of SWD in the entire cohort of
147
DAA patients compared to the entire cohort of PA patients (p=0.01, RR=2.7). In comparing the two
148
approaches, we also identified higher rates of SWD in the non-obese DAA patients compared to non-
149
obese PA patients (Groups 1 and 3, p=0.0048); however there was no significant difference in the
150
obese cohorts (Groups 2 and 4, p=1.0). Previous authors have proposed explanations for the increased
151
rate of wound complications in obese patients following DAA THA, such as the overhanging
152
abdominal pannus and propinquity to the genitalia, raising concerns about hygiene and contamination
153
[4, 11]. We identified a RR of 2.7 for developing SWD following DAA compared to PA, regardless of
154
BMI, and a similar RR of 2.4 for developing SWD in obese compared to non-obese patients, regardless
155
of approach. The overlying abdominal pannus may play a role in the increased rate of wound
156
complications following DAA.[12] Our data suggests that the presence of obesity is likely a main
157
contributing factor, resulting in systemic immune dysfunction and subsequent wound healing
158
complications, regardless of surgical approach used, as explained by the high rate of complications in
159
obese PA patients.
SC
M AN U
TE D
EP
AC C
160
RI PT
144
Similar to the greater incidence of wound complications in the entire DAA cohort in the
161
current study, previous studies have also reported on increased rates of reoperations for wound healing
162
following DAA THA due to the thinner skin and higher shear forces. [4, 5, 7] Christensen et al.
163
reported a 1.4% reoperation rate for wound healing issues following DAA, compared to 0.2%
164
reoperation rate in PA patients, however this was a review of only 500 DAA patients and 1200 PA
165
patients [5]. In contrast, we identified a 0.5% reoperation rate in the entire DAA cohort, compared to 8
ACCEPTED MANUSCRIPT 166
0.4% in the entire PA group. Although ‘wound complications’ may require a wide variety of
167
conservative treatments, a broader consensus occurs on patients requiring a return to the operating
168
room for revision of wound closure. These reoperations gave a clearer representation of ‘wound
169
complications’ given the increased risk and morbidity imparted on the patient. Weaknesses inherent to any retrospective review hold true for this study. Patient-specific
RI PT
170
medical co-morbidities are not routinely recorded in our institutional database and therefore,
172
differences in co-morbidities between the groups cannot be gauged. Secondly, as the THAs in this
173
study were performed by multiple surgeons, the threshold for wound treatments may differ and may
174
explain differences in the rates of complications. As with many arthroplasty studies evaluating rates of
175
infection, the current study may be subject to Type II error, potentially contributing to our inability to
176
find differences in infection and SWD between the obese patient cohorts. However in groups that were
177
significantly different the power was over 80%. Lastly, BMI is a continuous variable, affecting
178
arthroplasty outcomes while as low as 28 kg/m.[13] We chose a cutoff value of 35 kg/m2 in
179
distinguishing outcomes in non-obese and obese patients based on a recent study by Lubbeke et al.
180
which found that patients with a BMI below 35 kg/m2 had similar infection rates and cited a threshold
181
of greater than 35 kg/m2 for being a significant risk factor for the development of infection.[14]
M AN U
TE D
In conclusion, this is the largest study to directly compare the rates of postoperative infection
EP
182
SC
171
and wound complications among obese and non-obese hip patients with two of the most common
184
surgical approaches. In accordance with the literature, this study also found higher rates of superficial
185
wound complications following DAA hip arthroplasty regardless of BMI. However, no significant
186
difference in deep infection rates between the two approaches was shown. In the subset of patients
187
with BMI above 35 kg/m2, there was no increased risk of deep infection or SWD in DAA patients
188
compared to their PA counterparts. Regardless of surgical approach used, obese patients had higher
189
rates of deep infection and SWD compared to non-obese patients. Obesity was a greater risk factor for
190
deep infection and wound complication than surgical approach. Obese DAA patients are at no higher
AC C
183
9
ACCEPTED MANUSCRIPT 191
risk of reoperation for infection or SWD compared to obese PA patients. Prospective studies are
192
needed to examine the rates of infection and reoperation in obese patients following DAA and PA hip
193
replacement.
AC C
EP
TE D
M AN U
SC
RI PT
194
10
ACCEPTED MANUSCRIPT REFERENCES 195
1.
Bergin, P.F., et al., Comparison of minimally invasive direct anterior versus posterior total
196
hip arthroplasty based on inflammation and muscle damage markers. J Bone Joint Surg Am,
197
2011. 93(15): p. 1392-8. 2.
Barrett, W.P., S.E. Turner, and J.P. Leopold, Prospective randomized study of direct anterior
RI PT
198 199
vs postero-lateral approach for total hip arthroplasty. J Arthroplasty, 2013. 28(9): p. 1634-
200
8. 3.
