Disseminated Coxsackie A9 Infection Complicating Bone Marrow Transplantation

Opportunistic viral infections occur often in patients undergoing bone marrow transplantation (BMT) and can be a major cause of morbidity and mortality. Cytomegalovirus, Epstein-Barr virus, adenovirus, varicella-zoster virus and herpes simplex infections have been well-described in these patients.1 Enteroviral infections have rarely been described in patients undergoing bone marrow transplantation.2 Coxsackie A virus infection has been identified as a cause of gastroenteritis in BMT patients3; however, to our knowledge no cases of disseminated coxsackie A virus infection have been reported. We describe the clinical and pathologic findings in a child who developed meningitis and pneumonitis caused by coxsackie A9 infection. Case report. A 21-month-old Latin-American infant with recurrent acute lymphocytic leukemia (ALL) in second remission was referred for allogeneic BMT. No HLA identical donors were identified and an initial review of the volunteer donor registries revealed no suitable donors. An unrelated umbilical cord blood unit was found that was a one antigen serologic mismatch at the B locus. Typing was confirmed by high resolution DNA typing. At the time of evaluation for BMT, the patient was in complete remission from leukemia, had normal organ function and had no evidence of infection. The child was admitted to the bone marrow transplant unit in June, 1995. Four days before transplantation he began infection prophylaxis with intravenous acyclovir, amphotericin B (0.25 mg/kg/day), intravenous pentamidine and weekly intravenous immunoglobulin (IVIG; Sandoglobulin®; Sandoz Pharmaceuticals, East Hanover, NJ). The induction regimen consisted of total body irradiation at a total dose of 1350 cGy given in 9 fractions of 150 cGy, melphalan (60 mg/m2/day) for 3 days and anti-thymocyte globulin (Atagam®; Upjohn Pharmaceuticals, Kalamazoo, MI) in a dosage of 30 mg/kg/day for 3 days. Graft vs. host disease (GVHD) prophylaxis consisted of four doses of methotrexate, continuous infusion cyclosporine and methylprednisolone. The umbilical cord unit was thawed and infused without complication. Two days after transplant he developed fever and therapy was begun with ceftazidime and vancomycin. Fevers persisted and on Day 7 the amphotericin B dose was increased to 1 mg/kg/day. Blood and urine cultures were sterile. He achieved an absolute neutrophil count of 500/mm3 on Day 21 and an absolute neutrophil count of 1000/mm3 on Day 22. On Day 30 a bone marrow examination showed a hypocellular bone marrow with evidence of trilineage engraftment. Cytogenetics showed complete chimerism. The patient remained stable until Day 19 when he developed a cutaneous rash which on biopsy was shown to be acute GVHD, and treatment was started with 2 mg/kg/day of methylprednisolone. He also developed persistent vomiting. Endoscopy revealed esophageal stricture, and biopsies showed Grade III upper gastrointestinal GVHD. On the day after endoscopy he developed diarrhea. The dose of steroid was increased to 4 mg/kg/day. On Day 30 he developed bloody diarrhea and was treated with two pulse doses of methylprednisolone (500 mg/m2) with some improvement in his diarrhea. All stool cultures, including Clostridium difficile toxin A testing, were negative. He was again stable until Day 40 when he developed high fevers, tachypnea, irritability and altered mental status. A computerized tomogram of the head was negative and a chest roentgenogram showed diffuse interstitial infiltrates. No history of sick contacts was elicited. Lumbar puncture and bronchoalveolar lavage were performed. Cerebrospinal fluid (CSF) examination revealed 10 nucleated cells/mm3 (35% segmented neutrophils, 17% monocytes and 48% lymphocytes) and 933 red blood cells/mm3 with a glucose of 104 mg/dl and a protein of 22 mg/dl. Bacterial and fungal cultures were negative; however, viral cultures from the CSF and bronchial washings were both positive for coxsackie A9 virus. Two days later the child developed respiratory failure and was transferred to the intensive care unit for airway intubation and ventilatory support. The patient received intravenous gamma-globulin in a dosage of 1 g/kg/day for 5 days. The child improved clinically and was extubated 10 days later. Repeat CSF cultures were negative. The patient continued to have bloody diarrhea due to severe gastrointestinal graft vs. host disease and received antithymocyte globulin in a dosage of 20 mg/kg/day for 3 days. He subsequently developed septic shock on Day 64 with pulmonary edema and renal failure. Blood and CSF cultures were sterile. The patient developed progressive respiratory failure and died on Day 66. Postmortem cultures from blood, urine, CSF and lung tissue were negative for coxsackie A9 virus. Cytomegalovirus, bacterial and fungal cultures were sterile. The postmortem examination of the lungs showed extensive diffuse alveolar damage with hyaline membrane formation and bilateral patchy bronchopneumonia was seen. Examination of the heart showed no evidence of myocarditis. The gastrointestinal tract had diffuse ulceration and hemorrhage of the intestine consistent with grade III GVHD. No evidence of maternal lymphocyte contamination of the umbilical cord unit was found in the patient's bone marrow or gastrointestinal tract. Discussion. Coxsackie A virus is frequently recovered from the stools of healthy subjects without evidence of disease. Coxsackie viruses are most commonly associated with nonspecific febrile illnesses, and have also been identified as a cause of exanthems, pharyngitis and pneumonia. Coxsackie B viruses are associated with myocarditis, whereas type A viruses are rarely associated with myocarditis.4 Although our patient had multiorgan involvement, he had no evidence of myocarditis during life or on postmortem examination. Coxsackie viruses have also been associated with meningitis, encephalitis and acute transverse myelitis in immunocompetent hosts.5 Because the coxsackievirus was identified from a bronchoalveolar lavage specimen and not an open lung biopsy, it cannot be proved that the virus caused the interstitial pneumonia. Because this virus was the only organism recovered, it was presumed to be the etiologic agent of the pneumonia. Coxsackie viruses have been rarely associated with more serious infections in immunocompromised patients. McKinney et al.6 identified 42 patients with agammaglobulinemia with chronic enteroviral meningoencephalitis; 2 (5%) had coxsackie A virus isolated. Although rare, cases of coxsackie A infection have been described in BMT patients, and to our knowledge there are no cases of interstitial pneumonitis caused by coxsackie A virus reported in the literature. A fatal case of disseminated ECHO 11 virus has been described in the BMT setting,2 as has an outbreak of coxsackie A1 gastroenteritis in BMT recipients.3 Our patient should have been at low risk for severe GVHD based on the age of the recipient, typing and lack of maternal lymphocytes found on autopsy specimens. It is possible that the viremia associated with the coxsackie A infection may have predisposed him to severe GVHD. Several studies have hypothesized a correlation between viral and bacterial infections and severe GVHD. The mechanism is not known but may result from increased cytokine production induced by the response to the infection or by reactivation of latent virusinducing antigens on cell surfaces that become targets of alloreactivity.7 No specific therapy for Coxsackievirus infection is known. On the basis of several anecdotal reports describing the successful use of intravenous gamma-globulin for treatment of chronic enteroviral infections in immunocompromised patients,6, 8, 9 we elected to treat this child with high dose intravenous immunoglobulin. The mechanism of effectiveness of neutralizing antibodies has been postulated to result from a direct action of the antibody on the virus during its extracellular phase as the virus is transferred from cell to cell.9 Our patient improved clinically with the administration of IVIG. However, the patient's course without the IVIG is unknown, as is the titer of antibody to coxsackie A in the IVIG administered. We describe a patient undergoing bone marrow transplant who developed disseminated coxsackie A virus infection. Coxsackie A virus must be considered in the differential diagnosis of interstitial pneumonitis in the bone marrow transplant setting, especially in patients with concurrent central nervous system involvement. IVIG may have a role in its therapy. Acknowledgments. This work is supported by National Institutes of Health T32 Training Grant CA09640, the Children's Cancer Fund of Dallas and A Weekend to Wipe Out Cancer. Victor M. Aquino, M.D.; Roula A. Farah, M.D.; Michael C. Lee, M.D.; Eric S. Sandler, M.D.