Transfusions increase complications and infections after hip and knee arthroplasty: An analysis of 2760 cases

Abstract

BACKGROUND:

Knee and hip replacement surgery are still the mainstay therapy for osteoarthritis. In spite of the improvement of techniques and implants, anemia is a relatively common complication, with transfusion rates of up to 23% in some centers.

OBJECTIVE:

The purpose of the study was to determine a correlation of transfusions to complications including infection since this topic is still being debated or even disputed in the literature.

METHODS:

This is a level III, single center retrospective observational cohort study of 2760 unilateral primary knee and hip replacements. Preoperative assessment, the number of transfusions and the occurrence of complications were collected and the correlations were analyzed using analysis of variance and logistic regression.

RESULTS:

Fifteen percent of all patients developed at least one complication. Transfusion rate was 9%. Risk factors for receiving a transfusion were female gender, hip replacement, American Society of Anesthesiologists’ Score (ASA) $>$ III, history of myocardial infarction, chronic cardiac disease, diabetes mellitus, chronic kidney disease, and length of surgery. The risk factors for developing a complication were: ASA score, presence of chronic renal insufficiency, and transfusion during hospital stay. Transfusion increases the risk of complications and infection rate. Complication rate with transfusion was 34.7% and without transfusion 13.2%. Infection rate without transfusion was 0.4% and with transfusion 2.82%.

CONCLUSIONS:

The complication rate is higher in transfused patients. The number of complications rises linearly with the number of transfusions. Infection rate is also higher after a transfusion. Efforts should be made to reduce the transfusion rate.

Keywords

Arthroplasty transfusion complication infection

1. Introduction

Total knee (TKA) and hip arthroplasty (THA) are therapeutic mainstays for advanced osteoarthritis, and their number is on the rise [1, 2, 3] while the complication rate is falling [4]. Germany is the world leader in the number of total hip and knee replacements at 295 per 100,000 people [5] or 375,000 yearly, with an estimated market value of €2.6 billion. Anemia due to blood loss is a relatively common complication after both hip and knee arthroplasty, and its major form is treated with blood transfusion from an allogenic donor [6]. Authors report transfusion rates after these procedures to be between 5% and 26% [6, 7, 8, 9]. Browne et al. suggested that transfusion rate during or after THA is increasing in the United States and other countries [10]. Blood transfusions have been deemed safe after rigorous screening for hepatitis and human immunodeficiency virus.

However, data shows that blood transfusion still carries risks [11]. Even though studies from orthopedics and other surgical fields show an increased complication rate in patients receiving transfusion [7, 12], there is still an ongoing debate on the validity of this data [6].

The purpose of this study was to determine risk factors for receiving a transfusion, determining if there is a possibility of reduction of transfusion using blood collecting systems and to determine if there is a correlation between transfusions and the overall complication rate. The hypothesis was that the number of transfusions increases the number of complications and the infection rate.

2. Patients and methods

2.1 Patients

The data for our level III, single center, retrospective observational cohort study was collected in house, where, with the exception of open spine procedures, the full spectrum of orthopedic surgery is performed, mainstay being elective surgery. The most frequent surgery is joint replacement surgery, particularly TKA and THA.

At the time of the study, there were 5 senior surgeons in the department and all operations were performed by them or under their immediate supervision. The surgeries have been performed in three operating rooms.

The data was recorded for postoperative in-hospital stay in a cohort of $n=$ 2,760 patients who received unilateral TKA or THA from 1 January 2010 to 31 December 2014.

Figure 1.

Patient inclusion diagram.

Patients undergoing revision surgery, fracture and tumor prostheses, were excluded from this study. Patient inclusion flow chart is shown in Fig. 1. The average patient age was 69.6 years. The youngest patient was a 23-year-old male who had a dysplastic hip on both sides. The oldest patient was aged 94. The majority of patients $n=$ 1088 or 61% were female. Documentation of chronological development of the anemia, transfusion and development of any complication were collected. The $n=$ 1418 THAs were mainly implanted through an anterior approach (59%), using the AMIS system (Anterior Minimally Invasive Surgery©; Medacta, Castel San Pietro, Switzerland). All cups were cementless Versafit© cups (Medacta, Castel San Pietro, Switzerland) and primarily cementless Amistem or Quadra Stem (both Medacta, Castel San Pietro, Switzerland) have been used. Other 41% were implanted via the anterolateral approach. Here the R3 cementless cup and SBG cementless anatomical stem were primarily used. Cemented versions were used for only $n=$ 57 of all THA (4%) and had no statistical significance.

