Annals of Tropical Paediatrics (2001) 21, 59–65 Detection of adenovirus infection in children in Jordan MAMDOH M. MEQDAM, GYEYATH NASRALLAH* & ABDULLAH AL-SHURMAN† Departments of Applied Biology, *Clinical Laboratories Sciences and †Pediatrics, Princess Rahma Hospital, Jordan University of Science and Technology, Irbid, Jordan (Accepted August 2000) Summary Between November 1997 and May 1998, 350 nasopharyngeal aspirates (NPA) were obtained from children admitted to the Respiratory Disease Unit at Princess Rahma Hospital, northern Jordan who were clinically diagnosed as suffering from respiratory tract infections. NPA were investigated for the presence of adenovirus using shell vial (SV) culture assay, conventional culture (CC) assay, and direct immunouorescence assay (DFA). Of the 350 NPA, adenoviruses were detected in 54 (15.4%) by the combined techniques used. SV identiŽed 34 (63%), CC 48 (89%) and DFA 30 (56%). Most virus isolations were in children aged 1– , 5 years old and were associated with pneumonia in 39% and bronchopneumonia in 32%. SV assay showed a sensitivity and speciŽcity of 68.8% and 99.7%, respectively, for detecting adenovirus from NPA. These results emphasize that CC assay is still important for the diagnosis of adenovirus, although SV and DFA are superior diagnostic assays. Introduction World Health Organization (WHO) data indicate that respiratory tract infection (RTI) is one of the major causes of morbidity and mortality in infants and children.1–3 In many countries, including Jordan, acute respiratory infection (ARI) accounts for nearly one-third of all deaths in children less than 5 years of age.4,5 ARI kills an estimated four million children every year in developing countries and most of these deaths are caused by pneumonia.6 In those countries, the rate of infection is very high, particularly in infancy and early childhood.7 Adenovirus is known to infect persons by aerosols or eye contact, but it also spreads by Reprint requests to: Dr Mamdoh M. Meqdam, Department of Applied Biology, Jordan University of Science and Technology, PO Box. 3030, Irbid 22110, Jordan. Fax: 1 962 2 709 5019; e-mail: meqdam@just.edu.jo faecal/oral routes, especially where hygiene is poor. During an incubation period of 5–10 days, the virus replicates in pharynx, conjunctivae and small intestine. It is recognized that all adenovirus serotypes are endemic in the community and that for unknown reasons some can cause explosive outbreaks of disease, usually respiratory but also involving the eye.8 Adenoviruses are responsible for a signiŽcant number of clinical respiratory illnesses. Upper respiratory disease caused by adenoviruses includes colds, pharyngitis and tonsillitis and occurs mainly in infants and young children. Information about adenoviruses causing respiratory disease has been gathered in a series of reports to the WHO and evaluated over a 10-year period (1967–76),9 during which time adenoviruses accounted for 13% of a total 135,702 reported cases. This was second only to inuenza A, which accounts for 28% of reported cases. ISSN 0272-4936 print/ISSN 1465-3281/online/01/010059-07 Ó 2001 The Liverpool School of Tropical Medicine Carfax Publishing DOI: 10.1080/02724930020028939 60 M. M. Meqdam et al. StratiŽcation by age of reported adenovirus cases revealed the following: 22% , 1 year, 42% 1–4 years, 18% 5–14 years, 7% 25–29 years, 1% . 60 years. Other studies indicate that approximately 10% of childhood pneumonia is probably caused by adenoviruses.10 Other adenovirus-related lower respiratory illnesses include bronchitis and bronchiolitis.11 Adenoviruses are stable and do not need refrigeration during transport. They can be cultured and isolated in a variety of cell lines. Suitable cell lines include Hep-2, Hela, A549 and PMK.12–14 The infected cells become swollen, rounded and refractive and, clustering together, form grape-like structures which subsequently become detached from their support.10 ConŽrmation of infection is usually achieved by uorescent assay, enzyme immunoassay or haemagglutination-inhibition .14 The purpose of this study was to determine the incidence of RTI caused by adenovirus and to evaluate the different diagnostic assays, shell vial (SV) culture assay, conventional culture (CC) assay and direct immunouorescence assay (DFA), in terms of sensitivity, speciŽcity, predictive value and agreement. al.,15 nasopharyngeal aspirates (NPA) were collected using a sterile infant feeding