Otitis and Respiratory Distress Episodes Following a Respiratory Syncytial Virus Infection D

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

ORIGINAL ARTICLE

Otitis and respiratory distress episodes following a respiratory syncytial virus infection D. A. Kafetzis, H. Astra, M. Tsolia, G. Liapi, J. Mathioudakis and K. Kallergi

Second Department of Pediatrics, University of Athens, P. & A. Kyriakou Children's Hospital, Athens, Greece

Objective To document, over two consecutive respiratory syncytial virus (RSV) seasons, the occurrence of acute otitis media (AOM) and recurrence of respiratory distress in children <2 years of age hospitalized for respiratory distress. Methods Patients were examined during hospitalization and at 6 weeks and 6 months after discharge. RSV testing was performed on all patients, and hospitalized patients were evaluated daily for the occurrence of AOM. Results In total, 347 children were enrolled; 54.8% were RSV positive, and 45.2% were RSV negative. Children were most frequently diagnosed as having bronchiolitis (71.9%) or asthmatic bronchitis (17.9%); other diagnoses included pneumonia, laryngitis, and rhinitis. During hospitalization, AOM was diagnosed in 16.8% of RSV-positive versus 8.3% of RSV-negative children (P < 0.05). Six weeks after discharge, AOM was reported in 10.4% of RSV-positive as compared with 5.8% of RSV-negative patients. Six months later, AOM was reported in 2.9% of the RSV-positive and 7.6% of the RSV-negative patients. A second episode of acute respiratory distress, which either required (9) or did not require (35) hospitalization, occurred in 18.4% of the total population, with similar proportions of RSV-positive and RSV-negative children (17% versus 18.6%). Conclusion We conclude that RSV appears to be an important contributing factor for the occurrence of AOM in young children hospitalized with respiratory distress. The occurrence of a second episode of acute respiratory distress did not appear to correlate with the previous RSV infection, but longer-term follow-up is required. Keywords RSV, otitis media, respiratory distress, bronchiolitis Accepted 25 November 2002 Clin Microbiol Infect 2003; 9: 1006±1010

be connected more commonly with RSV infections INTRODUCTION than with other viral infections [2±8]. Respiratory syncytial virus (RSV) is one of the To our knowledge, the short-term sequelae of major causes of lower as well as upper respiratory hospitalization for acute respiratory distress tract infection in children all over the world. (including RSV infection) have not been exten- Bronchiolitis is the primary lower respiratory tract sively studied [9±13]. We therefore prospectively infection caused by RSV, and is the main reason for examined the occurrence of AOM during hospi- hospitalization during infancy [1]. Acute otitis talization, and the reoccurrence of AOM and media (AOM) is also a common infection affecting respiratory distress after 6 weeks and 6 months mainly infants and young children. In retrospec- following hospitalization for respiratory distress tive studies, it has been observed that AOM might over two consecutive RSV seasons.

Corresponding author and reprint requests: D. A. Kafetzis, METHODS Second Department of Pediatrics, University of Athens, P. & Eligible participants were children between A. Kyriakou Children's Hospital, Athens 11527, Greece Tel/Fax: 30 1 6743381 2 weeks and 2 years of age who were admitted E-mail: [email protected] to the P. & A. Kyriakou children's hospital due to

