Answers are grounded in this paper and cite their sources.
Influenza viruses have occasionally been associated with severe manifestations of croup, but no comparative studies of different viral etiologies are available. In a retrospective study we compared the clinical courses of croup caused by influenza and parainfluenza viruses in hospitalized children. By several indicators the clinical picture of croup caused by influenza viruses was significantly more severe than that caused by parainfluenza viruses. Croup (laryngotracheobronchitis) is a common respiratory tract infection in children. The highest incidence, 3 to 5 cases per 100 children, occurs during the second year of life. 1 Croup accounts for ∼15% of all lower respiratory tract infections in children, and 35% of those occurring during the first 2 years of life. 1, 2 Although croup is a relatively mild disease in the majority of cases, the clinical presentation is occasionally severe, and hospitalization rates of 1.3 to 2.6% have been reported. 1, 2 In the United States croup has been estimated to cause 41 000 hospitalizations yearly. 3 Parainfluenza virus types 1, 2 and 3 are the most frequent etiologic agents in croup, accounting for approximately two-thirds of cases with known etiology. 1, 4, 5 Other etiologic agents include influenza viruses, respiratory syncytial virus, adenovirus and rhinovirus. In a 19-year study, 6 influenza A or B viruses were detected in 14% of all children hospitalized with croup, but during the peak months of 13 consecutive influenza epidemics, influenza A infection was documented in 67% of croup patients. In another study 7 croup was the clinical presentation in 8% of children hospitalized with influenza. Although some reports have associated influenza viruses with severe manifestations of croup, 8, 9 no comparison of the clinical presentation of croup caused by influenza and parainfluenza viruses is available. We sought to investigate whether the clinical picture of croup caused by influenza viruses differs from that caused by parainfluenza viruses in hospitalized children. Methods. This retrospective study included children hospitalized at the Department of Pediatrics, Turku University Hospital, during 1980 through 1999. The average number of children with croup as the discharge diagnosis was 95 per year. However, nasopharyngeal aspirates for viral detection were obtained from only a minority of croup patients as part of clinical practice, not according to any prespecified criteria. The medical records of all croup patients with a positive finding for either influenza or parainfluenza virus infection were reviewed. The diagnosis of croup was considered validated if inspiratory stridor or distress and/or hoarseness or barking cough were recorded, and no other explanation for the symptoms was available. Because some children with influenza croup might have been discharged with the diagnosis of influenza instead of croup, we also checked the medical records of all influenza-positive children regardless of the discharge diagnosis. In this way we identified 4 additional children with croup caused by influenza viruses. All children in both groups were managed according to generally accepted methods without any knowledge of this study. The clinical data collected from the medical records included age, gender, presence of chronic illnesses or other underlying conditions, duration of fever and other symptoms before hospitalization, occurrence of febrile convulsions, length of hospital stay, highest measured temperature and duration of fever in hospital, presence of acute otitis media or pneumonia, treatment with inhaled epinephrine or glucocorticoids (inhaled, oral or parenteral), need for supplemental oxygen or tracheal intubation and treatment at the intensive care unit. Only radiologically confirmed cases of pneumonia were included in this analysis. During the study years the primary method for the detection of influenza A and B and parainfluenza type 1, 2 and 3 viruses was antigen detection by radioimmunoassay, immunofluorescence, enzyme immunoassay or time-resolved fluoroimmunoassay. 10 In a few children the documentation of the viral etiology was based on viral culture or routine serologic methods. The Mann-Whitney U test, the t test and the chi square test were used to compare differences between the groups. Results. Twenty-nine children had croup caused by influenza viruses (influenza A, 23; influenza B, 3; type unknown, 3), and 88 children had croup caused by parainfluenza viruses (type 1, 33; type 2, 26; type 3, 29). No marked differences were observed among the children with parainfluenza type 1, 2, or 3 croup (data not shown). The demographic characteristics of children with influenza and parainfluenza croup were comparable (Table 1). Boys were affected more often than girls in both groups. The median duration of respiratory symptoms or fever before hospitalization was 1 day in both groups.Table 1: Demographic and clinical data of children with influenza or parainfluenza croupThe median duration of hospitalization was significantly longer in children with influenza croup (4 days) than in those with parainfluenza croup (2 days;P = 0.001). After the initial discharge from the hospital, children with influenza croup had a 4-fold higher rate of readmission because of relapsing inspiratory distress during the next few days than did children with parainfluenza croup (21%vs. 5%;P = 0.02). During the hospital stay inhaled epinephrine was used more frequently in children with influenza (48%) than in those with parainfluenza croup (11%;P < 0.001). Children with influenza croup received glucocorticoids more commonly and for a longer period and needed supplemental oxygen more frequently than children with parainfluenza infection. Tracheal intubation was rarely needed in either group. With respect to clinical findings unrelated to the severity of croup, acute otitis media was diagnosed in almost one-half (45%) of the children with influenza croup, compared with 17% of those with parainfluenza croup (P = 0.005). Also pneumonia occurred more frequently in children with influenza croup. During the hospital stay 66% of children with influenza croup had fever ≥39.0°C, compared with 40% of children with parainfluenza croup (P = 0.03). The mean highest measured temperatures in children with influenza and parainfluenza croup were 39.2°C and 38.6°C, respectively (P = 0.002). All results remained essentially unchanged even when children with underlying conditions were excluded from the analyses. Discussion. This study demonstrates that the clinical picture of croup caused by influenza viruses is substantially more severe than that caused by parainfluenza viruses in hospitalized children. Children with influenza croup stayed longer in the hospital, and after initial discharge the risk of readmission was substantially greater in children with influenza infection. These findings could be partly explained by the well-known features of influenza infection that are not directly related to croup: high and long-lasting fever; and frequent development of pneumonia or other complications. However, all readmissions occurred because of relapsing laryngeal symptoms, not because of development of bacterial complications. In the retrospective setting we were unable to obtain reliable data about the degree of dyspnea. However, inhaled epinephrine and glucocorticoids are clearly used for the management of inspiratory airway obstruction, and the more frequent use of these drugs in children with influenza than with parainfluenza croup provides evidence for the severity of the symptoms of croup in children with influenza infection. The typical clinical presentation of croup including inspiratory stridor, barking cough and hoarseness is easy to distinguish, which increases the reliability of the discharge diagnoses in the medical records. Even in the case of influenza croup, most children had the discharge diagnosis of croup instead of influenza. When interpreting the results of this study, it should be noted that virologic samples were not taken systematically from all children with croup, and it is probable that children with more severe manifestations of the disease were overrepresented in the study population. However, this should not bias the comparison between children with positive findings for influenza or parainfluenza viruses. The nasopharyngeal specimens were obtained on clinical grounds alone, without any knowledge of this analysis at that time. Therefore we consider our main findings reliable in the setting of hospitalized children. However, it is not known whether there are differences between influenza and parainfluenza croup in the majority of children who are managed as outpatients. Antigen detection of influenza and parainfluenza viruses is possible within 1 day in the laboratory, and for influenza rapid point-of-care tests are currently available. Although the treatment of croup with glucocorticoids and eventually with inhaled epinephrine is similar irrespective of the etiology, detection of the causing virus might be beneficial for the clinical management of the children. If influenza is diagnosed within 48 h of the onset of symptoms, specific antiviral therapy with a neuraminidase inhibitor could be considered. A recent study demonstrated that oral oseltamivir treatment shortened the duration of symptoms of influenza and reduced the development of acute otitis media as a complication in children. 11 Influenza epidemics are associated with substantial numbers of excess hospitalizations and outpatient visits in young children. 12, 13 The recently developed intranasal, live attenuated influenza vaccine 14 could be anticipated to lower the threshold of influenza vaccination in children. Although croup is a relatively rare manifestation of influenza infection, the results of the present study indicate that the severity of croup caused by influenza viruses is greater than average, which lends support to the idea of more widespread use of influenza vaccine in children.
No takes yet. Share an insight, caveat, or question.
Peltola et al. (2002) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: