Journal Pre-proof

The Effect of Preoperative Anemia on Complications Following Total Hip Arthroplasty

Matthew J. Grosso, MD, Venkat Boddapati, MD, H. John Cooper, MD, Jeffrey A.
Geller, MD, Roshan P. Shah, MD, JD, Alexander L. Neuwirth, MD

PII: S0883-5403(20)30030-9
DOI: https://doi.org/10.1016/j.arth.2020.01.012
Reference: YARTH 57720

To appear in: The Journal of Arthroplasty

Received Date: 22 November 2019

Revised Date: 26 December 2019
Accepted Date: 8 January 2020

Please cite this article as: Grosso MJ, Boddapati V, Cooper HJ, Geller JA, Shah RP, Neuwirth AL,
The Effect of Preoperative Anemia on Complications Following Total Hip Arthroplasty, The Journal of
Arthroplasty (2020), doi: https://doi.org/10.1016/j.arth.2020.01.012.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.

© 2020 Elsevier Inc. All rights reserved.

TITLE:
The Effect of Preoperative Anemia on Complications Following Total Hip Arthroplasty

AUTHORS:
Matthew J. Grosso, MD1; Venkat Boddapati, MD1; H. John Cooper, MD1; Jeffrey A. Geller,
MD1; Roshan P. Shah, MD, JD1; Alexander L. Neuwirth, MD1
1
Center for Hip and Knee Replacement
Department of Orthopedic Surgery
Columbia University Irving Medical Center
622 West 168th Street PH 1138
New York, NY 10032

