論文ID: CJ-24-0496
Takayasu arteritis (TAK) is classified as a large vessel vasculitis and often causes vascular stenosis, occlusion, and aneurysm formation. Although the principal treatment for TAK involves suppressing inflammation with glucocorticoids, the emergence of biological disease-modifying antirheumatic drugs has considerably changed the treatment landscape of TAK in recent years. Several biological disease-modifying antirheumatic drugs, such as tocilizumab (TCZ), have shown promising effects on TAK in clinical studies. Cardiologists and cardiovascular surgeons encounter patients receiving these drugs who require catheterization, endovascular treatment, or cardiovascular surgery. However, in patients treated with glucocorticoids and TCZ, there needs to be greater awareness of more complications than usual after surgery, such as delayed wound healing, systemic infection, and surgical site infection. In addition, in patients receiving TCZ, inflammatory markers, such as C-reactive protein, may not increase when complications arise from infection. Unfortunately, there are no guidelines or solid evidence that have clearly defined the optimal perioperative treatment strategy for patients with TAK who require cardiovascular surgery. This article reviews the evidence and our recent experience supporting the perioperative use of TCZ, and proposes a protocol that can reduce complications in patients with TAK undergoing invasive cardiovascular treatment.
Takayasu arteritis (TAK), which often affects young women, is a rare disease characterized by the inflammation of large blood vessels, resulting in vascular stenosis, occlusion, and aneurysm formation.1–6 Since 2000, the age at diagnosis of TAK has become younger because of advances in imaging tools, such as computed tomography (CT) angiography, magnetic resonance imaging and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET).2 Glucocorticoids (GCs) and immunosuppressant doses have increased, but the proportion of patients requiring surgical treatment has remained the same at 22%.2 Therefore, the opportunities for cardiovascular surgery in young patients with TAK may have increased. TAK often causes aortic dilatation/aneurysm and stenosis/occlusion of the coronary arteries. Therefore, aortic replacement/repair, aortic valve replacement, coronary artery bypass graft (CABG) surgery, and endovascular treatment are often required in patients with TAK.7–13
GCs are the principal treatment for TAK.14,15 However, the emergence of biological disease-modifying antirheumatic drugs (bDMARDs), represented by tocilizumab (TCZ), has markedly changed the treatment landscape of TAK.16–21 The 2017 Japanese Circulation Society (JCS) guidelines for the management of vasculitis syndrome indicated TCZ as a Class I treatment recommendation for TAK, whereas the use of a tumor necrosis factor inhibitor was a Class IIa recommendation.14
Perioperative TAK management in patients who have cardiovascular surgery and receive TCZ can be challenging because of concerns about an increased likelihood of surgical complications, such as systemic infection, delayed wound healing, and surgical site infection (SSI), due to the administration of GCs and immunosuppressants. Furthermore, if immunosuppressant drugs are discontinued before surgery, clinicians must be aware of the potential for TAK relapse after surgery. This dilemma concerns many clinicians involved with the perioperative management of patients with TAK. At present, there are no guidelines or solid evidence that clearly define the optimal timing of perioperative discontinuation of TCZ for patients with TAK who require cardiovascular surgery. In this article, we describe perioperative management for patients with TAK who require cardiovascular surgery, focusing on GCs and TCZ. Our aim is to help cardiologists and cardiovascular surgeons, who are occasionally involved in treating patients with TAK, to better understand this disease and perioperative treatment.
TCZ is a recombinant humanized anti-interleukin-6 receptor monoclonal antibody. Patients with TAK usually self-administer 162 mg TCZ subcutaneously once a week. Current and future promising immunosuppressive therapies for TAK are listed in Table 1.22 In the 2017 JCS guidelines, only TCZ and tumor necrosis factor inhibitor are recommended as treatments for TAK; other bDMARDs are not described in the guidelines.14 Among the bDMARDs, the evidence for treating TAK is accumulating, especially for TCZ, and the results of several large-scale clinical studies investigating the efficacy of TCZ in TAK have been published.23–26 Furthermore, randomized controlled trials have suggested TCZ treatment enables the dose of GC to be reduced.17,27 Other trials have reported on the long-term safety and efficacy outcomes for TCZ in patients with TAK refractory to other conventional treatments.28,29
Current and Future Promising Immunosuppressive Therapies for TAK
Immunosuppressant | JCS recommendation | Clinical evidence |
---|---|---|
Tocilizumab | Class I | RCT17,28 |
TNF inhibitors | Class II a | RCT52 |
JAK inhibitors | Not listed | Prospective study53,54 |
Secukinumab | Not listed | Prospective study55 |
Rituximab | Not listed | Case reports56 |
JAK, Janus kinase; JCS, Japanese Circulation Society; RCT, randomized controlled trial; TAK, Takayasu arteritis; TNF, tumor necrosis factor.
TCZ has been commercially available for use in the treatment of TAK since 2017, but has been available in Japan for the treatment of patients with rheumatoid arthritis (RA) since 2008. Although most surgeries for patients with RA are orthopedic surgeries, which differ from cardiovascular surgeries in patients with TAK, the perioperative management of patients with RA can provide guidance for the perioperative management of patients with TAK treated with TCZ.30 Specifically, perioperative withdrawal of TCZ and its resumption can be considered for patients with RA. According to the 2022 American College of Rheumatology guidelines, patients on once-weekly subcutaneous TCZ injections for RA should have treatment withdrawn 2 weeks before elective total hip and knee arthroplasty.31,32 In the absence of wound healing problems, SSI, and systemic infection, TCZ can be resumed a minimum of 2 weeks after surgery.
In a previous retrospective multicenter study, Saadoun et al. analyzed the outcomes of 166 vascular procedures to manage arterial complications in patients with TAK.33 In that study, a multivariate analysis revealed that biological inflammation at the time of revascularization was independently associated with the occurrence of arterial complications after the vascular procedure.33 Other studies have also reported that active disease of TAK markedly increases the risk of postoperative complications and results in poor outcomes after surgery.34,35
The long-term outcomes of coronary artery revascularization for patients with TAK have also been reported, with CABG and percutaneous coronary intervention having a poor prognosis for revascularization during the active vs. stable stage of TAK.36 Another study reported that initial disease activity was a predictor of long-term cardiovascular outcomes in patients with TAK and severe coronary stenosis.37 Therefore, cardiovascular surgery for patients with TAK should be performed after suppressing inflammation as much as possible if the patient’s condition allows.
An association between the dosage of GCs and wound healing has been reported.38 The rate of wound complications may be increased 2- to 5-fold in patients taking chronic GCs for at least 30 days before surgery compared with those not taking GCs. In addition, complication rates may vary depending on the dose and duration of GC use, comorbidities, and type of surgery. Another study reported that the use of GCs was a risk factor for SSI, with increased doses increasing the risk.39 Finally, the use of GCs has been reported to increase the hazard ratio for postoperative death within 1 year after orthopedic surgery.40
Delayed wound healing under TCZ administration has been reported. For example, Momohara et al. reported that delayed wound healing was observed after surgery in 20 of 161 RA patients who were receiving intravenous TCZ once every 4 weeks.41 Delayed wound healing was more likely to occur after spinal surgery, but there was no significant difference in the GC dose administered.41 The time from final TCZ infusion to the surgical procedure did not differ significantly between patients with and without delayed wound healing.41 Multivariate analysis revealed that the use of prednisolone (PSL), a foot operation, and spinal surgery were factors associated with delayed wound healing.41 In a recent meta-analysis, the use of bDMARDs was a risk factor for SSI, but not for delayed wound healing and postoperative death after orthopedic surgery in patients with RA.39 These results suggest that the use of GCs and complex surgery are closely associated with delayed wound healing.
A relapse of inflammatory disease due to discontinuing TCZ is another concern. Momohara et al. reported that 36 of 161 patients experienced a relapses of RA symptoms after surgery.41 Compared with the non-relapse group, the relapse group was more likely to have a history of tumor necrosis factor inhibitor administration and a longer duration from final TCZ infusion to surgery.41 These results suggest that TCZ withdrawal for a longer period of time preoperatively may cause postoperative relapse.
We previously reported on a patient who experienced postoperative TAK relapse after a long preoperative TCZ withdrawal period.42 Briefly, our patient was a 23-year-old woman who was diagnosed with TAK. CT angiography showed severe stenosis in the ostia of the left main coronary artery. PSL was initiated at 45 mg (1 mg/kg/day). During the PSL tapering process, TCZ was administered every week because of an increase in C-reactive protein concentrations, which were thought to indicate an asymptomatic relapse of TAK. Thereafter, the dose of PSL was reduced to 10 mg, and TCZ was suspended for 5 weeks. Minimally invasive cardiac surgery (MICS)-CABG was performed because the incidence of wound infections and septic complications was expected to be lower with a thoracotomy than with a median sternotomy.43–45 In this case, because the preoperative withdrawal period was 5 weeks, TCZ had to be resumed 1 week after surgery owing to a relapse of TAK before wound healing was complete.42 After restarting TCZ, left pleural effusion increased. A thoracentesis culture was negative; therefore, the pleural effusion was thought to be due to delayed wound healing associated with the left intercostal incision by MICS-CABG. The healing process was slower than normal, and the pleural effusion took 4 weeks after surgery to disappear. In this case, wound healing was thought to be delayed because of the early resumption of TCZ.42 These findings suggest that TCZ should be restarted after the completion of wound healing.
Table 2 presents data for 8 consecutive patients with TAK who were prescribed TCZ and who underwent cardiovascular surgery at the National Cerebral and Cardiovascular Center between December 2019 and August 2023. The patients’ clinical information was collected after approval had been obtained from the research ethics committees of the National Cerebral and Cardiovascular Center (R21076-2). Informed consent was obtained in the form of opt-out on the National Cerebral and Cardiovascular Center website. Postoperative complications were observed in 3 of 8 patients; Patients 1–3 experienced postoperative complications, Patients 4–8 did not. None of the patients had elevated white blood cell counts or inflammatory markers (e.g., C-reactive protein concentrations and erythrocyte sedimentation rate). The mean PSL dose at the time of surgery was 8.7 mg/day. The mean time from the final administration of TCZ to surgery was 17.5 days, and the mean time from surgery to the resumption of TCZ was 17.0 days.
Clinical Information for 8 Consecutive Patients With TAK Who Were Prescribed TCZ and Underwent Cardiovascular Surgery at the NCVC
Patient no. | Age (years) |
Sex | Duration of TAK (years) |
Disease for surgery | Surgical procedure | Past surgical procedure | WBC (/μL) |
CRP (mg/dL) |
ESR (mm/h) |
Fibrinogen (mg/dL) |
---|---|---|---|---|---|---|---|---|---|---|
1 | 29 | F | 3 | TAA | TAR+Des Ao replacement, rt CCA reconstruction |
– | 8,730 | 0.04 | 3 | 180 |
2 | 29 | F | 2 | Severe AR, TAA, lt CCA aneurysm |
Bentall, TAR+CET, lt CCA reconstruction |
– | 3,290 | 0.01 | 7 | 169 |
3 | 57 | M | 6 | LCA pseudoaneurysm | LCA reconstruction | Bentall | 2,420 | 0.01 | NA | 120 |
4 | 78 | F | 40 | SMI | MICS-CABG | – | 8,590 | 0.02 | 5 | 305 |
5 | 26 | F | 6 | TAA | TAR+CET | – | 6,690 | 0.01 | 2 | 164 |
6 | 38 | F | 11 | AAE | Bentall | – | 6,750 | 0.01 | 3 | 195 |
7 | 60 | F | 32 | TAA | TEVAR | AAA (Y-grafting), TEVAR | 4050 | 0.01 | 3 | 147 |
8 | 53 | F | 27 | Prosthetic valve dysfunction | Re-Bentall | AVR | 5,440 | 0.03 | 4 | 195 |
Patient no. | PSL (mg) |
Immunosuppressant | Time from final TCZ administration to operation (days) |
Time from surgery to TCZ resumption (days) |
FDG uptake | Complications (excluding delayed wound healing and relapse of TAK) |
Delayed wound healing |
Relapse | ||
1 | 16 | MTX, AZA | 14 | 9 | + | Chest re-exploration for hemostasis | – | – | ||
2 | 13 | MTX | 19 | 14 | + | Chest re-exploration for hemostasis | + | – | ||
3 | 2 | MTX | 8 | 29 | NA | STEMI, bacteremia | + | – | ||
4 | 10 | – | 39 | 15 | + | – | – | – | ||
5 | 12.5 | CYA | 19 | 17 | – | – | – | – | ||
6 | 6 | – | 18 | 17 | – | – | – | – | ||
7 | 5 | – | 18 | 14 | NA | – | – | – | ||
8 | 5 | – | 5 | 21 | – | – | – | – |
Patients 1–3 experienced complications, Patients 4–8 did not. AAA, abdominal aortic aneurysm; AAE, annulo-aortic ectasia; AR, aortic regurgitation; AVR, aortic valve replacement; AZA, azathioprine; CABG, coronary artery bypass grafting; CCA, common carotid artery; CET, conventional elephant trunk; CRP, C-reactive protein; CYA, cyclosporine A; Des Ao, descending aorta; ESR, erythrocyte sedimentation rate; FDG, 18F-fluorodeoxyglucose; LCA, left coronary artery; lt, left; MICS, minimally invasive cardiac surgery; MTX, methotrexate; NA, not available; NCVC, National Cerebral and Cardiovascular Center; rt, right; SMI, silent myocardial ischemia; STEMI, ST-elevation myocardial infarction; TAA, thoracic aortic aneurysm; TAK, Takayasu arteritis; TAR, total arch replacement; TCZ, tocilizumab; TEVAR, thoracic endovascular aortic repair; WBC, white blood cell.
Patient 1 had extremely refractory TAK. Despite the administration of GCs and TCZ, the thoracic aortic aneurysm expanded. Then, immunosuppressant combination therapy with azathioprine and methotrexate (MTX) was administered. Even so, the diameter of the ascending aorta expanded to 58 mm×62 mm (Figure 1A) when the patient was referred to our hospital for evaluation of surgical indications. Although FDG-PET showed heterogeneous FDG accumulation in the ascending aorta that became clear in delayed scans (Figure 1B), the degree of accumulation had weakened compared with FDG accumulation prior to initiation of immunosuppressant combination therapy. Because there was a risk that the thoracic aortic aneurysm would rupture while TAK inflammation was being brought under control, surgery was performed. The day after surgery, the patient required chest re-exploration for hemostasis. However, it has been 4 years since Patient 1 was discharged, and the TAK inflammation has been controlled without the need for further surgery.
Chest computed tomography and 18F-fluorodeoxyglucose (FDG) positron emission tomography images of Patient 1. (A) The ascending aorta expanded to 58 mm×62 mm. (B) Heterogeneous FDG accumulation in the aorta.
Patient 2 also had refractory TAK. Despite treatment with GCs and MTX, TAK activity was not under control. TCZ administration improved FDG-PET uptake, but it was still evident (Figure 2A) at the time when Patient 2 was referred to our hospital. Moreover, the ascending aortic aneurysm had expanded to 57 mm×53 mm (Figure 2B), and the patient developed severe aortic regurgitation (Figure 2C) and heart failure symptoms, so surgery was performed. The day after surgery, Patient 2 required chest re-exploration for hemostasis. The postoperative course was uneventful. Three years have passed since the surgery, and there has been no recurrence of TAK and no symptoms of heart failure.
18F-Fluorodeoxyglucose positron emission tomography (FDG-PET) images, chest computed tomography, and echocardiography images of Patient 2. (A) FDG-PET uptake remained in the aorta. (B) The ascending aortic aneurysm expanded to 57 mm×53 mm. (C) Echocardiography showing severe aortic regurgitation.
Patient 4 is a newly diagnosed TAK patient. Coronary angiography revealed complete occlusion of the right coronary artery and 90% stenosis of the left anterior descending artery (Figure 3A,B). Although immediate CABG was considered, GCs and TCZ were administered based on evidence that it is better to suppress inflammation before surgery. After 4 months treatment with GCs and TCZ, FDG-PET uptake was still observed but was showing signs of improvement (Figure 3C,D). MICS-CABG was performed without any related complications.
Coronary angiography and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) images of Patient 4. (A) There is 100% occlusion the right coronary artery (right anterior oblique view). Good collateral flow is observed from the conus branch to the posterolateral branch. (B) There is a severe (90%) stenosis in the proximal of the left coronary artery (cranial view). (C) A strong FDG-PET signal is observed in the aorta before medical therapy. (D) After initiating treatment with glucocorticoids and tocilizumab, FDG-PET uptake is still observed but is showing signs of improvement.
At present, it is difficult to accurately define remission of TAK.46 In previous studies, FDG-PET had the potential to monitor disease activity in TAK.47–51 Of our 3 patients with postoperative complications (Patients 1–3; Table 2), 1 did not undergo preoperative FDG-PET and the other 2 showed FDG-PET uptake in the aorta and its major branch arteries. Of our 5 patients without postoperative complications (Patients 4–8; Table 2), 1 did not undergo preoperative FDG-PET and only 1 of the others showed FDG-PET uptake. These results suggest that complications are more likely to occur in patients whose preoperative TAK activity is not sufficiently controlled. In addition, the 3 patients with complications were characterized by the concomitant use of immunosuppressants (e.g., MTX) other than TCZ and a short disease duration. Patients 1 and 2 had rapidly enlarging aortic aneurysms, and Patient 3 was undergoing reoperation for a coronary artery aneurysm after a Bentall procedure. The complications in these patients were thought to be due to the extreme difficulty in controlling inflammation even with the use of MTX and TCZ. Delayed wound healing was observed in 2 patients with complications, but was not observed in any patients without complications. In addition, there did not appear to be any noticeable difference in the time of preoperative discontinuation or the number of days before restarting TCZ between patients with and without complications. A relapse of TAK was not observed in any of the patients.
In summary, the following patients with TAK are more likely to experience postoperative complications:
• those who undergo surgery without adequate suppression of TAK inflammation
• those who receive high doses of GCs before surgery, causing delayed wound healing and SSI
• those who do not discontinue TCZ, which causes SSI
• those with a long TCZ withdrawal period before surgery, causing postoperative relapse of TAK
• those who restart TCZ before wounds have healed, resulting in delayed wound healing.
Therefore, to reduce perioperative complications, clinicians should perform the following procedures before cardiac surgery in patients with TAK:
• administer TCZ in the early stage of the disease and reduce the GC dose as much as possible
• perform FDG-PET before surgery to confirm the absence of active inflammation
• suspend TCZ 2–3 weeks before cardiovascular surgery and resume TCZ after wounds have healed
• consider MICS to reduce the extent of surgical wounds and the risk of infection.
Based on our experience, we propose a treatment protocol during the perioperative period as shown in Figure 4. Please note that this proposal is a suggestion only for cardiovascular surgery, and there is no consensus regarding the TCZ withdrawal period for minimally invasive catheter procedures such as percutaneous coronary intervention, percutaneous transluminal renal angioplasty, and endovascular treatment. Patients with TAK undergoing cardiovascular surgery should be administered TCZ and the dose of GCs should be reduced as fast as possible. After confirming the induction of TAK remission in patients, TCZ should be discontinued 2–3 weeks before surgery. After confirming postoperative wound healing, TCZ should be resumed 2–3 weeks after surgery. We believe this protocol will be able to reduce postoperative complications and TAK relapse associated with the discontinuation of TCZ.
Proposed treatment protocol for Takayasu arteritis (TAK). In patients with TAK undergoing cardiovascular surgery, the dose of glucocorticoids (GCs) should be reduced as much as possible. The use of tocilizumab (TCZ) should be suspended for 2–3 weeks before surgery. After cardiovascular surgery, wound healing should be confirmed and TCZ should be resumed 2–3 weeks after surgery.
The authors thank Ellen Knapp, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
This work was supported by grants from the Ministry of Health, Labour and Welfare, Japan (23FC1019).
Y.N. has received lecture fees and research grants from Chugai, and research grants from AbbVie. The other authors have no conflict of interests to disclose.
Approval to collect patient information was obtained from the research ethics committees of the National Cerebral and Cardiovascular Center (R21076-2).