Total knee arthroplasty (TKA) and total hip arthroplasty (THA) are 2 of the most commonly performed surgical procedures in the United States. It is estimated that by the year 2030 the number of primary TKA and THA performed annually will reach 4 million.1 This number reflects an increase of 673% for TKA and 176% for THA. Not unexpectedly, the number of revision procedures is estimated to increase by a similar percentage.2 Surgical-site infection (SSI) is one of the most challenging and costly complications associated with total joint arthroplasty. The relative incidence of periprosthetic joint infection (PJI) ranges from 2.0% to 2.4% for TKA and THA, with a corresponding cost to US hospitals of >$566 million dollars.3,4 As the number of total joint arthroplasties increases, the cost of SSI and PJI is projected to exceed $1.62 billion by 2020.
A previous large, case-controlled study of patients undergoing THA at the Mayo Clinic reported that the 4 risk factors most predictive of PJI were (1) postoperative SSI (odds ratio [OR], 35.9); (2) a National Nosocomial Infection Surveillance score >2.0 (OR, 3.9); (3) concurrent malignancy (OR, 3.1); and (4) previous arthroplasty (OR, 2.0). Other identified risk factors include rheumatoid arthritis, diabetes mellitus, obesity, and steroid therapy.5 A review of a large health insurance database in Taiwan concluded that TKA and THA performed under general anesthesia were associated with a greater risk of SSI within the first 30 days compared with the same procedure performed under neuraxial (spinal or epidural) anesthesia.6 Since this first publication by Chang et al.,6 other investigators have published similar studies yielding conflicting results.7–9 The primary aim of this case-controlled study was to compare the risk of SSI for patients undergoing primary TKA or THA and revision TKA or THA under general anesthesia versus neuraxial anesthesia. Our secondary aim was to determine which patient, anesthetic, and surgical variables influence the risk of SSI. We hypothesized that patients who undergo neuraxial anesthesia may have a lesser risk of SSI compared with those who had a general anesthetic. In addition, we hypothesized that the postoperative analgesia achieved with peripheral nerve blocks may have a beneficial effect on the incidence of SSI that may be protective even in patients undergoing general anesthesia.
This retrospective study was conducted at the Mayo Clinic in Rochester, Minnesota, a tertiary care academic medical center in the United States. The study was approved by the Mayo Clinic IRB. Patients who denied research authorization were excluded according to Minnesota State government statute. We conducted a single-center, retrospective, case-control study of patients undergoing primary or revision TKA or THA at the Mayo Clinic between January 1, 1998, and December 31, 2008, and who were subsequently diagnosed with SSI within a year postoperatively. All patients at least 18 years of age who underwent a primary or revision TKA or THA at Mayo Clinic from January 1, 1998, to December 31, 2008, were identified retrospectively via the Mayo Clinic Total Joint Registry, the master diagnostic index, and the clinical microbiology laboratory. The Mayo Clinic Total Joint Registry is a previously validated10 and comprehensive repository of data collected for each joint replacement surgery performed at Mayo Clinic since 1969. Data included within the registry have been defined prospectively and are collected by manual chart review, written patient questionnaire, and follow-up telephone surveys conducted by full-time research assistants unaware of the study hypotheses. Follow-up is performed twice in the first postsurgical year and then once every 5 years thereafter. When follow-up examination at the Mayo Clinic was not possible, the patient completed a standardized data collection form, and roentgenograms of the joint prosthesis were sent to the Mayo Clinic for review. By using this system, >90% of this cohort has been followed.
The primary outcome variable was the diagnosis of SSI occurring early (within 30 days of surgery) or late (>30 days and ≤1 year) after arthroplasty. These times were defined by the Centers for Disease Control definition of SSI, superficial SSI as those reported within 30 days of procedure, as well as deep and organ space infections being those reported within 30 days of procedure and at 1 year if implant was in place.11 Potential SSI cases were identified using the Mayo Clinic Total Joint Registry and the infection control SSI institutional database. Members of the Anesthesia Clinical Research Unit subsequently verified all identified cases using manual chart review. Using a 1:2 matched case/control design, each case was matched with 2 controls based on type of procedure (primary, revision, bilateral), joint site (TKA, THA), sex (male, female), surgical duration (<3 hours, ≥3 hours), and ASA physical status (I and II versus III, IV, and V). In addition to control for the change in practice that likely occurred during the 11-year study period, the surgical dates of the cases and controls were matched within 1 year.
Patient demographics, including age, sex, ethnicity, and ASA physical status, were recorded. Patient comorbidities, including body mass index (BMI), underlying joint disease, diabetes mellitus, rheumatoid arthritis, malignancy, liver cirrhosis, chronic renal disease, steroid use, and current tobacco use, were identified from the patient’s medical record. Details of the anesthetic management also were recorded, including type of anesthesia (general, epidural, spinal); postoperative pain management (single injection versus continuous peripheral nerve block, epidural analgesia, and patient-controlled opioid analgesia); temperature at the beginning, middle, and end of the surgical procedure; and antibiotic administration. Details of the surgical procedure also were documented, including type of surgery (primary or revision TKA or THA), surgeon, urgency (elective versus emergency), unilateral or bilateral arthroplasty, total tourniquet time, and duration of surgery. The National Nosocomial Infection Surveillance system surgical patient risk scores were calculated from 0 to 3 points, with 1 point each for ASA physical status score >3; operating room time >3 hours; and wound class ≥3.5 Details of the postoperative period also were documented, including the use of anticoagulant medications; presence of a superficial, deep, or organ space SSI (Centers for Disease Control definition; Table 1); and the microbiology.
In a review of a large health insurance database, Chang et al.6 reported that the use of general anesthesia more than doubled the odds (OR = 2.21) for SSI after TKA and THA compared with epidural or spinal anesthesia. The decision to select 2 controls for each case for the present investigation was made on the assumption that approximately 200 cases would be identified and that general anesthesia is used for approximately 60% of the given patient population at our institution. Under these assumptions, we determined that a case-control study with 200 cases and 400 controls would provide statistical power (2-tailed, α = 0.05) of approximately 80% to detect an OR of 1.6 when assessing whether general anesthesia is associated with an increased risk of SSI compared with neuraxial techniques.
Data were analyzed using conditional logistic regression making use of the 1:2 matched set study design. In addition to assessing the primary risk factor of interest (neuraxial versus general anesthesia), a number of secondary risk factors were assessed. These secondary risk factors included patient age, BMI, smoking status, blood transfusions, peripheral nerve blockade, and preoperative antibiotic use. Both univariate and multivariate analyses were performed. Because cases and controls were matched on ASA physical status, specific patient comorbidities were not assessed as risk factors but were included as additional covariates in the multivariable model. In addition to the overall analysis, supplemental analyses were performed separately for superficial and deep/organ space infections and for early (≤30 days) and late (31–365 days) infections. In all cases, findings are summarized using ORs and corresponding 95% confidence intervals (CIs). Two-tailed P values <0.05 were considered statistically significant.
Between January 1, 1998, and December 31, 2008, 202 SSIs (74 superficial, 16 deep, and 112 organ space) were identified in 202 patients undergoing primary or revision TKA and THA at Mayo Clinic. Demographics are listed in Table 2. Of the infections identified, 115 (57%) occurred within the first 30 days (early) and 87 (43%) occurred between 31 and 365 days postoperatively (late). The variables analyzed as potential risk factors for SSI are listed in Table 3 along with the results of the overall conditional logistic regression analyses. From both univariate and multivariable analyses, no significant association was found between the use of central neuraxial anesthesia and SSI (univariate OR = 0.92, 95% CI 0.63–1.34, P = 0.651; multivariable OR = 1.10, 95% CI 0.72–1.69, P = 0.664). The use of peripheral nerve block also was not found to influence the risk of SSI (univariate OR = 1.41, 95% CI 0.84–2.37, P = 0.193; multivariable OR = 1.35, 95% CI 0.75–2.44, P = 0.312). The factors that were found to be associated with SSI in multivariable analysis included current smoking (OR = 5.10, 95% CI 2.30–11.33) and greater BMI (OR = 2.68, 95% CI 1.42–5.06 for BMI ≥ 35 kg/m2 compared with those with BMI < 25 kg/m2). Similar findings were observed when the analyses were performed separately for superficial and deep/organ space infections (Table 4) and also when analyses were performed separately for early and late infections (Table 5).
Despite advances in infection control practices, including sterilization methods, barriers, surgical technique, operating room ventilation, and antibiotic prophylaxis, SSI continues to result in major postoperative morbidity and mortality. The National Healthcare Safety Network data for 2006–2008 reported an overall rate of SSI to be 1.9%,12 which makes SSI the most common cause of hospital-acquired infections, accounting for 31% of all hospital-acquired infections in hospitalized patients.13
The prevention of SSI is a multidisciplinary effort, and many of the prevention strategies are initiated in the operating room. The anesthesiologist is involved directly with the timing of antibiotic administration, perioperative hyperoxia, normothermia, glucose management, blood transfusion, and the use of regional anesthesia (neuraxial blocks ± peripheral nerve blocks) versus general anesthesia. Recent studies designed to determine the relationship between the type of anesthesia (general versus neuraxial) and the rates of SSI have yielded conflicting results. A review of 3081 patients sampled from the Longitudinal Heath Insurance Database of Taiwan published by Chang et al.6 was the first to report a greater risk of SSI in patients undergoing a TKA or THA under general anesthesia when compared with epidural or spinal anesthesia. A subsequent review of The American College of Surgeons, National Surgical Quality Improvement Program, involving 14,502 cases published by Pugely et al.14 demonstrated that patients who underwent primary TKA and were managed with general anesthesia had more superficial wound infections and a statistically significant increase in overall complications compared with those undergoing spinal anesthesia. Two additional studies using large administrative databases (National Surgical Quality Improvement Program; Premier Perspectives Database) were unable to confirm the findings specific to SSI risk published by Chang et al., although they did conclude that the use of neuraxial anesthesia compared with general anesthesia may be associated with a decreased incidence of overall systemic infection (SSI, sepsis, urinary tract infection, pneumonia, etc.) in patients undergoing total joint arthroplasty.6–8,14 Two other large retrospective studies, one using a large joint registry and the other using the Premier Database, were unable to document a statistically significant difference in the risk of SSI based on the choice of primary anesthesia.15,16 Similarly, this case-controlled study was unable to support the hypothesis that the type of primary anesthetic could influence the incidence of SSI in patients undergoing total joint arthroplasty.
Studies using large administrative databases have several limitations inherent to their use of secondary data. First, detailed information regarding perioperative management (e.g., blood loss, hypothermia, antibiotic administration) that is known to influence the rate of SSI is not included within the databases. Second, the definition of comorbidities (i.e., diabetes mellitus, obesity, and smoking), surgical pathology, and complications are based on the International Classification of Diseases and Injuries, version 9 coding system, which may lead to coding bias, despite the quality checks performed by the database management. Third, most databases only capture SSIs if they occur during the index admission8 or within 30 days postoperatively (6, 7, and 15). The current case-controlled study design is well suited to investigate rare outcomes or outcomes with a long latency period, such as SSI because subjects are selected from the beginning based on their long-term outcome status. Thus, compared with large database cohort studies, case-control studies require fewer subjects and allow for multiple exposures or risk factors to be assessed for one outcome.
The mechanism by which neuraxial anesthesia may decrease the risk of SSI remains speculative, although 3 mechanisms have been proposed: (1) the autonomic response incited from severe postoperative pain causes vasoconstriction and subsequently decreases peripheral perfusion. Severe surgical pain has been shown to decrease tissue oxygenation by approximately 15 mm Hg.17,18 Therefore, providing excellent postoperative analgesia with neuraxial or peripheral nerve blocks may reduce the risk of SSI. (2) Neuraxial anesthesia may increase tissue oxygenation via the resultant vasodilatation. Several studies comparing epidural and general anesthesia have documented a small increase (10 mm Hg) in tissue oxygenation19–21 and one reported no effect.22 These studies investigated the degree of vasodilatation with epidural anesthesia compared with general anesthesia. It is theoretically possible that the degree of vasodilatation may be greater in patients undergoing spinal anesthesia. (3) Finally, neuraxial anesthesia has been shown to reduce the inflammatory response to surgery by decreasing the generalized pain response. By reducing these nonspecific responses, the immune system is better able to focus on the task of fighting bacteria. In addition, volatile anesthetics and opioids have been shown to impair neutrophil, macrophage, dendritic cell, T-cell, and natural killer cell functions, which may decrease the host’s defense mechanism. One could speculate that through similar mechanisms, and an overall reduction in the inflammatory response because of better analgesia, peripheral nerve blocks could be protective as well. To our knowledge, this is the first study that specifically evaluated the impact of peripheral nerve blocks on SSI.
