Respiratory Viral Detection in the Pediatric Hematopoietic Stem Cell Transplant Population

Respiratory viral infections (RVI) are a frequent cause of morbidity and mortality in hematopoietic stem cell transplant (HSCT) patients. We examined clinical characteristics and respiratory viral detection in asymptomatic pediatric HSCT pre-transplant patients and symptomatic post-transplant patients. Coxsackie/echovirus (most common virus detected pre and post-transplant), rhinovirus, and coronavirus were detected pre-transplant and at the first post-transplant event suggesting persistent detection. None of the clinical characteristics examined were associated with viral detection and there was no increase in mortality noted with asymptomatic viral detection.


Introduction
Infections, including respiratory viral infections (RVI), are one of the leading causes of morbidity and mortality in hematopoietic stem cell transplant recipients as a direct result of the viral infection and non-infectious sequelae including graft versus host disease and allo-immune lung syndromes [1][2][3]. Few studies in pediatric HSCT patients have examined the presence of viruses in the immediate pre-transplant period through the follow-on post-transplant period and it has been suggested that patients with asymptomatic viral detection do not have increased mortality and those with symptomatic viral detection do have increased mortality [4]. Screening for respiratory viral infections may impact a physician's decision of timing for HSCT and transplant may be delayed until viral clearance is documented [5,6].
Studies on respiratory viruses in the pediatric HSCT transplant population have identified rhinovirus as a common virus found in the pre-transplant period and rhinovirus, coronavirus, parainfluenza virus, respiratory syncytial virus and adenovirus in the post-transplant period [4, [7][8][9][10][11]. We evaluated asymptomatic pediatric HSCT patients for the presence of respiratory virus in the immediate pre-transplant period and symptomatic patients for up to 90 days post-transplant to examine epidemiology and clinical characteristics associated with viral detection.

Study Design
Allogeneic and autologous hematopoietic stem cell transplant (HSCT) candidates between 1 month and 17 years of age were enrolled in this single-site, prospective study. There were two enrolment periods; Phase

Routine clinical virology testing
All patients scheduled to undergo HSCT had a nasopharyngeal wash (NPW) for viral testing as part of clinical care prior to hospital admission. During the Phase 1 of the study, viral testing for clinical care was performed using standard tissue culture and antigen detection techniques (respiratory syncytial virus, influenza, parainfluenza, and adenovirus).
Due to the improved sensitivity of newer diagnostic techniques, during Phase 2 of the study, the routine clinical testing transitioned to a multiplex RT-PCR assay capable of detecting eight viruses (adenovirus, human meta-pneumovirus, parainfluneza 1, 2, 3, influenza A and B, and respiratory syncytial virus).
After transplant, when patients presented with fever and/or symptomatic respiratory events, NPW samples were collected for viral testing as part of routine clinical care. Signs and symptoms of respiratory events included: fever, cough, tachypnea, headache, malaise, sore throat, nasal congestion, poor feeding, abnormal lung exam and/or increase in oxygen requirement. Study patients were followed up to day 90 after HSCT. As with the pre-transplant specimens, routine clinical testing during Phase 1 was performed using viral culture and antigen detection and during Phase 2 testing was performed using the commercial 8-pathogen multiplex RT-PCR.

Supplemental Virology Testing
Patients were enrolled in the study within 72 h of hospital admission for HSCT and additional baseline NPW samples were collected for an expanded target, multiplex viral RT-PCR. In addition to clinical testing described above, samples collected during respiratory events were also stored for later testing using the expanded testing.
The research samples were processed using the BioRobot EZ1 and the EZ1 Virus minikit (Qiagen, Hilden, Germany). Molecular testing was performed using the research Qiagen Resplex II V2.0 RT-PCR kit. The kit detects: respiratory syncytial virus A and B, influenza A and B, parainfluenza virus 1, 2, 3, 4, human metapneumovirus; coxsackie/echovirus, rhinovirus, coronavirus 229E, OC43, NL63, HKU1, bocavirus, and adenovirus B and E. Both clinical and research testing results were included in the analysis. If either testing modality was positive, it was considered a positive result. Clinical information was collected from the chart by study personnel. Analysis was performed using JMP PRO®, Version 11 SAS Institute Inc., Cary, NC.

Results
A total of 45 subjects were enrolled over both Phase 1 (N=20) and Phase 2 (N=25). The mean age of participants was 6.8 years (range: 0.7-17.8), with a median of 5.5 years. Patients were split equally by gender (51.1% male) and were predominantly Caucasian (62.2%), with a large African American minority (28.9%). Study patients were enrolled in the study for a mean of 106 days and 43 of 45 patients (95.6%) were followed 90 days post-HSCT.
The most common underlying conditions leading to HSCT were inborn errors of metabolism (31.1%) and hematologic malignancy (26.7%), followed by other conditions (20%), solid tumors (13.3%), and myelodysplastic syndrome (8.9%). All patients were to receive their 1 st HSCT, including 28 unrelated cord blood transplants, 10 autologous transplants, and 6 matched, related bone marrow transplants. Mean time of hospitalization was 60 days ± 29 days (continuous days from pretransplant admission to post-transplant discharge).
All 45 patients had baseline samples collected at the time of enrollment. Of these, 10  from respiratory failure attributed to adenovirus, aspergillus, and toxoplasmosis. Table 1: Acute respiratory event signs and symptoms compared for those who had a positive viral detection compared those who had a negative viral detection during an acute respiratory event.

Discussion
RVI's are one of the leading causes of morbidity in HSCT recipients and may result in death; with one recent study demonstrating a case-fatality rate of 10% in the pediatric HSCT population [9]. Clinical clues to diagnose respiratory infection can be unreliable in immunocompromised patients, although clinical scoring systems have been proposed [12]. Our study did not identify any specific demographic, sign, or symptom associated with viral detection in pre and post-transplant patients. We identified coxsackie/enterovirus as the most common viral etiology detected both in the asymptomatic pretransplant period and the symptomatic post-transplant period. Bocavirus was also frequently identified during the posttransplant period. Other studies examining the pediatric HSCT population have identified rhinovirus as a common virus found in the pre-transplant period and rhinovirus, coronavirus, parainfluenza, respiratory syncytial virus, and adenovirus in the post-transplant period [4, [7][8][9][10][11]. It is interesting that these studies identified patients with respiratory syncytial virus and influenza in their study population but neither virus was found in patients in our study despite enrollment during typical respiratory seasons.
In our study, a number of viruses (coxsackie/echovirus, rhinovirus and coronavirus NL63) were detected at baseline and repeatedly during acute respiratory events, suggesting the possibility of asymptomatic infection pre-HSCT resulting in prolonged shedding, perhaps as a result of immunosuppression. Furthermore, subclinical respiratory viral infection has been documented and is considered transmissible [13][14][15]. A recent study by Campbell et al., looking at clinical outcomes in patients with pre-HSCT viral detection and post-transplant viral detection found no increased mortality with asymptomatic viral detection including those with asymptomatic pre-transplant viral detection but did find increased mortality with symptomatic viral detection including those with symptomatic pre-transplant viral detection [4]. We also found no increased mortality associated with asymptomatic pre-transplant viral detection.
The major limitation of the study was the small sample size. It would have also been interesting to perform viral sequencing and continue to test pre-transplant patients with positive viral detection on an on-going basis to determine the duration of viral persistence and more closely examine its impact on patient outcome. However, important strengths are the prospective nature of this study, as well as the use of internal controls during RT-PCR testing that verified the integrity of the archival specimens.
While we can now identify these respiratory viruses in this population, their role in ultimate clinical outcome and the degree of nosocomial transmission remains unclear. There are increasing numbers of patients undergoing HSCT with increasing complexity. However, there also are newer drugs being studied for the treatment of viral infections (i.e., adenovirus, parainfluneza virus, and respiratory syncytial virus) in this population and they offer a potential to suppress or eliminate these viruses throughout the peri-transplant continuum [16][17][18]. Further prospective studies will be needed to determine the clinical impact and optimal prevention and treatment strategies for RVI's in this population.