ß 2003 Copyright by the European Society of Clinical Microbiology and Infectious Diseases Kafetzis et al Otitis, respiratory distress episodes and RSV infection 1007 respiratory distress during the RSV season and incubated in a humid chamber for 30 min at 35± who were not under RSV prophylaxis. The parents 37 8C. After incubation, the slide was carefully of the patients were informed, and written consent washed in PBS, air-dried, and examined micro- was obtained from each of them. The study was scopically (¯uorescence microscope) under Â400 approved by the Ethical Committee of our hospi- magni®cation. A specimen showing one or more tal. Study months included 1 January 1999 to 30 intact columnar epithelial cells with the character- June 1999, and 1 December 1999 to 30 June 2000. istic apple-green cytoplasmic ¯uorescence was During the months of July to November 1999, no considered to be positive for RSV. The typical patients were enrolled because of the low inci- staining morphology included small to large cyto- dence of RSV in Greece during this season. All plasmic inclusions or ®ne particulate ¯uorescence. patients had symptoms and signs of respiratory Specimens with at least 100 columnar epithelial distress, such as tachypnea, intercostal or subcos- cells per well and not exhibiting any speci®c posi- tal retractions, and low O2 saturation (<95%). tive ¯uorescence were considered to be negative Respiratory diagnoses included bronchiolitis, for RSV. Specimens with fewer than 100 cells/well asthmatic bronchitis, laryngitis, rhinitis and pneu- were considered to be inadequate, and the proce- monia. Bronchiolitis was de®ned as an acute lower dure was repeated to exclude the possibility of respiratory tract illness characterized by cough, poor ®xing. When the specimen did not have a signs of respiratory distress, a prolonged expira- satisfactory number of cells, due to poor ®xing, the tory phase, and the presence of wheezing and/or procedure was repeated. If the specimen contin- `crackles'. Only children hospitalized with a ®rst ued to have an unsatisfactory number of cells, it episode of wheezing were classi®ed as having was excluded in order to avoid a false-negative bronchiolitis. If one or more episodes of wheezing result. had occurred previously, the diagnosis of asth- The immuno¯uorescence techniques for detect- matic bronchitis was used. Laryngitis was de®ned ing RSV antigen are usually very sensitive and as an upper respiratory tract infection with char- speci®c (sensitivity as high as 97% and speci®city acteristic `croupy' cough and signs of respiratory greater than 90%) in comparison to culture. The distress. Rhinitis was characterized by nasal dis- actual sensitivity and speci®city are probably charge and nasal obstruction, which in young higher than reported, because culture itself is infants was noted with signs of respiratory distress not 100% sensitive [14,15]. but without wheezing or `crackles'. Pneumonia Upon admission, all patients had a complete was de®ned by the presence of consolidations clinical examination; examinations were repeated on chest X-ray in patients with localized `crackles' and recorded daily during hospitalization. AOM in the affected area. was de®ned when bulging and/or opacity of the Nasopharyngeal washings were obtained from tympanic membrane was observed and accompa- each patient within the ®rst 24 h of hospitalization, nied by one or more signs of acute infection, such and were tested for RSV antigen with a direct as: ear pain, including unaccustomed tugging or ¯uorescence assay (DFA) (Mono¯uo Screen RSV, rubbing of the ear, and marked redness of the Sano® Diagnostics, Pasteur, Marnes la Coguette, tympanic membrane. AOM was also diagnosed France). The DFA procedure followed the recom- when purulent otorrhea was present. The diagno- mendations of the supplier. Each specimen was sis of AOM was made by an ear, nose and throat resuspended in 2±3 mL of phosphate-buffered (ENT) specialist of our hospital who was blinded saline (PBS) and vortexed for about 10 s; this to the diagnosis of RSV. Following hospital dis- was followed by centrifugation at 1500 rev/min charge, all parents who lived in Athens were asked for 5 min, to pellet cellular material. The cell pellet to contact study investigators if any signs or symp- was washed twice with PBS, or until the specimen toms of AOM or respiratory disease occurred, and was virtually free of mucus. The cells were resus- they were referred to either a blinded ENT spe- pended in 0.3 mL of PBS, and 25 mL of the suspen- cialist of our hospital or to us, respectively. For sion was applied to a 13-mm-diameter well of a patients who lived in other cities, parents were multiwell microscopic slide. The slide was air- asked to visit an ENT specialist or a pediatrician in dried, ®xed with cold acetone for 10 min, rinsed their own city, with whom we made contact. in distilled water, and air-dried before staining. Approximately 6±8 weeks following hospital dis- The slide was overlaid with the DFA stain, and charge, the parents of each child were contacted by

ß 2003 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 9, 1006±1010 1008 Clinical Microbiology and Infection, Volume 9 Number 10, October 2003 telephone. This time period was considered suf®- Table 1 Characteristics of study participants cient for short-term follow-up, because an episode Characteristics RSV RSVÀ P-value of AOM during hospitalization should be resolved within that period of time [16]. Parents were ques- Age (months) <39150<0.005 tioned about the health status of the child, whether 3±6 48 27 <0.005 an episode of AOM or respiratory disease had 6±12 38 43 0.12 occurred within that period, and if they had visited 12±24 13 37 <0.005 an ENT specialist or pediatrician but for some Sex Male 126 98 reason had not contacted us. The speciality of the Female 64 59 physician who made the diagnosis (family physi- Days of hospitalization 6.57 4.53 cian, pediatrician, or ENT specialist) was recorded, Smoking of parents 62.7% 77.6% 0.029 and the physician was contacted by one of the Smoking during pregnancy 19.8% 29.4% 0.051 investigators in order to record any episode. A Breast-fed 57% 45.7% 0.036 Center of attendance 2.1% 6.3% 0.042 second phone call was made 6 months after hos- Family history of atopy 31.6% 40.6% 0.080 pital discharge; the same questions were asked, Older siblings 58.7% 65.3% 0.183 and the same procedure was followed. This time period was considered suf®cient to recognize new infections. Patients who were not found or never with AOM, as compared to 13 (8.3%) of 157 RSV- communicated with us, patients who developed negative children (P < 0.05). RSV-positive symptoms and were not referred to any physician, patients diagnosed with AOM had an associated and patients whose parents refused collaboration, diagnosis of bronchiolitis in 75% and of asthmatic were considered as `lost to follow±up'. bronchitis in 18.8%. Comparison of proportions between the groups The ®rst outpatient communication at 6 weeks was made by the standard chi-square test and was established in 239 (68.9%) of the study sub- Fisher's exact test. A P-value of less than 0.05 jects, of whom 135 (56.5%) had a primary diagnosis was considered to indicate statistical signi®cance. of RSV at hospitalization. These ®gures are repre- sentative of the entire study population. Occur- rence of an AOM episode was recorded in 20 RESULTS patients (8.3%), of whom seven had a ®rst AOM During the two RSV seasons, 347 children, 64.5% diagnosed during hospitalization (®ve RSV posi- males and 35.5% females, were enrolled. Their tive versus two RSV negative). Of 135 RSV-posi- median age was 4.3 months (14 days to tive patients, 14 (10.4%) had an AOM episode, as 24.7 months), and the median duration of hospi- compared to six (5.8%) of 104 RSV-negative chil- talization for acute respiratory distress was 4 days dren; these differences were not statistically sig- (1±36 days). Of these, 190 patients (54.8%) were ni®cant (P 0.149) (Table 3). Of the 14 RSV- RSV positive, and 157 patients (45.2%) were RSV positive patients, 13 had a primary diagnosis of negative. The demographics of these patients are bronchiolitis and one of asthmatic bronchitis, shown in Table 1. whereas of the six RSV-negative patients, four The diagnosis of bronchiolitis was made in 248 had bronchiolitis, one asthmatic bronchitis, and (71.5%) of all children enrolled, and speci®cally in one pneumonia. 164 (86.3%) of all RSV-positive children; these patients had a two-fold higher prevalence Table 2 Diagnosis of the hospitalized patients (66.1%) of bronchiolitis as compared to RSV-nega- Diagnosis Patients hospitalized tive children (33.9%). The diagnoses of the patients are shown in Table 2. Asthmatic bronchitis was RSV RSVÀ Total diagnosed in 62 (17.9%) patients, pneumonia in 16 Bronchiolitis 164 (66.1%)Ã 84 (33.9%) 248 (4.3%), and laryngitis and rhinitis in 21. Asthmatic bronchitis 21 (33.9%)Ã 41 (66.1%) 62 The occurence of AOM and respiratory distress Pneumonia 1 (6.7%)Ã 14 (93.3%) 15 episodes at follow-up are shown in Table 3. Dur- Laryngitis 3 (21.4%)Ã 11 (78.6%) 14 ing hospitalization, AOM was diagnosed in 45 Rhinitis 1 (14.3%)ÃÃ 6 (85.7%) 7 patients (13% of the total population). Of 190 Total 190 157 347 RSV-positive children, 32 (16.8%) were diagnosed ÃP < 0.005. ÃÃP 0.023.

ß 2003 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 9, 1006±1010 Kafetzis et al Otitis, respiratory distress episodes and RSV infection 1009

Table 3 Occurrence of AOM during hospitalization and 6 weeks to 6 months after discharge; occurrence of a second respiratory distress episode within 6 months

AOM during AOM within AOM within Respiratory hospitalization 6 weeks 6 months distress RSV 16.8% (32)Ã 10.4% (14) 2.9% (4) 17% (23) Total 190 Total 135 RSVÀ 8.3% (13) 5.8% (6) 7.6% (8) 18.6% (19) Total 157 Total 98

ÃP < 0.05.

During the second telephone communication of hospitalized patients. However, it is apparent (6 months after discharge), the same number of that younger infants suffering from RSV bronch- patients was found (239). A new AOM episode iolitis are in greater need of hospitalization. In this was reported in 12 children (5%). None of the study, smoking by parents, smoking during preg- children had a repeat episode of AOM. Of 135 nancy, breast-feeding, day-care center attendance RSV-positive patients, four (2.9%) had an AOM and family history of atopy did not seem to in¯u- episode diagnosed for the ®rst time, as compared ence the incidence of RSV infection. to eight (7.6%) of 104 RSV-negative children Acute respiratory distress requiring hospitaliza- (Table 3). Again, these differences were not statis- tion was frequently associated with AOM in this tically signi®cant (P 0.505). study. It appears that RSV may play a prominent A repeat episode of acute respiratory distress, role in the etiology of AOM, both during and de®ned as occurrence of tachypnea, retractions, or immediately following an RSV hospitalization. wheezing, was reported in 44 (18.4%) of the study In our study population, RSV-positive patients children during the 6-month period after hospita- were more commonly diagnosed with AOM lization. Of these children, nine (20.4%) required during hospitalization than were RSV-negative hospitalization. No signi®cant difference in patients. These ®ndings correlated with those respiratory distress was observed between RSV- reported in the Finnish Otitis Media Cohort positive and RSV-negative children (17% versus Study, in which the association of RSV with 18.6%, P 0.864). concurrent AOM was relatively higher than that In the total study population, there were four of other viruses [17]. Also, higher percentages children with serious underlying diseases. Two of AOM in children with bronchiolitis were cases with congenital heart disease were operated reported [2]. on, and during the 6-month follow-up period they At the 6-week follow-up, there was a trend did not develop AOM or respiratory distress; one towards an increased number of AOM episodes case with aplasia of the right lung was operated on in the RSV group. The incidence of AOM was not at 5 months, and one case, operated on for tra- very high, and our numbers were not very large, so cheoesophageal ®stula, never developed AOM these differences did not reach statistical signi®- during follow-up. cance. While this difference appeared to be a clinically important observation, it may be that a study with a larger sample size is necessary in DISCUSSION order for this difference to be statistically evident. In our study, RSV was found to be the cause of In contrast to the ®ndings at 6 weeks, 6 months 71.5% of cases of bronchiolitis in hospitalized following hospitalization the incidence of AOM children less than 2 years of age. In patients with episodes was low in both RSV-positive and RSV- other diagnoses, the incidence of RSV was lower, negative patients. The occurrence of a second because they were admitted to the study only if episode of respiratory distress was comparable they suffered respiratory distress. The incidence of between RSV-positive and RSV-negative patients RSV was particularly high among infants under of all age groups at the 6-month follow-up. In this 3 months of age. It appears that, in our study, RSV small, prospective study, our data do not demon- positivity is associated with a younger age than strate a higher incidence of recurrent respiratory has been reported in other studies. This could be distress in the ®rst 6 months after hospitalization explained by the fact that we studied a population due to RSV infection. A long-term increased risk of

ß 2003 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 9, 1006±1010 1010 Clinical Microbiology and Infection, Volume 9 Number 10, October 2003 recurrent wheezing after RSV disease in infancy 7. Heikkinen T, Thint M, Chonmaitree T. Prevalence has previously been shown [18]. of various respiratory viruses in the middle ear We conclude that serious RSV infections are during acute otitis media. N Engl J Med 1999; 340: frequently accompanied by AOM during and 260±4. 8. Anderson LJ. Respiratory syncytial virus vaccines immediately following hospitalization. The use for otitis media. Vaccine 2000; 19(suppl 1): S59±65. of preventative treatment for infants at high risk 9. Maseda E, Llorente JL, Melon S et al. Identification for severe RSV infection might also prevent such of respiratory syncytial virus and adenovirus in episodes. However, this remains to be proved. In children with chronic serous otitis. Acta Otorrinolar- this study, the occurrence of an AOM episode ingol Esp 2000; 51(8): 687±90. 6 months after hospitalization for an RSV infec- 10. Eriksson M, Bennet R, Nilsson A. Pediatr Allergy tion did not appear to be more common than in Immunol 2000; 11(3): 193±7. children hospitalized with RSV-negative respira- 11. Tuffaha A, Gern JE, Lemanske RF Jr. The role of tory distress. respiratory viruses in acute and chronic asthma. Clin Chest Med 2000; 21(2): 289±300. 12. Kneyber MCJ, Steyeberg EW, de Groot R, Moll HA. ACKNOWLEDGMENTS Long-term effects of respiratory syncytial virus (RSV) bronchiolitis in infants and young children: We thank Katerina Samara for her technical assis- a quantitative review. Acta Paediatr 2000; 89(6): tance. We also thank Jessy Grouthious and Marize 654±60. Markopoulou for their valuable assistance in writ- 13. Dutau G, Micheau P, Rittie JL, Juchet A, Rance F, ing this paper. Bremont F. Relationship between respiratory syncytial virus bronchiolitis and asthma. Arch Pediatr 2000; 7(suppl 3): 536s±43s. REFERENCES 14. Landy LM. Rapid viral diagnosis. In: Rose NR, ed. Manual of clinical laboratory immunology, 5th edn. 1. Cantani A. Bronchiolitis in infants. Eur Rev Med Washington DC: American Society for Microbiol- Pharmacol Sci 1999; 3(4): 195±6. ogy, 1997: 608±17. 2. Andrade AM, Hoberman A, Glustein J, Paradise LJ, 15. Tristram AD, Welliver CR. Respiratory syncytial Wald RE. Acute otitis media in children with virus. In: Murray PR et al., eds. Manual of clinical bronchiolitis. Pediatrics 1998; 101: 617±19. microbiology, 6th edn. Washington DC: American 3. Klein SB, Dollete FR, Yolken RH. The role of Society for Microbiology, 1995: 932±9. respiratory syncytial virus and other viral patho- 16. Daly K, Hunter L, Giebink S. Chronic otitis media gens in acute otitis media. J Pediatr 1982; 101: 16±20. with effusion. Pediatrics 1999; 20(3): 85±93. 4. Uhari M, Hietala J, Tuokko H. Risk of acute otitis 17. Vesa S, Kleemola M, Blomqvist S, Takala A, Kilpi T, media in relation to the viral etiology of infections Hovi T. Epidemiology of documented viral respira- in children. Clin Infect Dis 1995; 20: 521±4. tory infections and acute otitis media in a cohort of 5. Ruuskanen O, Arola M, Putto-Laurila A et al. Acute children followed from two to twenty-four months otitis media and respiratory virus infections. Pediatr of age. Pediatr Infect Dis 2001; 20(6): 574±81. Infect Dis J 1989; 8: 94±9. 18. Noah TL, Ivins SS, Murphy P, Kazachkova I, 6. Chonmaitree T, Owen JM, Patel AJ, Hedgpeth D, Moats±Staats B, Henderson FW. Chemokines and Horlick D, Howie MV. Effect of viral respiratory inflammation in the nasal passages of infants with tract infection on outcome of acute otitis media. respiratory syncytial virus bronchiolitis. Clin Im- J Pediatr 1992; 120: 856±62. munol 2002; 104(1): 86±95.

ß 2003 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 9, 1006±1010