CORRESPONDING AUTHOR:
Alexander L. Neuwirth, MD
Center for Hip and Knee Replacement
Department of Orthopedic Surgery
Columbia University Irving Medical Center
622 West 168th Street PH 1138
New York, NY 10032
Phone: 212-305-5974
Fax: 212-305-4024
Email: aln2137@cumc.columbia.edu
 1 The Effect of Preoperative Anemia on Complications Following Total Hip Arthroplasty
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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48 Abstract
49 Introduction:
50 Current literature suggests that preoperative hematocrit levels may play an important role in
51 determining risk for complications following THA. The purpose of this study was to determine
52 the role of preoperative anemia status on 30-day complications following total hip arthroplasty.
53
54 Methods:
55 Using the National Surgical Quality Improvement Program registry from 2006-2016, we
56 identified all patients who underwent primary THA. Patients were placed into 3 cohorts based
57 on preoperative hematocrit levels (Normal >36% [N=166,538], Mild Anemia 27%-36%
58 [N=13,214], Severe Anemia <27% [N=541]). Differences in 30-day postoperative medical
59 complications and readmission rates were compared using bivariate and multivariate analyses.
60
61 Results:
62 Multivariate logistic regression analysis identified mild anemia compared to normal hematocrit
63 as a significant risk factor for total complications (OR 1.46, p<0.001), mortality (OR 2.06,
64 p<0.001), renal complications (OR 2.59, p<0.001), respiratory complications (OR 1.89,
65 p<0.001), sepsis (OR 2.01, p<0.001), wound infection (OR 1.36, p<0.001), and urinary tract
66 infection (OR 1.44, p<0.001). Severe anemia was also risk factor, with a higher odds ratio, for
67 total complications (OR 1.99, p<0.001). Both mild and severe anemia were significant risk
68 factors for increased rates of perioperative blood transfusion (mild: OR 4.04, severe: OR 5.58),
69 non-home discharge (OR 1.74, OR 1.64), and unplanned hospital readmissions (OR 1.42, OR
70 1.66).
71
72 Conclusions:
73 Preoperative anemia is a significant risk for perioperative complications following primary THA.
74 Even mild anemia can lead to significantly increased risks of mortality, medical complications,
75 and unplanned hospital readmissions in THA. This study further supports the need for screening
76 and preoperative intervention for patients in this at risk group.
77
78 Keywords:
79 Hip arthroplasty; anemia; complications
80
81
82
83
84
85
86
87
88
89
90
91
92
 93
94 Introduction:
95
96 Complications following primary total hip arthroplasty (THA) lead to increased patient
97 and economic burdens. In addition, with over 600,000 THA procedures performed annually in
98 the United States, these complications can have significant impacts on the healthcare system [1].
99 Surgeons can help reduce rates of complications by identifying common preoperative risk factors
100 associated with complications following THA.
101 Anemia is a common condition that increases in prevalence with advancing age, affecting
102 up to 1/3 of patients who undergo total hip arthroplasty [2–4]. Numerous studies have associated
103 anemia with increased risks of complications in surgical populations. For example, a number of
104 studies have identified preoperative anemia, defined as hematocrit <36%, as a significant
105 predictor of increased morbidity and mortality in patients who undergo surgery following hip
106 fracture [5–7]. In addition, a number of smaller or institutional based studies have examined the
107 relationship between pre-operative anemia and post-operative complications in total joint
108 arthroplasty, showing increased risks of infection, length of stay, blood transfusion requirements,
109 and mortality [8–10]. Potential explanations for inferior outcomes associated with preoperative
110 anemia in arthroplasty, include its association with chronic conditions, such as renal disease and
111 nutritional deficiency, as well as a predictor for increased rates of allogenic blood transfusions,
112 an independent risk factor for complications, increased length of stay, and increased cost of care
113 [11-12].
114 To our knowledge, no registry based studies have the examined the effects of pre-
115 operative anemia on adverse outcomes following primary total hip arthroplasty. Therefore, the
116 purpose of this study was to use an updated, large, nationwide database to determine the
117 association between pre-operative anemia and post-operative 30-day morbidity and mortality
118 following primary THA. We hypothesize that pre-operative anemia will be associated with
119 increased 30-day complications following primary THA.
120
121 Methods:
122 This retrospective cohort study examined the American College of Surgeons National
123 Surgical Quality Improvement Program (ACS-NSQIP) data from 2006-2016. NSQIP is a
124 prospectively collected registry which includes patient data collected by trained clinical
125 reviewers from >600 institutions both in the community and academic settings across the United
126 States. The NSQIP database includes data from more than 800,000 unique surgical encounters.
127 Rigorous collection of data and auditing of the data by the ACS have led to high inter-relater
