Pericardial effusion and cardiac tamponade in pediatric stem ... - Nature

Bone Marrow Transplantation (2005) 36, 139–144 & 2005 Nature Publishing Group All rights reserved 0268-3369/05 $30.00

www.nature.com/bmt

Pericardial effusion and cardiac tamponade in pediatric stem cell transplant recipients M Rhodes1, T Lautz1, A Kavanaugh-Mchugh2, B Manes1, C Calder1, T Koyama3, M Liske2, D Parra2 and H Frangoul1 1 Pediatric Stem Cell Transplant Program, Vanderbilt Children’s Hospital, Nashville, TN, USA; 2Department of Pediatric Cardiology, Vanderbilt Children’s Hospital, Nashville, TN, USA; and 3Department of Biostatistics, Vanderbilt University, Nashville, TN, USA

Summary: Pericardial effusion and cardiac tamponade is a rarely reported complication following stem cell transplant (SCT). The incidence among pediatric SCT recipients is not well defined. To assess the frequency of clinically significant pericardial effusions, we retrospectively examined clinically significant cardiac effusions at our center. Between January of 1993 and August 2004, clinically significant pericardial effusions were identified in nine of 205 patients (4.4%). The median age at the time of transplant was 9 years (range 0.6–18 years) and seven received an allogeneic transplant. All nine had normal cardiac function prior to transplant. The effusion developed at a median of 30 days (range 18–210 days). All allogeneic recipients had acute or clinically extensive graft-versus-host disease (GVHD) at the time the effusion was diagnosed. Seven patients (78%) required pericardiocentesis or surgical creation of a pericardial window. No patient died as a complication of the effusion or the therapeutic procedures. Clinically significant pericardial effusions are more common than previously reported in pediatric SCT recipients. Acute and chronic GVHD is an associated factor. Bone Marrow Transplantation (2005) 36, 139–144. doi:10.1038/sj.bmt.1705023 Published online 23 May 2005 Keywords: effusion; tamponade; pediatric

Stem cell transplant (SCT) is an established therapy for various malignant and nonmalignant disorders in children. Despite significant improvement in supportive care, SCT continues to be associated with significant morbidity and mortality. Although acute cardiac complications occur in less than 2% of patients, they can be life threatening.1,2 Progressive heart failure, acute ventricular fibrillation and cardiac tamponade are among the reported cardiac Correspondence: Dr H Frangoul, Vanderbilt University, 397 PRB, Nashville, TN 37232-2573, USA; E-mail: [email protected] Received 31 January 2005; accepted 12 April 2005 Published online 23 May 2005

complications of SCT. Sudden cardiac tamponade has been described as a complication affecting 2% of pediatric thalassemia patients undergoing allogeneic BMT.3 Thalassemia patients often suffer from iron overload and decreased cardiac function prior to SCT, and are thus at higher risk for cardiac complications.3 Children who undergo SCT for malignancy have usually been pre-treated with chemotherapy and radiation. Cardiac damage is a known complication of mantle radiation, high-dose cyclophosphamide4–6 and anthracycline chemotherapy,7 but abnormal results of pretransplant cardiac evaluation have not always been associated with an increased risk of cardiac complications during SCT in nonthalassemia patients.1 There are few case reports in the literature describing cardiac tamponade in nonthalassemia pediatric patients undergoing SCT.1,8–13 The incidence of pericardial effusion and cardiac tamponade following SCT in pediatric patients is not known. We report on nine children who developed pericardial effusion and cardiac tamponade following SCT at our center and review the literature of all reported pediatric cases.

Materials and methods The BMT database at Vanderbilt Children’s Hospital contains prospectively collected data on all patients transplanted at our center. Database records on 205 patients receiving a blood or marrow transplant between January 1993 and August 2004 were screened for any cardiac complications. For these patients, all available medical records, notes, correspondence and autopsy reports were reviewed to identify, in detail, the nature of the event and/or prior cardiac disease or its evolution. Patients with clinically significant pericardial effusion or cardiac tamponade were identified. An extensive chart review was performed to determine age, gender, underlying diagnosis, type of transplant, conditioning therapy, graftversus-host disease (GVHD) (prophylaxis, grade, treatment), infections, other clinical complications as well as the onset, evolution and clinical outcome of pericardial effusion. We collected additional data including cumulative doses of previous anthracycline therapy or mediastinal radiation. The cumulative dose of anthracycline therapy given as doxorubicin, daunorubicin, idarubicin and

Pericardial effusion in pediatric stem cell transplant M Rhodes et al

Bone Marrow Transplantation

201 210 155 MRD Auto MMURD Cy, TBI CVM Cy, TBI 48 56 58 None 225 None 2/M 3/M 17/F 7 8 9

MDS NB MDS

19 25 30 19 MRD MMURD MMURD MURD Cy TBI TBI Flu, ATG Bu, Cy, Cy, Bu, 65 68 59 56 160 340 225 160 AML ALL ALL ALL 18/M 14/M 7/M 9/F 3 4 5 6

NB ¼ neuroblastoma; BuMel ¼ busulfan and melphalan; CVM ¼ carboplatin, etoposide and melphalan; CBV ¼ cytoxan, BCNU and etoposide; URCB ¼ unrelated cord blood; MRD ¼ matched related donor; MMURD ¼ mismatched unrelated donor; MMRD ¼ mismatched related donor; N/A ¼ not applicable.

Died day 232 chronic GVHD Died of relapsed disease Died day 195 chronic GVHD Yes N/A Yes

Alive 2.5 years Died day 500 relapsed disease Died day 829 of relapse Alive day 100 Yes Yes Yes Yes

N/A Yes

Pericardiocentesis (day 29) Pericardiocentesis (day 69) Fenestration (day 138) Fenestration (day 35) No procedure Pericardiocentesis (days 30) Pericardiocentesis (day 19) Fenestration (days 55) Pericardiocentesis (day 205) Fenestration (day 225) No procedure 28 34 Auto URCB CBV Bu, Mel, ATG 69 63 210 75 Hodgkin’s ALL

Age (years)/ Diagnosis sex Patient

14/F 0.6/F

Patient characteristics Table 1

This detailed review identified nine SCT patients who experienced clinically significant pericardial effusion or tamponade post-SCT. Pericardial effusion was diagnosed by echocardiogram performed due to clinical suspicion rather than routinely scheduled echocardiograms. The frequency of this serious cardiac complication was nine of 205 receiving SCT (4.4%). The characteristics of the nine patients are summarized in Table 1. The median age was 9 years (ranged 0.6–18 years); four were female and five male. Underlying diagnoses consisted of acute lymphocytic leukemia (n ¼ 4), myelodysplastic syndrome (n ¼ 2), acute myelogenous leukemia, neuroblastoma and Hodgkin’s disease (one each). Two patients had no anthracycline exposure and the rest had a dose ranging from 75 to 340 mg/m2. All patients had a left ventricular ejection fraction of 445% prior to transplant. Conditioning regimens consisted of chemotherapy alone in five patients and cyclophosphamide plus TBI in four patients. Two patients had prior radiation exposure: patient 1 had 3500 cGy mediastinal radiation and patient 9 had previous TBI (1320 cGy) a year prior to this transplant. Seven of the nine received allogeneic transplants (two related and five unrelated) and two had autologous PBSC transplant. All

Prior anthracycline (mg/m2)

Results

1 2

Pre-BMT Conditioning LVEF (%)

Donor

Onset of effusion Therapy (days post-BMT)

GVHD

Outcome

mitoxantrone was calculated using previously reported maximum cumulative doses.14 The actual cumulative dose of idarubicin was multiplied by 3.0, and that of mitoxantrone by 2.8, to obtain a doxorubicin equivalent anthracycline dose. For daunorubicin conversion, doxorubicin is computed as two-thirds of the daunorubicin dose with respect to cardiotoxicity.14 As part of their pre-transplant evaluation, all patients underwent a cardiac evaluation with an electrocardiogram and an echocardiogram or radionuclide ventriculography to evaluate left ventricular ejection fraction. During the transplant admission, the patients had weekly chest X-rays. Additional studies were carried out if the patients developed fever or respiratory symptoms. Echocardiography was routinely performed at day 100 post transplant or as clinically indicated. The effusions were assessed by echocardiography and clinically by measurement of pulses paradoxus. Echocardiographic signs of potential or impending tamponade physiology included right atrial collapse, right ventricular collapse and respiratory variation in mitral inflow velocity with accompanying increase in isovolumic relaxation time. Clinical confirmation of tamponade was defined as the presence of pulsus paradoxus 410 mmHg.15 Percutaneous pericardiocentesis or surgical creation of a window were performed in the setting of significant effusions with echocardiographic or clinical evidence of tamponade. The frequencies of pericardial effusions between the patients who received allogeneic SCT and those who received autologous SCT were compared by a Fisher’s exact test. The same method was used to compare the recipients of allogeneic SCT who did and did not develop GVHD. The retrospective review of patients’ medical records was approved by the Institutional Review Board of Vanderbilt University Medical Center.

Alive 7 years Died day 546 chronic GVHD

140


141

allogeneic transplants were non-T-cell depleted and four were from mismatched donors (Table 1). GVHD prophylaxis consisted of cyclosporine and methotrexate in all allogeneic transplants, except for patient 2 who received cyclosporine and prednisone and patient 9 who received cyclosporine and mycophenelate. Significant pericardial effusion developed between 19 and 210 days (median 30) post transplant. None of the effusions were diagnosed on routine echocardiograms but rather when patients developed symptoms. All patients who received allogeneic transplants had associated acute GVHD (n ¼ 5) or clinically extensive chronic GVHD (n ¼ 2). The patients who developed clinically significant pericardial effusion were more commonly recipients of unrelated donor transplants (44.4%) compared to related (33.3%) or autologous transplants (22.2%). Further analysis suggests a higher risk among recipients of allogeneic SCT of developing a clinically significant pericardial effusion as compared to recipients of autologous SCT, P ¼ 0.08. Among patients who received allogeneic SCT (n ¼ 85),there was a trend towards higher risk of pericardial effusion among patients who developed acute or chronic GVHD (seven of 58) compared to those with no GVHD (0 of 25) (P ¼ 0.07). Six patients developed the pericardial effusion within 100 days of transplant: Patient 1 had received mediastinal radiation therapy with 3500 cGy prior to transplant; she required pericardiocentesis on day þ 29. She had also required thoracentesis for subpulmonic effusion, which appeared on day þ 21. Patient 2 first developed pericardial effusion day þ 34 that required pericardiocentesis on day þ 69. The effusion reaccumulated and a pericardial window was performed on day þ 138 when she developed evidence of tamponade physiology. This was in the setting of steroid-refractory GVHD. Patient 3 required pericardiocentesis day þ 19 and creation of a pericardial window on day þ 35 due to recurrent effusion with evidence of tamponade. Patient 4 developed pericardial effusion day þ 25 coinciding with the onset of grade III acute GVHD. Despite echocardiographic evidence of tamponade physiology, drainage was deferred due to refractory thrombocytopenia. His effusion resolved after initiating prednisone therapy for GVHD. Patient 5 required pericardiocentesis and placement of external drain on day þ 30. He had associated acute GVHD and steroid therapy was initiated with resolution of the drainage and no evidence of reaccumulation. Patient 6 received reduced-intensity conditioning 1 year following a fully ablative transplant due to disease recurrence. She developed her effusion 19 days post transplant and required pericardiocentesis in the setting of acute GVHD. The effusion recurred on day 55 requiring surgical creation of pericardial window, while tapering prednisone and was associated with flare up of GVHD. The remaining three patients developed pericardial effusions between day þ 155 and day þ 210. Patient 7 developed pericardial effusion day þ 201 associated with extensive chronic GVHD; he had pericardiocentesis performed day þ 205 and þ 210 post transplant. Patient 8 had transplant-related chronic renal failure and was receiving 13-cis retinoic acid therapy; he developed tamponade on day þ 210 and had fenestration performed on day þ 225.

Patient 9 developed a large effusion with echocardiographic evidence of tamponade physiology on day þ 155 in the setting of extensive chronic GVHD and poor engraftment. She died on day þ 195 of bacterial sepsis during the hospitalization. Pericardial fluid analysis was performed on samples from eight patients. Glucose levels were normal, except for patient 2 who had an elevated level coinciding with an elevated serum glucose at the time of the pericardial fluid analysis. There was elevated total protein ranging from 2000 to 3800 mg/dl (median 3400). There were elevations in the number of WBC ranging from 18 to 1260 cells/ml (median 215). Eight of the nine patients had no evidence of malignancy or infection. Patient 9 had postmortem evaluation of the effusion that revealed the presence of Enterobacter cloacae. Blood cultures on this patient were also positive for the same organism at the time of death. Viral and bacterial cultures of the fluid on the remaining patients were negative. Cytology was performed on all eight samples and revealed no evidence of malignancy. Six patients died secondary to relapse (patients 4, 5 and 8) or as a result of complications related to chronic GVHD (patients 2, 7 and 9). None of these deaths were related to the pericardial effusions. All three surviving patients (patients 1, 3 and 6) have normal cardiac function and have had no recurrence of the pericardial effusion.

Discussion The goal of this study was to review the incidence of clinically significant pericardial effusion and cardiac tamponade in pediatric patients following SCT. The incidence of 4.4% is significantly higher than what has been previously reported in the literature. In a large study from the University of Minnesota, cardiac complications occurred in 26 of 2821 transplants (0.9%) and only five had pericardial effusion or cardiac tamponade (0.2%).2 Review of the literature identified 18 reported cases of pericardial effusions in children undergoing SCT (Table 2). The time to developing an effusion ranged from 3 days prior to transplant to 600 days post transplant (median 34).1,8–13,16 Nine cases were reported in patients undergoing matched related donor transplant for thalassemia.3,16 All of those patients had the effusion develop in the first month post transplant and it was fatal in six of the patients. The overall incidence of pericardial effusion in thalassemia transplants was 2% among 400 consecutive children. Although the etiology was not clear, it was thought to be related to toxicity from the conditioning regimen and possible iron overload.3,16 Eight of the remaining nine cases were reported in patients undergoing allogeneic transplant from related donors.1,8–13 Acute or chronic GVHD was present in 62% of those patients at the time of the diagnosis of the pericardial effusion. Although the etiology of cardiac toxicity following SCT is not well known, some factors have been implicated. High-dose cyclophosphamide therapy has been noted to be a potential risk factor for developing cardiac toxicity.4,6,10 Steinerz et al5 reported on 40 children who received more than 80 mg/kg of cyclophosphamide and 21 patients Bone Marrow Transplantation

142


Reported cases of pericardial effusion

Table 2

Conditioning

Transplant type

Onset of effusion (degree)

Therapy (day)

9/M

SAA

Cy, TLI

Allo

Day 74 (tamponade)

Fenestration

1

11/M

ALL

TBI, VP16, Cy, melphalan, TLI

MRD

20 months

Pericardiocentesis

Yes

Angelucci et al16

1

11/M

Thalassemia

Bu, Cy

MRD

Day 17 (tamponade)

Pericardiocentesis

—

Alive

Angelucci et al3

8

15/F 15/M 9/M 13/M 4/F 4/M 11/F 9/F

Thalassemia

Bu, Cy

MRD

Day Day Day Day Day Day Day Day

None None None None None None Pericardiocentesis Pericardiocentesis

—

Died Died Died Died Died Died Alive Alive

Seber et al11

3

8 10 12

ALL ALL SAA

Cy Cy Cy

MMRD MRD MRD

Day 63 Day 77 Day 96

Pihkala et al10

2

3/M

Neuroblastoma

Melphalan

Auto

5/F

ALL

Ara C

Allo

6 months (tamponade) 3.5 months

No Procedure

1

10/F

ALL

Cy, TBI, Et

MRD

Day 216

—

Yes

Died of Aspergillus pneumonia 183 days

1

7/F

Diamond Blackfan Syndrome

Bu, Cy

MRD

Day 28

—

—

Died at 1 month because of effusion

N

Age (years)/ sex

Murdych and Weisdoef2

1

Toren and Nagler12

Ueda et al

13

Saunders et al9

N/A ¼ not applicable.

–3 41 21 3 8 23 0 6

Pericardiocentesis

Associated GVHD —

Outcome

Tamponade resolved, but died of sepsis Died (sepsis)

Yes Yes Yes

Died day 89 Alive day 405 Alive day 1168

N/A

Alive 3.5 years

—

Alive 7 months


Diagnosis

Reference


143

developed cardiac abnormalities. Pericardial effusion was the most common observed abnormality, occurring in 15 of the 21. Receiving anthracycline doses of X100 mg/m2 prior to cyclophosphamide was identified as a significant risk factor. Six of our nine patients received a cyclophosphamide dose X120 mg/kg and four of those patients received anthracycline doses of 4100 mg/m2. Reduced ejection fraction prior to SCT17,18 has been associated with increased risk of cardiac complication post transplant, although others found no association.1 Our patients all underwent cardiac evaluation prior to transplant and none had decreased cardiac function. Infection has not been reported as a potential risk factor. One patient in our experience had positive cultures from the pericardial fluid post mortem (patient 5) in the setting of positive blood cultures. The effusion in that patient was identified in association with exacerbation of chronic GVHD 1 month prior to her death, so the infection might be a secondary finding rather than the primary cause of her effusion. The only consistent observation in our recipients of allogeneic transplant is the association with acute or chronic GVHD. In one patient (patient 6), pericardiocentesis was not carried out due to refractory thrombocytopenia. The effusion resolved after initiating prednisone therapy, and importantly, the resolution of the effusion coincided with the resolution of GVHD. The association of acute and chronic GVHD with pericardial effusions has also been observed in pediatric and adult patients.11,12,19,20 If pericardial effusion is detected prior to the development of tamponade, initiation of steroid therapy or intensifying immune suppression should be considered especially in patients with active GVHD. There is only one reported pediatric case of pericardial effusion following autologous transplant; a patient with neuroblastoma developed pericardial effusion 6 months post transplant.10 In our experience, we observed two cases in recipients of autologous SCT. One patient had prior anthracycline exposure and mediastinal irradiation (3000 cGy) prior to transplant (patient 1). The second patient developed the effusion while receiving 13-cis retinoic acid therapy post transplant for neuroblastoma. Therapy with 13-cis retinoic acid has been associated with pleural effusion21,22 but not pericardial effusion. It is not clear whether the therapy with 13-cis retinoic acid contributed to the development of this patient’s pericardial effusion. Among reported cases of pericardial effusion in the literature, effusion was the cause of death in six of nine (67%) patients with thalassemia and two of nine (22%) with other diagnoses. A third of the cases required pericardiocentesis or fenestration. None of our patients died as a result of the effusion, but seven (78%) required either fenestration or pericardiocentesis. Three required multiple procedures for recurrent and refractory effusions. The procedures were tolerated without any complications. None of the patients had cardiac dysfunction on follow-up evaluation. We conclude that pericardial effusion and cardiac tamponade is more common among pediatric patients undergoing SCT than has been previously reported. Prompt intervention can prevent death related to this complication. Pediatric SCT physicians should

monitor for this complication, especially in patients with acute and chronic GVHD.

Acknowledgements We gratefully acknowledge the excellent care provided by the nursing and support staff.

References 1 Hertenstein B, Stefanic M, Schmeiser T et al. Cardiac toxicity of bone marrow transplantation: predictive value of cardiologic evaluation before transplant. J Clin Oncol 1994; 12: 998–1004. 2 Murdych T, Weisdorf DJ. Serious cardiac complications during bone marrow transplantation at the University of Minnesota, 1977–1997. Bone Marrow Transplant 2001; 28: 283–287. 3 Angelucci E, Mariotti E, Lucarelli G et al. Sudden cardiac tamponade after chemotherapy for marrow transplantation in thalassaemia. Lancet 1992; 339: 287–289. 4 Appelbaum F, Strauchen JA, Graw Jr RG et al. Acute lethal carditis caused by high-dose combination chemotherapy. A unique clinical and pathological entity. Lancet 1976; 1: 58–62. 5 Steinherz LJ, Steinherz PG, Mangiacasale D et al. Cardiac changes with cyclophosphamide. Med Pediatr Oncol 1981; 9: 417–422. 6 Gottdiener JS, Appelbaum FR, Ferrans VJ et al. Cardiotoxicity associated with high-dose cyclophosphamide therapy. Arch Intern Med 1981; 141: 758–763. 7 Von Hoff DD, Rozencweig M, Layard M et al. Daunomycininduced cardiotoxicity in children and adults. A review of 110 cases. Am J Med 1977; 62: 200–208. 8 Veys PA, McAvinchey R, Rothman MT et al. Pericardial effusion following conditioning for bone marrow transplantation in acute leukaemia. Bone Marrow Transplant 1987; 2: 213–216. 9 Saunders EF, Olivieri N, Freedman MH. Unexpected complications after bone marrow transplantation in transfusion-dependent children. Bone Marrow Transplant 1993; 12 (Suppl. 1): 88–90. 10 Pihkala J, Saarinen UM, Lundstrom U et al. Effects of bone marrow transplantation on myocardial function in children. Bone Marrow Transplant 1994; 13: 149–155. 11 Seber A, Khan SP, Kersey JH. Unexplained effusions: association with allogeneic bone marrow transplantation and acute or chronic graft-versus-host disease. Bone Marrow Transplant 1996; 17: 207–211. 12 Toren A, Nagler A. Massive pericardial effusion complicating the course of chronic graft-versus-host disease (cGVHD) in a child with acute lymphoblastic leukemia following allogeneic bone marrow transplantation. Bone Marrow Transplant 1997; 20: 805–807. 13 Ueda T, Manabe A, Kikuchi A et al. Massive pericardial and pleural effusion with anasarca following allogeneic bone marrow transplantation. Int J Hematol 2000; 71: 394–397. 14 Crossley RJ. Clinical safety and tolerance of mitoxantrone. Semin Oncol 1984; 11 (Suppl. 1): 54–58. 15 Chandraratna PA. Echocardiography and Doppler ultrasound in the evaluation of pericardial disease. Circulation 1991; 84 (Suppl. I): I-303–I-310. 16 Angelucci E, Mariotti E, Lucarelli G et al. Cardiac tamponade in thalassemia. Bone Marrow Transplant 1994; 13: 827–829. Bone Marrow Transplantation


144 17 Bearman SI, Petersen FB, Schor RA et al. Radionuclide ejection fractions in the evaluation of patients being considered for bone marrow transplantation: risk for cardiac toxicity. Bone Marrow Transplant 1990; 5: 173–177. 18 Brockstein BE, Smiley C, Al-Sadir J et al. Cardiac and pulmonary toxicity in patients undergoing high-dose chemotherapy for lymphoma and breast cancer: prognostic factors. Bone Marrow Transplant 2000; 25: 885–894. 19 Silberstein L, Davies A, Kelsey S et al. Myositis, polyserositis with a large pericardial effusion and constrictive pericarditis as manifestations of chronic graft-versus-host disease after non-


myeloablative peripheral stem cell transplantation and subsequent donor lymphocyte infusion. Bone Marrow Transplant 2001; 27: 231–233. 20 Cereda M, Trocino G, Pogliani EM, Schiavina R. A case of cardiac localization of graft-versus-host disease after allogenic bone marrow transplantation. Ital Heart J 2003; 4: 60–63. 21 Bunker CB, Sheron N, Maurice PD et al. Isotretinoin and eosinophilic pleural effusion. Lancet 1989; 1: 435–436. 22 Milleron BJ, Valcke J, Akoun GM, Mayaud CM. Isotretinoinrelated eosinophilic pleural effusion. Chest 1996; 110: 1128.