Author Affiliations: From the University of Michigan Health System, Ann Arbor, Michigan. The authors report that there are no disclosures or conflicts of interest relevant to this publication. Manuscript submitted March 3, 2008, provisional acceptance given April 22, 2008, manuscript accepted June 30, 2008. Address for Correspondence: Dee Dee Wang, MD, University of Michigan Health System,3116G TC SPC 5368, 1500 East Medical Center Drive, Ann Arbor, MI 48109-5368. E-mail: firstname.lastname@example.org
ABSTRACT: Spontaneous hemopericardium is extremely rare in patients with hemophilia and has never been reported in hemophilia B-factor IX deficiency. We report a patient with hemophilia B who presented with spontaneous hemopericardium and tamponade requiring emergent pericardiocentesis. As advances in therapies improve the longevity and quality of life of patients with hemophilia, it is likely that adult interventional cardiologists will increasingly encounter these patients in the cardiac catheterization laboratory. We review the clinical challenges raised by this group of patients undergoing cardiac procedures and present a stepwise approach to optimal management during the periprocedural period.
J INVASIVE CARDIOL 2008;20:E296–E300
Spontaneous hemopericardium is extremely rare in patients with hemophilia and has never been reported in a patient with hemophilia B-factor IX deficiency.1–4 This case presentation describes a patient with hemophilia B who presented with cardiac tamponade due to spontaneous hemopericardium following recent diagnosis and treatment for community-acquired pneumonia. Here, we review the clinical challenges raised by patients with hemophilia undergoing procedures in the cardiac catheterization laboratory. Recent advances in therapies, including the use of recombinant factor replacements, have improved the long-term prognosis of these patients making it increasingly likely that they will be encountered by interventional cardiologists.
Case Presentation. A 47 year-old male with a history of hemophilia B-factor IX deficiency presented to the hospital with anterior, nonradiating chest pain, aggravated with cough and deep respirations. The chest pain was worse in the supine position and improved with leaning forward. He also complained of worsening dyspnea on exertion and orthopnea. He denied any recent trauma. One week earlier he had been discharged following a brief hospitalization for community-acquired right-lower lobe pneumonia. This had been treated initially with intravenous ceftriaxone and azithromycin, and subsequently with a 10-day course of oral levofloxacin.
His past medical history also included HIV, hepatitis B and hepatitis C, all resulting from multiple blood product transfusions in the 1980s. In the past few years, he had intermittent episodes of hemoarthroses in both hips and left wrist, which were treated with intermittent recombinant factor IX infusions. Three months prior to this presentation, maintenance intravenous recombinant factor IX (3,000 units twice a week) was started to prevent recurrent hemoarthroses. Additional medications included an antiretroviral regimen of nevirapine, lamivudine and tenofovir.
On examination, he was hypotensive with a blood pressure of 96/81 mmHg and tachycardic (135 beats/minute). He was saturating normally on room air. An examination for pulsus paradoxus demonstrated a 20 mmHg drop in systolic blood pressure with inspiration. Prominent jugular venous distension (~12 cm above the sternal angle at 45 degrees) was noted in his neck examination. His cardiac examination revealed normal heart sounds with no gallops or pericardial rub. He had no peripheral edema. Electrocardiography showed sinus tachycardia with PR depressions in II, III, aVF, V4, V5, V6, and PR elevation in aVR as well as decreased voltage compared to previous electrocardiograms (ECGs) (Figure 1). Laboratory values showed normal cardiac biomarkers, hemoglobin of 11.3, with a platelet count of 598 K, a prothrombin time of 16.3 (with an international normalized ratio of 1.7), a partial thromboplastin time of 53.6, a factor IX assay of 4% of normal, fibrinogen of 566, AST of 583, ALT of 380 and alkaline phosphatase of 107. Chest radiography showed interval enlargement of the cardiac silhouette compared with a prior exam 1 week earlier (Figure 2). Bedside echocardiography revealed a large pericardial effusion with evidence of right atrial and right ventricular diastolic collapse, with Doppler evidence of significant respiratory variation in tricuspid and mitral inflow velocities (Figure 3).
Given evidence of hemodynamic compromise, an emergent pericardiocentesis was performed in the cardiac catheterization laboratory under fluoroscopy through a subxiphoid approach. He was given 5,000 units of intravenous recombinant factor IX one hour prior to needle access of the pericardial space. A 5 Fr pigtail catheter was placed in the pericardial space, draining 1 liter of bloody fluid. No drain was left in place. An echocardiogram performed immediately after the procedure showed significant improvement and minimal residual pericardial effusion (Figure 4). His heart rate decreased from 135 to 105 bpm, and his blood pressure increased to 119/77 mmHg. He tolerated the procedure well without complications. He received an additional dose of 2,500 units of recombinant factor IX 12 hours post procedure to raise his factor IX levels between 80–100% with intravenous infusions. Forty-eight hours following his procedure, repeat echocardiography showed no evidence of fluid reaccumulation. He was subsequently placed on 3,000 units of intravenous recombinant factor IX every 12 hours. His hospital course was otherwise uncomplicated, and he was discharged home 3 days after admission. He was continued on intravenous recombinant factor IX 3,000 units daily for 1 week, followed by a maintenance dose of 3,000 units twice weekly.
The precise etiology of his pericardial effusion remains unclear. Pericardial fluid analysis showed a cell count of 1,798,000 RBC/cmm, with 6,200 white blood cells, with a differential of 59% segmented neutrophils, 25% lymphocytes, 15% histiocytes, 1% eosinophil and a hematocrit of 22%. The fluid showed no evidence of bacterial, fungal or acid-fast bacilli growth, and cytology revealed no evidence of malignancy. Viral respiratory antigen screen and blood cultures were negative. Assays for rheumatoid factor and antinuclear antibody were negative. His CD4 count was 374 cells/cm3, unchanged compared to his prior baseline, with an undetectable HIV-1 quantitative RNA load. He tested positive for the hepatitis C antibody with a negative HCV RNA. His hepatitis B viral load was also undetectable, with a positive HBsAg and a nonreactive IgM anti-HBc antibody. A respiratory influenza virus type A antigen was indeterminate due to an inadequate specimen size.
One month after his initial presentation, the patient denied any further symptoms and repeat studies revealed a normal chest X-ray, ECG and echocardiogram without any evidence of recurrence of his pericardial effusion. His transaminitis also subsequently resolved and was thought to be related to passive liver congestion secondary to cardiac tamponade.
Discussion. Hemophilia A and B are sex-linked disorders resulting from deficiencies in the production of coagulation factors VIII and IX, respectively. Its overall incidence is approximately 1 in 5,000 male births, with the clinical severity dictated by the extent of factor level deficiency present. Severe hemophilia is associated with factor levels below 1%, moderate with factor levels between 1 and 5%, and mild with levels above 5%. Although bleeding can occur anywhere in patients with hemophilia, the most commonly affected sites are muscles, joints and the gastrointestinal tract. Prior studies suggest that hemophilia B is less clinically severe than hemophilia A and is associated with fewer hospital admissions.5–7
Hemopericardium has been previously reported in patients with hemophilia after surgery, trauma or injury to the chest or great vessels.8 Two cases of hemopericardium have been reported in patients with hemophilia B as complications of cardiac surgery and ascending aortic dissection, respectively.9,10 Although both of these patients had mild deficiencies of factor IX (i.e., > 5% of normal), they eventually required drainage of the pericardial effusions. This supports earlier observations that even patients with mild deficiencies may be at risk for hemopericardium after surgery or trauma.9–11 While our patient had slightly lower factor IX levels at baseline (4% of normal), he had no preceding history of surgical intervention, aortic dissection or trauma.
Another important issue to consider in this case was the patient’s HIV infection. Pericardial effusions associated with HIV are common manifestations of HIV cardiac involvement, with an incidence of 11% per year, and are closely associated with the stage of HIV infection.12–14 The majority of HIV-associated pericardial effusions are small and asymptomatic, while the incidence of cardiac tamponade for AIDS patients is 1% per year.13 In a case series of 185 documented cases of cardiac tamponade in patients with HIV disease, the mean CD4 cell count was 98 ± 95 cells/cm3, with opportunistic infections being the most common etiology such as mycobacterial infection (42%), followed by bacterial origins (11%), lymphoma (8%), Kaposi sarcoma (7%), fungal (3%) and viruses (2%).15 However, our patient had a CD4 count of 374 cells/cm3, his HIV-1 RNA viral load was undetectable and he had no history of AIDS-associated opportunistic infections. While it is unlikely that the pericardial effusion was solely due to HIV, it is possible that a small pericardial effusion in that location from the infection may have transformed due to hemorrhage. Similarly, hepatitis B and C have rarely been associated with the development of acute pericarditis and pericardial effusions.16,17 Our patient, however, had undetectable levels of both HBV DNA and HCV RNA.
Comprehensive analysis of the pericardial fluid did not identify any specific etiology in this case. We hypothesize that our patient most likely suffered from a spontaneous hemopericardium despite his maintenance infusions of factor IX. A possible precipitating factor for the bleeding may have been pericardial inflammation associated with a parapneumonic process from his preceding community-acquired pneumonia or a baseline pericardial effusion associated with his HIV infection. In this case, a small and benign pericardial effusion may have rapidly converted into hemopericardium, resulting in cardiac tamponade. There are two previously reported cases of spontaneous hemopericardium associated with viral prodromes, both of which occurred in hemophilia A patients.1,3
Not surprisingly, invasive treatments are frequently challenging in patients with hemophilia. Any procedure has the potential for bleeding complications, and up until the 1960s, patients with hemophilia underwent only emergent surgeries suffering a mortality rate of up to 50%.18 Recent advances in therapies, particularly the development of coagulation factor concentrates and recombinant factors have improved outcomes, but the potential for postoperative complications remains significant.19 Most data on the safety of procedures in these patients are limited to orthopedic, dental and gastrointestinal surgery.19–21 There is a paucity of evidence regarding the safety of cardiac procedures in these patients.
Currently, there is little consensus on the optimal management of patients with hemophilia and moderate or severe factor level deficiencies who require invasive procedures.22–24 Replacement products for factor VIII and IX include cryoprecipitate and recombinant factors. The choice between these two types of replacement products often requires consultation with a hematologist, since there is a need to balance many competing issues related to the products including safety, purity and costs. In our patient, recombinant factor IX was used due to his prior exposure with this replacement product and his likelihood of requiring maintenance infusions.
The required dose for replacement products is primarily driven by the clinical scenario.25 In our patient with a life-threatening bleed requiring an emergent procedure, the goal was to achieve immediate correction of his factor IX levels to 80–100%, with a calculated recombinant factor IX dose of 5,000 units. This was followed, in the postprocedural phase, by twice daily doses adjusted based on the patient’s serum factor IX levels to a goal of 80–100% and continued for 48 hours until there was no evidence of recurrent bleeding. The patient was subsequently discharged on an additional 7 days of subcutaneous daily recombinant factor IX doses. Table 1 summarizes a stepwise approach for the management of our patient.
Few data are available on hemophilia patients specifically undergoing procedures in the cardiac catheterization laboratory. For example, there have been only a handful of case reports of successful percutaneous coronary interventions in patients with hemophilia.26,27 These studies used a variety of approaches including both bare-metal and drug-eluting stents as well as different regimens for anticoagulation and antiplatelet therapies.26–33 In patients with severe hemophilia, bare-metal stent implantation may be preferred over drug-eluting stents given the shorter duration of required dual antiplatelet therapy and the subsequent decrease in the risk of hemorrhagic events.29 In one case, a radial approach was favored over femoral access allowing for more efficient mechanical hemostasis, whereas a closure device (StarClose Occluder, Abbott Vascular, Abbott Park, Illinois) was successfully used in another case.30,31
In summary, hemopericardium is a rare but life-threatening complication in patients with hemophilia. These events are typically precipitated by trauma, but spontaneous bleeding can also occur or be triggered by underlying pericardial inflammation or disease. As advances in therapies improve the longevity and quality of life of these individuals, it is likely that adult interventional cardiologists will increasingly encounter patients with hemophilia in the cardiac catheterization laboratory. Familiarity with a stepwise approach can lead to optimal management during the periprocedural period.