Arthritis Care & Research Vol. 64, No. 8, August 2012, pp 1256 –1260 DOI 10.1002/acr.21670 © 2012, American College of Rheumatology

BRIEF REPORT

Cardiac Valve Replacement in Patients With Antiphospholipid Syndrome JOSE´-GABRIEL ERDOZAIN,1 GUILLERMO RUIZ-IRASTORZA,1 MARIA-ISABEL SEGURA,2 MARY-CARMEN AMIGO,3 GERARD ESPINOSA,4 JOSE-LUIS POMAR,4 IGNACIO PEREZ-VALERO,5 OIER ATEKA-BARRUTIA,5 AND MUNTHER A. KHAMASHTA5 Objective. To analyze the results of cardiac valve replacement in a multicenter cohort of patients with antiphospholipid syndrome (APS) and to identify prognostic factors of poor outcome. Methods. We performed a retrospective analysis of clinical manifestations (cardiac involvement and APS characteristics), operative and early postoperative courses, and long-term followup. All of the patients fulfilled the Sapporo criteria for APS. Logistic regression analyses were performed to identify those variables associated with adverse outcomes. Results. Between 1981 and 2008, 33 valvular replacements were carried out in 32 patients with APS. The mean ⴞ SD age at the time of surgery was 43.09 ⴞ 14.08 years. Thirty patients were women. Primary APS was present in 21 patients. The median followup time after surgery was 33.5 months (range 0 –192 months). The mitral valve was the most frequently replaced (22 of 33). Mechanical valve replacement was performed in 23 patients (71.9%). The mortality rate was 12.5% (1 cardiogenic shock, 1 septic shock, 1 following renal transplantation, and 1 hemorrhagic stroke). Fourteen patients experienced 20 complications (8 major bleeding, 5 thrombotic events, 2 valvular deteriorations, 2 third-degree atrioventricular block, 1 endocarditis, 1 cardiac tamponade, and 1 cardiac failure). Fifty percent of the patients had an uneventful outcome. Conclusion. Morbidity and mortality were high in APS patients undergoing valve replacement surgery. Most complications were related to thrombosis and bleeding. Anticoagulation must be carefully monitored to prevent hemorrhagic and thrombotic complications.

Introduction Antiphospholipid syndrome (APS) is defined by thrombosis (venous, arterial, or small vessel) and/or pregnancy morbidity (recurrent miscarriage, fetal loss, or placental insufficiency) occurring in the presence of persistently positive antiphospholipid antibodies (aPL): lupus anti1 Jose´-Gabriel Erdozain, MD, Guillermo Ruiz-Irastorza, MD, PhD: Hospital Universitario Cruces, Barakaldo, and Universidad del Paı´s Vasco/Euskal Herriko Unibertsitatea, Bizkaia, Spain; 2Maria-Isabel Segura, MD: Instituto Nacional de Cardiologı´a I. Cha´vez, Distrito Federal, Mexico; 3 Mary-Carmen Amigo, MD: Instituto Nacional de Cardiologı´a I. Cha´vez and Centro Medico ABC, Distrito Federal, Mexico; 4Gerard Espinosa, MD, PhD, Jose-Luis Pomar, MD: Hospital Clinic, Barcelona, Spain; 5Ignacio Perez-Valero, MD, Oier Ateka-Barrutia, MD, Munther A. Khamashta, MD, FRCP, PhD: The Rayne Institute, St Thomas’ Hospital, London, UK. Dr. Pomar has received consultant fees, speaking fees, and/or honoraria (more than $10,000 each) from Edwards Lifesciences and Medtronic. Address correspondence to Jose´-Gabriel Erdozain, MD, Servicio de Medicina Interna, Hospital de Cruces, Plaza Cruces s/n, 48903-Bizkaia, Spain. E-mail: [email protected]. Submitted for publication September 2, 2011; accepted in revised form March 5, 2012.

1256

coagulant (LAC) test, anticardiolipin antibodies (aCL), and anti–␤2-glycoprotein I antibodies (1). APS can occur as an isolated condition or can be associated with connective tissue diseases, most commonly systemic lupus erythematosus (SLE). The cardiac manifestations of APS include heart valve disease (valve thickening, vegetations, and valvular dysfunction), coronary thrombosis, ventricular hypertrophy and dysfunction, intracardiac thrombi, and pulmonary hypertension (2). Valvular involvement has a prevalence of greater than 80% if highly sensitive techniques such as transesophageal echocardiography are used. The mitral valve is the most frequently involved (3,4). Most patients are asymptomatic, but cerebrovascular accidents are reported to be more prevalent among patients with significant valve lesions (5,6). Despite this high prevalence, only 4 – 6% of aPL-positive patients with valve disease will require surgical treatment (3). Therefore, published data on the outcome of APS patients after heart valve surgery are scarce (7–9). In this study, we aim to describe the outcome and complications of heart valve replacement surgery in patients with APS and to identify prognostic factors associated with adverse outcomes.

APS Patients With Cardiac Valve Replacement

Significance & Innovations ●

Cardiac valve surgery in patients with antiphospholipid syndrome (APS) is infrequent, with all previously published case series including no more than 10 patients. Our study is the largest case series reported (APS), analyzing the course of 32 patients attending 4 international tertiary hospitals with experience in the management of this syndrome.

●

We report a mortality rate higher than that seen in the general population, but lower than those previously published in other APS cohorts. On the other hand, our data contrast with several published uneventful case reports. Our results are probably closer to the real outcome of patients with APS treated in referral centers.

●

Unfortunately, we were not able to identify variables associated with adverse outcomes. However, given the high frequency of both thrombotic and bleeding complications, very careful management of anticoagulation after surgery seems essential to improve the prognosis of heart valve surgery in patients with APS.

1257 and death was analyzed within 1 month after surgery (early complications) and afterward (late complications) (11). A descriptive analysis of the baseline clinical characteristics of the patients as well as of the early and late complications was carried out, using proportions for categorical variables and medians and ranges for continuous variables. Univariate logistic regression was performed for each variable potentially associated with poor outcome, the latter being defined as 4 different dependent variables (death, thrombosis, major bleeding, or global complications, which included any of the previous or valvular infection, reoperation, or atrioventricular [AV] block). The following independent variables were tested: age at the time of surgery, NYHA functional class, date of surgery, type of prosthetic valve, presence of pulmonary hypertension, presence of SLE, and positivity for LAC. In the case of finding independent variables with a P value of less than 0.1 in the univariate analysis, multivariate logistic regression models would be constructed in order to identify those variables with statistically independent associations. Sex and history of thrombosis were not analyzed, since more than 90% of patients were women and had experienced previous thrombotic events.

Results Patients and Methods Patients with APS having valve replacement surgery between 1981 and 2008 were identified in 4 centers: Lupus Research Unit, The Rayne Institute, St Thomas’ Hospital, London, UK; Instituto Nacional de Cardiologı´a I. Cha´vez, Distrito Federal, Mexico; Hospital Universitario Cruces, Barakaldo, Spain; and Hospital Clinic, Barcelona, Spain. Cases were identified from the specific databases of each participating center. All of the patients fulfilled the Sapporo criteria for the classification of APS (10), although 5 patients were actually diagnosed with APS during the followup time after surgery: one of them was operated on in 1981 and the other 4 were operated on between 1995 and 2002. Data were collected using the same protocol in all of the participating centers. The following variables were retrospectively extracted from the medical notes: age, date of surgery (as a 3-category variable: from 1981–1990, 1991– 2000, and 2001–2008), APS duration (from diagnosis to surgery), previous APS features (clinical and immunologic), the presence of associated SLE, functional class according to the New York Heart Association (NYHA) scale, echocardiographic findings, pathologic findings, and type of prosthetic valve implanted (mechanical or biologic). The presence of pulmonary hypertension was established by echocardiography as an estimated systolic pulmonary arterial pressure higher than 30 mm Hg. Values between 30 and 40 mm Hg were labeled as mild, between 40 and 60 mm Hg were labeled as moderate, and greater than 60 mm Hg were labeled as severe. The occurrence of complications (thrombosis, bleeding, structural or nonstructural prosthetic valve deterioration, infection, other)

Demographic data. Thirty-two patients were included in the study: 16 from London, 7 from Distrito Federal, 5 from Barakaldo, and 4 from Barcelona. Thirty patients (93.7%) were women. The median age at the time of surgery was 39 years (range 25–73 years). Twenty-four patients (75%) were white, 7 (21.9%) were Hispanic, and 1 (3.1%) was African Caribbean. APS was secondary to SLE in 11 patients (34.4%). The median APS duration up to the time of surgery was 6 years (range 0 –34 years). The median followup time between surgery and the last documented visit was 33.5 months (range 0 –192 months). Clinical features. Prior to surgery, 19 patients were classified in NYHA functional classes III and IV. Pulmonary hypertension was found in 13 patients (40.6%), with 10 being classified as moderate/severe. Coronary artery angiography before surgery was available in 15 patients, with normal results in 12 cases (80%). The mitral valve was the most frequently affected, with 17 patients (53.1%) having isolated involvement and an additional 6 patients (18.7%) having combined valve disease. Regurgitation was more frequent than stenosis in all of the affected valves (Table 1). No patient had a previous diagnosis of rheumatic heart disease. APS features. aCL IgG at medium to high titers were positive in 24 patients (75%). Seven patients (21.8%) had aCL IgM. LAC was positive in 26 patients (81.2%). Twenty-five patients (78%) had experienced previous arterial and/or venous thrombosis. Twenty-four arterial events and 12 venous thromboses were recorded. Only 7 patients had not experienced previous thrombosis. Obstetric APS features were present in 14 patients (Table 2).

1258

Erdozain et al

Table 1. Cardiac features* No. of patients (n ⴝ 32) NYHA score Class I or II Class III or IV NR Coronary angiography Normal result Abnormal result NR Echocardiographic findings Valves involved Mitral valve Aortic valve Mitral and aortic valves Mitral and tricuspid valves Mitral, aortic, and pulmonary valves Tricuspid valve Valvular dysfunction Mitral regurgitation Mitral stenosis Aortic regurgitation Aortic stenosis Tricuspid regurgitation Tricuspid stenosis Pulmonary regurgitation Pulmonary hypertension Moderate/severe

only (2 warfarin, 5 acenocoumarol, and 1 heparin), and 3 patients were treated with aspirin plus oral anticoagulation. Nine patients were not taking any antithrombotic drugs and treatment was not available in 7 cases. Pathologic findings. Data from the pathologic evaluation were available for 20 valves. Chronic inflammation and/or calcification was present in 12 cases. Thrombosis with or without fibrosis was seen in 8 cases.

7 19 6 12 3 17

Outcome. Seven patients (21.8%) experienced early complications: 4 major hemorrhages, 1 atrial clot, 1 cardiac tamponade (with reoperation required), and 2 thirddegree AV blocks, both requiring a pacemaker. One patient died of cardiogenic shock in the surgical theater. The late mortality rate was 9.6%: 1 patient each died of septic shock, hemorrhagic stroke, and thrombosis of renal graft after renal transplantation. Other late complications were present in 25.8% of cases: 4 major bleeding events, 4 thrombotic events (in 3 patients), 2 episodes of heart failure, 1 nonstructural valve deterioration, and 1 valve infection. Two patients required additional valve replacements due to postoperative valve thrombosis and severe valvular deterioration. One patient required a second valve replacement due to Staphylococcus aureus endocarditis (Table 3).

17 8 3 2 1 1 18 10 11 5 2 1 1 13 10

* NYHA ⫽ New York Heart Association; NR ⫽ not reported.

Table 3. Early and late outcomes after valvular surgery*

Surgical procedure and antithrombotic treatments. Thirty-three valve replacement procedures were registered, with 23 mechanical and 10 biologic valves implanted. One patient had a double (mitral and aortic) valve replacement. Prior to surgery, 5 patients were treated with aspirin, 8 patients were treated with anticoagulant drugs Table 2. Previous antiphospholipid syndrome–related clinical and immunologic features* Value (n ⴝ 32) Antiphospholipid antibodies aCL alone (IgG and/or IgM) aCL (IgG and/or IgM) plus LAC LAC alone Thrombosis, no. of patients Arterial Stroke Valvular thrombosis Peripheral thromboembolism Venous No thrombosis Gynecologic Miscarriages (ⱖ3) Fetal loss (ⱖ1) Preeclampsia (ⱖ1)

6 (18.7) 19 (59.3) 7 (21.8) 24 19 4 1 12 7 5 (15.6) 9 (28.1) 2 (6.2)

* Values are the number (percentage) of patients unless otherwise indicated. aCL ⫽ anticardiolipin antibodies; LAC ⫽ lupus anticoagulant.

Value Early outcome† Mortality, no./total (%) patients Cardiogenic shock Complications, no./total (%) patients Major bleeding Atrial thrombosis Cardiac tamponade Third-degree AV block Late outcome‡ Mortality, no./total (%) patients Septic shock After renal transplantation Hemorrhagic stroke Complications, no./total (%) patients Major bleeding Thrombosis TIA Peripheral thromboembolic disease Valvular thrombosis Heart failure Nonstructural valvular deterioration Prosthetic valve infection Global outcome, no./total (%) patients Mortality and/or complications Uneventful

1/32 (3) 1 7/32 (21.8) 4 1 1 2 3/31 (9.6) 1 1 1 8/31 (25.8) 4 4 2 1 1 2 1 1 16/32 (50) 16/32 (50)

* Values are the number of occurrences unless otherwise indicated. Some patients experienced several complications. AV ⫽ atrioventricular; TIA ⫽ transient ischemic attack. † Within 1 month after surgery. ‡ More than 1 month after surgery.

APS Patients With Cardiac Valve Replacement Variables associated with adverse outcomes. Age at the time of surgery, NYHA functional class, date of surgery, type of prosthetic valve, presence of pulmonary hypertension, presence of SLE, and positivity for LAC were tested for associations with adverse outcomes (death, thrombosis, bleeding, or global complication; see Patients and Methods). The presence of global complications was not associated with any of the independent variables. Likewise, none of these variables was statistically associated with the occurrence of thrombosis or bleeding. The date of surgery was the only variable associated with mortality (1 of 2 patients in the period 1980 –1990, 3 of 12 patients in the period 1991–2000, and 0 of 18 patients in the period 2001–2008; P ⫽ 0.037). Despite the lack of statistical association, all 3 patients with thrombotic complications and all 4 patients who eventually died had primary APS.

Discussion The pathogenesis of APS valvulopathy is unclear. Deposits of aCL and complement have been found in the subendothelial connective tissue of the affected valves (12). Farzaneh-Far et al (13) suggested that aPL may initiate an inflammatory cascade, inducing endothelial cell activation and monocyte recruitment in vulnerable sites and leading to valvular damage. Surgery increases the risk of thrombosis in patients with APS due to a number of factors: withdrawal of oral anticoagulants, induction of a transient hypercoagulability state, even despite ongoing anticoagulant therapy, or catastrophic exacerbation of APS. Bleeding complications can also occur during the perioperative period due to excessive anticoagulation and/or thrombocytopenia (14). Berkun et al (7) reported a mortality rate of 40%, with complications in an additional 40% of patients, in a small series of 10 cases. Comorbidity and advanced valvular disease were suggested as possible predictors of adverse outcomes. Hegde et al (8) published a case series of 10 patients with different types of major cardiovascular procedures, with 3 of them having heart valve replacement. All 3 experienced complications, albeit not directly related to APS (heparin-induced thrombocytopenia, AV block, and sepsis). A meta-analysis of heart valve surgery in patients with APS by Gorki et al (9) showed a high mortality rate, with 7% early and 12% late deaths. Only 42% of patients had an uneventful recovery. More recently, Colli et al reported a series of 9 patients (15). Two patients died in the early postoperative period due to thrombotic complications. Nonfatal thrombotic and bleeding events were the most frequent early and late complications. In this retrospective study of 32 patients, we observed a 12.5% mortality rate, more than 2-fold the expected mortality for this type of surgery (7). Forty-three percent of the patients experienced complications, most of them major bleeding or thrombosis. There is no consensus regarding the optimal perioperative management of anticoagulation in APS. Erkan et al (14) recommended keeping to an absolute minimum the periods without anticoagulation and the use of physical methods such as intermittent venous compression. They also proposed different ways of monitoring heparin during cardiopulmonary bypass surgery, such as doubling the

1259 baseline activated coagulation time (ACT) level, obtaining heparin concentrations by protamine titration rather than following ACT levels, and performing heparin ACT titration curves preoperatively to determine patient-specific target ACT levels. The choice between mechanical and biologic valves in these patients is not clear cut. Berkun et al (7) recommend the use of mechanical valves, given that patients are usually young and the common need of anticoagulation for previous thromboembolic disease. On the other hand, Colli et al suggest the use of biologic valves, given their lower potential for associated thrombosis (15). In this study, no differences in outcome were demonstrated between patients with mechanical and biologic valves. This retrospective study has several substantial limitations that weaken the results. Important missing data included information regarding previous surgery angiograms and hemostatic treatment used during surgical procedures. In addition, the surgical and medical management of patients was heterogeneous, given that surgery was undertaken in 4 different hospitals and within a broad time span. The lack of power, a consequence of the small study size, probably precluded the identification of clinically meaningful predictors of adverse outcomes. In particular, although all of the postoperative thrombosis and deaths happened in patients with primary APS, a statistically significant association could not be established. Likewise, the high prevalence of LAC positivity (26 [81.2%] of 32 patients) reduced the size of the LAC-negative group and thus the power to detect differences. Accordingly, a worse prognosis of LAC-positive patients, as in other settings of APS, cannot be excluded. On the other hand, our results suggest a better prognosis in patients having surgery within more recent years. Despite these serious limitations, this study highlights several clinically meaningful issues. Heart valve surgery is a very high-risk procedure in patients with APS, and must be accordingly undertaken only after a thorough and joined evaluation of the patient’s clinical condition and therapeutic alternatives. Both bleeding and thrombotic complications are the main causes for morbidity and mortality; therefore, strict and coordinated perioperative control of the anticoagulant therapy must be accomplished. In this setting, it is not yet clear whether the use of biologic valves could be associated with a lower frequency of early complications. Fortunately, our data also point to a better prognosis of patients having surgery in recent years, which confirms our better management of this complex situation within a complicated clinical condition. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Erdozain had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Erdozain, Amigo, Perez-Valero, Khamashta. Acquisition of data. Erdozain, Segura, Amigo, Espinosa, Pomar, Perez-Valero, Ateka-Barrutia, Khamashta. Analysis and interpretation of data. Erdozain, Ruiz-Irastorza, Amigo, Khamashta.

1260

Erdozain et al

REFERENCES 1. Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295–306. 2. Tenedios F, Erkan D, Lockshin MD. Cardiac involvement in the antiphospholipid syndrome. Lupus 2005;14:691– 6. 3. Nesher G, Ilany J, Rosenmann D, Abraham AS. Valvular dysfunction in antiphospholipid syndrome: prevalence, clinical features, and treatment. Semin Arthritis Rheum 1997;27:27– 35. 4. Erdogan D, Goren MT, Diz-Kucukkaya R, Inanc M. Assessment of cardiac structure and left atrial appendage functions in primary antiphospholipid syndrome: a transesophageal echocardiographic study. Stroke 2005;36:592– 6. 5. Cervera R, Khamashta MA, Font J, Reyes PA, Vianna JL, Lopez-Soto A, et al. High prevalence of significant heart valve lesions in patients with the ‘primary’ antiphospholipid syndrome. Lupus 1991;1:43–7. 6. Krause I, Lev S, Fraser A, Blank M, Lorber M, Stojanovich L, et al. Close association between valvular heart disease and central nervous system manifestations in the antiphospholipid syndrome. Ann Rheum Dis 2005;64:1490 –3. 7. Berkun Y, Elami A, Meir K, Mevorach D, Naparstek Y. Increased morbidity and mortality in patients with antiphospholipid syndrome undergoing valve replacement. J Thorac Cardiovasc Surg 2004;127:414 –20. 8. Hegde VA, Vivas Y, Shah H, Haybron D, Srinivasan V, Dua A, et al. Cardiovascular surgical outcomes in patients with the

9. 10.

11.

12.

13.

14.

15.

antiphospholipid syndrome: a case-series. Heart Lung Circ 2007;16:423–7. Gorki H, Malinovski V, Stanbridge RD. The antiphospholipid syndrome and heart valve surgery. Eur J Cardiothorac Surg 2008;33:168 – 81. Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999;42:1309 –11. Edmunds LH, Clark RE, Cohn LH, Grunkemeier GL, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1996;62:932–5. Ziporen L, Goldberg I, Arad M, Hojnik M, Ordi-Ros J, Afek A, et al. Libman-Sacks endocarditis in the antiphospholipid syndrome: immunopathologic findings in deformed heart valves. Lupus 1996;5:196 –205. Farzaneh-Far A, Roman MJ, Lockshin MD, Devereux RB, Paget SA, Crow MK, et al. Relationship of antiphospholipid antibodies to cardiovascular manifestations of systemic lupus erythematosus. Arthritis Rheum 2006;54:3918 –25. Erkan D, Leibowitz E, Berman J, Lockshin MD. Perioperative medical management of antiphospholipid syndrome: hospital for special surgery experience, review of the literature, and recommendations. J Rheumatol 2002;29:843–9. Colli A, Mestres CA, Espinosa G, Plasin MA, Pomar JL, Font J, et al. Heart valve surgery in patients with the antiphospholipid syndrome: analysis of a series of nine cases. Eur J Cardiothorac Surg 2010;37:154 – 8.