The $n=$ 1342 TKAs were all implanted using a medial parapatellar approach. Tourniquet was routinely used only for the cementing at 350 mmHg and was opened for wound closure. A total of 4 systems have been used, PFC Sigma (DePuy, Raynham, MA, USA), Genesis II (Smith and Nephew, London, UK), Conformis (Billerica, MA, USA) and MyKnee (Medacta, Castel San Pietro, Switzerland). All systems were implanted in the fully cemented technique using the standardized instruments for each system. PFC and Genesis II are conventional instruments systems, MyKnee is a patient-specific instruments system and Conformis is a customized implant knee system General anesthesia has been performed in $n=$ 1003 patients and $n=$ 1757 received regional anesthesia.

Tranexamic acid was not used routinely at that time in our clinic. For thromboembolic prophylaxis enoxaparine (Clexane©; Sanofi, Paris, France) was used. Patients without oral anticoagulation $n=$ 2,535 received 40 mg administered subcutaneously on a daily basis in the evening. Patients with oral anticoagulation $n=$ 225 were bridged with enoxaparine according to their weight.

Ethical committee approval was obtained (study number F-2015-010).

2.2 Methods

We collected blood intraoperatively and within 6 hours postoperatively, using the intraoperatively applied drainages with the CellSaver System (Haemonetics, Braintree, MA, USA). All patients received a drainage. Contraindication for the use of the CellSaver systems were all types of malignant diseases in the patient history and religious grounds. The retransfusion was done when at least 200 mL of blood could have been retransfusioned. The hemoglobin level was documented preoperatively, after 24 and 48 hours as well as on the 5th day and on the 10th day, if applicable. The drainages were removed after 48 h. Complications were defined as by Dindo et al. [13]. Transfusion trigger were the guidelines of the German Transfusion Society which recommends a single blood transfusion in patients who are symptomatic and have a hemoglobin level lower than 8 mg/dL, or asymptomatic with hemoglobin less than 7 mg/dL [14]. Body mass index (BMI) was calculated from the height and the weight which was collected by the anesthesiologist during preparation. Length of surgery has been documented. Further predictors that were collected are: age, gender, arterial hypertension, usage of anticoagulation and the type, history of myocardial infarction, chronic heart and renal disease, diabetes, chronic obstructive pulmonary disease (COPD) and American Society of Anesthesiologists’ (ASA) score. Lab results that were collected preoperatively were C-reactive protein level and number of leukocytes. The predictors are shown in Table 1.

Table 1
Predictors for calculating the correlation of transfusion to complications

	Mean $\pm$ stand. dev. (range)	Prevalence (%)
Female gender	1672 (60.5)
Age (years)	69.6 $\pm$ 7.4 (23–94)
Hemoglobin preop (mg/dL)	14.2 $\pm$ 1.3 (9.5–19.8)
BMI	28.8 $\pm$ 5.3 (15.9–63.6)
ASA score	2.2 $\pm$ 0.5 (1–4)
CRP preop (mg/L)	4.8 $\pm$ 6.4 (0.0–121.5)
Leukocytes preop (10 ${}^{9}$ /L)	7.0 $\pm$ 2.1 (2.7–40.1)
Arterial hypertension		1936 (70.1)
Usage of ASS only		405 (14.7)
Usage of oral anticoagulation		225 (8.1)
History of myocardial infarction		167 (6)
Chronic heart disease		507 (18.4)
Chronic renal disease		498 (18)
Diabetes		421 (15.3)
COPD		226 (8.2)

BMI – body mass index; ASA – American Society of Anesthesiologists; ASS – acetylsalicylic acid; COPD – chronic obstructive pulmonary diease.

2.3 Statistical analysis

Statistical software SPSS 22 (SPSS, IBM, Armonk, NY, USA; licensed to University of Marburg) was used to determine the frequency of single factors and complications. Analysis of variance was used for calculating the correlations. Comparison of mean and median showed a normal distribution of the data. The odds ratios were calculated with a confidence interval of 95% using binary logistic regression and $\chi^{2}$ test. Statistical significance was set at $P<$ 0.05.