Zawadsky, M.W., et al., Early outcome comparison between the direct anterior approach
SC
201
and the mini-incision posterior approach for primary total hip arthroplasty: 150 consecutive
203
cases. J Arthroplasty, 2014. 29(6): p. 1256-60.
204
4.
Watts, C.D., et al., High Risk of Wound Complications Following Direct Anterior Total Hip Arthroplasty in Obese Patients. J Arthroplasty, 2015. 30(12): p. 2296-8.
205 206
M AN U
202
5.
Christensen, C.P., T. Karthikeyan, and C.A. Jacobs, Greater prevalence of wound complications requiring reoperation with direct anterior approach total hip arthroplasty. J
208
Arthroplasty, 2014. 29(9): p. 1839-41. 6.
7.
8.
9.
218
Parvizi, J., et al., New definition for periprosthetic joint infection: from the Workgroup of the Musculoskeletal Infection Society. Clin Orthop Relat Res, 2011. 469(11): p. 2992-4.
216 217
Purcell, R.L., et al., Severely Obese Patients Have a Higher Risk of Infection After Direct Anterior Approach Total Hip Arthroplasty. J Arthroplasty, 2016.
214 215
Jewett, B.A. and D.K. Collis, High complication rate with anterior total hip arthroplasties on a fracture table. Clin Orthop Relat Res, 2011. 469(2): p. 503-7.
212 213
EP
Total Hip Arthroplasty. J Arthroplasty, 2015. 30(8): p. 1384-7.
210 211
Russo, M.W., et al., Increased Complications in Obese Patients Undergoing Direct Anterior
AC C
209
TE D
207
10.
Dowsey, M.M. and P.F. Choong, Obesity is a major risk factor for prosthetic infection after primary hip arthroplasty. Clin Orthop Relat Res, 2008. 466(1): p. 153-8. 11
ACCEPTED MANUSCRIPT 11.
Arthroplasty. J Arthroplasty, 2016.
220 12.
222 223
p. 294-9. 13.
226
Wagner, E.R., et al., Effect of Body Mass Index on Complications and Reoperations After Total Hip Arthroplasty. J Bone Joint Surg Am, 2016. 98(3): p. 169-79.
224 225
Nieman, D.C., et al., Influence of obesity on immune function. J Am Diet Assoc, 1999. 99(3):
RI PT
221
Jahng, K.H., et al., Risk Factors for Wound Complications After Direct Anterior Approach Hip
14.
Lubbeke, A., et al., Body mass and weight thresholds for increased prosthetic joint infection rates after primary total joint arthroplasty. Acta Orthop, 2016. 87(2): p. 132-8.
SC
219
AC C
EP
TE D
M AN U
227
12
ACCEPTED MANUSCRIPT
Acknowledgements:
AC C
EP
TE D
M AN U
SC
RI PT
The authors would like to acknowledge Inova Health System for institutional research support.
ACCEPTED MANUSCRIPT
Table 1: Demographics of the four study groups
Average AGE (years) age range N male percent male
63 20-96
Group 4: PA BMI >35 294 (13% of total) 61 28-85
63 19-95
62.9 19-96
888 42%
118 40%
934 48%
112 38%
2052 44%
26.7 15.6-34.9
39 35-53.3
27.4 16.3-34.9
AC C
EP
TE D
M AN U
Average BMI (kg/m2) BMI range
Group 3: PA BMI<35 1933
Total 4651
RI PT
N
Group 2: DAA BMI>35 297 (12% of total) 60 29-84
SC
Group 1: DAA BMI<35 2127
Study Group
39.7 35-60
28.6 15.2-60
ACCEPTED MANUSCRIPT
Table 2: Rates of infection and wound complications in each study group
Group 4: PA BMI >35 294 (13% of total)
all DAA 2424
Revised for Infection
N=28
13
6
7
15
7
8
% Wound complications (Re-OP in parentheses)
0.60%
0.54%
0.28%
2.35%
0.67%
N=35 (22)
26 (13)
21 (11)
%
0.75%
1.07%
1%
N= 63
39
27
1.35%
1.60%
1.27%
0.36%
p-values Group 2 vs. 4
0.78
0.8
0.0048
1
0.037
1
2.70%
5 (2)
9 (9)
5 (5)
4 (4)
1.68%
0.40%
0.26%
1.40%
12
24
12
12
4.04%
1.07%
0.62%
4.08%
TE D
p-values Group 1 vs. 3
RI PT
all PA 2227
Group 3: PA BMI<35 1933
SC
Total 4651
Either of above
Group 2: DAA BMI>35 297 (12% of total)
M AN U
Study Group N
Group 1: DAA BMI<35 2127
AC C
EP
Bolded p-values are <0.05 and indicate statistically significant differences between the groups.