Previously identified risk factors for SSI include male gender, minority race, diabetes mellitus, malignancy, rheumatoid arthritis, steroid use, current smoking, and younger age.5,23–25 The prevalence of obesity is reaching epidemic proportions in the United States, with more than one-third of adults having a BMI >30. At least half of the patients who undergo TKA and one-third of patients who undergo THA are obese.26,27 It has become exceedingly clear that the risk of SSI increases as a patient’s BMI increases.28–30 We similarly concluded that as BMI increased, so did the risk of developing an SSI. Patients with a BMI of ≥35 had a 2.68 times greater risk (OR, 2.68; 95% CI, 1.42–5.06) of developing an infection compared with patients with a BMI <25. Obesity is a proinflammatory state that is associated with a low-grade inflammatory response.31 A recent review identified enhanced cytokine activity in obese patients who underwent THA and concluded that obesity may affect the postoperative immune response and increase the risk of SSI.32 In addition, obese patients have more comorbidities, and it is likely that these comorbidities ultimately increase the risk of complications, including SSI.
In this study, active smokers undergoing THA or TKA were 5 times more likely to develop an SSI than nonsmokers. A recent meta-analysis of 107 studies by Grønkjær et al.,33 including 11 with orthopedic surgeries, revealed that smokers were more likely to develop wound healing complications than nonsmokers (Relative Risk [RR] = 2.15; 95% CI, 1.87–2.49). It is postulated that smoking may impair wound healing by decreasing tissue oxygenation, reduction in inflammatory cell chemotactic responsiveness, and oxidative bactericidal mechanisms, as well as downregulating collagen synthesis and deposition. Although our study did not assess the impact of smoking cessation on the risk of SSI, it is well established that smoking cessation restores tissue oxygenation and inflammatory cell response within 4 weeks.34
This study was retrospective and therefore subject to inherent limitations. In any study involving perioperative mortality, it is important to consider the source and integrity of the database to minimize bias. The Mayo Clinic Total Joint Registry has been validated previously, so it is likely that all cases of infection were identified. Mayo Clinic is a tertiary referral center and despite the advantage of consistent data collection, the results may not be applicable to the general population. Likewise, the statistical power for assessing the association of a specific patient/procedural factor with SSI is dependent on the prevalence of the risk factor among patients who experienced the SSI. As a result, nonstatistically significant findings should be interpreted with caution when the number of patients with/without the risk factor is small.
Despite the many other potential benefits of regional anesthesia (peripheral blocks and neuraxial blocks), this retrospective, case-controlled study found no difference in the incidence of SSI in patients who undergo total joint arthroplasty under general versus neuraxial anesthesia. We also reported that the use of peripheral nerve blocks for postoperative analgesia does not influence the incidence of SSI. Increased BMI and current smoking were found to significantly increase the risk of SSI. Further research and ideally prospective studies are needed to help answer this very important question.
Name: Sandra L. Kopp, MD.
Contribution: This author helped design the study, conduct the study, collect the data, analyze the data, and prepare the manuscript.
Attestation: Sandra L. Kopp approved the final manuscript. Sandra L. Kopp attests to the integrity of the original data and the analysis reported in this manuscript. Sandra L. Kopp is the archival author.
Name: Elie F. Berbari, MD.
Contribution: This author helped design the study, conduct the study, collect the data, analyze the data, and prepare the manuscript.
Attestation: Eli F. Berbari approved the final manuscript. Eli F. Berbari attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Douglas R. Osmon, MD, MPH.
Contribution: This author helped design the study, conduct of the study, and prepare the manuscript.
Attestation: Douglas R. Osmon approved the final manuscript. Douglas R. Osmon attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Darrell R. Schroeder, MS.
Contribution: This author helped design the study, collect the data, analyze the data, and prepare the manuscript.
Attestation: Darrell R. Schroeder approved the final manuscript. Darrell R. Schroeder attests to the integrity of the original data and the analysis reported in this manuscript.
Name: James R. Hebl, MD.
Contribution: This author helped design the study and prepare the manuscript.
Attestation: James R. Hebl approved the final manuscript.
Name: Terese T. Horlocker, MD.
Contribution: This author helped design the study and prepare the manuscript.
Attestation: Terese T. Horlocker approved the final manuscript.
Name: Arlen D. Hanssen, MD.
Contribution: This author helped design the study, conduct the study, and prepare the manuscript.
Attestation: Arlen D. Hanssen approved the final manuscript.
RECUSE NOTE Dr. Terese Horlocker is the Section Editor for Regional Anesthesia for Anesthesia & Analgesia. This manuscript was handled by Dr. Sorin J. Brull, and Dr. Horlocker was not involved in any way with the editorial process or decision.
1. Kurtz SM, Ong KL, Schmier J, Mowat F, Saleh K, Dybvik E, Kärrholm J, Garellick G, Havelin LI, Furnes O, Malchau H, Lau E. Future clinical and economic impact of revision total hip and knee arthroplasty. J Bone Joint Surg Am. 2007;89(suppl 3):144–51
2. Kurtz S, Lau E, Halpern M, Ong K. Trend shows growing orthopedic surgery case load. Will surgeons be able to keep up? Mater Manag Health Care. 2006;15:61–2
3. Kurtz SM, Lau E, Watson H, Schmier JK, Parvizi J. Economic burden of periprosthetic joint infection in the United States. J Arthroplasty. 2012;27:61–5.e1
4. Lentino JR. Prosthetic joint infections: bane of orthopedists, challenge for infectious disease specialists. Clin Infect Dis. 2003;36:1157–61
5. Berbari EF, Hanssen AD, Duffy MC, Steckelberg JM, Ilstrup DM, Harmsen WS, Osmon DR. Risk factors for prosthetic joint infection: case-control study. Clin Infect Dis. 1998;27:1247–54
6. Chang CC, Lin HC, Lin HW, Lin HC. Anesthetic management and surgical site infections in total hip or knee replacement: a population-based study. Anesthesiology. 2010;113:279–84
7. Liu J, Ma C, Elkassabany N, Fleisher LA, Neuman MD. Neuraxial anesthesia decreases postoperative systemic infection risk compared with general anesthesia in knee arthroplasty. Anesth Analg. 2013;117:1010–6
8. Poultsides LA, Ma Y, Della Valle AG, Chiu YL, Sculco TP, Memtsoudis SG. In-hospital surgical site infections after primary hip and knee arthroplasty—incidence and risk factors. J Arthroplasty. 2013;28:385–9
9. Pugely AJ, Martin CT, Gao Y, Mendoza-Lattes SA. Outpatient surgery reduces short-term complications in lumbar discectomy: an analysis of 4310 patients from the ACS-NSQIP database. Spine (Phila Pa 1976). 2013;38:264–71
10. Mantilla CB, Horlocker TT, Schroeder DR, Berry DJ, Brown DL. Frequency of myocardial infarction, pulmonary embolism, deep venous thrombosis, and death following primary hip or knee arthroplasty. Anesthesiology. 2002;96:1140–6
11. Horan TC, Gaynes RP, Martone WJ, Jarvis WR, Emori TG. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol. 1992;13:606–8
12. Mu Y, Edwards JR, Horan TC, Berrios-Torres SI, Fridkin SK. Improving risk-adjusted measures of surgical site infection for the national healthcare safety network. Infect Control Hosp Epidemiol. 2011;32:970–86
13. Magill SS, Hellinger W, Cohen J, Kay R, Bailey C, Boland B, Carey D, de Guzman J, Dominguez K, Edwards J, Goraczewski L, Horan T, Miller M, Phelps M, Saltford R, Seibert J, Smith B, Starling P, Viergutz B, Walsh K, Rathore M, Guzman N, Fridkin S. Prevalence of healthcare-associated infections in acute care hospitals in Jacksonville, Florida. Infect Control Hosp Epidemiol. 2012;33:283–91
14. Pugely AJ, Martin CT, Gao Y, Mendoza-Lattes S, Callaghan JJ. Differences in short-term complications between spinal and general anesthesia for primary total knee arthroplasty. J Bone Joint Surg Am. 2013;95:193–9
15. Namba RS, Inacio MC, Paxton EW. Risk factors associated with deep surgical site infections after primary total knee arthroplasty: an analysis of 56,216 knees. J Bone Joint Surg Am. 2013;95:775–82
16. Stundner O, Chiu YL, Sun X, Mazumdar M, Fleischut P, Poultsides L, Gerner P, Fritsch G, Memtsoudis SG. Comparative perioperative outcomes associated with neuraxial versus general anesthesia for simultaneous bilateral total knee arthroplasty. Reg Anesth Pain Med. 2012;37:638–44
17. Sessler DI. Neuraxial anesthesia and surgical site infection. Anesthesiology. 2010;113:265–7
18. Akça O, Melischek M, Scheck T, Hellwagner K, Arkiliç CF, Kurz A, Kapral S, Heinz T, Lackner FX, Sessler DI. Postoperative pain and subcutaneous oxygen tension. Lancet. 1999;354:41–2
19. Buggy DJ, Doherty WL, Hart EM, Pallett EJ. Postoperative wound oxygen tension with epidural or intravenous analgesia: a prospective, randomized, single-blind clinical trial. Anesthesiology. 2002;97:952–8
20. Kabon B, Fleischmann E, Treschan T, Taguchi A, Kapral S, Kurz A. Thoracic epidural anesthesia increases tissue oxygenation during major abdominal surgery. Anesth Analg. 2003;97:1812–7
21. Treschan TA, Taguchi A, Ali SZ, Sharma N, Kabon B, Sessler DI, Kurz A. The effects of epidural and general anesthesia on tissue oxygenation. Anesth Analg. 2003;96:1553–7
22. Rosenberg J, Pedersen U, Erichsen CJ, Vibits H, Moesgaard F, Kehlet H. Effect of epidural blockade and oxygen therapy on changes in subcutaneous oxygen tension after abdominal surgery. J Surg Res. 1994;56:72–6
23. Malinzak RA, Ritter MA, Berend ME, Meding JB, Olberding EM, Davis KE. Morbidly obese, diabetic, younger, and unilateral joint arthroplasty patients have elevated total joint arthroplasty infection rates. J Arthroplasty. 2009;24:84–8
24. Masgala A, Chronopoulos E, Nikolopoulos G, Sourlas J, Lallos S, Brilakis E, Lazarettos J, Efstathopoulos N. Risk factors affecting the incidence of infection after orthopaedic surgery: the role of chemoprophylaxis. Cent Eur J Public Health. 2012;20:252–6
25. Singh JA, Houston TK, Ponce BA, Maddox G, Bishop MJ, Richman J, Campagna EJ, Henderson WG, Hawn MT. Smoking as a risk factor for short-term outcomes following primary total hip and total knee replacement in veterans. Arthritis Care Res (Hoboken). 2011;63:1365–74
26. Batsis JA, Naessens JM, Keegan MT, Huddleston PM, Wagie AE, Huddleston JM. Body mass index and the impact on hospital resource use in patients undergoing total knee arthroplasty. J Arthroplasty. 2010;25:1250–7.e1
27. Namba RS, Paxton L, Fithian DC, Stone ML. Obesity and perioperative morbidity in total hip and total knee arthroplasty patients. J Arthroplasty. 2005;20:46–50
28. Bozic KJ, Ward DT, Lau EC, Chan V, Wetters NG, Naziri Q, Odum S, Fehring TK, Mont MA, Gioe TJ, Della Valle CJ. Risk factors for periprosthetic joint infection following primary total hip arthroplasty: a case control study. J Arthroplasty. 2014;29:154–6
29. Everhart JS, Altneu E, Calhoun JH. Medical comorbidities are independent preoperative risk factors for surgical infection after total joint arthroplasty. Clin Orthop Relat Res. 2013;471:3112–9
30. Kerkhoffs GM, Servien E, Dunn W, Dahm D, Bramer JA, Haverkamp D. The influence of obesity on the complication rate and outcome of total knee arthroplasty: a meta-analysis and systematic literature review. J Bone Joint Surg Am. 2012;94:1839–44
31. Galic S, Oakhill JS, Steinberg GR. Adipose tissue as an endocrine organ. Mol Cell Endocrinol. 2010;316:129–39
32. Wolff AS, Sarkadi AK, Maródi L, Kärner J, Orlova E, Oftedal BE, Kisand K, Oláh E, Meloni A, Myhre AG, Husebye ES, Motaghedi R, Perheentupa J, Peterson P, Willcox N, Meager A. Anti-cytokine autoantibodies preceding onset of autoimmune polyendocrine syndrome type I features in early childhood. J Clin Immunol. 2013;33:1341–8
33. Grønkjær M, Eliasen M, Skov-Ettrup LS, Tolstrup JS, Christiansen AH, Mikkelsen SS, Becker U, Flensborg-Madsen T. Preoperative smoking status and postoperative complications: a systematic review and meta-analysis. Ann Surg. 2014;259:52–71
34. Sørensen LT. Wound healing and infection in surgery: the pathophysiological impact of smoking, smoking cessation, and nicotine replacement therapy: a systematic review. Ann Surg. 2012;255:1069–79