Table 2
Results

	Transfused	Non-transfused
( $n$ ) of patients	248 2512
( $n$ ) of transfusion units	495
Hemoglobin 24 hours post op (mg/dL)	9.87 $\pm$ 1.23 (6.6–13.7)	11.48 $\pm$ 1.32 (7–16.6)
Age	76.4 $\pm$ 7.4 (52–94)	68.9 $\pm$ 9.8 (23–92)
Length of surgery (min)	65.3 $\pm$ 18.4 (25–127)	62.9 $\pm$ 18.4 (25–150)
Female $n$ (%)	186 (75%)	1486 (59%)

3. Results

Table 2 shows the results. A total of $n=$ 248 patients (9%) received at least one allogenic blood transfusion. A total of 495 blood units have been transfused. The highest number of transfusions in a single patient was 18. The transfusion rate after TKA was 6.91% $n=$ 93 patients and 10.94% $n=$ 155 patients after THA. Patients receiving a transfusion had an average ASA of 2.59 ( $\pm$ 0.55) and patients without a transfusion had an average ASA of 2.18 ( $\pm$ 0.54) Average patient age in the transfusion group was 76.31 ( $\pm$ 7.41) and without transfusion 68.90 ( $\pm$ 9.79).

Gender did not play a role in the development of a postoperative complication, but female gender was a risk factor for receiving a transfusion ( $P<$ 0.01).

The risk factors for receiving a transfusion were the following: an ASA score of III and IV, age more than 70 years, history of myocardial infarction, chronic cardiac disease, diabetes mellitus, chronic kidney disease, and length of surgery ( $>$ 60 minutes).

Preoperative hemoglobin level showed no correlation to the number of transfusions ( $P=$ 0.241). The use of CellSaver system had no statistical correlation to development of complication, nor did it reduce the probability of allogenic blood transfusion.

Figure 2.

Scattered plots of the number of complications (y axis) and number of transfusions (x axis).

Four hundred and seventeen patients (15%) developed a complication. Complication rate was 34.7% ( $n=$ 86) in the transfused group and 13.2% ( $n=$ 332) in the non-transfused group. Postoperative infection occurred in $n=$ 23 patients; $n=$ 11 of these received a transfusion, including seven who were transfused in the first 72 hours, before revision surgery or specific symptoms of inflammation or sepsis. Figure 2 shows the scattered plot of the number of transfusions and the number of complications.

4. Discussion

Transfusion increases the complication rate and the risk of wound infection. This confirms the hypothesis of this study.

The general risks associated with allogenic transfusion have been well known for some time and include immunological reactions [15], disease transmission, transfusion-induced coagulopathy, thromboembolic complications [16], transfusion-induced sepsis, and acute lung injury [17]. Effective preoperative management programs that reduce the need for transfusion already exist [18].

This study is not the first in this field that had these observations. Two recent studies [7, 12] had similar results to this study. Their findings included that the prediction factors for receiving an allogenic transfusion were hemoglobin level, age, female gender, BMI, high creatinine levels, TKA, operating room time, and blood loss. The transfusion rates reported were 9.27% for TKA and 26.6% for THA. In their study the rate of surgical site infection rose from 0.5% in non-transfused patients to 2.4% in transfused patients. These findings correlate almost to the decimal with the findings of this study. Interestingly, in the Frisch study, high hemoglobin was protective against an allogenic transfusion, which we did not observe. The lack of protection against transfusion has been observed in other studies [19]. One additional study [20] had similar results, but the chronology of events has not been cleared and it is a multicentric cohort. It still adds to the argumentation of the negative role of the transfusions.

The difference in the infection rate between transfused and non-transfused patients has also been recently observed in Europe [21]. Also, increased mortality rates, higher rates of pneumonia, and higher rates of cardiovascular and cerebral complications have already been documented [22].

One of the main motivations for doing this study was our contribution to a still ongoing scientific debate whether these correlations are real or completely due to confounding. Since ethical boundaries prevent a double blinded, placebo controlled study, especially on a scale large enough to prove or disprove these correlations, levels II and III studies are the only scientific sources.

