Various nonspecific and specific factors contribute to resistance to infection and recovery from infection. These mechanisms include the mucosal barrier, fever, local inflammatory response, α and β interferon, nonspecific phagocytosis, and the humoral and cellular immune response. A number of immune deficiencies that contribute to a greater susceptibility to respiratory infection have been described after bone marrow transplant (BMT). The high degree of immunodeficiency in these patients because of their chemotherapy leads to mucositis, with loss of integrity in the mucosa and immunoglobulin (Ig)A impairment. The low production of serum antibodies increases the risk of infection in the lower airways. Deficient cytotoxic T-cell function reduces the capacity for viral inactivation, resulting in increased magnitude and duration of the infection, with a greater probability of evolving to viral pneumonia (1). Some features distinctive among immunocompetent and immunocompromised patients with respiratory viral infections are high frequencies of nosocomial acquisition, persistence of infection beyond usual time periods, and high frequency of pneumonia and death in association with infection. The role of the virus infection in causing pneumonia and death in immunocompromised patients is uncertain because the patients often have multiple infections. Thus, the immune deficiencies can account for a high frequency of viral respiratory infections as well as an increased severity of the illness and a significant risk of death. The variability of the frequency and severity of the infections in immunocompromised persons should reflect the degree of immune deficiency (2,3).
Respiratory infections caused by the respiratory syncytial virus (RSV), adenovirus, influenza A and B, and parainfluenza are frequent in immunocompromised patients. Such infections can be acquired by nosocomial transmission or through community outbreaks. In this population of patients, these infections are associated with high morbidity and mortality rates (4).
Clinical manifestations can be limited to infection of the upper airways, but these viruses are also responsible for a serious infection of the lower airways, which can be fatal (5). RSV infection in BMT patients appears to evolve independently of the patient's age, with characteristic signs of upper respiratory involvement such as rhinorrhea and cough preceding the development of signs and symptoms in the lower airways such as tachypnea, rales, and hypoxemia. The development of pneumonia is usually a preterminal event (6).
Pneumonia is one of the most common complications in patients submitted to BMT (7). The viral agent most frequently described as responsible for this disease has been cytomegalovirus (CMV). However, there has been an increase in reports of community respiratory viruses causing many of the pneumonias classified until recently as idiopathic (8).
The involvement of these respiratory viruses in BMT patients in our country has been discussed but, up to now, the importance of these for the success or failure of BMT has not been analyzed nor has the seasonality of these viruses in a country with a significant climatic variability such as Brazil.
To evaluate the importance of these viruses in upper and lower respiratory tract diseases in patients submitted to BMT, a retrospective study was conducted in patients with positive samples for respiratory viruses analyzed at the Virology Laboratory of the University Hospital of the Federal University in the state of Paraná, Brazil.
MATERIALS AND METHODS
Respiratory secretions that had been sent to the Virology Laboratory from March 1993 to August 1999, from hospitalized and ambulatory patients submitted to BMT and who presented acute respiratory disease or coryza, were reviewed. Records of patients with reactive samples were evaluated retrospectively, with the following being observed: demographic data, underlying disease, date and type of transplant, clinical signs and symptoms (fever, cough), radiographic or tomographic findings, and clinical evolution.
Upper respiratory tract disease was defined as the presence of rhinorrhea, nasal or sinus congestion, otitis media, pharyngitis, or cough with a normal chest radiograph. Pneumonia was defined as acute respiratory disease in association with a new radiographic infiltrate (8).
Samples consisted of nasopharyngeal aspirate (NPA) or bronchoalveolar lavage (BAL) collected in all BMT patients with upper respiratory infection during the first days of symptoms. The NPA was obtained through a catheter attached to a suction device and was sent to the laboratory in coolers and processed within 6 hours. Detection of viral antigens was performed by using indirect immunofluorescence (IIF) with specific monoclonal antibodies for RSV, adenovirus, influenza A, influenza B, and the parainfluenza group (which detects types 1, 2, and 3) (Chemicon International, Inc., Temecula, CA). A background control was routinely used.
During the study, 810 samples collected from 722 patients that had been sent to the Virology Laboratory by the BMT unit were analyzed. In some patients, the samples were taken more than once to watch for negative conversion of the viral secretion. Of that total, 136 samples were reactive for respiratory viruses in 62 patients. Sixty-nine percent of the patients who presented with viral respiratory infection after BMT had chronic myelogenous leukemia (CML) or severe aplastic anemia (SAA) as their underlying disease. The majority of patients had received allogenic consanguineous transplants, and the infections occurred with equal frequency during the first 30 days, as compared with more than 90 days postBMT. Of the 62 patients studied, 71% were male. Median age was 23 years (Table 1).
Respiratory Viruses in Bone Marrow Transplantation
Respiratory viruses were found in 136 of 810 (17%) samples, with RSV being the most frequently detected and adenovirus the least. One patient had been infected concomitantly by two viruses, influenza B and parainfluenza.
Pneumonia occurred in 19 of 62 (31%) patients, all of whom began with symptoms of upper respiratory infection. Pneumonia was more frequent in patients infected by RSV and adenovirus.
Thirty-seven percent (23/62) of patients infected with these respiratory viruses died, 26% (8/30) of cases infected with RSV and 57% (8/14) with influenza A (Table 2).
The distribution of patients infected with respiratory viruses during the different postBMT periods showed that the infection occurred principally during the first month in 22 of 62 (35.5%) patients and after 3 months postBMT in 26 of 62 (42%) patients. The influenza B virus was involved in the largest number of deaths during the period between 60 and 90 days posttransplantation. On the whole, there is no significant difference in the mortality rate between the different periods of the infection. Of the eight deaths that occurred in the patients infected by RSV and influenza A, four occurred in a period shorter than 30 days, one between 30 and 60 days, and 3 after 90 days (Table 3).
Seasonal Characteristics of Respiratory Viral Infections inBone Marrow Transplantation
Figure 1 shows the proportion of respiratory virus detection in relation to the number of the samples collected. An increase in the incidence of respiratory viruses was observed in 1998 resulting from a higher number of RSV cases.
As had happened with community infection, RSV and influenza A occurred in these patients during the winter (Fig. 2). The frequency of influenza B, parainfluenza, and adenovirus was very low, which made their seasonal analysis difficult.
All the patients presented with upper respiratory symptoms such as rhinorrhea, nasal congestion, or pharyngitis. The other frequent clinical manifestations were fever in 36 of 62 (59%) infected patients and cough in 32 of 62 (52%).
Pneumonia associated with infection by RSV, adenovirus, influenza B, and parainfluenza was diagnosed in some patients. It was described in 15 of 30 (50%) patients infected by RSV and 1 of 2 (50%) infected by adenovirus. No cases of viral pneumonia were detected in patients infected with influenza A virus (Table 2). Radiographic and tomographic tests were conducted in 25 of 62 (41%) and 7 of 62 (12%) patients, respectively. Alveolar condensation was the most frequently observed abnormality followed by interstitial infiltrate.
The principal underlying disease in patients undergoing BMT in this study was SAA, but infection by respiratory viruses occurred more frequently in patients whose underlying disease was CML, which may result from the need for greater immunosuppression for transplantation in the latter, making them more susceptible to infection.
Until recently, infections with community respiratory viruses were not adequately recognized, but the results of this study suggest that community respiratory viruses are important causes of potentially serious acute respiratory illnesses in BMT recipients, which is consistent with reports by other centers (4, 5, 8–11).
Detection of viral antigens using IIF is commonly performed in many medical centers because this method is highly sensitive, specific, rapid, and low cost (12). Speed and accuracy in diagnosing these viruses, particularly RSV, have important implications for patient care because they allow for early treatment, prevent the unnecessary use of antibiotics, possibly delay the beginning of chemotherapy, and orient hospital infection control measures (13). In adults, IIF assays are probably too insensitive to detect some respiratory virus infections, and the test's sensitivity increases when performed on BAL as compared with NPA samples (13, 14) so that in high-risk patients, such as those submitted to transplantation, bronchoscopy may be indicated as early as possible when there are respiratory symptoms, for investigating the etiologic agent in cases of negative NPA.
This study showed 17% (136/810) of positive respiratory secretion samples from BMT patients, which is similar to the results published by other authors (10). This study demonstrated that the types of viruses infecting BMT patients varied over the years, with the overlapping of viral types during some seasons, RSV being the predominant respiratory pathogen (10, 11) and followed by influenza A, influenza B, parainfluenza group, and adenovirus. It was observed that in some years the frequency of the viruses was very low, around 5%. When we compared this with pediatric inpatients, it became evident that the epidemics in those years was less severe and involved a low number of patients. On the other hand, in other years, such as 1998, we observed a high percentage of positive samples, around 25%, that could be observed in pediatric inpatients.
As reported previously, the influenza A and influenza B viruses appeared as the second and third most frequent causes, respectively, of respiratory viral disease (10). However, studies demonstrate a very low frequency of the influenza virus in some European centers (some 2.7%), probably the result of the recommended immunization of family contacts and health care professionals, thereby preventing infection of the immunosuppressed patient (11). The low influenza immunization rate in Brazil explains the greater frequency of these viruses in our patients. A high mortality rate (8/14, 57%) among patients infected by influenza A was observed, but this could not be clearly related to this virus: 2 of 8 for recurrence of disease, 2 of 8 for pulmonary infection by CMV and Aspergillus fumigatus, 2 of 8 for graft-versushost disease (GvHD), 1 of 8 for massive pulmonary hemorrhage, and 1 of 8 patients who probably presented with GvHD, was not submitted to necropsy. Forty percent (12/30) of RSV infections occurred in inpatients, probably through nosocomial transmission. Because the BMT outpatients come to the hospital daily to receive drugs, it is possible that other infections could be nosocomial.
In contrast with the RNA respiratory viruses, adenoviruses are capable of establishing latent infection, primarily involving lymphoid cells. Therefore, although the adenovirus infection can occur by inhalation of aerosolized virus and is reported to be frequently occurring from fall to spring, it is likely that the infections of BMT patients are primarily caused by endogenous reactivation rather than of exogenous origin (3).
With regard to the posttransplantation period in which the respiratory viral infection occurred, the critical periods were the first 30 days and after 90 days postBMT. The majority of respiratory viral infections occurred in this period. This may reflect the fact that the first 30 days postBMT is when the patient undergoes the most intense immunosuppression and is thus most prone to infection. During the 30 to 90 day period, patients gradually recover their immune status and have little contact with community infections. The number of respiratory viral infections increases again after 90 days postBMT because patients gradually return to contact with the community and are more exposed to infection by circulating viral agents.
In relation to seasonal distribution, respiratory viruses were concentrated in given periods of the year. They occurred predominantly in the winter months in the state of Paraná (Southern Hemisphere), with the highest peak in the month of July and no cases of respiratory viral infection in the months of January and December. Seasonal circulation of these viruses has been observed by our unit in nonimmunocompromised patients (unpublished data), which emphasizes the need for surveillance of these viral infections in groups of patients at high risk for severe infection.
Fever and cough were the most frequently observed clinical manifestations in these patients. Pneumonia is the main complication of respiratory viral infection (3). Contrary to those patients with CMV infection, all the patients that evolved to pneumonia initially presented with upper respiratory tract disease. The presence of these symptoms should call attention to the diagnosis of infection with community respiratory viruses (15). Pneumonia was reported in 50% of the patients with RSV or adenovirus infection. Mortality caused by RSV infection occurred in 8 of 30 patients (26%), a lower rate than in earlier reports on this infection in BMT (16), probably because of current therapeutic measures including early specific antiviral therapy (11).
No case of viral pneumonia was reported in the group of patients with influenza A infection, a result that merits attention and further study. However, a high mortality rate was observed in the influenza A patient group, probably because of the bacterial infections associated with this viral agent, as reported by other authors (10, 15, 17). Parainfluenza viruses were detected in 7 of 62 (11%) patients, a higher rate than in previous reports (5%) (9). Pulmonary complications in our patients occurred in only 2 of 7 cases (28%), with both evolving to death.
Radiographic or tomographic studies were analyzed in 26 of 62 (41%) patients. The radiologic studies indicated that alveolar condensations were more frequent than interstitial infiltrate. Note that although interstitial infiltrate is characteristic of viral respiratory infection, other radiologic alterations merit the tracking of such infections. However, chest computerized tomography is now performed earlier in patients with these respiratory viruses, even when they have limited symptoms, and in some cases, the presence of interstitial infiltrate has been observed in patients with a normal chest radiograph.
This study shows that respiratory viral infections are frequent in BMT patients and are associated with high morbidity and mortality rates. Demonstration of the occurrence of such infections during the same period as community viral outbreaks and the use of rapid and sensitive diagnostic methods for their detection are of crucial importance, which can help to intensify surveillance measures and implementation of prophylactic measures such as immunization of family members and health care personnel, or even empirical therapy for patients with acute disease, until a definitive diagnosis is available (18). Such factors can help decrease the overall mortality among BMT patients.
1. Couch RB, Englun JA, Whimbey E. Respiratory viral infections in immunocompetent and immunocompromised persons. Am J Med 1997; 102(3A): 2.
2. Williamson EC, Millar MR, Steward CG, et al. Infections in adults undergoing unrelated donor bone marrow transplantation. Br J Haematol 1999; 104: 560.
3. Shelhamer J, Pizzo PA, Pamillo JE, et al. Respiratory disease in the immunocompromised host. Philadelphia, JB Lippincott 1996.
4. Raad I, Abbas J, Whimbey E. Infection control of nosocomial respiratory viral disease in the immunocompromised host. Am J Med 1997; 102(3A): 48.
5. Wendt CH, Hertz M. Respiratory syncytial virus and parainfluenza virus infections in the immunocompromised host. Semin Respir Infect 1995; 10: 224.
6. McCarthy AJ, Kingman HM, Taylor GS, et al. The outcome of 26 patients with respiratory syncytial virus infection following allogeneic stem cell transplantation. Bone Marrow Transplant 1999; 24: 1315.
7. Quabeck K. The lung as a critical organ in marrow transplantation. Bone Marrow Transplant 1994; 14(suppl 4): S19-S28.
8. Whimbey E, Champlin RE, Couch RB, et al. Community respiratory virus infection among hospitalized adult bone marrow transplant recipients. Clin Infect Dis 1996; 22: 778.
9. Lewis VA, Champlin R, Englund J, et al. Respiratory disease due to parainfluenza virus in adult bone marrow transplant recipients. Clin Infect Dis 1996; 23: 1033.
10. Bowden RA. Respiratory virus infections after marrow transplant: the Fred Hutchinson Cancer Research Center experience. Am J Med 1997; 102(3A): 27.
11. Ljungman P. Respiratory virus infections in bone marrow transplant recipients: the European perspective. Am J Med 1997; 102(3A): 44.
12. Puthavathan P, Wasi C, Kositanont U, et al. A hospital-based study of acute viral infections of the respiratory tract in Thai children, with emphasis on laboratory diagnosis. Rev Infect Dis 1990; 12(suppl 8): S988.
13. Englund JÁ, Piedra PA, Jewell A, et al. Rapid diagnosis of respiratory syncytial virus infections in immunocompromised adults. J Clin Microbiol 1996; 34: 1649.
14. Huaringa AJ, Leyva FJ, Signes-Costa J, et al. Bronchoalveolar lavage in the diagnosis of pulmonary complications of bone marrow transplant patients. Bone Marrow Transplant 2000; 25: 975.
15. Whimbey E, Elting LS, Couch RB, et al. Influenza A virus infections among hospitalized adult bone marrow transplant recipients. Bone Marrow Transplant 1994; 13: 437.
16. Englund JA, Sullivan CJ, Jordan C, et al. Respiratory syncytial infections in immunocompromised adults. Ann Intern Med 1988; 1: 203.
17. Hayden FG. Prevention and treatment of influenza in immunocompromised patients. Am J Med 1997; 102(3A): 55.
18. Whimbey E, Englund JA, Couch RB. Community respiratory virus infections in immunocompromised patients with cancer. Am J Med 1997; 102(3A): 10.