128 reliability.[13] Data includes preoperative demographical information and risk factors,
129 perioperative events and complications, and postoperative events up to 30 days after surgery,
130 regardless of discharge date. The database does not include orthopedic specific complications,
131 such as dislocation, or periprosthetic fractures.
132 Using Current Procedural Terminology code 27130 for primary THA, 3 patient cohorts
133 were identified in this study based on preoperative hematocrit status: (1) Normal >36%, (2) Mild
134 Anemia 27%-36%, (3) Severe Anemia <27%. These groups were defined base on guidelines
135 from prior studies [14]. Patients missing preoperative laboratory values were excluded from the
136 study (5,671 patients), leaving a total of 130,293 patients. Baseline patient and operative
137 characteristics were assessed for all patients, which included age, gender, body mass index,
138 comorbidities, anesthesia type, dependent functional status, and American Society of
139 Anesthesiologists class.
140 Outcome variables in this study were 30-day complications, perioperative blood
141 transfusion, non-home discharge, and unplanned readmissions. Complications analyzed included
142 mortality, cardiac complications, renal complications, respiratory complications, deep vein
143 thrombosis, stroke/cerebrovascular accident, sepsis, wound infection, wound dehiscence, and
144 urinary tract infection.
145 Baseline patient characteristics were compared using Pearson’s chi-squared analysis.
146 Complications were compared by bivariate and multivariate analyses that controlled for baseline
147 patients’ characteristics, including age, gender, body mass index, medical comorbidities,
148 anesthesia type, functional status, and American Society of Anesthesiologists class. Multivariate
149 analysis was chosen over propensity score matching due to the large number of events per
150 confounder as a result of the large sample size.[15] Using a baseline statistical significance
151 defined as p<0.05, Bonferroni correction was applied to correct for multiple group comparisons.
152
153
154
155 Results:
156 Patient Demographics:
157 A total of 130,293 patients were identified who underwent primary THA between 2006
158 and 2016 in the NSQIP registry. Of this total, 13,214 patients (10%) were in the mild anemia
159 cohort, and 541 patients (0.4%) were in the severe anemia cohort (Table 1). There was a higher
160 proportion of older patients (>80 years old) in the mild and severe anemia cohorts. In addition,
161 patients with anemia were more likely to be female, have a lower BMI, and have increased rates
162 of comorbidities of diabetes mellitus, COPD, preoperative corticosteroid use, and hypertension
163 (Table 1), compared to normal hematocrit level patients.
164
165 Bivariate Analysis of Adverse Outcomes
166 Bivariate analysis of the three cohorts showed a statistically significant increased rate of
167 any complication following THA from normal (3.17%), to mild (5.94%), to severe (7.76%)
168 anemia (p<0.001) (Table 2). In addition there was an increasing rate of complications from
169 normal to mild to severe, for mortality, cardiac complications, renal complications, respiratory
170 complications, and wound infection (Table 2). In addition, sepsis and urinary tract infection
171 were increased in the mild and severe group compared to normal, although the mild group had
172 higher rates compared to the severe cohort.
173 There was a large and statistically significant increase in perioperative blood transfusion
174 rate between the normal (8.49%), mild (33.3%), and severe (40.3%) groups (p<0.001) (Table 2).
175 Both mild (41.1%) and severe (36.8%) anemia had higher rates of non-home discharge compared
176 to the normal hematocrit cohort (20.4%, p<0.001). In addition, unplanned hospital readmissions
177 were significantly lower for the normal hematocrit group (3.31% p<0.001) compared to the mild
178 (5.88%) and severe (6.88%) cohorts (Table 2).
179
180 Multivariate Analysis of Adverse Outcomes
181 Multivariate logistic regression analysis identified mild anemia compared to normal
182 hematocrit as a significant risk factor for total complications (OR 1.46, p<0.001), mortality (OR
183 2.06, p<0.001), renal complications (OR 2.59, p<0.001), respiratory complications (OR 1.89,
184 p<0.001), sepsis (OR 2.01, p<0.001), wound infection (OR 1.36, p<0.001), and urinary tract
185 infection (OR 1.44, p<0.001) (Table 3). Mild anemia was also a significant risk factor for
186 increased transfusion rate (OR 4.04, p<0.001), non-home discharge (OR 1.74, p<0.001), and
187 unplanned hospital readmission (OR 1.42, p<0.001).
188 When compared severe anemia to normal hematocrit, severe anemia had higher OR
189 (greater risk) for every complication identified in the mild cohort (Table 3) However, likely
190 secondary to the smaller cohort size and Bonferroni correction, only total complications was
191 statistically significant (OR 1.99, p<0.001). In addition, there were statistically significant
192 increased rates of blood transfusion (OR 5.58, p<0.001), non-home discharge (OR 1.64,
193 p<0.001), and unplanned hospital readmission (OR 1.66, p<0.001) in the severe versus normal
194 hematocrit cohorts.
195
196 Discussion:
197
198 Anemia prior to primary total hip arthroplasty is an identifiable risk factor that can lead to
199 increased adverse events. The extent of these effects on perioperative complications and
200 unplanned readmissions following primary THA is unclear. In this study, by examining a large,
201 nationwide database, we concluded that preoperative anemia is a significant risk factor for
202 medical complications. Even mild anemia, with hematocrit between 27%-36%, leads to
203 significantly increased risks of mortality, medical complications, and unplanned hospital
204 readmissions in THA.
205 A number of previous have examined the effects of anemia on post-operative morbidity
206 following orthopedic surgery. For example, a number of studies have looked at the effects of
207 anemia on patients with hip fractures. Gruson et al. associated anemia with longer length of
208 hospital stay and higher rates of 6 month mortality following surgery for hip fracture [7].
209 Similarly Halm et al., reported longer length of stay and higher odds of death and readmission
210 within 60 days of discharge for patients with hip fractures and preoperative anemia [6]. A few
211 studies have also looked at the effect anemia on arthroplasty patients. In a large institutional
212 based study, Viola et al. examined the effect of preoperative anemia on postoperative
213 complications and mortality on a cohort of total hip and total knee arthroplasty patients.[8] They
214 reported a higher rate of complications (OR 2.11) for the anemia cohort (defined Hgb <12 g/dL
215 for females, <13 g/dL for males), with cardiovascular and genitourinary complications being the
216 most significant contributors. Unlike in our study, Viola et al. did not find a statistically
217 significant difference in mortality between the two cohorts. Similar to our study, Lu et al.,
218 utilized the NSQIP database, but they chose to examine the effect of anemia on revision total
219 joint arthroplasty [10]. They reported similar findings to our study looking at primary THA,
220 with increased post-operative complications (OR 1.45), deep infection (OR 1.68), and mortality
221 (OR 2.18).
222 Our study corroborated previous findings of increased adverse events following primary
223 total hip arthroplasty in this population. As expected, there were significant differences in
224 preoperative characteristics between the anemia and non-anemia cohorts. Patients with mild and
225 severe anemia, were older, and had increased rates of comorbidities, such as DM, COPD, and
226 hypertension. In addition, these patients were more likely to receive general versus regional
227 anesthesia, and have a higher ASA class. Multivariate analysis was utilized to demonstrate that
228 even when accounting for these differences in preoperative characteristics, anemia patients were
229 still at risk for 30-day complications and unexpected readmissions. The statistically significant
230 variables in the univariate analysis were treated as potential confounders and analyzed in the
231 multivariate analysis. Perhaps, most clinically relevant, were the significant differences seen
232 comparing the mild anemia cohort to the normal hematocrit cohort. Mild anemia was present in
233 ~10% of the total primary THA patients in this study, and has been reported up to 30% in other
234 studies. Therefore, these equate to significant effects on the over 600,000 primary THAs
235 performed annually. Mild anemia patients were 50% more likely to have a 30-day medical
236 complication, and twice as likely to have a complication of mortality. In addition, they were 4x
237 as likely to receive a blood transfusion, and had a 40% higher risk of unplanned hospital
238 readmission.
239 Preoperative anemia presents in a variety of scenarios in patients with varying medical
240 comorbidities. This study demonstrates higher rates of complications in the anemia cohorts,
241 which suggests that it may be beneficial to improve hematocrit levels prior to surgery, when
242 possible. Presentations, such as iron-deficiency and vitamin deficiency anemia, can often be
243 addressed with preoperative supplementation strategies. Bisbe et al. demonstrate a potential
244 algorithmic approach to correcting preoperative anemia prior to major orthopedic surgery [16].
245 In other disease states, such as sickle cell anemia, hematocrit levels are more difficult to correct.
246 While the preoperative anemia complication rates are high compared to a normal hematocrit
247 group, it is important to note that the overall complication rate is still acceptable in these patients.
248 In our analysis, 94% of patients with mild anemia will have no medical complications. If the pre-
249 operative anemia level is unable to be corrected, we believe total hip arthroplasty should still be
250 considered in these patients.
251 This national registry-based study has inherent limitations. First, ACS-NSQIP
252 complications are limited to medical complications 30 days postoperatively, so we do not have
253 data on subacute or delayed complications, or orthopedic-specific perioperative complications.
254 Second, since this patient population was obtained from 2006-2016, there was significant change
255 in blood-loss related protocols including the popularization of tranexamic acid, increased use of