Opposite findings to ours regarding the role of transfusion is also well documented in literature [6, 10] with authors blaming confounding. Authors from other surgical fields, such as cardiac surgery report similar correlational findings [15, 23] or claim that transfusions increase mortality and morbidity even in low hematocrit patients [24]. General surgery offers very similar conclusions [25, 26]. There are, however, help and alternatives. Widespread use of tranexamic acid is proven useful [27, 28, 29]. Another well documented alternative is Epoetin alfa but the widespread use is still missing [30, 31, 32].

A very large cohort within one center is a strength of this study in an environment of vast registry data from dozens of clinics that is often used in these types of studies.

The main limitation is the not-controlled retrospective nature of the study. The use of tranexamic acid was very limited, one has to assume that the transfusion levels would be even lower, had it had been used routinely. Additionally, preoperative optimization of hemoglobin is not routinely done since it is in the primary domain of the general practitioner. This would probably impact the reduction of the transfusion rate and consequently the complications eve further.

Level II observational studies have been conducted in the area of hip fracture implants [20]. However, the comorbidity level and average age of these patients does not overlap with elective hip and knee surgery and can therefore not be directly compared.

To conclude the study, both increased ASA score and age were risk factors for a complication. However, the complication rate still is higher in transfused patients and the individual odds-ratios for complication categories have been shown. Also, the number of complications rises with the number of transfusions. Infection rate is also higher. Efforts should be made to reduce the transfusion rate.

Footnotes

Acknowledgments

We thank Prof. Albert Benzing and PD Thomas Leonhard for their support of the study.

Conflict of interest

None to report.

References

Kurtz

Ong

Lau

Mowat

Halpern

. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007; 89(4): 780-5.

Derman

Fabricant

David

. The role of overweight and obesity in relation to the more rapid growth of total knee arthroplasty volume compared with total hip arthroplasty volume. J Bone Joint Surg Am 2014; 96(11): 922-8.

Sueyoshi

Meding

Davis

Lackey

Malinzak

Ritter

. Clinical predictors for possible failure after total hip arthroplasty. Hip Int 2016; 26(6): 531-6.

Berstock

Beswick

Lenguerrand

Whitehouse

Blom

. Mortality after total hip replacement surgery: A systematic review. Bone Joint Res 2014; 3(6): 175-82.

Mcpherson

Gon

Scott

. International variations in a selected number of surgical procedures. OECD 2013; Contract No.: 61.

Hart

Khalil

Carli

Huk

Zukor

Antoniou

. Blood transfusion in primary total hip and knee arthroplasty. Incidence, risk factors, and thirty-day complication rates. J Bone Joint Surg Am 2014; 96(23): 1945-51.

Frisch

Wessell

Charters

Jeffries

Silverton

. Predictors and complications of blood transfusion in total hip and knee arthroplasty. J Arthroplasty 2014; 29(9 Suppl): 189-92.

Mähringer-Kunz

Efe

Fuchs-Winkelmann

Schüttler

Paletta

Heyse

. Bleeding in TKA: posterior stabilized vs. cruciate retaining. Arch Orthop Trauma Surg 2015; 135(6): 867-70.

Klasan

Dworschak

Heyse

Lahner

Malcherczyk

Efe

, et al. Patient-specific instruments’ routine use over conventional total knee arthroplasty remains inconclusive; analysis of 961 cases. Technol Health Care 2018.

10.

Browne

Adib

Brown

Novicoff

. Transfusion rates are increasing following total hip arthroplasty: Risk factors and outcomes. J Arthroplasty 2013; 28(8 Suppl): 34-7.

11.

Suh

Nho

Choi

Park

Koo

. A protocol avoiding allogeneic transfusion in joint arthroplasties. Arch Orthop Trauma Surg 2016; 136(9): 1213-26.

12.

Yoshihara

Yoneoka

. Predictors of allogeneic blood transfusion in total hip and knee arthroplasty in the United States, 2000-2009. J Arthroplasty 2014; 29(9): 1736-40.

13.

Dindo

Demartines

Clavien

. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240(2): 205-13.

14.

German national guidelines for therapy with blood components and plasma derivatives Querschnitts-Leitlinien (BÄK) zur Therapie mit Blutkomponenten und Plasmaderivaten. 2008.

15.

