Prolonged Rhinovirus Shedding in a Patient with Hodgkin Disease

To the Editor—Respiratory viral pathogens (RVPs) have been increasingly identified as a serious concern in immunocompromised patients. In this population, RVPs cause more lower-respiratory tract infections (LRIs), leading to increased mortality and morbidity. Prolonged viral shedding of RVP can become an infection control problem and has been implicated in at least 1 hospital outbreak. With respect to the hematopoietic stem cell transplant (HSCT) population, most publications have studied more virulent RVPs, whereas data on the nontransplant immunocompromised population with less virulent RVP are lacking altogether. Compared with other RVPs, rhinoviruses (RVs) cause proportionately fewer LRIs in the healthy population, but RVs are more prevalent than other RVPs and infect 22.3% of HSCT recipients within 100 days of transplantation. In a small retrospective study of immunocompromised patients and without inferring causation, RVs were associated with the same mortality as the 2009 H1N1 influenza. We report a patient with relapsed Hodgkin’s Disease (HD) without a transplant who was found to have prolonged RV shedding of 96 days with LRI. Our patient was a 37-year-old man with prior lung injury from acute respiratory distress syndrome, CD4 lymphopenia with recurrent pneumonia, and relapsed HD after treatment with bleomycin, adriamycin, vinblastine, and dacarbazine, treated with brentuximab. He experienced intermittent fever beginning in September 2014 and presented in late October 2014 with progressive dyspnea, continuing intermittent fever, and a nonproductive cough. He was hypoxemic on admission. Chest CT showed bilateral ground-glass opacities. Bronchoalveolar lavage (BAL) performed on October 29, 2014, was RT-PCR positive for RV/ enterovirus (EV). Other infectious disease testing was negative. Intravenous immunoglobulin was given with tapering prednisone for bronchospasm. He improved and was discharged a few days later. He remained afebrile with continued dry cough and dyspnea during November and December. In January, he began having afternoon fevers (38.9–39.5°C [102–103°F]), dyspnea, productive cough of whitish to yellow sputum, weight loss, drenching night sweats, and lymphadenopathy. He was readmitted in late January 2015 with severe sepsis and hypoxemia. Another chest CT showed progression of interstitial and airspace opacities. A nasopharyngeal swab was collected on January 31, 2015, and BAL was performed on February 2, 2015; both were RT-PCR positive for RV/EV; adenovirus PCR was also positive on the BAL. The patient was transitioned to comfort care after a repeat biopsy showed progression of HD, and he died February 5, 2015. Sanger-sequencing and bioinformatic analyses of clinical specimens from October 29, 2014, January 31, 2015, and February 2, 2015, identified RV-A51. Prolonged viral shedding, seen in immunocompromised patients, is dependent on the host’s immune status, virus species and strain, lung injury, and other risk factors, all of which are still poorly understood. This patient had at least 96 days of RV-A51 shedding, but because his symptoms started in September, viral shedding possibly started earlier than documented. Rhinovirus, which causes common cold, is a common etiology of respiratory infections. The normal host clears the infection in a short period, limiting the duration of infection and viral shedding. In a study of hospitalized patients with respiratory complaints, the mean duration of RV shedding was 10.1 days in adults with no known immunocompromising condition. In the HSCT population, the median duration of shedding was 3 weeks (range, 0–49 weeks), and in patients with hypogammaglobulinemia, the median duration of shedding was 40.9 days (range, 26.4–55.4 days). Due to this variability and our inability to predict the duration of viral shedding in immunocompromised patients, it may be necessary to test for RV or other RVP negativity before isolation precautions are removed. The incidence of RV LRI is unclear. In a prospective study of 215 HSCT recipients followed for 100 days, the incidence was 4% among the RV-infected recipients. However, in a retrospective chart review of HSCT recipients with RV infection, 43% subsequently had proven or possible RV-associated pneumonia, but more than half (60%) had at least 1 additional respiratory pathogen detected, confounding the attribution of the pneumonia. On his second admission, our patient was coinfected with adenovirus, which possibly aggravated the patient’s pulmonary condition either by itself or in combination with the recurrent HD. Whether a primary pathogen or a copathogen, RV infection has the potential to negatively affect the survival of immunocompromised patients. Establishing duration of viral shedding defines the course of infection, infectivity, and need for preventive strategies. Unfortunately, factors that predict duration of viral shedding have not been determined. Patient isolation and precautions for infection control should probably bemaintained until RVP testing becomes negative to avoid hospital transmission.