256 neuraxial anesthesia, and modified transfusion and perioperative protocols. These changes may
257 have a significant effect on perioperative complications in anemia patients which we are not able
258 to identify in this study. In addition, as discussed, there are significant differences in pre-
259 operative characteristics between the cohorts. The multivariate analysis attempts to account for
260 these differences, however, there is always a risk that the analysis missed important confounding
261 factors not included in the model. Lastly, there was a small portion of primary THA patients in
262 the registry that did not have preoperative laboratory values available (4%). Excluding a portion
263 of the population can lead to selection bias, although we felt this small percentage was
264 acceptable for this study.
265 Preoperative anemia prior to primary total hip arthroplasty is a potentially modifiable risk
266 factor. In this national registry based study, we identified both mild and severe anemia as
267 significant risk factors for 30-day perioperative medical complications and unplanned hospital
268 readmissions. This study further supports the need for screening and preoperative intervention
269 for patients in this at risk group.
270
271
272
273
274 References
275
276 [1] Sloan M, Premkumar A, Sheth NP. Projected Volume of Primary Total Joint Arthroplasty
277 in the U.S., 2014 to 2030. J Bone Joint Surg Am 2018;100:1455–60.
278 https://doi.org/10.2106/JBJS.17.01617.
279 [2] Bierbaum BE, Callaghan JJ, Galante JO, Rubash HE, Tooms RE, Welch RB. An analysis of
280 blood management in patients having a total hip or knee arthroplasty. J Bone Joint Surg Am
281 1999;81:2–10.
282 [3] Spahn DR. Anemia and patient blood management in hip and knee surgery: a systematic
283 review of the literature. Anesthesiology 2010;113:482–95.
284 https://doi.org/10.1097/ALN.0b013e3181e08e97.
285 [4] Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Woodman RC. Prevalence of anemia
286 in persons 65 years and older in the United States: evidence for a high rate of unexplained
287 anemia. Blood 2004;104:2263–8. https://doi.org/10.1182/blood-2004-05-1812.
288 [5] Hagino T, Ochiai S, Sato E, Maekawa S, Wako M, Haro H. The relationship between
289 anemia at admission and outcome in patients older than 60 years with hip fracture. J Orthop
290 Traumatol 2009;10:119–22. https://doi.org/10.1007/s10195-009-0060-8.
291 [6] Halm EA, Wang JJ, Boockvar K, Penrod J, Silberzweig SB, Magaziner J, et al. The effect
292 of perioperative anemia on clinical and functional outcomes in patients with hip fracture. J
293 Orthop Trauma 2004;18:369–74.
294 [7] Gruson KI, Aharonoff GB, Egol KA, Zuckerman JD, Koval KJ. The relationship between
295 admission hemoglobin level and outcome after hip fracture. J Orthop Trauma 2002;16:39–
296 44.
297 [8] Viola J, Gomez MM, Restrepo C, Maltenfort MG, Parvizi J. Preoperative anemia increases
298 postoperative complications and mortality following total joint arthroplasty. J Arthroplasty
299 2015;30:846–8. https://doi.org/10.1016/j.arth.2014.12.026.
300 [9] Greenky M, Gandhi K, Pulido L, Restrepo C, Parvizi J. Preoperative anemia in total joint
301 arthroplasty: is it associated with periprosthetic joint infection? Clin Orthop Relat Res
302 2012;470:2695–701. https://doi.org/10.1007/s11999-012-2435-z.
303 [10] Lu M, Sing DC, Kuo AC, Hansen EN. Preoperative Anemia Independently Predicts 30-Day
304 Complications After Aseptic and Septic Revision Total Joint Arthroplasty. The Journal of
305 Arthroplasty 2017;32:S197–201. https://doi.org/10.1016/j.arth.2017.02.076.
306 [11] Hart A, Khalil JA, Carli A, Huk O, Zukor D, Antoniou J. Blood transfusion in primary total
307 hip and knee arthroplasty. Incidence, risk factors, and thirty-day complication rates. J Bone
308 Joint Surg Am 2014;96:1945–51. https://doi.org/10.2106/JBJS.N.00077.
309 [12] Kim JL, Park J-H, Han S-B, Cho IY, Jang K-M. Allogeneic Blood Transfusion Is a
310 Significant Risk Factor for Surgical-Site Infection Following Total Hip and Knee
311 Arthroplasty: A Meta-Analysis. J Arthroplasty 2017;32:320–5.
312 https://doi.org/10.1016/j.arth.2016.08.026.
313 [13] Trickey AW, Wright JM, Donovan J, Reines HD, Dort JM, Prentice HA, et al. Interrater
314 Reliability of Hospital Readmission Evaluations for Surgical Patients. American Journal of
315 Medical Quality 2017;32:201–7. https://doi.org/10.1177/1062860615623854.
316 [14] Gabriel RA, Clark AI, Nguyen AP, Waterman RS, Schmidt UH. The Association of
317 Preoperative Hematocrit and Transfusion with Mortality in Patients Undergoing Elective
318 Non-cardiac Surgery. World J Surg 2018;42:1939–48. https://doi.org/10.1007/s00268-017-
319 4359-y.
320 [15] Cepeda MS. Comparison of Logistic Regression versus Propensity Score When the Number
321 of Events Is Low and There Are Multiple Confounders. American Journal of Epidemiology
322 2003;158:280–7. https://doi.org/10.1093/aje/kwg115.
323 [16] Bisbe E, Basora M, Colomina MJ, Spanish Best Practice in Peri-operative Anaemia
324 Optimisation Panel. Peri-operative treatment of anaemia in major orthopaedic surgery: a
325 practical approach from Spain. Blood Transfus 2017;15:296–306.
326 https://doi.org/10.2450/2017.0177-16.
Table I:
Table II:
Table III: