EXPERIMENTAL AND CLINICAL TRANSPLANTATION

A 53-year-old woman with myelodysplastic syndrome received a cord blood transplant because she had frequent episodes of febrile neutropenia. As a conditioning regimen for transplant, she received 12 Gy total body irradiation, intravenous cytosine arabinoside 3 g/m² every 12 hours on days -5 and -4, and cyclophosphamide 60 mg/kg/day on days -3 and -2. She received tacrolimus and short-term methotrexate treatment as prophylaxis for graft-versus-host disease. Her cardiac function was normal before transplant. She developed acute heart failure with a mild pericardial effusion 11 days after transplant, but her failure improved with a diuretic, vasodilator, and inotropic agent. She complained of dyspnea, and chest auscultation revealed pericardial friction rubs on day 28. Massive pericardial effusion was detected by echocardiography and pericarditis was diagnosed. The pericardial space was drained by pericardiocentesis. The pericardial fluid was exudative, but no bacteria or fungi were cultured. On viral polymerase chain reaction examination, human herpesvirus-6 was detected at a level of 3 × 10⁴ copies/mL in the pericardial effusion, but not in the peripheral blood. With conservative treatment alone, that did not include antiviral therapy, her symptoms disappeared on day 56. We conclude that human herpesvirus-6 reactivation may have been associated with her pericarditis.

Key words : Cord blood transplant, Cardiac tamponade, Human herpes virus-6 (HHV-6)

Large pericardial effusions, pericarditis, and cardiac tamponade are uncommon complications after hematopoietic stem cell transplant (HSCT). Although rarely encountered, these complications can be life-threatening.^1,2 Possible causes of cardiac damage in these patients have been reported to be cyclophosphamide,³ cytosine arabinoside,⁴ total body irradiation (TBI),⁵ engraftment syndrome,⁶ graft-versus-host disease (GVHD),⁷ and infection.⁸ However, cardiac complications associated with human herpesvirus-6 (HHV-6) have been reported in just 2 cases.⁹ We report a case of pericarditis associated with HHV-6 reactivation, which occurred soon after unrelated cord blood transplant.

The patient was a 53-year-old woman with pancy-topenia. She had taken titanium silicate-1 as adjuvant chemotherapy after surgical resection of colon cancer. Bone marrow aspiration showed 2.2% blast cells and multilineage dysplasia. A del (13)(q?) karyotype was present in 3 of 20 mitoses as the sole chromosomal abnormality, using G banding. We diagnosed refractory cytopenia with multilineage dysplasia and decided to perform a cord blood transplant, because the patient required regular blood transfusions and had frequent episodes of febrile neutropenia. Before transplant, she had no remarkable symptoms, and a physical examination was normal. Laboratory studies revealed a white blood cell count of 1.3 × 10⁹/L (band forms 2%, segmented neutrophils 8%, lymphocytes 84%, blast 0%), hemoglobin concentration of 8.3 g/dL, platelet count of 64 × 109/L, serum ferritin level 1861 ng/mL and NT-proBNP 30 pg/mL (normal range, 0-125 pg/mL). Other laboratory data were normal. Transthoracic echocardiography showed a left ventricular ejection fraction of 66% with no abnormalities of the valves or ventricular motion, and no pericardial effusion. Her body weight was 55 kg before transplant.

She received a myeloablative transplant on February 28, 2011. The conditioning regimen consisted of 12 Gy of TBI from days -8 to -6, intravenous cytosine arabinoside 3 g/m² every 12 hours on days -5 and -4, and cyclophosphamide 60 mg/kg/day on days -3 and -2. She received tacrolimus from day -1 and methotrexate 15 mg/m² intravenously on day 1, followed by 10 mg/m on days 3 and 6 as prophylaxis for GVHD. The donor source was unrelated cord blood with matches at 4 of 6 human leukocyte antigen loci. Granulocyte-colony stimulating factor infusion was started on day 1. Her body weight was maintained at 55 kg until day 8, but weight gain and leg edema appeared on day 9. Her body weight reached 61 kg and she complained of dyspnea on day 11. Her peripheral capillary oxygen saturation decreased to 88% on ambient air. Chest radiographs revealed a left pleural effusion, congestion, and cardiomegaly. Echocardiography showed a left ventricular ejection fraction of 58%, a 5 mm pericardial effusion and a 17 mm inferior vena cava. We diagnosed acute cardiac failure. Her symptoms and pericardial effusion disappeared with administration of carperitide, furosemide, and dopamine. She developed acute cutaneous GVHD (stage 1, grade 1) on day 23, but symptoms resolved with topical corticosteroid.

On day 28, she again complained of dyspnea and her peripheral capillary oxygen saturation decreased. Her blood pressure was 124/74 mm Hg, and her heart rate was 120 bpm. Pericardial friction rubs was detected on chest auscultation. Laboratory studies showed a white blood cell count of 0.9 × 10⁹/L (neutrophils 51%), NT-proBNP 853 pg/mL, creatine kinase 16 IU/L, total protein 4.5 g/dL, lactate dehydrogenase 266 IU/L. Echocardiography revealed left ventricular ejection fraction 55%, a 17 mm peri-cardial effusion, and a 22 mm inferior vena cava. Computed tomography scan showed a left pleural effusion and a pericardial effusion (Figure 1). Electro-cardiography showed no ST segment elevation. Based on these results, we made a diagnosis of pericarditis with cardiac tamponade. Pericardiocentesis and pericardial drainage were performed on day 29. Bloody, exudative pericardial fluid was drained.

Laboratory studies of the fluid revealed a white blood cell count of 0.3 × 10⁹/L, a red blood cell count of 0.87 × 10¹²/L, hemoglobin concentration 2.6 g/dL, total protein 4.5 g/dL, albumin 2.8 g/dL, and lactate dehydrogenase 973 IU/L. No malignant cells or inclusion bodies were detected on cytologic examination. No pathogen was cultured, including bacteria, acid-fast bacilli, and fungi. Human herpesvirus-6 variant B DNA was identified in the pericardial fluid by polymerase chain reaction, at a level of 3 × 10⁴ copies/mL, but no other viruses were detected despite screening for cytomegalovirus, Epstein-Barr virus, herpes simplex virus, varicella zoster virus, human herpesvirus-7, human herpesvirus-8, or adenovirus. Human herpesvirus-6 reactivation was not detected in the peripheral blood. Without administering any antiviral drug, the amount of pericardial and pleural fluid decreased after administering carperitide and furosemide; the drainage tube was removed on day 31. Her body weight decreased to 56 kg, and treatment of heart failure was discontinued on day 56. Neutrophil engraftment occurred on day 30. Acute GVHD of the upper gastrointestinal tract developed on day 72 and resolved with systemic corticosteroid therapy. She was discharged from hospital on day 106.

We report pericarditis and cardiac tamponade associated with human herpesvirus-6 reactivation shortly after an unrelated cord blood transplant. Various articles describe series of patients with complications of large pericardial effusions after HSCT. In these reports, large pericardial effusions occurs in 4.4% to 16.9% of children¹ and 0.8% to 1.0% of adult patients.² An allogeneic donor source, GVHD, infection, relapsed malignant disease, high-risk status, myeloablative conditioning, and TBI have been reported as risk factors for developing pericardial effusion.1 The causes of pericardial effusion after HSCT have been reported as cyclophosphamide,³ cytosine arabinoside,⁴ TBI,⁵ engraftment syndrome,⁶ GVHD,⁷ and infection.⁸ Cardiac complications associated with HHV-6 were reported in just 2 cases.⁹

Cardiac damage because of cyclophosphamide usually occurs several days to 20 days after drug administration. Pathological examination reveals hemorrhagic myocardial necrosis, serosanguineous pericardial effusions, and fibrinous pericarditis.³ Conversely, cytosine arabinoside-induced cardiac damage and pericarditis are considered to result from anaphylactic reactions, usually seen during administration of the drug. Even low-dose cytosine arabinoside can induce pericarditis.⁴ A few cases of TBI-related pericarditis have been reported as a late-onset complication after irradiation.⁵ There also is a case report of pericarditis related to engraftment syndrome, which appeared during the engraftment period and resolved with corticosteroid therapy.⁶ In addition, some investigators have reported pericardial effusion associated with chronic GVHD, which occurred several months after HSCT.⁷

In our case, acute heart failure developed twice, and a pleural and a pericardial effusion were detected during both events. The first episode could be explained by cardiac damage because of cyclophosphamide or cytosine arabinoside, as it developed on day 11. The pleural effusion during both events was thought to be due to cardiac failure. However, the second episode differed from the first one with the development of pericarditis and cardiac tamponade, and detection of HHV-6 in the pericardial effusion. It is not likely that cyclophosphamide or cytosine arabinoside caused this episode of pericarditis, because the onset of pericarditis was 4 weeks after HSCT. Furthermore, she also did not meet the criteria for engraftment syndrome. Graft-versus-host disease-mediated pericarditis, too, is unlikely owing to the occurrence in the early phase after HSCT. We attributed the pericarditis to local HHV-6 reactivation in the pericardium because HHV-6 was detected in the pericardial effusion without HHV-6 viremia. Accurate diagnostic methods for HHV-6 associated diseases have not been established, because histologic examination by a biopsy for accurate diagnosis is frequently difficult early after HSCT because of disturbance of coagulation, low platelet counts, and thereby bleeding tendency. As a fallback position, detection of viral DNA methods by using a quantitative real-time polymerase chain reaction has become a useful option for diagnosing active HHV-6 infection.

Suminoe and associates have reported 2 cases of HHV-6–associated pleurisy.¹⁰ Similar to our case, elevation of HHV-6 DNA was detected only in pleural effusion. Two cases of HHV-6–associated pericarditis have been reported previously, however, diagnostic processes, treatment, and results were not described in detail.⁹ Human herpesvirus-6 is considered to have some affinity for the myocardium, because HHV-6 is increasingly found in endomyo-cardial biopsy samples in a patients with viral myocarditis.¹¹

Human herpesvirus-6 reactivation has been reportedly associated with allele-mismatched donors, umbilical cord blood transplant, myeloablative conditioning, and corticosteroid administration.^12,13 It occurred 2 to 4 weeks after HSCT, but often disappeared spontaneously. In patients who received HSCT from an allele-mismatch donor, an impaired immune response because of GVHD itself and immunosuppressive treatment against GVHD may, in part, contribute to HHV-6 reactivation.12 Elevated IL-6 level in the blood are considered to be associated with HHV-6 reactivation and development of HHV-6 encephalitis.^14,15In our case, acute GVHD involving the skin and neutrophil engraftment had already appeared before development of pericarditis. Hypercytokinemia derived from acute GVHD and engraftment may affect HHV-6 reactivation. We speculate that HHV-6 reactivation was suppressed by reconstitution of innate immunity as provided by natural killer cells, and that the pericarditis thereby improved without antiviral treatment.

Although we could not exclude other possibilities because of a lack of pericardial or myocardial biopsy specimens, HHV-6 might play a crucial role in developing pericarditis. In conclusion, HHV-6 should be considered a possible cause of pericardial effusion soon after HSCT.

Volume: 15 Issue: 2 April 2017

FULL TEXT