- Email: [email protected]
Preoperative Anemia Is Associated With Failure of Open Debridement Polyethylene Exchange in Acute and Acute Hematogenous Prosthetic Joint Infection
Accepted Manuscript Preoperative Anemia Is Associated With Failure Of Open Debridement Polyethylene Exchange In Acute And Acute Hematogenous Prosthetic Joint Infection Richard D. Swenson, MD, James Butterfield, BS, Timothy J. Irwin, MD, John J. Zurlo, MD, Charles M. Davis, III, MD/PhD PII:
S0883-5403(18)30080-9
DOI:
10.1016/j.arth.2018.01.042
Reference:
YARTH 56363
To appear in:
The Journal of Arthroplasty
Received Date: 27 November 2017 Revised Date:
8 January 2018
Accepted Date: 12 January 2018
Please cite this article as: Swenson RD, Butterfield J, Irwin TJ, Zurlo JJ, Davis III CM, Preoperative Anemia Is Associated With Failure Of Open Debridement Polyethylene Exchange In Acute And Acute Hematogenous Prosthetic Joint Infection, The Journal of Arthroplasty (2018), doi: 10.1016/ j.arth.2018.01.042. 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
PREOPERATIVE ANEMIA IS ASSOCIATED WITH FAILURE OF OPEN DEBRIDEMENT POLYETHYLENE EXCHANGE IN ACUTE AND ACUTE HEMATOGENOUS PROSTHETIC JOINT INFECTIONS
Please address all correspondence to: Richard D Swenson, MD Department of Orthopaedic Surgery 500 University Drive Hershey, PA 17033 USA
M AN U
2. Infectious Disease Penn State College of Medicine Hershey, Pa 17033
SC
1. Department of Orthopaedic Surgery Penn State College of Medicine Hershey, Pa 17033
[email protected] [email protected] [email protected] [email protected] [email protected]
RI PT
Richard D. Swenson, MD1 James Butterfield, BS1 Timothy J. Irwin, MD1 John J. Zurlo, MD2 Charles M. Davis III, MD/PhD1
TE D
Phone: 801-971-9610 Email: [email protected]
EP
Investigation performed at Penn State Department of Orthopaedic Surgery
AC C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
1
ACCEPTED MANUSCRIPT
1
Abstract: Background: Acute and acute hematogenous Periprosthetic Joint Infections (PJIs) are
3
often treated with Open Debridement Polyethylene Exchange (ODPE) in an effort to save the
4
prosthesis, decrease morbidity and reduce costs. However, failure of ODPE may compromise a
5
subsequent two-stage treatment. The purpose of this study is to identify patient factors that
6
impact the success of ODPE for acute and acute hematogenous PJIs.
Methods: A retrospective review examined comorbidities, preoperative labs, and patient
SC
7
RI PT
2
history for patients with successful and failed ODPE treatment for acute peri-operative or acute
9
hematogenous periprostheic hip or knee joint infections. Successful treatment was defined as
M AN U
8
10
retaining a well-fixed implant without the need for additional surgery for a minimum of 6 month
11
follow up with or without lifelong oral maintenance antibiotics.
12
Results: 53/72 patients (73.6%) underwent successful ODPE. Of the 19 failures, 14 completed 2-stage revision with one subsequent known failure for recurrent infection. Patients
14
with a Staph aureus infection were more likely to fail ODPE (48.3% vs 11.6% p=0.0012), odds
15
ratio 7.1 (2.3-25.3). Patients with a preoperative hematocrit <=32.1 were also more likely to fail
16
ODPE (55% vs 16% p=0.0013), odds ratio 6.7 (2.2-22.4). When neither risk factor was present,
17
97.1% of peri-prosthetic joint infections were successfully treated with ODPE.
EP
Conclusion: Staph aureus infection and preoperative hematocrit <=32.1 are independent
AC C
18
TE D
13
19
risk factors for ODPE failure. ODPE is a safe alternative to 2-stage revision in patients without
20
preoperative anemia and without Staphylococcus aureus infection. Two-thirds of patients with a
21
failed ODPE were successfully treated with a two-stage re-implantation.
22
1
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
PREOPERATIVE ANEMIA IS ASSOCIATED WITH FAILURE OF OPEN DEBRIDEMENT
2
POLYETHYLENE EXCHANGE IN ACUTE AND ACUTE HEMATOGENOUS
3
PROSTHETIC JOINT INFECTION
EP AC C
4
TE D
1
1
ACCEPTED MANUSCRIPT
5
Abstract: Background: Acute and acute hematogenous Periprosthetic Joint Infections (PJIs) are
7
often treated with Open Debridement Polyethylene Exchange (ODPE) in an effort to save the
8
prosthesis, decrease morbidity and reduce costs. However, failure of ODPE may compromise a
9
subsequent two-stage treatment. The purpose of this study is to identify patient factors that
11
impact the success of ODPE for acute and acute hematogenous PJIs.
Methods: A retrospective review examined comorbidities, preoperative labs, and patient
SC
10
RI PT
6
history for patients with successful and failed ODPE treatment for acute peri-operative or acute
13
hematogenous periprostheic hip or knee joint infections. Successful treatment was defined as
14
retaining a well-fixed implant without the need for additional surgery for a minimum of 6 month
15
follow up with or without lifelong oral maintenance antibiotics.
16
M AN U
12
Results: 53/72 patients (73.6%) underwent successful ODPE. Of the 19 failures, 14 completed 2-stage revision with one subsequent known failure for recurrent infection. Patients
18
with a Staph aureus infection were more likely to fail ODPE (48.3% vs 11.6% p=0.0012), odds
19
ratio 7.1 (2.3-25.3). Patients with a preoperative hematocrit <=32.1 were also more likely to fail
20
ODPE (55% vs 16% p=0.0013), odds ratio 6.7 (2.2-22.4). When neither risk factor was present,
21
97.1% of peri-prosthetic joint infections were successfully treated with ODPE.
EP
Conclusion: Staph aureus infection and preoperative hematocrit <=32.1 are independent
AC C
22
TE D
17
23
risk factors for ODPE failure. ODPE is a safe alternative to 2-stage revision in patients without
24
preoperative anemia and without Staphylococcus aureus infection. Two-thirds of patients with a
25
failed ODPE were successfully treated with a two-stage re-implantation.
26
Level of Evidence: III – Retrospective Cohort Study
27
2
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
Prosthetic Joint Infection; ODPE; DAIR; Outcomes; Infection; Treatment
EP
29
Keywords:
AC C
28
3
ACCEPTED MANUSCRIPT
30 31
Introduction: Prosthetic hip and knee replacement is one of the most successful elective surgeries providing long term pain relief and improved function to patients with joint pain. The success of
33
these procedures in combination with an aging population has led to over 1 million procedures
34
performed in the United States in 2010 with a projected growth to nearly 3.5 million procedures
35
annually by 2030 [1].
Unfortunately, approximately 1-2% of total joint replacements will become infected, and
SC
36
RI PT
32
the burden will increase with the increasing number of primary and revision procedures[1-3].
38
Identified risk factors for prosthetic joint infection (PJI) include prior surgery, anemia, blood
39
transfusions, liver disease, drug abuse, diabetes, obesity and others [4-7].
40
M AN U
37
The gold standard for eradicating prosthetic joint infections is a two stage revision[8, 9]; however, the disability to the patient [10], increased mortality [11, 12], and the costs are quite
42
high [13, 14]. Additionally, the reported success rates are modest and a significant number of
43
patients are not candidates for a re-implantation and thus are left with significant, permanent
44
disability [15, 16].
Infections less than 3-4 weeks old (acute peri-operative or acute hematogenous) are often
EP
45
TE D
41
managed with open debridement with polyethylene exchange (ODPE) and adjuvant antibiotic
47
therapy. Due to lower morbidity and improved function, the ODPE pathway is less demanding
48
on patients than a two stage revision, leading some authors to conclude that ODPE “is the
49
obvious treatment of acute PJI, with good success rates in selected patients.” [17] Other authors,
50
however, have noted less successful results with ODPE [18],[19].
AC C
46
51
Over the last few years, several groups have raised concerns about the appropriateness of
52
ODPE observing that 2-stage revisions following failed ODPE are less successful compared with
4
ACCEPTED MANUSCRIPT
2-stage revision without a trial of hardware retention (59-66% vs. 89 %)[18], [19]. Failed ODPE
54
that delays definitive 2-stage revision may extend the period of disability for the patient and may
55
predispose the patient to failure of 2-stage revision. A better understanding of which patients
56
will likely benefit from ODPE and which patients are better managed with an immediate two-
57
stage treatment would improve the management of peri-prosthetic infection.
RI PT
53
Efforts to identify patient risk factors for failed ODPE have consistently found
59
Staphylococcus aureus infections to be the strongest risk factor for failure [4, 5, 7, 18, 20-22].
60
Other risk factors such as arthroscopic washout [5], less than 4 weeks of IV antibiotics [5], more
61
than 90 days from index procedure, elevated ASA scores [4], purulence in the joint [4], elevated
62
inflammatory markers[7, 20], and bacteremia[4, 20] have been reported, but have not been
63
universally observed. Buller attempted to synthesize these risk factors into a nomogram tool to
64
help predict failure [7]. The nomogram relies most heavily on the infecting organism. Several
65
authors have reported the association of anemia and primary periprosthetic infection [23];
66
however, anemia at the time of infection presentation has not been connected to the success or
67
failure of ODPE.
M AN U
TE D
The purpose of this work is to re-evaluate reported risk factors for success or failure of
EP
68
SC
58
ODPE in our population and examine the hypothesis that anemia and blood transfusions will be
70
associated with failed infection treatment in appropriate ODPE candidates.
71 72 73
AC C
69
Methods:
Following IRB approval, CPT codes were used to identify patients who underwent open
74
debridement with either femoral head and polyethylene liner exchange for hips or polyethylene
75
exchange for knees between 1/1/2007 to 12/31/2013. This list was cross-referenced with a
5
ACCEPTED MANUSCRIPT
clinical antibiotic database used by the Infectious Disease department to manage antibiotics
77
during PJIs. The subsequent list was used to perform a retrospective chart review. Inclusion
78
criteria for the chart review included patients who were diagnosed with an acute peri-operative
79
infection (<4 weeks from index procedure) or an acute hematogenous infection (<4 week history
80
of new onset symptoms concerning for infection). Only patients who received ODPE with
81
adjunct IV antibiotics were included. Patients with less than 6 months follow-up were excluded
82
unless they had failed.
SC
83
RI PT
76
Periprosthetic joint infections were identified by the attending surgeon after considering patient history, clinical exam findings, aspiration results and laboratory results. Musculoskeletal
85
Infection Society (MSIS) criteria were applied when these became available. Clinical findings of
86
infection included pain, erythema, swelling, drainage from acute surgical wounds, and purulence.
87
Preoperative laboratory evaluation included serum CBC/ ESR/CRP and synovial fluid cell count
88
with differential. Preoperative cultures obtained via aspiration of the joint as well as tissue and
89
fluid obtained during surgery were also used to confirm infection. When the infecting organism
90
was not isolated, the clinical exam findings along with laboratory values were used to diagnosis
91
infection and to direct treatment with IV antibiotics. Parenteral antibiotic therapy was managed
92
by the Bone and Joint Infectious Disease team for at least 6 weeks followed by routine lifelong
93
oral suppression. When antibiotics were stopped by outside providers or when not tolerated by
94
the patient, they were not restarted.
TE D
EP
AC C
95
M AN U
84
Successful ODPE was defined as retaining prosthetic joint implants with no additional
96
surgery of the affected joint for infection after discharge. Patient charts were reviewed for
97
patient characteristics including age, gender, BMI, ASA, tobacco use, diabetes,
98
immunosuppression, rheumatoid arthritis, kidney disease, HIV, the presence of a primary joint
6
ACCEPTED MANUSCRIPT
99
replacement vs revision, and the time from implantation. Charts were also used to extract information regarding the patient status at the time of infection diagnosis including preoperative
101
Systemic Inflammatory Response Syndrome (SIRS) criteria, joint aspiration data, culture data,
102
and pre- and post-operative lab values. Details from the surgery and hospital stay including
103
duration of stay and perioperative blood product transfusion were also noted.
104
RI PT
100
RStudio software[24] was used for all calculations including statistical values. Pearson’s Chi-squared test was used for categorical data while Student’s t-Test was used for continuous
106
data. Linear regression was employed for analysis of the interaction of anemia and Staph aureus
107
infection using the GLM function with binomial family within RStudio.
M AN U
SC
105
108 109 110
Results:
83 patients were identified who were treated with ODPE. Eleven patients had less than 6 months follow-up but are not known to have failed. One patient failed at less than 6 months.
112
The study group then consisted of 71 patients with > 6 month follow-up and 1 patient who failed
113
within 6 months. The average follow-up was 2.89 years with a range of 6 months to 6.3 years.
114
Nineteen patients retained their hardware and oral antibiotics were stopped (26.4%), 34 patients
115
retained their hardware and suppressive antibiotics were continued (47.2%), and 19 patients
116
failed ODPE (26.4%). Implant retention with respect to time is shown in Figure 1A. Fourteen
117
patients who failed ODPE were treated with 2 stage revision. Seven of these patients have been
118
seen recently and are without infectious symptoms, 1 patient with severe hemophilia and history
119
of multiple line infections underwent a 2 stage revision that developed recurrent infection and
120
was explanted 4 years after his replant. Five patients are no longer followed by our practice (2
121
requested pcp follow up only as they were doing well at one year, one requested to follow up at
AC C
EP
TE D
111
7
ACCEPTED MANUSCRIPT
122
the VA 3 yrs after the replant, and 2 stopped coming to visits for unknown reasons one after the
123
10 wk appointment and the other after one year). 4 joints were treated with explantation without
124
replantation and 1 knee was fused following infected extensor mechanism allograft.
126 127
RI PT
125
Patient Characteristics
There were no statistical differences between the groups with respect to age, gender, BMI, ASA, tobacco use, the presence of a revision prostheses, or comorbid conditions including
129
Diabetes, Rheumatoid Arthritis, ESRD, or HIV (Table 1). Vital signs and laboratory values prior
130
to ODPE were not significantly different for CRP, ESR, aspirate cell count and %PMN, platelets,
131
glucose, or for number of SIRS criteria (WBC >12k or <4k, Temp >38C or <36C, respiratory
132
rate > 20, or tachycardia >90) (Table 1). One patient had all 4 SIRS criteria and was able to
133
successfully retain hardware. Other combinations of SIRS criteria likewise failed to show
134
statistical significant predictive ability. The complication rates during antibiotic therapy
135
including 30-day readmission, line infection, Clostridium difficile infection and deep vein
136
thrombosis were not significantly different between the groups (Table 1). The average length of
137
stay for the failure group was about twice as long as the success group (14.32 vs 7.32 days,
138
p=0.0926); however, two PJIs in the failure group were associated with one patient who was in
139
the hospital for 60 days. The medians for the two groups were much closer (8 vs 6 days)
140
indicating that these outliers may have skewed the averages in the failure group. Likewise, there
141
was a trend for more patients in the successful group to have been discharged home following
142
ODPE (38/53 (71.7%) vs 10/19 (52.6%) p=0.1138) (table1).
143 144
AC C
EP
TE D
M AN U
SC
128
Fifty patients had symptoms for less than one week and the remaining 22 had symptoms for up to 4 weeks. There was no difference in the success rate for infections treated within one
8
ACCEPTED MANUSCRIPT
week of symptoms with a success rate of 37/50 (74.0%) verses 1-4 weeks of symptoms 16/22
146
(72.7% (p=1.000)). Likewise, there was no difference in the success vs failure groups with
147
respect to time from implantation (3.15 vs 2.13 years p=0.2466) or acute peri-operative vs acute
148
hematogenous infections (13/53 (24.5%) vs 3/19 (15.8%) p=0.6160).
149 150
RI PT
145
The only factors which correlated with failure of ODPE were Staph aureus infection and presenting HCT <=32.1 % (Table 1).
153
Impact of Organism
Twenty-nine of the seventy-two patients (29/72 (40.3%)) presented with an isolated
M AN U
152
SC
151
Staph aureus infection or a multi-organismal infection including at least one positive culture for
155
Staph aureus. Fourteen of these patients (14/29 (48.3%)) failed ODPE. Forty-three patients
156
(43/72 (59.7%)) presented with a non-Staph aureus infection and 5 failed ODPE (15/43
157
(11.6%)). This difference was highly significant with a p value of p=0.0012 and an odds ratio of
158
7.1 (95% confidence interval of (2.3-25.3)). Implant retention with respect to time is shown in
159
Figure 1B. The average time to failure was 16.0 months for the Staph aureus group and 3.8
160
months for the non-Staph aureus group (p=0.0968) (Table 2). Twelve patients had culture
161
negative infections and all underwent successful ODPE; however, the incidence of culture
162
negative infection was not high enough to generate statistically significant differences (Table 1).
163
There were no significant differences for other observed bacteria or for polymicrobial vs
164
monomicrobial infections (Table 1).
166
EP
AC C
165
TE D
154
Impact of Presenting HCT
9
ACCEPTED MANUSCRIPT
Patients who failed ODPE had a lower average preoperative hematocrit (31.39 vs 34.40
168
p=0.0104) (Figure 2 & Table 1). Twenty patients who underwent ODPE had a presenting HCT
169
<=32.1% and 52 patients had a presenting HCT >32.1%. Fifty-five percent of patients in the low
170
HCT group failed ODPE versus 15.4%of patients in the high HCT group (p=0.0013). Implant
171
retention with respect to time is shown in Figure 1C. There was no difference in the time to
172
failure between the groups (Table 2).
There was also a disproportionate incidence of Staph aureus infection in the low
SC
173
RI PT
167
hematocrit group. Sixty percent of patients with HCT <=32.1 were infected with Staph aureus
175
while seventeen of the fifty-two patients (17/52 (32.7%)) with HCT >32.1 were infected with
176
Staph aureus (p=0.0646). Eleven of the 12 patients with culture negative infections had a
177
presenting HCT > =32.1.
178
M AN U
174
During this study, we had an overall transfusion rate of 33.3% for packed red blood cells and 40.2% for blood products of any kind (platelets, fresh frozen plasma, or packed red blood
180
cells). There were no statistical differences between the groups for the number of patients
181
receiving transfusions of packed red blood cells (8/19 (42.1%) vs 16/53(30.2%) p=0.5081) or the
182
number receiving blood products of any kind (9/19 (47.4%) vs 20/53 (37.7%), p=0.6442)
183
(Table1). Thirty-one percent (9/19) of patients who received blood products of any kind failed
184
ODPE while 10/43 (23.3%) who did not receive blood failed ODPE (p=0.6442).
186 187
EP
AC C
185
TE D
179
Combined Effect of Organism and Presenting HCT Linear regression modeling demonstrated that both low HCT and Staph aureus were
188
independent risk factors for failed ODPE (p=0.0082 and p=0.0057). The different combinations
189
of these risk factors were then analyzed (Table 2). Strikingly, of the thirty-five patients without
10
ACCEPTED MANUSCRIPT
Staph aureus infection and with a HCT >32.1, only one (2.9%) failed ODPE (Figure 1D, Table
191
2). The remaining groups were much less successful and the reported odds ratios for failure are
192
with respect to these thirty-five patients. Twelve patients presented with a Staph aureus infection
193
and a low HCT with 58.3% failing ODPE and an odds ratio for failure of 47.6 (95% confidence
194
interval (6.7-997.2)) when compared to the group without Staph aureus and with a HCT >32.1.
195
Seventeen patients were infected with Staph aureus and presented with a HCT >32.1 with 41.1%
196
failing ODPE and an odds ratio for failure of 34.0 (3.9-765.4). Eight patients presented with low
197
HCT without Staph aureus infection. 50.0% of these patients failed ODPE yielding an odds ratio
198
of failing of 23.8 (3.6-473.7).
200
Discussion:
M AN U
199
SC
RI PT
190
Peri-prosthetic infection after hip and knee arthroplasty remains a very serious
202
complication with high morbidity, mortality and costs, and is one of the most common reasons
203
for revision surgeries[25]. Two stage revision has been the most consistently successful
204
approach to eradicating these infections but the morbidity, cost and rate of poor functional
205
outcomes are high [10-16]. As a result, many surgeons have turned to ODPE for selected
206
patients with acute peri-prosthetic infections due to the lower morbidity and improved early
207
return of function. In general, the indications for ODPE have included acute infections (<3-4
208
weeks) in a patient with well-fixed components, an infecting organism responsive to oral
209
antibiotics, and absence of a draining sinus. Reported outcomes of this technique have been
210
varied ranging from 16 % to 83 % infection-free survival[17, 26, 27]. Recently there has been
211
some concern that failed ODPE may be associated with a higher rate of subsequent 2-stage
212
revision failure[18, 19].
AC C
EP
TE D
201
11
ACCEPTED MANUSCRIPT
213
If ODPE fails, a two-stage procedure would typically be the salvage option. Sherrell et al suggest that 2-stage revisions in the absence of prior ODPE have a near 90% success rate
215
while ODPE has close to a 40% success rate leading some authors to discourage ODPE[18, 19].
216
However, more recent studies have suggested that 2-stage revision outcomes may not be as
217
successful as once thought. Gomez et al reported that about 20% of patients who have had their
218
joints explanted never progress to the second stage[15]. These patients are often excluded from
219
the success rates for 2-stage revisions as rates are usually reported as the number of successfully
220
reconstructed 2-stage revisions. Berend et al. report a higher reimplantation rate in their series of
221
hip infections at 92%[11]. They report a success rate of 157/189 (83%) for reimplanted joints.
222
When all patients who were explanted are included and those lost to follow up are excluded, the
223
success rate drops to 144/189 (76%). This result for 2-stage revision is strikingly close to our
224
success rate of 75.3% with ODPE. While the important distinction of chronic infection vs acute
225
infection cannot be ignored, when failure to reimplant due to comorbidities or death are included
226
as failures of 2-stage revision as they would be in ODPE, the resulting success rates of ODPE
227
and 2-stage revision are much closer than previously reported. Our data would suggest that
228
failed ODPE may simply select for more difficult to treat infections and/or a less healthy host.
229
The ability to identify these at risk patients may allow surgeons to escalate care to improve
230
outcomes.
SC
M AN U
TE D
EP
AC C
231
RI PT
214
In this study we have identified two markers for poor outcomes with ODPE in acute peri-
232
prosthetic infection. The presence of a Staph aureus infection or a pre-operative HCT <=32.1
233
predict a success rate of about 56% with the success rate falling to 42% when both factors are
234
combined. When neither factor is present (about half of our patients) the success rate was 97%.
235
The ability to identify patients who will do well with ODPE allows surgeons to more confidently
12
ACCEPTED MANUSCRIPT
236
recommend ODPE. For this group, ODPE appears to be safe and effective sparing them from
237
the challenges of a 2-stage revision.
238
This study is not without limitations. This work is a retrospective evaluation of a relatively small number of patients. This challenge of small sample size is fairly consistent
240
within the literature for ODPE candidates. This is largely due to the low rates of infection
241
overall but is compounded by lower rates of infections at high volume joints centers[28, 29] such
242
as ours. Further, while all infections are acute or acute hematogenous at one point, the majority
243
of patients do not present from outside surgeons until they have developed a chronic infection.
244
Strategies to increase numbers include including expanding encatchment time or the number of
245
treating surgeons. Increasing the encatchment time can introduce temporal biases highlighting
246
the need for data sharing between large centers either in the form of collaborative studies, meta-
247
analysis, or database creation. Despite the challenge of small sample size in this work, our results
248
were highly statistically significant with p-values of 0.0012 and 0.0013. Multiple surgeons and
249
infectious disease specialists participated in the management of these patients. Preoperative
250
ESR, CRP, CBC and cultures were the most consistently ordered tests. Information concerning
251
nutritional status at the time of surgery was not available. Culture negative infections that were
252
treated with ODPE were included in this study; however, the data were recalculated excluding
253
culture negative infections and the statistical significance of the anemia and bacterial species
254
remained. Patients who had retained their components but had a poorly functioning knee would
255
be considered a success in this study.
257
SC
M AN U
TE D
EP
AC C
256
RI PT
239
Conclusion:
13
ACCEPTED MANUSCRIPT
258
ODPE is an attractive treatment modality for management of acute and acute hematogenous periprosthetic infections in selected circumstances. This study once again
260
confirms that the infecting organism is an important determinant of outcome. Staphylococcus
261
aureus infection greatly decreases the success of ODPE while non-staph aureus infection was
262
much more successful. Culture negative infections had a 100% success rate. We further present
263
the finding that a hematocrit at or below 32.1 is also strongly associated with ODPE failure.
264
Patients with a hematocrit above 32.1 and who were not infected with Staph aureus were
265
associated with a 97% successful ODPE treatment. For this subset of patients, ODPE is safe and
266
effective. For patients that failed ODPE in this study, we did not observe high rates of failure of
267
subsequent 2-stage revision. Further work should be done to identify additional markers for
268
success and further examine the applicability of these markers to larger patient populations.
AC C
EP
TE D
M AN U
SC
RI PT
259
14
ACCEPTED MANUSCRIPT
269
References:
270
1.
Kurtz, S., Projections of Primary and Revision Hip and Knee Arthroplasty in the United States from 2005 to 2030. The Journal of Bone and Joint Surgery (American), 2007.
272
89(4): p. 780. 2.
The Journal of Arthroplasty, 2008. 23(7): p. 984-991.
274 275
Kurtz, S.M., et al., Infection Burden for Hip and Knee Arthroplasty in the United States.
3.
Wilson, M.G., K. Kelley, and T.S. Thornhill, Infection as a complication of total knee-
SC
273
RI PT
271
replacement arthroplasty. Risk factors and treatment in sixty-seven cases. The Journal of
277
Bone & Joint Surgery, 1990. 72(6): p. 878-883.
278
4.
M AN U
276
Azzam, K.A., et al., Irrigation and Debridement in the Management of Prosthetic Joint
279
Infection: Traditional Indications Revisited. The Journal of Arthroplasty, 2010. 25(7): p.
280
1022-1027. 5.
Byren, I., et al., One hundred and twelve infected arthroplasties treated with 'DAIR'
TE D
281 282
(debridement, antibiotics and implant retention): antibiotic duration and outcome.
283
Journal of Antimicrobial Chemotherapy, 2009. 63(6): p. 1264-1271. 6.
Cordero-Ampuero, J. and M. de Dios, What Are the Risk Factors for Infection in
EP
284
Hemiarthroplasties and Total Hip Arthroplasties? Clinical Orthopaedics and Related
286
Research®, 2010. 468(12): p. 3268-3277.
287
7.
Arthroplasty, 2012. 27(6): p. 857-64.e1-4.
289
291
Buller, L.T., et al., The preoperative prediction of success following irrigation and debridement with polyethylene exchange for hip and knee prosthetic joint infections. J
288
290
AC C
285
8.
Parvizi, J., B. Zmistowski, and B. Adeli, Periprosthetic Joint Infection: Treatment Options. Orthopedics, 2010.
15
ACCEPTED MANUSCRIPT
292
9.
Osmon, D.R., et al., Diagnosis and Management of Prosthetic Joint Infection: Clinical
293
Practice Guidelines by the Infectious Diseases Society of America. Clinical Infectious
294
Diseases, 2012. 56(1): p. e1-e25. 10.
Oussedik, S.I.S., M.B. Dodd, and F.S. Haddad, Outcomes of revision total hip
RI PT
295 296
replacement for infection after grading according to a standard protocol. Journal of
297
Bone and Joint Surgery - British Volume, 2010. 92-B(9): p. 1222-1226. 11.
Berend, K.R., et al., Two-stage Treatment of Hip Periprosthetic Joint Infection Is
SC
298
Associated With a High Rate of Infection Control but High Mortality. Clinical
300
Orthopaedics and Related Research®, 2012. 471(2): p. 510-518.
301
12.
M AN U
299
Zmistowski, B., et al., Periprosthetic Joint Infection Increases the Risk of One-Year
302
Mortality. The Journal of Bone and Joint Surgery-American Volume, 2013. 95(24): p.
303
2177-2184.
18(9): p. 871-873.
305
14.
Br, 2012. 94(5): p. 619-623.
307 308
15.
2012. 50(11): p. 3501-8.
310
312 313
Gomez, E., et al., Prosthetic joint infection diagnosis using broad-range PCR of biofilms dislodged from knee and hip arthroplasty surfaces using sonication. J Clin Microbiol,
309
311
Vanhegan, I., et al., A financial analysis of revision hip arthroplasty. J Bone Joint Surg
EP
306
Sculco, T.P., The economic impact of infected joint arthroplasty. Orthopedics, 1995.
TE D
13.
AC C
304
16.
Kheir, M.M., et al., Patients With Failed Prior Two-Stage Exchange Have Poor Outcomes After Further Surgical Intervention. The Journal of Arthroplasty, 2017. 32(4): p. 1262-1265.
16
ACCEPTED MANUSCRIPT
314
17.
Review of current concepts. World Journal of Orthopedics, 2014. 5(5): p. 667.
315 316
Kuiper, J.W.P., Treatment of acute periprosthetic infections with prosthesis retention:
18.
Gardner, J., T.J. Gioe, and P. Tatman, Can This Prosthesis Be Saved?: Implant Salvage Attempts in Infected Primary TKA. Clinical Orthopaedics and Related Research®, 2010.
318
469(4): p. 970-976.
319
19.
RI PT
317
Sherrell, J.C., et al., The Chitranjan Ranawat Award: Fate of Two-stage Reimplantation After Failed Irrigation and Débridement for Periprosthetic Knee Infection. Clinical
321
Orthopaedics and Related Research®, 2010. 469(1): p. 18-25. 20.
Vilchez, F., et al., Outcome and predictors of treatment failure in early post-surgical
M AN U
322
SC
320
323
prosthetic joint infections due to Staphylococcus aureus treated with debridement. Clin
324
Microbiol Infect, 2011. 17(3): p. 439-44. 21.
of Arthroplasty, 2011. 26(6): p. 114-118.
326 327
Odum, S.M., et al., Irrigation and Debridement for Periprosthetic Infections. The Journal
22.
TE D
325
Deirmengian, C., et al., Limited success with open debridement and retention of components in the treatment of acute Staphylococcus aureus infections after total knee
329
arthroplasty. The Journal of arthroplasty, 2003. 18: p. 22-26. 23.
Mortality Following Total Joint Arthroplasty. The Journal of Arthroplasty, 2015. 30(5):
331
p. 846-848.
332 333 334
Viola, J., et al., Preoperative Anemia Increases Postoperative Complications and
AC C
330
EP
328
24.
Team, R., RStudio: integrated development for R. RStudio, Inc., Boston, MA URL http://www. rstudio. com, 2015.
17
ACCEPTED MANUSCRIPT
335
25.
Koh, C.K., et al., Periprosthetic Joint Infection Is the Main Cause of Failure for Modern
336
Knee Arthroplasty: An Analysis of 11,134 Knees. Clinical Orthopaedics and Related
337
Research®, 2017. 475(9): p. 2194-2201. 26.
infected total knee arthroplasty. The Journal of arthroplasty, 1997. 12(4): p. 426-433.
339 340
Mont, M.A., et al., Multiple irrigation, debridement, and retention of components in
RI PT
338
27.
Bradbury, T., et al., The fate of acute methicillin-resistant Staphylococcus aureus
periprosthetic knee infections treated by open debridement and retention of components.
342
The Journal of arthroplasty, 2009. 24(6): p. 101-104. 28.
outcomes of total knee replacement. JBJS, 2004. 86(9): p. 1909-1916.
344 345
Katz, J.N., et al., Association between hospital and surgeon procedure volume and the
M AN U
343
SC
341
29.
Katz, J.N., et al., Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population. Jbjs, 2001.
347
83(11): p. 1622-1629.
EP AC C
348
TE D
346
18
ACCEPTED MANUSCRIPT
Tables and Figures:
2
Table 1: Comparison of demographics, lab data, and vital signs between successful and failure
3
groups. Only differences in blood counts and infecting organism were statistically significant.
4
Figure 1: Kaplan Meier Curves for implant survivability for all patients (A), with respect to
5
Staph aureus infection (B), with respect to preoperative anemia (C), and with respect to
6
combined anemia and Staph aureus infection (D).
7
Figure 2: Histogram illustrating the difference in distribution of preoperative blood counts
8
between successful and failure groups.
9
Table 2: Chi Square analysis for ODPE failure with respect to Staph aureus infection alone or
M AN U
SC
RI PT
1
10
with respect to low HCT alone and Linear Regression analysis for combinations of low HCT and
11
Staph aureus infection combined risk factors. Odds ratios are with respect to a patient without
12
the examined risk factor(s).
AC C
EP
TE D
13
ACCEPTED MANUSCRIPT
Hematocrit (%) HCT > 32.1% Hemoglobin WBC Platelets Glucose CRP ESR
Failure 88085.07 87.50
Preop Labs Success 34.40 44 (83.0%) 11.27 10.13 293.15 124.77 11.83 64.19
Failure 31.39 8 (42.1%) 10.43 11.78 284.89 124.05 15.25 58.09
Postop Labs Success 29.67 9.55 9.59 284.26 142.83 Organism Success 15 (28.3%) 12 (22.6%) 3 (5.7%) 2 (3.8%) 6 (11.3%) 16 (30.2%) 8 (15.1%) 3 (5.7%) 12 (22.6%) 29 (54.7%) 12 (22.6%)
SC
Staph aureus MSSA MRSA MSSE MRSE Strep Corynebacterium Other Culture Negative Monomicrobial Polymicrobial
1.0000 0.6160 0.2466
p-value 0.6963 0.5177 0.5863 0.6324
Failure 25.34 8.29 14.40 299.43 142.71
RI PT
Hematocrit Hemoglobin WBC Platelets glucose
Complications Success 11 (20.8%) 2 (3.8%) 3 (5.7%) 1 (1.9%)
p-value 0.0014 0.0034 0.8494 1 1 0.1647 0.9145 1.0000 0.0557 0.1131 1.0000
Failure 4 (2.1%) 1 (5.3%) 1 (5.3%) 1 (5.3%)
p-value 1.0000 1.0000 1.0000 1.0000
p-value 0.3217 0.9827
p-value 0.0104 0.0018 0.0239 0.2647 0.8514 0.9464 0.2486 0.5763
Hospital Course Success Failure HGB Drop (g/dL) 1.62 1.07 HCT Drop (% points) 4.48 2.34 Length of Stay (days) 7.32 14.32 Surgical Duration (min) 104.77 134.37 # Debridements 1.26 1.21 Discharged Home 38 (71.7%) 10 (52.6%) pRBC 16 (30.2%) 8 (42.1%) Any Blood Products 20 (37.7%) 9 (47.4%)
1
p-value 0.0031 0.0004 0.1860 0.8528 0.9958
Failure 14 (73.7%) 12 (63.2%) 2 (10.5%) 1 (5.3%) 2 (10.5%) 2 (10.5%) 2 (10.5%) 1 (5.3%) 0 15 (78.9%) 4 (21.1%)
30 Day Readmission Line Infection C diff Infection DVT
TE D
Joint Aspiration Success 54231.53 87.61
AC C
Cell Count %PMN
Failure 0.00 2 (10.5%) 5 (26.3%) 2 (10.5%)
EP
Temp >38C or <36C WBC >12k or <4k Tachycardia Tachypnea
Preop Vitals Success 3 (5.7%) 11 (20.8%) 9 (17.0%) 10 (18.9%)
p-value 0.0660 1.0000 0.8979 0.4935 0.8452 0.5557 0.6869 0.5166 0.5166
M AN U
Demographics Success Failure Age 67.11 60.90 Gender(Female) 27 (50.9%) 10 (52.6%) BMI 35.50 35.14 ASA 2.75 2.84 ASA >=3 38 (71.7%) 15 (78.9%) Tobacco Use 4 (7.5%) 3 (15.8%) DM 15 (28.3%) 7 (36.8%) Immunosupressed 4 (7.5%) 0 RA 4 (7.5%) 0 ESRD 0 0 HIV 0 0 Replant 10 (18.8%) 4 (21.1%) Acute Peri-Operative 13 (24.5%) 3 (15.8%) Time From Implantation 3.15 2.13
p-value 0.0456 0.0862 0.0926 0.2531 0.6653 0.2191 0.5081 0.6442
ACCEPTED MANUSCRIPT
Failure Based on Staph aureus Odds Ratio
95% CI
14/29 (48.3%) 5/43 (11.6%)
0.0012 --
7.1
(2.3-25.3)
--
--
Failure Based on HCT HCT<=32.1 HCT>32.1
Failure
p-value
Odds Ratio
95% CI
11/20 (55.0%) 8/52 (15.4%)
0.0013 --
6.7
(2.2-22.4)
--
--
Failure Based on Risk Factors p-value
7/12 (58.2%) 4/8 (50.0%) 7/17 (41.1%) 1/35 (2.9%)
0.0010 0.0050 0.0044 --
Odds Ratio
95% CI
47.6 34.0 23.8 --
(6.7-997.2) (3.6-473.7) (3.9-765.4) --
M AN U
Failure
AC C
EP
TE D
HCT <= 32.1 and Staph aureus HCT <= 32.1 and not Staph aureus HCT > 32.1 and Staph aureus HCT > 32.1 and not Staph aureus
RI PT
p-value
SC
Staph aureus No-Staph aureus
Failure
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0883-5403(18)30080-9
DOI:
10.1016/j.arth.2018.01.042
Reference:
YARTH 56363
To appear in:
The Journal of Arthroplasty
Received Date: 27 November 2017 Revised Date:
8 January 2018
Accepted Date: 12 January 2018
Please cite this article as: Swenson RD, Butterfield J, Irwin TJ, Zurlo JJ, Davis III CM, Preoperative Anemia Is Associated With Failure Of Open Debridement Polyethylene Exchange In Acute And Acute Hematogenous Prosthetic Joint Infection, The Journal of Arthroplasty (2018), doi: 10.1016/ j.arth.2018.01.042. 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
PREOPERATIVE ANEMIA IS ASSOCIATED WITH FAILURE OF OPEN DEBRIDEMENT POLYETHYLENE EXCHANGE IN ACUTE AND ACUTE HEMATOGENOUS PROSTHETIC JOINT INFECTIONS
Please address all correspondence to: Richard D Swenson, MD Department of Orthopaedic Surgery 500 University Drive Hershey, PA 17033 USA
M AN U
2. Infectious Disease Penn State College of Medicine Hershey, Pa 17033
SC
1. Department of Orthopaedic Surgery Penn State College of Medicine Hershey, Pa 17033
[email protected] [email protected] [email protected] [email protected] [email protected]
RI PT
Richard D. Swenson, MD1 James Butterfield, BS1 Timothy J. Irwin, MD1 John J. Zurlo, MD2 Charles M. Davis III, MD/PhD1
TE D
Phone: 801-971-9610 Email: [email protected]
EP
Investigation performed at Penn State Department of Orthopaedic Surgery
AC C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
1
ACCEPTED MANUSCRIPT
1
Abstract: Background: Acute and acute hematogenous Periprosthetic Joint Infections (PJIs) are
3
often treated with Open Debridement Polyethylene Exchange (ODPE) in an effort to save the
4
prosthesis, decrease morbidity and reduce costs. However, failure of ODPE may compromise a
5
subsequent two-stage treatment. The purpose of this study is to identify patient factors that
6
impact the success of ODPE for acute and acute hematogenous PJIs.
Methods: A retrospective review examined comorbidities, preoperative labs, and patient
SC
7
RI PT
2
history for patients with successful and failed ODPE treatment for acute peri-operative or acute
9
hematogenous periprostheic hip or knee joint infections. Successful treatment was defined as
M AN U
8
10
retaining a well-fixed implant without the need for additional surgery for a minimum of 6 month
11
follow up with or without lifelong oral maintenance antibiotics.
12
Results: 53/72 patients (73.6%) underwent successful ODPE. Of the 19 failures, 14 completed 2-stage revision with one subsequent known failure for recurrent infection. Patients
14
with a Staph aureus infection were more likely to fail ODPE (48.3% vs 11.6% p=0.0012), odds
15
ratio 7.1 (2.3-25.3). Patients with a preoperative hematocrit <=32.1 were also more likely to fail
16
ODPE (55% vs 16% p=0.0013), odds ratio 6.7 (2.2-22.4). When neither risk factor was present,
17
97.1% of peri-prosthetic joint infections were successfully treated with ODPE.
EP
Conclusion: Staph aureus infection and preoperative hematocrit <=32.1 are independent
AC C
18
TE D
13
19
risk factors for ODPE failure. ODPE is a safe alternative to 2-stage revision in patients without
20
preoperative anemia and without Staphylococcus aureus infection. Two-thirds of patients with a
21
failed ODPE were successfully treated with a two-stage re-implantation.
22
1
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
PREOPERATIVE ANEMIA IS ASSOCIATED WITH FAILURE OF OPEN DEBRIDEMENT
2
POLYETHYLENE EXCHANGE IN ACUTE AND ACUTE HEMATOGENOUS
3
PROSTHETIC JOINT INFECTION
EP AC C
4
TE D
1
1
ACCEPTED MANUSCRIPT
5
Abstract: Background: Acute and acute hematogenous Periprosthetic Joint Infections (PJIs) are
7
often treated with Open Debridement Polyethylene Exchange (ODPE) in an effort to save the
8
prosthesis, decrease morbidity and reduce costs. However, failure of ODPE may compromise a
9
subsequent two-stage treatment. The purpose of this study is to identify patient factors that
11
impact the success of ODPE for acute and acute hematogenous PJIs.
Methods: A retrospective review examined comorbidities, preoperative labs, and patient
SC
10
RI PT
6
history for patients with successful and failed ODPE treatment for acute peri-operative or acute
13
hematogenous periprostheic hip or knee joint infections. Successful treatment was defined as
14
retaining a well-fixed implant without the need for additional surgery for a minimum of 6 month
15
follow up with or without lifelong oral maintenance antibiotics.
16
M AN U
12
Results: 53/72 patients (73.6%) underwent successful ODPE. Of the 19 failures, 14 completed 2-stage revision with one subsequent known failure for recurrent infection. Patients
18
with a Staph aureus infection were more likely to fail ODPE (48.3% vs 11.6% p=0.0012), odds
19
ratio 7.1 (2.3-25.3). Patients with a preoperative hematocrit <=32.1 were also more likely to fail
20
ODPE (55% vs 16% p=0.0013), odds ratio 6.7 (2.2-22.4). When neither risk factor was present,
21
97.1% of peri-prosthetic joint infections were successfully treated with ODPE.
EP
Conclusion: Staph aureus infection and preoperative hematocrit <=32.1 are independent
AC C
22
TE D
17
23
risk factors for ODPE failure. ODPE is a safe alternative to 2-stage revision in patients without
24
preoperative anemia and without Staphylococcus aureus infection. Two-thirds of patients with a
25
failed ODPE were successfully treated with a two-stage re-implantation.
26
Level of Evidence: III – Retrospective Cohort Study
27
2
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
Prosthetic Joint Infection; ODPE; DAIR; Outcomes; Infection; Treatment
EP
29
Keywords:
AC C
28
3
ACCEPTED MANUSCRIPT
30 31
Introduction: Prosthetic hip and knee replacement is one of the most successful elective surgeries providing long term pain relief and improved function to patients with joint pain. The success of
33
these procedures in combination with an aging population has led to over 1 million procedures
34
performed in the United States in 2010 with a projected growth to nearly 3.5 million procedures
35
annually by 2030 [1].
Unfortunately, approximately 1-2% of total joint replacements will become infected, and
SC
36
RI PT
32
the burden will increase with the increasing number of primary and revision procedures[1-3].
38
Identified risk factors for prosthetic joint infection (PJI) include prior surgery, anemia, blood
39
transfusions, liver disease, drug abuse, diabetes, obesity and others [4-7].
40
M AN U
37
The gold standard for eradicating prosthetic joint infections is a two stage revision[8, 9]; however, the disability to the patient [10], increased mortality [11, 12], and the costs are quite
42
high [13, 14]. Additionally, the reported success rates are modest and a significant number of
43
patients are not candidates for a re-implantation and thus are left with significant, permanent
44
disability [15, 16].
Infections less than 3-4 weeks old (acute peri-operative or acute hematogenous) are often
EP
45
TE D
41
managed with open debridement with polyethylene exchange (ODPE) and adjuvant antibiotic
47
therapy. Due to lower morbidity and improved function, the ODPE pathway is less demanding
48
on patients than a two stage revision, leading some authors to conclude that ODPE “is the
49
obvious treatment of acute PJI, with good success rates in selected patients.” [17] Other authors,
50
however, have noted less successful results with ODPE [18],[19].
AC C
46
51
Over the last few years, several groups have raised concerns about the appropriateness of
52
ODPE observing that 2-stage revisions following failed ODPE are less successful compared with
4
ACCEPTED MANUSCRIPT
2-stage revision without a trial of hardware retention (59-66% vs. 89 %)[18], [19]. Failed ODPE
54
that delays definitive 2-stage revision may extend the period of disability for the patient and may
55
predispose the patient to failure of 2-stage revision. A better understanding of which patients
56
will likely benefit from ODPE and which patients are better managed with an immediate two-
57
stage treatment would improve the management of peri-prosthetic infection.
RI PT
53
Efforts to identify patient risk factors for failed ODPE have consistently found
59
Staphylococcus aureus infections to be the strongest risk factor for failure [4, 5, 7, 18, 20-22].
60
Other risk factors such as arthroscopic washout [5], less than 4 weeks of IV antibiotics [5], more
61
than 90 days from index procedure, elevated ASA scores [4], purulence in the joint [4], elevated
62
inflammatory markers[7, 20], and bacteremia[4, 20] have been reported, but have not been
63
universally observed. Buller attempted to synthesize these risk factors into a nomogram tool to
64
help predict failure [7]. The nomogram relies most heavily on the infecting organism. Several
65
authors have reported the association of anemia and primary periprosthetic infection [23];
66
however, anemia at the time of infection presentation has not been connected to the success or
67
failure of ODPE.
M AN U
TE D
The purpose of this work is to re-evaluate reported risk factors for success or failure of
EP
68
SC
58
ODPE in our population and examine the hypothesis that anemia and blood transfusions will be
70
associated with failed infection treatment in appropriate ODPE candidates.
71 72 73
AC C
69
Methods:
Following IRB approval, CPT codes were used to identify patients who underwent open
74
debridement with either femoral head and polyethylene liner exchange for hips or polyethylene
75
exchange for knees between 1/1/2007 to 12/31/2013. This list was cross-referenced with a
5
ACCEPTED MANUSCRIPT
clinical antibiotic database used by the Infectious Disease department to manage antibiotics
77
during PJIs. The subsequent list was used to perform a retrospective chart review. Inclusion
78
criteria for the chart review included patients who were diagnosed with an acute peri-operative
79
infection (<4 weeks from index procedure) or an acute hematogenous infection (<4 week history
80
of new onset symptoms concerning for infection). Only patients who received ODPE with
81
adjunct IV antibiotics were included. Patients with less than 6 months follow-up were excluded
82
unless they had failed.
SC
83
RI PT
76
Periprosthetic joint infections were identified by the attending surgeon after considering patient history, clinical exam findings, aspiration results and laboratory results. Musculoskeletal
85
Infection Society (MSIS) criteria were applied when these became available. Clinical findings of
86
infection included pain, erythema, swelling, drainage from acute surgical wounds, and purulence.
87
Preoperative laboratory evaluation included serum CBC/ ESR/CRP and synovial fluid cell count
88
with differential. Preoperative cultures obtained via aspiration of the joint as well as tissue and
89
fluid obtained during surgery were also used to confirm infection. When the infecting organism
90
was not isolated, the clinical exam findings along with laboratory values were used to diagnosis
91
infection and to direct treatment with IV antibiotics. Parenteral antibiotic therapy was managed
92
by the Bone and Joint Infectious Disease team for at least 6 weeks followed by routine lifelong
93
oral suppression. When antibiotics were stopped by outside providers or when not tolerated by
94
the patient, they were not restarted.
TE D
EP
AC C
95
M AN U
84
Successful ODPE was defined as retaining prosthetic joint implants with no additional
96
surgery of the affected joint for infection after discharge. Patient charts were reviewed for
97
patient characteristics including age, gender, BMI, ASA, tobacco use, diabetes,
98
immunosuppression, rheumatoid arthritis, kidney disease, HIV, the presence of a primary joint
6
ACCEPTED MANUSCRIPT
99
replacement vs revision, and the time from implantation. Charts were also used to extract information regarding the patient status at the time of infection diagnosis including preoperative
101
Systemic Inflammatory Response Syndrome (SIRS) criteria, joint aspiration data, culture data,
102
and pre- and post-operative lab values. Details from the surgery and hospital stay including
103
duration of stay and perioperative blood product transfusion were also noted.
104
RI PT
100
RStudio software[24] was used for all calculations including statistical values. Pearson’s Chi-squared test was used for categorical data while Student’s t-Test was used for continuous
106
data. Linear regression was employed for analysis of the interaction of anemia and Staph aureus
107
infection using the GLM function with binomial family within RStudio.
M AN U
SC
105
108 109 110
Results:
83 patients were identified who were treated with ODPE. Eleven patients had less than 6 months follow-up but are not known to have failed. One patient failed at less than 6 months.
112
The study group then consisted of 71 patients with > 6 month follow-up and 1 patient who failed
113
within 6 months. The average follow-up was 2.89 years with a range of 6 months to 6.3 years.
114
Nineteen patients retained their hardware and oral antibiotics were stopped (26.4%), 34 patients
115
retained their hardware and suppressive antibiotics were continued (47.2%), and 19 patients
116
failed ODPE (26.4%). Implant retention with respect to time is shown in Figure 1A. Fourteen
117
patients who failed ODPE were treated with 2 stage revision. Seven of these patients have been
118
seen recently and are without infectious symptoms, 1 patient with severe hemophilia and history
119
of multiple line infections underwent a 2 stage revision that developed recurrent infection and
120
was explanted 4 years after his replant. Five patients are no longer followed by our practice (2
121
requested pcp follow up only as they were doing well at one year, one requested to follow up at
AC C
EP
TE D
111
7
ACCEPTED MANUSCRIPT
122
the VA 3 yrs after the replant, and 2 stopped coming to visits for unknown reasons one after the
123
10 wk appointment and the other after one year). 4 joints were treated with explantation without
124
replantation and 1 knee was fused following infected extensor mechanism allograft.
126 127
RI PT
125
Patient Characteristics
There were no statistical differences between the groups with respect to age, gender, BMI, ASA, tobacco use, the presence of a revision prostheses, or comorbid conditions including
129
Diabetes, Rheumatoid Arthritis, ESRD, or HIV (Table 1). Vital signs and laboratory values prior
130
to ODPE were not significantly different for CRP, ESR, aspirate cell count and %PMN, platelets,
131
glucose, or for number of SIRS criteria (WBC >12k or <4k, Temp >38C or <36C, respiratory
132
rate > 20, or tachycardia >90) (Table 1). One patient had all 4 SIRS criteria and was able to
133
successfully retain hardware. Other combinations of SIRS criteria likewise failed to show
134
statistical significant predictive ability. The complication rates during antibiotic therapy
135
including 30-day readmission, line infection, Clostridium difficile infection and deep vein
136
thrombosis were not significantly different between the groups (Table 1). The average length of
137
stay for the failure group was about twice as long as the success group (14.32 vs 7.32 days,
138
p=0.0926); however, two PJIs in the failure group were associated with one patient who was in
139
the hospital for 60 days. The medians for the two groups were much closer (8 vs 6 days)
140
indicating that these outliers may have skewed the averages in the failure group. Likewise, there
141
was a trend for more patients in the successful group to have been discharged home following
142
ODPE (38/53 (71.7%) vs 10/19 (52.6%) p=0.1138) (table1).
143 144
AC C
EP
TE D
M AN U
SC
128
Fifty patients had symptoms for less than one week and the remaining 22 had symptoms for up to 4 weeks. There was no difference in the success rate for infections treated within one
8
ACCEPTED MANUSCRIPT
week of symptoms with a success rate of 37/50 (74.0%) verses 1-4 weeks of symptoms 16/22
146
(72.7% (p=1.000)). Likewise, there was no difference in the success vs failure groups with
147
respect to time from implantation (3.15 vs 2.13 years p=0.2466) or acute peri-operative vs acute
148
hematogenous infections (13/53 (24.5%) vs 3/19 (15.8%) p=0.6160).
149 150
RI PT
145
The only factors which correlated with failure of ODPE were Staph aureus infection and presenting HCT <=32.1 % (Table 1).
153
Impact of Organism
Twenty-nine of the seventy-two patients (29/72 (40.3%)) presented with an isolated
M AN U
152
SC
151
Staph aureus infection or a multi-organismal infection including at least one positive culture for
155
Staph aureus. Fourteen of these patients (14/29 (48.3%)) failed ODPE. Forty-three patients
156
(43/72 (59.7%)) presented with a non-Staph aureus infection and 5 failed ODPE (15/43
157
(11.6%)). This difference was highly significant with a p value of p=0.0012 and an odds ratio of
158
7.1 (95% confidence interval of (2.3-25.3)). Implant retention with respect to time is shown in
159
Figure 1B. The average time to failure was 16.0 months for the Staph aureus group and 3.8
160
months for the non-Staph aureus group (p=0.0968) (Table 2). Twelve patients had culture
161
negative infections and all underwent successful ODPE; however, the incidence of culture
162
negative infection was not high enough to generate statistically significant differences (Table 1).
163
There were no significant differences for other observed bacteria or for polymicrobial vs
164
monomicrobial infections (Table 1).
166
EP
AC C
165
TE D
154
Impact of Presenting HCT
9
ACCEPTED MANUSCRIPT
Patients who failed ODPE had a lower average preoperative hematocrit (31.39 vs 34.40
168
p=0.0104) (Figure 2 & Table 1). Twenty patients who underwent ODPE had a presenting HCT
169
<=32.1% and 52 patients had a presenting HCT >32.1%. Fifty-five percent of patients in the low
170
HCT group failed ODPE versus 15.4%of patients in the high HCT group (p=0.0013). Implant
171
retention with respect to time is shown in Figure 1C. There was no difference in the time to
172
failure between the groups (Table 2).
There was also a disproportionate incidence of Staph aureus infection in the low
SC
173
RI PT
167
hematocrit group. Sixty percent of patients with HCT <=32.1 were infected with Staph aureus
175
while seventeen of the fifty-two patients (17/52 (32.7%)) with HCT >32.1 were infected with
176
Staph aureus (p=0.0646). Eleven of the 12 patients with culture negative infections had a
177
presenting HCT > =32.1.
178
M AN U
174
During this study, we had an overall transfusion rate of 33.3% for packed red blood cells and 40.2% for blood products of any kind (platelets, fresh frozen plasma, or packed red blood
180
cells). There were no statistical differences between the groups for the number of patients
181
receiving transfusions of packed red blood cells (8/19 (42.1%) vs 16/53(30.2%) p=0.5081) or the
182
number receiving blood products of any kind (9/19 (47.4%) vs 20/53 (37.7%), p=0.6442)
183
(Table1). Thirty-one percent (9/19) of patients who received blood products of any kind failed
184
ODPE while 10/43 (23.3%) who did not receive blood failed ODPE (p=0.6442).
186 187
EP
AC C
185
TE D
179
Combined Effect of Organism and Presenting HCT Linear regression modeling demonstrated that both low HCT and Staph aureus were
188
independent risk factors for failed ODPE (p=0.0082 and p=0.0057). The different combinations
189
of these risk factors were then analyzed (Table 2). Strikingly, of the thirty-five patients without
10
ACCEPTED MANUSCRIPT
Staph aureus infection and with a HCT >32.1, only one (2.9%) failed ODPE (Figure 1D, Table
191
2). The remaining groups were much less successful and the reported odds ratios for failure are
192
with respect to these thirty-five patients. Twelve patients presented with a Staph aureus infection
193
and a low HCT with 58.3% failing ODPE and an odds ratio for failure of 47.6 (95% confidence
194
interval (6.7-997.2)) when compared to the group without Staph aureus and with a HCT >32.1.
195
Seventeen patients were infected with Staph aureus and presented with a HCT >32.1 with 41.1%
196
failing ODPE and an odds ratio for failure of 34.0 (3.9-765.4). Eight patients presented with low
197
HCT without Staph aureus infection. 50.0% of these patients failed ODPE yielding an odds ratio
198
of failing of 23.8 (3.6-473.7).
200
Discussion:
M AN U
199
SC
RI PT
190
Peri-prosthetic infection after hip and knee arthroplasty remains a very serious
202
complication with high morbidity, mortality and costs, and is one of the most common reasons
203
for revision surgeries[25]. Two stage revision has been the most consistently successful
204
approach to eradicating these infections but the morbidity, cost and rate of poor functional
205
outcomes are high [10-16]. As a result, many surgeons have turned to ODPE for selected
206
patients with acute peri-prosthetic infections due to the lower morbidity and improved early
207
return of function. In general, the indications for ODPE have included acute infections (<3-4
208
weeks) in a patient with well-fixed components, an infecting organism responsive to oral
209
antibiotics, and absence of a draining sinus. Reported outcomes of this technique have been
210
varied ranging from 16 % to 83 % infection-free survival[17, 26, 27]. Recently there has been
211
some concern that failed ODPE may be associated with a higher rate of subsequent 2-stage
212
revision failure[18, 19].
AC C
EP
TE D
201
11
ACCEPTED MANUSCRIPT
213
If ODPE fails, a two-stage procedure would typically be the salvage option. Sherrell et al suggest that 2-stage revisions in the absence of prior ODPE have a near 90% success rate
215
while ODPE has close to a 40% success rate leading some authors to discourage ODPE[18, 19].
216
However, more recent studies have suggested that 2-stage revision outcomes may not be as
217
successful as once thought. Gomez et al reported that about 20% of patients who have had their
218
joints explanted never progress to the second stage[15]. These patients are often excluded from
219
the success rates for 2-stage revisions as rates are usually reported as the number of successfully
220
reconstructed 2-stage revisions. Berend et al. report a higher reimplantation rate in their series of
221
hip infections at 92%[11]. They report a success rate of 157/189 (83%) for reimplanted joints.
222
When all patients who were explanted are included and those lost to follow up are excluded, the
223
success rate drops to 144/189 (76%). This result for 2-stage revision is strikingly close to our
224
success rate of 75.3% with ODPE. While the important distinction of chronic infection vs acute
225
infection cannot be ignored, when failure to reimplant due to comorbidities or death are included
226
as failures of 2-stage revision as they would be in ODPE, the resulting success rates of ODPE
227
and 2-stage revision are much closer than previously reported. Our data would suggest that
228
failed ODPE may simply select for more difficult to treat infections and/or a less healthy host.
229
The ability to identify these at risk patients may allow surgeons to escalate care to improve
230
outcomes.
SC
M AN U
TE D
EP
AC C
231
RI PT
214
In this study we have identified two markers for poor outcomes with ODPE in acute peri-
232
prosthetic infection. The presence of a Staph aureus infection or a pre-operative HCT <=32.1
233
predict a success rate of about 56% with the success rate falling to 42% when both factors are
234
combined. When neither factor is present (about half of our patients) the success rate was 97%.
235
The ability to identify patients who will do well with ODPE allows surgeons to more confidently
12
ACCEPTED MANUSCRIPT
236
recommend ODPE. For this group, ODPE appears to be safe and effective sparing them from
237
the challenges of a 2-stage revision.
238
This study is not without limitations. This work is a retrospective evaluation of a relatively small number of patients. This challenge of small sample size is fairly consistent
240
within the literature for ODPE candidates. This is largely due to the low rates of infection
241
overall but is compounded by lower rates of infections at high volume joints centers[28, 29] such
242
as ours. Further, while all infections are acute or acute hematogenous at one point, the majority
243
of patients do not present from outside surgeons until they have developed a chronic infection.
244
Strategies to increase numbers include including expanding encatchment time or the number of
245
treating surgeons. Increasing the encatchment time can introduce temporal biases highlighting
246
the need for data sharing between large centers either in the form of collaborative studies, meta-
247
analysis, or database creation. Despite the challenge of small sample size in this work, our results
248
were highly statistically significant with p-values of 0.0012 and 0.0013. Multiple surgeons and
249
infectious disease specialists participated in the management of these patients. Preoperative
250
ESR, CRP, CBC and cultures were the most consistently ordered tests. Information concerning
251
nutritional status at the time of surgery was not available. Culture negative infections that were
252
treated with ODPE were included in this study; however, the data were recalculated excluding
253
culture negative infections and the statistical significance of the anemia and bacterial species
254
remained. Patients who had retained their components but had a poorly functioning knee would
255
be considered a success in this study.
257
SC
M AN U
TE D
EP
AC C
256
RI PT
239
Conclusion:
13
ACCEPTED MANUSCRIPT
258
ODPE is an attractive treatment modality for management of acute and acute hematogenous periprosthetic infections in selected circumstances. This study once again
260
confirms that the infecting organism is an important determinant of outcome. Staphylococcus
261
aureus infection greatly decreases the success of ODPE while non-staph aureus infection was
262
much more successful. Culture negative infections had a 100% success rate. We further present
263
the finding that a hematocrit at or below 32.1 is also strongly associated with ODPE failure.
264
Patients with a hematocrit above 32.1 and who were not infected with Staph aureus were
265
associated with a 97% successful ODPE treatment. For this subset of patients, ODPE is safe and
266
effective. For patients that failed ODPE in this study, we did not observe high rates of failure of
267
subsequent 2-stage revision. Further work should be done to identify additional markers for
268
success and further examine the applicability of these markers to larger patient populations.
AC C
EP
TE D
M AN U
SC
RI PT
259
14
ACCEPTED MANUSCRIPT
269
References:
270
1.
Kurtz, S., Projections of Primary and Revision Hip and Knee Arthroplasty in the United States from 2005 to 2030. The Journal of Bone and Joint Surgery (American), 2007.
272
89(4): p. 780. 2.
The Journal of Arthroplasty, 2008. 23(7): p. 984-991.
274 275
Kurtz, S.M., et al., Infection Burden for Hip and Knee Arthroplasty in the United States.
3.
Wilson, M.G., K. Kelley, and T.S. Thornhill, Infection as a complication of total knee-
SC
273
RI PT
271
replacement arthroplasty. Risk factors and treatment in sixty-seven cases. The Journal of
277
Bone & Joint Surgery, 1990. 72(6): p. 878-883.
278
4.
M AN U
276
Azzam, K.A., et al., Irrigation and Debridement in the Management of Prosthetic Joint
279
Infection: Traditional Indications Revisited. The Journal of Arthroplasty, 2010. 25(7): p.
280
1022-1027. 5.
Byren, I., et al., One hundred and twelve infected arthroplasties treated with 'DAIR'
TE D
281 282
(debridement, antibiotics and implant retention): antibiotic duration and outcome.
283
Journal of Antimicrobial Chemotherapy, 2009. 63(6): p. 1264-1271. 6.
Cordero-Ampuero, J. and M. de Dios, What Are the Risk Factors for Infection in
EP
284
Hemiarthroplasties and Total Hip Arthroplasties? Clinical Orthopaedics and Related
286
Research®, 2010. 468(12): p. 3268-3277.
287
7.
Arthroplasty, 2012. 27(6): p. 857-64.e1-4.
289
291
Buller, L.T., et al., The preoperative prediction of success following irrigation and debridement with polyethylene exchange for hip and knee prosthetic joint infections. J
288
290
AC C
285
8.
Parvizi, J., B. Zmistowski, and B. Adeli, Periprosthetic Joint Infection: Treatment Options. Orthopedics, 2010.
15
ACCEPTED MANUSCRIPT
292
9.
Osmon, D.R., et al., Diagnosis and Management of Prosthetic Joint Infection: Clinical
293
Practice Guidelines by the Infectious Diseases Society of America. Clinical Infectious
294
Diseases, 2012. 56(1): p. e1-e25. 10.
Oussedik, S.I.S., M.B. Dodd, and F.S. Haddad, Outcomes of revision total hip
RI PT
295 296
replacement for infection after grading according to a standard protocol. Journal of
297
Bone and Joint Surgery - British Volume, 2010. 92-B(9): p. 1222-1226. 11.
Berend, K.R., et al., Two-stage Treatment of Hip Periprosthetic Joint Infection Is
SC
298
Associated With a High Rate of Infection Control but High Mortality. Clinical
300
Orthopaedics and Related Research®, 2012. 471(2): p. 510-518.
301
12.
M AN U
299
Zmistowski, B., et al., Periprosthetic Joint Infection Increases the Risk of One-Year
302
Mortality. The Journal of Bone and Joint Surgery-American Volume, 2013. 95(24): p.
303
2177-2184.
18(9): p. 871-873.
305
14.
Br, 2012. 94(5): p. 619-623.
307 308
15.
2012. 50(11): p. 3501-8.
310
312 313
Gomez, E., et al., Prosthetic joint infection diagnosis using broad-range PCR of biofilms dislodged from knee and hip arthroplasty surfaces using sonication. J Clin Microbiol,
309
311
Vanhegan, I., et al., A financial analysis of revision hip arthroplasty. J Bone Joint Surg
EP
306
Sculco, T.P., The economic impact of infected joint arthroplasty. Orthopedics, 1995.
TE D
13.
AC C
304
16.
Kheir, M.M., et al., Patients With Failed Prior Two-Stage Exchange Have Poor Outcomes After Further Surgical Intervention. The Journal of Arthroplasty, 2017. 32(4): p. 1262-1265.
16
ACCEPTED MANUSCRIPT
314
17.
Review of current concepts. World Journal of Orthopedics, 2014. 5(5): p. 667.
315 316
Kuiper, J.W.P., Treatment of acute periprosthetic infections with prosthesis retention:
18.
Gardner, J., T.J. Gioe, and P. Tatman, Can This Prosthesis Be Saved?: Implant Salvage Attempts in Infected Primary TKA. Clinical Orthopaedics and Related Research®, 2010.
318
469(4): p. 970-976.
319
19.
RI PT
317
Sherrell, J.C., et al., The Chitranjan Ranawat Award: Fate of Two-stage Reimplantation After Failed Irrigation and Débridement for Periprosthetic Knee Infection. Clinical
321
Orthopaedics and Related Research®, 2010. 469(1): p. 18-25. 20.
Vilchez, F., et al., Outcome and predictors of treatment failure in early post-surgical
M AN U
322
SC
320
323
prosthetic joint infections due to Staphylococcus aureus treated with debridement. Clin
324
Microbiol Infect, 2011. 17(3): p. 439-44. 21.
of Arthroplasty, 2011. 26(6): p. 114-118.
326 327
Odum, S.M., et al., Irrigation and Debridement for Periprosthetic Infections. The Journal
22.
TE D
325
Deirmengian, C., et al., Limited success with open debridement and retention of components in the treatment of acute Staphylococcus aureus infections after total knee
329
arthroplasty. The Journal of arthroplasty, 2003. 18: p. 22-26. 23.
Mortality Following Total Joint Arthroplasty. The Journal of Arthroplasty, 2015. 30(5):
331
p. 846-848.
332 333 334
Viola, J., et al., Preoperative Anemia Increases Postoperative Complications and
AC C
330
EP
328
24.
Team, R., RStudio: integrated development for R. RStudio, Inc., Boston, MA URL http://www. rstudio. com, 2015.
17
ACCEPTED MANUSCRIPT
335
25.
Koh, C.K., et al., Periprosthetic Joint Infection Is the Main Cause of Failure for Modern
336
Knee Arthroplasty: An Analysis of 11,134 Knees. Clinical Orthopaedics and Related
337
Research®, 2017. 475(9): p. 2194-2201. 26.
infected total knee arthroplasty. The Journal of arthroplasty, 1997. 12(4): p. 426-433.
339 340
Mont, M.A., et al., Multiple irrigation, debridement, and retention of components in
RI PT
338
27.
Bradbury, T., et al., The fate of acute methicillin-resistant Staphylococcus aureus
periprosthetic knee infections treated by open debridement and retention of components.
342
The Journal of arthroplasty, 2009. 24(6): p. 101-104. 28.
outcomes of total knee replacement. JBJS, 2004. 86(9): p. 1909-1916.
344 345
Katz, J.N., et al., Association between hospital and surgeon procedure volume and the
M AN U
343
SC
341
29.
Katz, J.N., et al., Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population. Jbjs, 2001.
347
83(11): p. 1622-1629.
EP AC C
348
TE D
346
18
ACCEPTED MANUSCRIPT
Tables and Figures:
2
Table 1: Comparison of demographics, lab data, and vital signs between successful and failure
3
groups. Only differences in blood counts and infecting organism were statistically significant.
4
Figure 1: Kaplan Meier Curves for implant survivability for all patients (A), with respect to
5
Staph aureus infection (B), with respect to preoperative anemia (C), and with respect to
6
combined anemia and Staph aureus infection (D).
7
Figure 2: Histogram illustrating the difference in distribution of preoperative blood counts
8
between successful and failure groups.
9
Table 2: Chi Square analysis for ODPE failure with respect to Staph aureus infection alone or
M AN U
SC
RI PT
1
10
with respect to low HCT alone and Linear Regression analysis for combinations of low HCT and
11
Staph aureus infection combined risk factors. Odds ratios are with respect to a patient without
12
the examined risk factor(s).
AC C
EP
TE D
13
ACCEPTED MANUSCRIPT
Hematocrit (%) HCT > 32.1% Hemoglobin WBC Platelets Glucose CRP ESR
Failure 88085.07 87.50
Preop Labs Success 34.40 44 (83.0%) 11.27 10.13 293.15 124.77 11.83 64.19
Failure 31.39 8 (42.1%) 10.43 11.78 284.89 124.05 15.25 58.09
Postop Labs Success 29.67 9.55 9.59 284.26 142.83 Organism Success 15 (28.3%) 12 (22.6%) 3 (5.7%) 2 (3.8%) 6 (11.3%) 16 (30.2%) 8 (15.1%) 3 (5.7%) 12 (22.6%) 29 (54.7%) 12 (22.6%)
SC
Staph aureus MSSA MRSA MSSE MRSE Strep Corynebacterium Other Culture Negative Monomicrobial Polymicrobial
1.0000 0.6160 0.2466
p-value 0.6963 0.5177 0.5863 0.6324
Failure 25.34 8.29 14.40 299.43 142.71
RI PT
Hematocrit Hemoglobin WBC Platelets glucose
Complications Success 11 (20.8%) 2 (3.8%) 3 (5.7%) 1 (1.9%)
p-value 0.0014 0.0034 0.8494 1 1 0.1647 0.9145 1.0000 0.0557 0.1131 1.0000
Failure 4 (2.1%) 1 (5.3%) 1 (5.3%) 1 (5.3%)
p-value 1.0000 1.0000 1.0000 1.0000
p-value 0.3217 0.9827
p-value 0.0104 0.0018 0.0239 0.2647 0.8514 0.9464 0.2486 0.5763
Hospital Course Success Failure HGB Drop (g/dL) 1.62 1.07 HCT Drop (% points) 4.48 2.34 Length of Stay (days) 7.32 14.32 Surgical Duration (min) 104.77 134.37 # Debridements 1.26 1.21 Discharged Home 38 (71.7%) 10 (52.6%) pRBC 16 (30.2%) 8 (42.1%) Any Blood Products 20 (37.7%) 9 (47.4%)
1
p-value 0.0031 0.0004 0.1860 0.8528 0.9958
Failure 14 (73.7%) 12 (63.2%) 2 (10.5%) 1 (5.3%) 2 (10.5%) 2 (10.5%) 2 (10.5%) 1 (5.3%) 0 15 (78.9%) 4 (21.1%)
30 Day Readmission Line Infection C diff Infection DVT
TE D
Joint Aspiration Success 54231.53 87.61
AC C
Cell Count %PMN
Failure 0.00 2 (10.5%) 5 (26.3%) 2 (10.5%)
EP
Temp >38C or <36C WBC >12k or <4k Tachycardia Tachypnea
Preop Vitals Success 3 (5.7%) 11 (20.8%) 9 (17.0%) 10 (18.9%)
p-value 0.0660 1.0000 0.8979 0.4935 0.8452 0.5557 0.6869 0.5166 0.5166
M AN U
Demographics Success Failure Age 67.11 60.90 Gender(Female) 27 (50.9%) 10 (52.6%) BMI 35.50 35.14 ASA 2.75 2.84 ASA >=3 38 (71.7%) 15 (78.9%) Tobacco Use 4 (7.5%) 3 (15.8%) DM 15 (28.3%) 7 (36.8%) Immunosupressed 4 (7.5%) 0 RA 4 (7.5%) 0 ESRD 0 0 HIV 0 0 Replant 10 (18.8%) 4 (21.1%) Acute Peri-Operative 13 (24.5%) 3 (15.8%) Time From Implantation 3.15 2.13
p-value 0.0456 0.0862 0.0926 0.2531 0.6653 0.2191 0.5081 0.6442
ACCEPTED MANUSCRIPT
Failure Based on Staph aureus Odds Ratio
95% CI
14/29 (48.3%) 5/43 (11.6%)
0.0012 --
7.1
(2.3-25.3)
--
--
Failure Based on HCT HCT<=32.1 HCT>32.1
Failure
p-value
Odds Ratio
95% CI
11/20 (55.0%) 8/52 (15.4%)
0.0013 --
6.7
(2.2-22.4)
--
--
Failure Based on Risk Factors p-value
7/12 (58.2%) 4/8 (50.0%) 7/17 (41.1%) 1/35 (2.9%)
0.0010 0.0050 0.0044 --
Odds Ratio
95% CI
47.6 34.0 23.8 --
(6.7-997.2) (3.6-473.7) (3.9-765.4) --
M AN U
Failure
AC C
EP
TE D
HCT <= 32.1 and Staph aureus HCT <= 32.1 and not Staph aureus HCT > 32.1 and Staph aureus HCT > 32.1 and not Staph aureus
RI PT
p-value
SC
Staph aureus No-Staph aureus
Failure
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT