- Email: [email protected]
Serum concentrations of interleukin-6 in patients following unilateral versus bilateral total knee arthroplasty
J Orthop Sci (2009) 14:437–442 DOI 10.1007/s00776-009-1344-9
Original article Serum concentrations of interleukin-6 in patients following unilateral versus bilateral total knee arthroplasty HAJIME KUGISAKI, MOTOKI SONOHATA, MITSUNORI KOMINE, KENJI TSUNODA, SHINSUKE SOMEYA, HIDEFUMI HONKE, MASAAKI MAWATARI, and TAKAO HOTOKEBUCHI Department of Orthopedic Surgery, Faculty of Medicine, Saga University, 5-1-1 Nabeshima, Saga 849-8501, Japan
Abstract Background. Surgical stress is known to affect body temperature, white blood cell (WBC) count, C-reactive protein (CRP), and interleukin-6 (IL-6). The aim of the present study was to investigate which parameter is most suitable for quantitative analysis of surgical stress. Methods. Unilateral total knee arthroplasty (U-TKA) and bilateral TKA (B-TKA) were selected for the subjects of this study because the B-TKA creates approximately double the surgical stress of the U-TKA. The temperature, WBC count, CRP, and IL-6 in the blood were measured pre- and postoperatively in both groups. The IL-6 in the drainage fluid was also measured after the operation. Results. The temperature, WBC count, CRP, and IL-6 in the blood significantly increased on the first day after the operation in both groups. There were significant differences between the two groups in the WBC count (P < 0.05) and the IL-6 level in the blood (P < 0.05) on the first day after the surgery. There were no significant differences between the two groups for the CRP and IL-6 levels in the drainage fluid. The relative proportions — (B-TKA/U-TKA) × 100 (%) — were 170.4% for the operating time, 219.4 % for total blood loss, 200.0% for blood transfusion, 100.3% for temperature, 128.9% for WBC count, 127.4% for CRP, and 246.5% for the IL-6 level in the blood. Conclusions. The serum IL-6 level may best reflect surgical stress and could therefore be a quantitative marker of surgical stress.
Introduction Surgical stress can elicit a characteristic response involving increased circulating concentrations of stress hormones (e.g., cortisol and catecholamines), synthesis and release of various humoral mediators (e.g., proinflammatory cytokines), and induction of synthesis and release of acute-phase proteins such as C-reactive protein (CRP); it may also induce various metabolic Offprint requests to: M. Sonohata Received: October 29, 2008 / Accepted: March 4, 2009
changes (e.g., lipolysis or hyperglycemia).1 Cytokines and the acute-phase responses play an important role in controlling the human immune system. Especially, the proinflammatory cytokines — e.g., tumor necrosis factor-α (TNFα),2 interleukin-1β (IL-1β),3 interleukin-6 (IL-6),1,4 interleukin-8 (IL-8)4 — are considered important mediators of the pathological changes associated with surgery. Various studies have suggested that IL-6 is a major mediator of the acute-phase protein response and is recognized as a marker of postoperative tissue trauma.1,4,5 Minimally invasive surgery (MIS) procedures have recently become popular for their advantage of producing less tissue damage. Previously, the surgical trauma was mainly evaluated by the operating time, blood loss, body temperature, number of leukocytes and percentage of neutrophils, and the CRP level. However, several recent lines of evidence indicate that IL-6 is one of the earliest and most important mediators of the short-term phase response. Several authors have reported on the comparison of MIS surgery with conventional surgery.6,7 Previous investigations have found simply whether there are significant differences between the surgical procedures (e.g., conventional vs. scopic surgery).6,7 To our knowledge, however, no previous study has investigated the propriety of the quantitative relation of the response of serum IL-6 to surgery. The aim of the present study was to investigate the quantitative relation of the response of serum IL-6 to surgery. For the purpose of quantitative analysis, unilateral total knee arthroplasty (U-TKA) and bilateral TKA (B-TKA) were selected for the present study. Two types of surgery can provide an interesting model for analyzing the role of tissue damage during the acute phase response because the B-TKA requires twice the operative process as the U-TKA. The study protocol adhered to the ethical guidelines of the 1975 Declaration of Helsinki; and the institu-
438
H. Kugisaki et al.: Interleukin-6 in TKA
tional review board of the Faculty of Medicine, Saga University at Saga approved this study.
Materials and methods Between August 2004 and April 2005, 48 patients underwent TKA in our institution. We excluded any patients with inflammatory arthritis, a systemic inflammatory or autoimmune disorder, or a history of any type of cancer or chronic illness. We enlisted the remaining 34 patients (1 man, 33 women) in our study. Finally, the 33 women were enrolled in the study. The indication for each patient was osteoarthritis of the knee. Ten patients underwent U-TKA and 23 patients B-TKA. Table 1 shows the patient characteristics, including sex, age, weight, height, and body mass index [BMI = weight (kg) ÷ height (m)2 ], at the time of the operation. There were no significant differences between the two groups regarding these characteristics. Strict inclusion criteria for TKA included clinically significant osteoarthritic changes in the knee that were symptomatic at the time of surgery and failure of conservative treatment for osteoarthritis of the knee. The decision regarding whether to perform a U-TKA or B-TKA was based on both knees being equally symptomatic. In addition, all patients in both groups were classified according to the American Society of Anesthesiologists (ASA) physical status as grade 1 or 2. The clinical findings at the time of surgery were evaluated using the Japanese Orthopaedic Association (JOA) knee rating score. The JOA score consisted of four categories and a total 100 points as full marks: pain and walking (30 points), pain and ascending or descending stairs (25 points), range of motion (35 points), and joint effusion (10 points). In addition, the femorotibial angle (FTA) at the standing position was measured using plain radiography. Surgery was performed by a single surgeon via a medial parapatellar approach. Implants were all a cementless system (NexGen; Zimmer, Warsaw, IN, USA). Table 1. Patient characteristics Parameter Sex (F/M) Age (years) Weight (kg) Height (cm) BMI (kg/m2) JOA score FTA (°)
U-TKA (n = 10)
B-TKA (n = 23)
P
10/0 75.4 ± 6.8 53.8 ± 8.0 148.0 ± 5.1 26.7 ± 3.9 54.5 ± 9.8 186.6 ± 4.2