Murphy

Reeves

Rogers

Rizvi

Culliford

Angelini

. Increased mortality, postoperative morbidity, and cost after red blood cell transfusion in patients having cardiac surgery. Circulation 2007; 116(22): 2544-52.

16.

Hendrickson

Hillyer

. Noninfectious serious hazards of transfusion. Anesth Analg 2009; 108(3): 759-69.

17.

Vamvakas

Blajchman

. Transfusion-related mortality: The ongoing risks of allogeneic blood transfusion and the available strategies for their prevention. Blood 2009; 113(15): 3406-17.

18.

Alexander

Frew

. Preoperative optimisation of anaemia for primary total hip arthroplasty: A systematic review. Hip Int 2017; 27(6): 515-22.

19.

Schneider

Kendoff

Oloughlin

Hessling

Gehrke

Citak

. Effectiveness of autologous transfusion system in primary total hip and knee arthroplasty. Technol Health Care 2014; 22(1): 123-8.

20.

Friedman

Homering

Holberg

Berkowitz

. Allogeneic blood transfusions and postoperative infections after total hip or knee arthroplasty. J Bone Joint Surg Am 2014; 96(4): 272-8.

21.

Rosencher

Kerkkamp

Macheras

Munuera

Menichella

Barton

, et al. Orthopedic Surgery Transfusion Hemoglobin European Overview (OSTHEO) study: Blood management in elective knee and hip arthroplasty in Europe. Transfusion 2003; 43(4): 459-69.

22.

Pedersen

Mehnert

Overgaard

Johnsen

. Allogeneic blood transfusion and prognosis following total hip replacement: A population-based follow up study. BMC Musculoskelet Disord 2009; 10: 167.

23.

Koch

Duncan

Mihaljevic

Cosgrove

Loop

, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med 2006; 34(6): 1608-16.

24.

Shaw

Johnson

Ferrari

Zapolanski

Brizzio

Rioux

, et al. Balancing the benefits and risks of blood transfusions in patients undergoing cardiac surgery: A propensity-matched analysis. Interact Cardiovasc Thorac Surg 2013; 17(1): 96-102.

25.

Glance

Dick

Mukamel

Fleming

Zollo

Wissler

, et al. Association between intraoperative blood transfusion and mortality and morbidity in patients undergoing noncardiac surgery. Anesthesiology 2011; 114(2): 283-92.

26.

Bernard

Davenport

Chang

Vaughan

Zwischenberger

. Intraoperative transfusion of 1 U to 2 U packed red blood cells is associated with increased 30-day mortality, surgical-site infection, pneumonia, and sepsis in general surgery patients. J Am Coll Surg 2009; 208(5): 931-7, 7.e1-2; discussion 8-9.

27.

Aguilera

Martínez-Zapata

Hinarejos

Jordán

Leal

González

, et al. Topical and intravenous tranexamic acid reduce blood loss compared to routine hemostasis in total knee arthroplasty: A multicenter, randomized, controlled trial. Arch Orthop Trauma Surg 2015; 135(7): 1017-25.

28.

Shemshaki

Nourian

Dehghani

Mokhtari

Mazoochian

. One step closer to sparing total blood loss and transfusion rate in total knee arthroplasty: A meta-analysis of different methods of tranexamic acid administration. Arch Orthop Trauma Surg 2015; 135(4): 573-88.

29.

Hourlier

Reina

Fennema

. Single dose intravenous tranexamic acid as effective as continuous infusion in primary total knee arthroplasty: A randomised clinical trial. Arch Orthop Trauma Surg 2015; 135(4): 465-71.

30.

Tomeczkowski

Stern

Müller

, von Heymann

. Potential cost saving of Epoetin alfa in elective hip or knee surgery due to reduction in blood transfusions and their side effects: A discrete-event simulation model. PLoS One 2013; 8(9): e72949.

31.

Goodnough

Maniatis

Earnshaw

Benoni

Beris

Bisbe

, et al. Detection, evaluation, and management of preoperative anaemia in the elective orthopaedic surgical patient: NATA guidelines. Br J Anaesth 2011; 106(1): 13-22.

32.

Xie

Yue

Kang

Pei

. Combined use of intravenous and topical tranexamic acid following cementless total hip arthroplasty: A randomised clinical trial. Hip Int 2016; 26(1): 36-42.