23/0 74.0 ± 4.4 50.0 ± 8.1 147.7 ± 5.6 26.6 ± 3.4 51.2 ± 8.4 185.2 ± 5.4
0.40 0.92 0.86 0.95 0.28 0.46
Results are the number or the mean ± SD U-TKA, B-TKA, unilateral and bilateral total knee arthroplasty, respectively; JOA, Japanese Orthopaedic Association; FTA, femorotibial angle
Not all of the patellae were replaced by the implant. The B-TKA proceeded successively; when the skin closure of the first side was begun, the other side was started. An air tourniquet was used during all operations. Before wound closure, a 1/8-inch Silastic wound drain with vacuum suction was placed, which was removed on the second postoperative day. The anesthesia team was not involved in this study; however, all patients were administered spinal and/or epidural or general anesthesia as determined by the anesthesia team. The patients who received epidural anesthesia were provided with continuous epidural anesthesia on the first postoperative day. Flurbiprofen axetil, diclofenac sodium, pentazocine, and loxoprofen sodium were used based on the decision of the physician in charge. Venous blood samples were obtained before surgery, and on postoperative days (PODs) 1, 7, and 14 as a routine inspection in accordance with the regimen used at our institution. The IL-6 blood level was only measured before surgery and on PODs 1 and 7. The drain reservoirs were emptied 2 h before sample collection (22 h after surgery) to ensure that all fluid collected had drained from the wound at the specified point (24 h after surgery). In the B-TKA group, the drain fluid samples were collected only from the left side. There were no significant differences in the IL-6 level in the drainage fluid between the right and left knees (data not shown). The blood and drainage fluid samples were centrifuged for 7 min at 3500 rpm to separate the serum. The supernatant was stored at −25°C until the sample analysis was done. An enzyme-linked immunosorbent assay (ELISA) was used for determining the serum IL-6 and drainage fluid levels (BioSource Human IL-6 ELISA kit; BioSource, Camarillo, CA, USA). The ELISA was done according to the manufacturer’s instructions. The assay had a sensitivity of 2–500 pg/ml for IL-6. The CRP concentration was measured by immunonephelometry (Hitachi 7600 modular system; Hitachi, Tokyo, Japan). The detection limit of the CRP assay was 0.01 mg/dl. All samples also tested for the total WBC count. We recorded the patients’ body temperature three times per day, and the maximum daily temperature was detected. The above factors (age, BMI, operating time, blood loss, units of blood transfused, body temperature, CRP, WBC, IL-6 in blood and drainage fluid) were analyzed regarding the relative proportions for the B-UKA and the U-TKA. The relative proportion was calculated as: (B-TKA/U-TKA) × 100 (%). The average values were determined for all the relative proportions. The average values of the body temperature, CRP, WBC, and blood IL-6 were selected as the mean value on POD 1. Each patient had a 1-year follow-up to rule out the development of a surgical-site infection.
H. Kugisaki et al.: Interleukin-6 in TKA
All numerical data were expressed as the mean and standard deviation. The comparison of the mean IL-6 levels, CRP levels, WBC counts, and maximum daily temperatures were performed using the one-way analysis of variance (ANOVA) with Fisher’s PLSD (IL-6 levels: before surgery vs. PODs 1 and 7; CRP levels and WBC counts: before surgery vs. PODs 1, 7, and 14; maximum daily temperatures: before surgery vs. PODs 1, 3, and 7). The mean variables in the U-TKA and BTKA were compared using Student’s t-test. Statistical significance was set at P < 0.05 for each test.
Results There were no anesthetic or surgical complications, including pneumonia, atelectasis, urinary tract infection, symptomatic thromboembolic event, or wound infection. There was no significant difference between the patients who underwent U-TKA and the patients who underwent B-TKA in terms of possible confounding factors, including age (P = 0.40), sex (all were female), weight (P = 0.93), height (P = 0.86), BMI (P = 0.95), JOA score (P = 0.28), and FTA (P = 0.46) (Table 1). Table 2 shows the perioperative data, including the type of anesthesia, operating time, total blood loss, and the units transfused. Significant differences were observed between the two groups regarding the operating time, total blood loss, and units transfused. All blood transfusions were done up to 12 h after the surgery. In the U-TKA group, all transfusions used autologous blood. In the B-UKA group, nine patients were transfused with only autologous blood and one was transfused both autologous and homologous blood. Epidural anesthesia was used for postoperative analgesia on the first day after the surgery, 0.2% ropivacaine
439
hydrochloride hydrate (4 ml/h) was used in two cases and 0.25% bupivacaine hydrochloride hydrate (2 ml/h) in one case in the U-TKA group. In the B-TKA group, 0.2% ropivacaine hydrochloride hydrate (4 ml/h) was used in two cases and 0.25% bupivacaine hydrochloride hydrate in three cases (4 ml/h in one case and 2 ml/h in two cases). The amounts of flurbiprofen axetil used on the day of the operation in the two groups were 200 ± 33 mg and 176 ± 37 mg (P = 0.09), diclofenac sodium 30 ± 42 mg and 33 ± 42 mg (P = 0.87), and pentazocine 9 ± 13 and 17 ± 14 (P = 0.13), respectively. All of the patients were given loxoprofen sodium orally180 mg per a day (60 mg × 3) from POD 1. There was no significant difference between the two groups in terms of the maximum daily temperature. The maximum daily temperature levels peaked on POD 1 in both groups (37.4° ± 0.36°C in the U-TKA group and 37.5° ± 0.32°C in the B-TKA group); in each group the peak levels were significantly higher than at other points (P < 0.01). The maximum daily temperature returned to the preoperative level on POD 3 (P = 0.08) in the U-TKA group, and on POD 7 (P = 0.32) in the B-TKA group (Fig. 1A). The WBC counts peaked on POD 1 in both groups (7080 ± 1406 /μl in the U-TKA group and 9126 ± 2524 /μl Table 2. Summary of perioperative data Parameter Anesthesia type Spinal Spinal and epidural General Operating time (min) Total blood loss (g) Units transfused (no.)
U-TKA (n = 10)
B-TKA (n = 23)
P
7 3 0 53.3 ± 9.2 480 ± 110 1.6 ± 0.8
17 5 1 90.8 ± 20.8 1053 ± 302 3.2 ± 1.2
<0.01 <0.01 <0.01
Results are the number or the mean ± SD
Fig. 1. Unilateral (U-TKA) and bilateral (B-TKA) total knee arthroplasties in regard to various parameters. A Temperature. B White blood cell (WBC) count. C C-reactive protein (CRP). D Interleukin-6 (IL-6) in blood.
440
in the B-TKA group); the peak levels were significantly higher than at any other point in each group (P < 0.01). The maximum WBC count returned to the preoperative level on POD 7 in both groups (P = 0.90 and P = 0.38, respectively) (Fig. 1B). The CRP levels peaked on POD 7 in the U-TKA group and on POD 1 in the B-TKA group (3.04 ± 2.14 mg/dl in the U-TKA group and 3.86 ± 2.22 mg/dl in the B-TKA group). In both groups, the CRP levels on PODs 1 and 7 were significantly higher than they were preoperatively and on POD 14 (P < 0.01); however, there were no significant differences between the CRP levels on PODs 1 and 7 in each group (P = 0.70 and P = 0.16, respectively). In the U-TKA group, the maximum CRP returned to the preoperative level on POD 14 (P = 0.81). In the B-TKA group, however, the maximum CRP did not return to the preoperative level by POD 14 (P < 0.05) (Fig. 1C). In both groups, the blood IL-6 level was significantly higher on POD 1 than preoperatively and on POD 7 (P < 0.01) (445.1 ± 377.5 pg/ml in the U-TKA group and 1097.9 ± 839.9 pg/ml in the B-TKA group). There were no significant differences in the circulating IL-6 before surgery and on POD 7 in each group (Fig. 1D). The IL-6 level in the drainage fluid on POD 1 was 413.5 ± 109.8 ng/ml in the U-TKA group and 403.9 ± 238.1 ng/ml in the B-TKA group. There was no significant difference between the groups (P = 0.93). The relative proportions for the operating time, total blood loss, units of blood transfused, and IL-6 level in blood were around 200% (Fig. 2). At the 1-year follow-up, no patient in either group had demonstrated any wound infection.
Fig. 2. Relative proportion of B-UKA and U-TKA × 100 (%), in regard to temperature, WBC, CRP, blood IL-6, and IL-6 in drainage fluid (drain). BMI, body mass index
H. Kugisaki et al.: Interleukin-6 in TKA
Discussion In the present study, we studied the cytokine and the acute-phase response to U-TKA and the B-TKA. The model has a major advantage in that the B-TKA is associated with twice as much surgical trauma as the U-TKA. This is the first clinical comparison of the response to the surgery to elucidate the relative proportion of surgical stress. Interleukin-6 is produced mainly by monocytes and macrophages after antigen stimulation, although other cells (i.e., T lymphocytes, endothelial cells, fibroblasts) may also produce it.8 Surgery may directly induce IL-6 release by these cells, or IL-6 release may be mediated by other locally released cytokines. For example, both IL-1 and TNFα can up-regulate IL-6 synthesis in fibroblasts.9 However, concentrations of TNFα, IL-1β, and IL-8 are too low to be detected. IL-1β and TNFα are frequently difficult to detect in the systemic circulation even in severely infected patients.10 IL-6 therefore appears to be a major endogenous protein mediator of fever and the acute-phase response following surgery.9 IL-6 also plays a role in the induction and control of the acute-phase protein response, especially that of CRP synthesis by human hepatocytes.1 Andres et al.11 reported that increased local production of IL-6 at the surgical site results in the release of IL-6 into the local circulation, thus leading to leukocytosis and fever. Kragsbjerg et al.5 showed that the peak IL-6 level differed among surgical procedures, and the peak serum IL-6 concentrations occurred 6–24 h after surgery for joint replacement surgery. IL-6 is produced locally and, through circulation in the blood, transmits local information to the whole system. It has been reported that there are significant differences in the local and systemic cytokine response to surgical trauma.12,13 The most prominent response was a 1000-fold increase of IL-6 locally, represented by drained fluid, in comparison to the systemic concentrations in arterial blood of patients operated on for thoracic scoliosis.13 Therefore, the cytokine response to surgical trauma is characterized by both a local and a systemic reaction.12,13 In a previous study, the peak serum IL-6 level was 172–400 pg/ml11,14,15; and in the present study, the serum IL-6 level on POD 1 was 445 pg/ml. A number of researchers have reported that serum IL-6 reflects the kinetics of surgical trauma.4–7,11,14,15 Vita et al. showed that cytokines in drainage fluid also demonstrated this.16 In our study, the Il-6 level in drainage fluid on POD 1 was also about 1000 times that of the systemic blood, and there was no significant difference between the U-TKA and the BTKA groups or the right and left sides. Blood transfusion induced production of cytokines and cortisol. However, the cytokines returned to preinfusion levels after blood transfusion.17 Bottner et al.18
H. Kugisaki et al.: Interleukin-6 in TKA
reported that the increased serum IL-6 concentration after TKA was related to the surgical intervention itself, and the amount of transfusion did not correlate with the increase in the patient’s serum IL-6. The development of postoperative fever is common after a surgical procedure.11,19 Andres et al.11 reported that elevated serum IL-6 levels after total joint arthroplasty caused fever after TKA. Another potential cause of fever is blood transfusion,20 but this finding has not yet been substantiated in other studies of postoperative fever in this population.19,21 In the current study, body temperature significantly increased after TKA in both groups, but there was no significant difference between the groups. Therefore, fever may not be suitable for the analysis of surgical trauma after TKA. The WBC count is well known to increase and remain after surgical trauma.22 The WBC count also increases after blood transfusion, and a higher WBC count due to blood transfusion is not a transient reaction.17 In the present study, the WBC count increased significantly after TKA in both groups, although there was a significant difference between the two groups, which might have been due to the synergy of surgical trauma and blood transfusion; therefore, further investigation is required to clarify the proportions. The surgical stress of the B-TKA was twice that of the U-TKA in the current model; however, the relative population of WBCs remained at 127.0%. Therefore, the surgical stress may be underestimated when the WBC count is used to assess surgical stress. Tissue damage at operation induces an elevated serum CRP level during the following 2–3 days. If no complications occur, however, it thereafter decreases rapidly.23 The CRP level is generally used for monitoring surgical stress, and it has been used to compare different surgical procedures in many studies.6,7,24 Elevated CRP levels are seen in association with a number of ailments including infection, inflammation, and malignancy as a response to tissue injury. In the present study, the subjects did not exhibit these conditions. Larsson et al.23 reported that the CRP peak level did not correlate with the quantity of transfused blood or the ischemic period caused by the tourniquet. As with the responses of temperature and WBC, the CRP increased significantly after the TKA in both groups, but there was no significant difference between the groups. In one study the CRP level reached a maximum 48–72 h after orthopedic surgery and regained its normal level 1–3 weeks after the surgery.23,24 In the current study, the CRP level 2 weeks after surgery showed a significant difference between the two groups. Severe surgical trauma may thus prolong normalization of the CRP level. The CRP level may be suitable for qualitative assays of surgical trauma but not for quantitative assays.
441
The acute-phase response after surgery is diverse and complicated. Serum IL-6 level, CRP level, WBC count, and fever are representative markers for surgical trauma. However, regarding temperature and CRP, no significant differences were observed between the UTKA and B-TKA groups on POD 1. There was a significant difference between the U-TKA and the B-TKA groups regarding the WBC counts; however, the relative proportion remained at 128.9%. U-TKA and the B-TKA were selected for the current study because the proportion of their surgical trauma was known. Therefore, an ideal parameter for surgical trauma should show around a 200% change in the relative proportion. The serum IL-6 level, among others, may be most suitable for quantitative analysis of surgical trauma. The serum IL-6 concentration proved to discriminate the severity of the surgical procedure better than temperature, WBC count, or CRP level. However, the response to surgical trauma naturally varies for each individual.3–7,10–16 Hence, the results of the current study might be suitable for a comparison of the amplitude in surgical techniques but not for individual reactions to surgery.
Conclusion We investigated the surgical stress for patients undergoing U-TKA and B-TKA using various parameters. Body temperature, WBC counts, CRP level, and IL-6 level in blood in both groups were found to increase significantly after the surgery; however, only the WBC count and blood IL-6 level showed a significant difference in the two groups. The IL-6 level in blood is therefore considered a more suitable marker for a quantitative analysis of surgical stress. Acknowledgments. The authors thank Dr. Jun Ito, Dr. Shuya Ide, Dr. Masaru Kitajima, Dr. Kenji Ogawa, Dr. Riki Tanaka, Dr. Tomonori Tajima, and Dr. Osamu Nozaki for their valuable contributions to this study.
References 1. Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J 1990;265:621–36. 2. Tracey KJ, Morgello S, Koplin B, Fahey TJ 3rd, Fox J, Aledo A, et al. Metabolic effects of cachectin/tumor necrosis factor are modified by site of production: cachectin/tumor necrosis factorsecreting tumor in skeletal muscle induces chronic cachexia, while implantation in brain induces predominantly acute anorexia. J Clin Invest 1990;86:2014–24. 3. Kaplan E, Dinarello CA, Gelfand JA. Interleukin-1 and the response to injury. Immunol Res 1989;8:118–29. 4. Ohzato H, Yoshizaki K, Nishimoto N, Ogata A, Tagoh H, Monden M, et al. Interleukin-6 as a new indicator of inflammatory status:
442
5.
6.
7.
8. 9.
10.
11.
12.
13.
H. Kugisaki et al.: Interleukin-6 in TKA detection of serum levels of interleukin-6 and C-reactive protein after surgery. Surgery 1992;111:201–9. Kragsbjerg P, Holmberg H, Vikerfors T. Serum concentrations of interleukin-6, tumour necrosis factor-alpha, and C-reactive protein in patients undergoing major operations. Eur J Surg 1995;161:17–22. Yoshida S, Ohta J, Yamasaki K, Kamei H, Harada Y, Yahara T, et al. Effect of surgical stress on endogenous morphine and cytokine levels in the plasma after laparoscopic or open cholecystectomy. Surg Endosc 2000;14:137–40. Grande M, Tucci GF, Adorisio O, Barini A, Rulli F, Neri A, et al. Systemic acute-phase response after laparoscopic and open cholecystectomy. Surg Endosc 2002;16:313–6. Kishimoto T. The biology of interleukin-6. Blood 1989;74:1– 10. Walther Z, May LT, Sehgal PB. Transcriptional regulation of the interferon-beta 2/B cell differentiation factor BSF-2/hepatocytestimulating factor gene in human fibroblasts by other cytokines. J Immunol 1988;140:974–7. Damas P, Reuter A, Gysen P, Demonty J, Lamy M, Franchimont P. Tumor necrosis factor and interleukin-1 serum levels during sever sepsis in humans. Crit Care Med 1989;17:975–8. Andres BM, Taub DD, Gurkan I, Wenz JF. Postopetrative fever after total knee arthroplasty: the role of cytokine. Clin Orthop 2003;415:221–31. Krohn CD, Reikerås O, Aasen AO. The cytokines IL-1beta and IL-1 receptor antagonist, IL-2 and IL-2 soluble receptor-alpha, IL-6 and IL-6 soluble receptor, TNF-alpha and TNF soluble receptor I, and IL10 in drained and systemic blood after major orthopaedic surgery. Eur J Surg 1999;165:101–9. Krohn CD, Reikerås O, Mollnes TE, Aasen AO. Complement activation and release of interleukin-6 and tumor necrosis factoralpha in drained and systemic blood after major orthopaedic surgery. Eur J Surg 1998;164:103–8.
14. Wirtz DC, Heller KD, Miltner O, Zilkens KW, Wolff JM. Interleukin-6: a potential inflammatory marker after total joint replacement. Int Orthop 2004;24:194–6. 15. Hall GM, Peerbhoy D, Shenkin A, Parker CJR, Salmon P. Hip and knee arthroplasty: a comparison and the endocrine, metabolic and inflammatory responses. Clin Sci 2000;98:71–9. 16. Vita GD, Patti R, D’Agostino P, Ferlazzo V, Angileri M, Sieli G, et al. Modifications in the production of cytokines and growth factors in drainage following mesh implantation after incisional hernia repair. Am J Surg 2006;191:785–90. 17. Krohn CD, Reikerås O, Aasen AO. Inflammatory cytokines and their receptors in arterial and mixed venous blood before, during and after infusion of drained untreated blood. Transfusion Med 1999;9:125–30. 18. Bottner F, Sheth N, Chimento G, Sculco T. Cytokine levels aftert transfusion of wash wound drainage in total knee arthroplasty. J Knee Surg 2003;16:93–7. 19. Guinn S, Castro FP Jr, Garcia R, Barrack RL. Fever following total knee arthroplasty. Am J Knee Surg 1999;12:161–4. 20. Kennedy JG, Rodgers WB, Zurakowski D, Sullivan R, Griffin D, Beardsley W, et al. Pyrexia after total knee replacement: a cause for concern? Am J Orthop 1997;26:549–52, 554. 21. Shaw JA, Chung R. Febrile response after knee and hip arthroplasty. Clin Orthop 1999;367:181–9. 22. Eriksson S, Olander B, Pira U, Granstöm L. White blood cell count, leucocyte elastase activity, and serum concentrations of interleukin-6 and C-reactive protein after open appendectomy. Eur J Surg 1997;163:123–7. 23. Larsson S, Thelander U, Freiberg S. C-reactive protein (CRP) levels after elective orthopaedic surgery. Clin Orthop 1992;275: 237–42. 24. White J, Kelly M, Dunsmuir R. C-reactive protein after total hip and total knee replacement. J Bone Joint Surg Br 1998;80: 909–11.
Original article Serum concentrations of interleukin-6 in patients following unilateral versus bilateral total knee arthroplasty HAJIME KUGISAKI, MOTOKI SONOHATA, MITSUNORI KOMINE, KENJI TSUNODA, SHINSUKE SOMEYA, HIDEFUMI HONKE, MASAAKI MAWATARI, and TAKAO HOTOKEBUCHI Department of Orthopedic Surgery, Faculty of Medicine, Saga University, 5-1-1 Nabeshima, Saga 849-8501, Japan
Abstract Background. Surgical stress is known to affect body temperature, white blood cell (WBC) count, C-reactive protein (CRP), and interleukin-6 (IL-6). The aim of the present study was to investigate which parameter is most suitable for quantitative analysis of surgical stress. Methods. Unilateral total knee arthroplasty (U-TKA) and bilateral TKA (B-TKA) were selected for the subjects of this study because the B-TKA creates approximately double the surgical stress of the U-TKA. The temperature, WBC count, CRP, and IL-6 in the blood were measured pre- and postoperatively in both groups. The IL-6 in the drainage fluid was also measured after the operation. Results. The temperature, WBC count, CRP, and IL-6 in the blood significantly increased on the first day after the operation in both groups. There were significant differences between the two groups in the WBC count (P < 0.05) and the IL-6 level in the blood (P < 0.05) on the first day after the surgery. There were no significant differences between the two groups for the CRP and IL-6 levels in the drainage fluid. The relative proportions — (B-TKA/U-TKA) × 100 (%) — were 170.4% for the operating time, 219.4 % for total blood loss, 200.0% for blood transfusion, 100.3% for temperature, 128.9% for WBC count, 127.4% for CRP, and 246.5% for the IL-6 level in the blood. Conclusions. The serum IL-6 level may best reflect surgical stress and could therefore be a quantitative marker of surgical stress.
Introduction Surgical stress can elicit a characteristic response involving increased circulating concentrations of stress hormones (e.g., cortisol and catecholamines), synthesis and release of various humoral mediators (e.g., proinflammatory cytokines), and induction of synthesis and release of acute-phase proteins such as C-reactive protein (CRP); it may also induce various metabolic Offprint requests to: M. Sonohata Received: October 29, 2008 / Accepted: March 4, 2009
changes (e.g., lipolysis or hyperglycemia).1 Cytokines and the acute-phase responses play an important role in controlling the human immune system. Especially, the proinflammatory cytokines — e.g., tumor necrosis factor-α (TNFα),2 interleukin-1β (IL-1β),3 interleukin-6 (IL-6),1,4 interleukin-8 (IL-8)4 — are considered important mediators of the pathological changes associated with surgery. Various studies have suggested that IL-6 is a major mediator of the acute-phase protein response and is recognized as a marker of postoperative tissue trauma.1,4,5 Minimally invasive surgery (MIS) procedures have recently become popular for their advantage of producing less tissue damage. Previously, the surgical trauma was mainly evaluated by the operating time, blood loss, body temperature, number of leukocytes and percentage of neutrophils, and the CRP level. However, several recent lines of evidence indicate that IL-6 is one of the earliest and most important mediators of the short-term phase response. Several authors have reported on the comparison of MIS surgery with conventional surgery.6,7 Previous investigations have found simply whether there are significant differences between the surgical procedures (e.g., conventional vs. scopic surgery).6,7 To our knowledge, however, no previous study has investigated the propriety of the quantitative relation of the response of serum IL-6 to surgery. The aim of the present study was to investigate the quantitative relation of the response of serum IL-6 to surgery. For the purpose of quantitative analysis, unilateral total knee arthroplasty (U-TKA) and bilateral TKA (B-TKA) were selected for the present study. Two types of surgery can provide an interesting model for analyzing the role of tissue damage during the acute phase response because the B-TKA requires twice the operative process as the U-TKA. The study protocol adhered to the ethical guidelines of the 1975 Declaration of Helsinki; and the institu-
438
H. Kugisaki et al.: Interleukin-6 in TKA
tional review board of the Faculty of Medicine, Saga University at Saga approved this study.
Materials and methods Between August 2004 and April 2005, 48 patients underwent TKA in our institution. We excluded any patients with inflammatory arthritis, a systemic inflammatory or autoimmune disorder, or a history of any type of cancer or chronic illness. We enlisted the remaining 34 patients (1 man, 33 women) in our study. Finally, the 33 women were enrolled in the study. The indication for each patient was osteoarthritis of the knee. Ten patients underwent U-TKA and 23 patients B-TKA. Table 1 shows the patient characteristics, including sex, age, weight, height, and body mass index [BMI = weight (kg) ÷ height (m)2 ], at the time of the operation. There were no significant differences between the two groups regarding these characteristics. Strict inclusion criteria for TKA included clinically significant osteoarthritic changes in the knee that were symptomatic at the time of surgery and failure of conservative treatment for osteoarthritis of the knee. The decision regarding whether to perform a U-TKA or B-TKA was based on both knees being equally symptomatic. In addition, all patients in both groups were classified according to the American Society of Anesthesiologists (ASA) physical status as grade 1 or 2. The clinical findings at the time of surgery were evaluated using the Japanese Orthopaedic Association (JOA) knee rating score. The JOA score consisted of four categories and a total 100 points as full marks: pain and walking (30 points), pain and ascending or descending stairs (25 points), range of motion (35 points), and joint effusion (10 points). In addition, the femorotibial angle (FTA) at the standing position was measured using plain radiography. Surgery was performed by a single surgeon via a medial parapatellar approach. Implants were all a cementless system (NexGen; Zimmer, Warsaw, IN, USA). Table 1. Patient characteristics Parameter Sex (F/M) Age (years) Weight (kg) Height (cm) BMI (kg/m2) JOA score FTA (°)
U-TKA (n = 10)
B-TKA (n = 23)
P
10/0 75.4 ± 6.8 53.8 ± 8.0 148.0 ± 5.1 26.7 ± 3.9 54.5 ± 9.8 186.6 ± 4.2
23/0 74.0 ± 4.4 50.0 ± 8.1 147.7 ± 5.6 26.6 ± 3.4 51.2 ± 8.4 185.2 ± 5.4
0.40 0.92 0.86 0.95 0.28 0.46
Results are the number or the mean ± SD U-TKA, B-TKA, unilateral and bilateral total knee arthroplasty, respectively; JOA, Japanese Orthopaedic Association; FTA, femorotibial angle
Not all of the patellae were replaced by the implant. The B-TKA proceeded successively; when the skin closure of the first side was begun, the other side was started. An air tourniquet was used during all operations. Before wound closure, a 1/8-inch Silastic wound drain with vacuum suction was placed, which was removed on the second postoperative day. The anesthesia team was not involved in this study; however, all patients were administered spinal and/or epidural or general anesthesia as determined by the anesthesia team. The patients who received epidural anesthesia were provided with continuous epidural anesthesia on the first postoperative day. Flurbiprofen axetil, diclofenac sodium, pentazocine, and loxoprofen sodium were used based on the decision of the physician in charge. Venous blood samples were obtained before surgery, and on postoperative days (PODs) 1, 7, and 14 as a routine inspection in accordance with the regimen used at our institution. The IL-6 blood level was only measured before surgery and on PODs 1 and 7. The drain reservoirs were emptied 2 h before sample collection (22 h after surgery) to ensure that all fluid collected had drained from the wound at the specified point (24 h after surgery). In the B-TKA group, the drain fluid samples were collected only from the left side. There were no significant differences in the IL-6 level in the drainage fluid between the right and left knees (data not shown). The blood and drainage fluid samples were centrifuged for 7 min at 3500 rpm to separate the serum. The supernatant was stored at −25°C until the sample analysis was done. An enzyme-linked immunosorbent assay (ELISA) was used for determining the serum IL-6 and drainage fluid levels (BioSource Human IL-6 ELISA kit; BioSource, Camarillo, CA, USA). The ELISA was done according to the manufacturer’s instructions. The assay had a sensitivity of 2–500 pg/ml for IL-6. The CRP concentration was measured by immunonephelometry (Hitachi 7600 modular system; Hitachi, Tokyo, Japan). The detection limit of the CRP assay was 0.01 mg/dl. All samples also tested for the total WBC count. We recorded the patients’ body temperature three times per day, and the maximum daily temperature was detected. The above factors (age, BMI, operating time, blood loss, units of blood transfused, body temperature, CRP, WBC, IL-6 in blood and drainage fluid) were analyzed regarding the relative proportions for the B-UKA and the U-TKA. The relative proportion was calculated as: (B-TKA/U-TKA) × 100 (%). The average values were determined for all the relative proportions. The average values of the body temperature, CRP, WBC, and blood IL-6 were selected as the mean value on POD 1. Each patient had a 1-year follow-up to rule out the development of a surgical-site infection.
H. Kugisaki et al.: Interleukin-6 in TKA
All numerical data were expressed as the mean and standard deviation. The comparison of the mean IL-6 levels, CRP levels, WBC counts, and maximum daily temperatures were performed using the one-way analysis of variance (ANOVA) with Fisher’s PLSD (IL-6 levels: before surgery vs. PODs 1 and 7; CRP levels and WBC counts: before surgery vs. PODs 1, 7, and 14; maximum daily temperatures: before surgery vs. PODs 1, 3, and 7). The mean variables in the U-TKA and BTKA were compared using Student’s t-test. Statistical significance was set at P < 0.05 for each test.
Results There were no anesthetic or surgical complications, including pneumonia, atelectasis, urinary tract infection, symptomatic thromboembolic event, or wound infection. There was no significant difference between the patients who underwent U-TKA and the patients who underwent B-TKA in terms of possible confounding factors, including age (P = 0.40), sex (all were female), weight (P = 0.93), height (P = 0.86), BMI (P = 0.95), JOA score (P = 0.28), and FTA (P = 0.46) (Table 1). Table 2 shows the perioperative data, including the type of anesthesia, operating time, total blood loss, and the units transfused. Significant differences were observed between the two groups regarding the operating time, total blood loss, and units transfused. All blood transfusions were done up to 12 h after the surgery. In the U-TKA group, all transfusions used autologous blood. In the B-UKA group, nine patients were transfused with only autologous blood and one was transfused both autologous and homologous blood. Epidural anesthesia was used for postoperative analgesia on the first day after the surgery, 0.2% ropivacaine
439
hydrochloride hydrate (4 ml/h) was used in two cases and 0.25% bupivacaine hydrochloride hydrate (2 ml/h) in one case in the U-TKA group. In the B-TKA group, 0.2% ropivacaine hydrochloride hydrate (4 ml/h) was used in two cases and 0.25% bupivacaine hydrochloride hydrate in three cases (4 ml/h in one case and 2 ml/h in two cases). The amounts of flurbiprofen axetil used on the day of the operation in the two groups were 200 ± 33 mg and 176 ± 37 mg (P = 0.09), diclofenac sodium 30 ± 42 mg and 33 ± 42 mg (P = 0.87), and pentazocine 9 ± 13 and 17 ± 14 (P = 0.13), respectively. All of the patients were given loxoprofen sodium orally180 mg per a day (60 mg × 3) from POD 1. There was no significant difference between the two groups in terms of the maximum daily temperature. The maximum daily temperature levels peaked on POD 1 in both groups (37.4° ± 0.36°C in the U-TKA group and 37.5° ± 0.32°C in the B-TKA group); in each group the peak levels were significantly higher than at other points (P < 0.01). The maximum daily temperature returned to the preoperative level on POD 3 (P = 0.08) in the U-TKA group, and on POD 7 (P = 0.32) in the B-TKA group (Fig. 1A). The WBC counts peaked on POD 1 in both groups (7080 ± 1406 /μl in the U-TKA group and 9126 ± 2524 /μl Table 2. Summary of perioperative data Parameter Anesthesia type Spinal Spinal and epidural General Operating time (min) Total blood loss (g) Units transfused (no.)
U-TKA (n = 10)
B-TKA (n = 23)
P
7 3 0 53.3 ± 9.2 480 ± 110 1.6 ± 0.8
17 5 1 90.8 ± 20.8 1053 ± 302 3.2 ± 1.2
<0.01 <0.01 <0.01
Results are the number or the mean ± SD
Fig. 1. Unilateral (U-TKA) and bilateral (B-TKA) total knee arthroplasties in regard to various parameters. A Temperature. B White blood cell (WBC) count. C C-reactive protein (CRP). D Interleukin-6 (IL-6) in blood.
440
in the B-TKA group); the peak levels were significantly higher than at any other point in each group (P < 0.01). The maximum WBC count returned to the preoperative level on POD 7 in both groups (P = 0.90 and P = 0.38, respectively) (Fig. 1B). The CRP levels peaked on POD 7 in the U-TKA group and on POD 1 in the B-TKA group (3.04 ± 2.14 mg/dl in the U-TKA group and 3.86 ± 2.22 mg/dl in the B-TKA group). In both groups, the CRP levels on PODs 1 and 7 were significantly higher than they were preoperatively and on POD 14 (P < 0.01); however, there were no significant differences between the CRP levels on PODs 1 and 7 in each group (P = 0.70 and P = 0.16, respectively). In the U-TKA group, the maximum CRP returned to the preoperative level on POD 14 (P = 0.81). In the B-TKA group, however, the maximum CRP did not return to the preoperative level by POD 14 (P < 0.05) (Fig. 1C). In both groups, the blood IL-6 level was significantly higher on POD 1 than preoperatively and on POD 7 (P < 0.01) (445.1 ± 377.5 pg/ml in the U-TKA group and 1097.9 ± 839.9 pg/ml in the B-TKA group). There were no significant differences in the circulating IL-6 before surgery and on POD 7 in each group (Fig. 1D). The IL-6 level in the drainage fluid on POD 1 was 413.5 ± 109.8 ng/ml in the U-TKA group and 403.9 ± 238.1 ng/ml in the B-TKA group. There was no significant difference between the groups (P = 0.93). The relative proportions for the operating time, total blood loss, units of blood transfused, and IL-6 level in blood were around 200% (Fig. 2). At the 1-year follow-up, no patient in either group had demonstrated any wound infection.
Fig. 2. Relative proportion of B-UKA and U-TKA × 100 (%), in regard to temperature, WBC, CRP, blood IL-6, and IL-6 in drainage fluid (drain). BMI, body mass index
H. Kugisaki et al.: Interleukin-6 in TKA
Discussion In the present study, we studied the cytokine and the acute-phase response to U-TKA and the B-TKA. The model has a major advantage in that the B-TKA is associated with twice as much surgical trauma as the U-TKA. This is the first clinical comparison of the response to the surgery to elucidate the relative proportion of surgical stress. Interleukin-6 is produced mainly by monocytes and macrophages after antigen stimulation, although other cells (i.e., T lymphocytes, endothelial cells, fibroblasts) may also produce it.8 Surgery may directly induce IL-6 release by these cells, or IL-6 release may be mediated by other locally released cytokines. For example, both IL-1 and TNFα can up-regulate IL-6 synthesis in fibroblasts.9 However, concentrations of TNFα, IL-1β, and IL-8 are too low to be detected. IL-1β and TNFα are frequently difficult to detect in the systemic circulation even in severely infected patients.10 IL-6 therefore appears to be a major endogenous protein mediator of fever and the acute-phase response following surgery.9 IL-6 also plays a role in the induction and control of the acute-phase protein response, especially that of CRP synthesis by human hepatocytes.1 Andres et al.11 reported that increased local production of IL-6 at the surgical site results in the release of IL-6 into the local circulation, thus leading to leukocytosis and fever. Kragsbjerg et al.5 showed that the peak IL-6 level differed among surgical procedures, and the peak serum IL-6 concentrations occurred 6–24 h after surgery for joint replacement surgery. IL-6 is produced locally and, through circulation in the blood, transmits local information to the whole system. It has been reported that there are significant differences in the local and systemic cytokine response to surgical trauma.12,13 The most prominent response was a 1000-fold increase of IL-6 locally, represented by drained fluid, in comparison to the systemic concentrations in arterial blood of patients operated on for thoracic scoliosis.13 Therefore, the cytokine response to surgical trauma is characterized by both a local and a systemic reaction.12,13 In a previous study, the peak serum IL-6 level was 172–400 pg/ml11,14,15; and in the present study, the serum IL-6 level on POD 1 was 445 pg/ml. A number of researchers have reported that serum IL-6 reflects the kinetics of surgical trauma.4–7,11,14,15 Vita et al. showed that cytokines in drainage fluid also demonstrated this.16 In our study, the Il-6 level in drainage fluid on POD 1 was also about 1000 times that of the systemic blood, and there was no significant difference between the U-TKA and the BTKA groups or the right and left sides. Blood transfusion induced production of cytokines and cortisol. However, the cytokines returned to preinfusion levels after blood transfusion.17 Bottner et al.18
H. Kugisaki et al.: Interleukin-6 in TKA
reported that the increased serum IL-6 concentration after TKA was related to the surgical intervention itself, and the amount of transfusion did not correlate with the increase in the patient’s serum IL-6. The development of postoperative fever is common after a surgical procedure.11,19 Andres et al.11 reported that elevated serum IL-6 levels after total joint arthroplasty caused fever after TKA. Another potential cause of fever is blood transfusion,20 but this finding has not yet been substantiated in other studies of postoperative fever in this population.19,21 In the current study, body temperature significantly increased after TKA in both groups, but there was no significant difference between the groups. Therefore, fever may not be suitable for the analysis of surgical trauma after TKA. The WBC count is well known to increase and remain after surgical trauma.22 The WBC count also increases after blood transfusion, and a higher WBC count due to blood transfusion is not a transient reaction.17 In the present study, the WBC count increased significantly after TKA in both groups, although there was a significant difference between the two groups, which might have been due to the synergy of surgical trauma and blood transfusion; therefore, further investigation is required to clarify the proportions. The surgical stress of the B-TKA was twice that of the U-TKA in the current model; however, the relative population of WBCs remained at 127.0%. Therefore, the surgical stress may be underestimated when the WBC count is used to assess surgical stress. Tissue damage at operation induces an elevated serum CRP level during the following 2–3 days. If no complications occur, however, it thereafter decreases rapidly.23 The CRP level is generally used for monitoring surgical stress, and it has been used to compare different surgical procedures in many studies.6,7,24 Elevated CRP levels are seen in association with a number of ailments including infection, inflammation, and malignancy as a response to tissue injury. In the present study, the subjects did not exhibit these conditions. Larsson et al.23 reported that the CRP peak level did not correlate with the quantity of transfused blood or the ischemic period caused by the tourniquet. As with the responses of temperature and WBC, the CRP increased significantly after the TKA in both groups, but there was no significant difference between the groups. In one study the CRP level reached a maximum 48–72 h after orthopedic surgery and regained its normal level 1–3 weeks after the surgery.23,24 In the current study, the CRP level 2 weeks after surgery showed a significant difference between the two groups. Severe surgical trauma may thus prolong normalization of the CRP level. The CRP level may be suitable for qualitative assays of surgical trauma but not for quantitative assays.
441
The acute-phase response after surgery is diverse and complicated. Serum IL-6 level, CRP level, WBC count, and fever are representative markers for surgical trauma. However, regarding temperature and CRP, no significant differences were observed between the UTKA and B-TKA groups on POD 1. There was a significant difference between the U-TKA and the B-TKA groups regarding the WBC counts; however, the relative proportion remained at 128.9%. U-TKA and the B-TKA were selected for the current study because the proportion of their surgical trauma was known. Therefore, an ideal parameter for surgical trauma should show around a 200% change in the relative proportion. The serum IL-6 level, among others, may be most suitable for quantitative analysis of surgical trauma. The serum IL-6 concentration proved to discriminate the severity of the surgical procedure better than temperature, WBC count, or CRP level. However, the response to surgical trauma naturally varies for each individual.3–7,10–16 Hence, the results of the current study might be suitable for a comparison of the amplitude in surgical techniques but not for individual reactions to surgery.
Conclusion We investigated the surgical stress for patients undergoing U-TKA and B-TKA using various parameters. Body temperature, WBC counts, CRP level, and IL-6 level in blood in both groups were found to increase significantly after the surgery; however, only the WBC count and blood IL-6 level showed a significant difference in the two groups. The IL-6 level in blood is therefore considered a more suitable marker for a quantitative analysis of surgical stress. Acknowledgments. The authors thank Dr. Jun Ito, Dr. Shuya Ide, Dr. Masaru Kitajima, Dr. Kenji Ogawa, Dr. Riki Tanaka, Dr. Tomonori Tajima, and Dr. Osamu Nozaki for their valuable contributions to this study.
References 1. Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J 1990;265:621–36. 2. Tracey KJ, Morgello S, Koplin B, Fahey TJ 3rd, Fox J, Aledo A, et al. Metabolic effects of cachectin/tumor necrosis factor are modified by site of production: cachectin/tumor necrosis factorsecreting tumor in skeletal muscle induces chronic cachexia, while implantation in brain induces predominantly acute anorexia. J Clin Invest 1990;86:2014–24. 3. Kaplan E, Dinarello CA, Gelfand JA. Interleukin-1 and the response to injury. Immunol Res 1989;8:118–29. 4. Ohzato H, Yoshizaki K, Nishimoto N, Ogata A, Tagoh H, Monden M, et al. Interleukin-6 as a new indicator of inflammatory status:
442
5.
6.
7.
8. 9.
10.
11.
12.
13.
H. Kugisaki et al.: Interleukin-6 in TKA detection of serum levels of interleukin-6 and C-reactive protein after surgery. Surgery 1992;111:201–9. Kragsbjerg P, Holmberg H, Vikerfors T. Serum concentrations of interleukin-6, tumour necrosis factor-alpha, and C-reactive protein in patients undergoing major operations. Eur J Surg 1995;161:17–22. Yoshida S, Ohta J, Yamasaki K, Kamei H, Harada Y, Yahara T, et al. Effect of surgical stress on endogenous morphine and cytokine levels in the plasma after laparoscopic or open cholecystectomy. Surg Endosc 2000;14:137–40. Grande M, Tucci GF, Adorisio O, Barini A, Rulli F, Neri A, et al. Systemic acute-phase response after laparoscopic and open cholecystectomy. Surg Endosc 2002;16:313–6. Kishimoto T. The biology of interleukin-6. Blood 1989;74:1– 10. Walther Z, May LT, Sehgal PB. Transcriptional regulation of the interferon-beta 2/B cell differentiation factor BSF-2/hepatocytestimulating factor gene in human fibroblasts by other cytokines. J Immunol 1988;140:974–7. Damas P, Reuter A, Gysen P, Demonty J, Lamy M, Franchimont P. Tumor necrosis factor and interleukin-1 serum levels during sever sepsis in humans. Crit Care Med 1989;17:975–8. Andres BM, Taub DD, Gurkan I, Wenz JF. Postopetrative fever after total knee arthroplasty: the role of cytokine. Clin Orthop 2003;415:221–31. Krohn CD, Reikerås O, Aasen AO. The cytokines IL-1beta and IL-1 receptor antagonist, IL-2 and IL-2 soluble receptor-alpha, IL-6 and IL-6 soluble receptor, TNF-alpha and TNF soluble receptor I, and IL10 in drained and systemic blood after major orthopaedic surgery. Eur J Surg 1999;165:101–9. Krohn CD, Reikerås O, Mollnes TE, Aasen AO. Complement activation and release of interleukin-6 and tumor necrosis factoralpha in drained and systemic blood after major orthopaedic surgery. Eur J Surg 1998;164:103–8.
14. Wirtz DC, Heller KD, Miltner O, Zilkens KW, Wolff JM. Interleukin-6: a potential inflammatory marker after total joint replacement. Int Orthop 2004;24:194–6. 15. Hall GM, Peerbhoy D, Shenkin A, Parker CJR, Salmon P. Hip and knee arthroplasty: a comparison and the endocrine, metabolic and inflammatory responses. Clin Sci 2000;98:71–9. 16. Vita GD, Patti R, D’Agostino P, Ferlazzo V, Angileri M, Sieli G, et al. Modifications in the production of cytokines and growth factors in drainage following mesh implantation after incisional hernia repair. Am J Surg 2006;191:785–90. 17. Krohn CD, Reikerås O, Aasen AO. Inflammatory cytokines and their receptors in arterial and mixed venous blood before, during and after infusion of drained untreated blood. Transfusion Med 1999;9:125–30. 18. Bottner F, Sheth N, Chimento G, Sculco T. Cytokine levels aftert transfusion of wash wound drainage in total knee arthroplasty. J Knee Surg 2003;16:93–7. 19. Guinn S, Castro FP Jr, Garcia R, Barrack RL. Fever following total knee arthroplasty. Am J Knee Surg 1999;12:161–4. 20. Kennedy JG, Rodgers WB, Zurakowski D, Sullivan R, Griffin D, Beardsley W, et al. Pyrexia after total knee replacement: a cause for concern? Am J Orthop 1997;26:549–52, 554. 21. Shaw JA, Chung R. Febrile response after knee and hip arthroplasty. Clin Orthop 1999;367:181–9. 22. Eriksson S, Olander B, Pira U, Granstöm L. White blood cell count, leucocyte elastase activity, and serum concentrations of interleukin-6 and C-reactive protein after open appendectomy. Eur J Surg 1997;163:123–7. 23. Larsson S, Thelander U, Freiberg S. C-reactive protein (CRP) levels after elective orthopaedic surgery. Clin Orthop 1992;275: 237–42. 24. White J, Kelly M, Dunsmuir R. C-reactive protein after total hip and total knee replacement. J Bone Joint Surg Br 1998;80: 909–11.