tube connected to a vacuum pump. The tube was introduced through each nostril, yielding about 0.5–1 ml of aspirate. The tube was placed into separate, tightly capped sterile containers labelled with the child’s name, identiŽcation number and the collection date and time. The container was transported to the laboratory in wet ice. Specimen processing Specimens were processed according to the method described by Isenberg,16 as follows. The contents of the infant feeding tube were emptied into an Ependorff tube by ushing with a mixture of 1.5 ml of PBS (pH 7.4) including 200 ll penicillin, 0.2 mg streptomycin and 10 lg amphotoricin B per ml. The specimen was vortexed for 30 s to dislodge cells from mucus and centrifuged at 400 3 g for 10 min using a Hettich centrifuge. The supernatant was aspirated and transported to another Ependorff tube for inoculation of cell culture. The pellet was used for rapid detection of adenovirus antigens by uorescence assay. Patients and methods Patients This study involved 350 children less than 13 years old admitted with respiratory tract illness to the Respiratory Diseases Unit in the Princess Rahma Hospital. Those with congenital and persistent respiratory disease were excluded. Princess Rahma Hospital is a Ministry of Health paediatric teaching hospital administered by the Faculty of Medicine of Jordan University of Science and Technology. It is the only paediatric hospital in northern Jordan and serves Irbid city and suburbs. All information concerning patients was collected on a special proforma. Preparation of cell pellet for immunouorescence staining The cell suspension was prepared from the pellet by washing the pellet twice in PBS (pH 7.4) for 10 min to decrease the mucus viscosity. The sediment was then resuspended in 0.5 ml PBS. Twenty ll of cell suspension was spotted per well on an eight multi-well slide (5-mm diameter). The slide was dried by hotplate at 40°C and Žxed with cold acetone at 4°C for 10 min. The slide was then stored in an airtight box at 2 70°C until used for staining. Direct immunouorescence staining Specimen collection Using the method described by Meziere et Ten ll of mouse monoclonal isothiocyanatelabelled anti-RSV antibody (Chemicon, USA) Detection of adenovirus in Jordan was added to each well. The slide was incubated at 37°C for 30 min in a humid chamber and then rinsed for 10–15 seconds in PBS (pH 7.4). After mounting in phosphatebuffered glycerol (20–80% v/v), the slide was read using a uorescence microscope (Nikon, Japan) at 3 20–40 magnŽcation. Specimens yielding less than an average of 10 columnar epithelial cells per well were rejected. RSVinfected and uninfected human epithelial cells (Hep-2) were stained as positive and negative controls, respectively, in each trial. The presence of one positive columnar epithelial cell was required to consider a specimen positive for adenovirus antigen. Virus isolation Conventional culture (CC) assay. Specimen inoculation was done according to the method described by Isenberg,16 as follows. A conventional culture tube with a semiconuent monolayer of Hep-2 cells was inoculated by aspirating the old growth medium (GM) and delivering 0.2 ml of specimen supernatant to each tube. After adsorption at 37°C for 60 min, 1 ml fresh maintenance medium (MM) was added and the incubation continued for up to 14 days. The culture was maintained by replacing the old MM with fresh MM every other day. Culture tubes were examined daily for evidence of viral replication and development of CPE. When seen, CPE were harvested and prepared for immunouorescence staining. Shell vial (SV) culture assay. Shell vial assay was performed according to the method described by Olsen et al.,13 as follows. A shell vial tube with semiconuent monolayers of Hep-2 cells was inoculated by aspirating the old GM and delivering 0.2 ml of specimen supernatant to each tube. The specimen was centrifuged (700 3 g) for 50 min (Hettich, Germany). After centrifugation, 1 ml of MM was added to each vial and incubated for 4 days. SV tubes were examined daily. When the SV tube showed toxic effects or contamination, the 61 specimen was decontaminated and the culture repeated. At the end of the incubation period (regardless of the presence of CPE), the Hep2 monolayers were harvested from the bottom of the SV tubes by scraping into a small volume of PBS, pH 7.4 by rubber policeman, and 20 ll of scraped cells were spotted onto each well of eight-well slides and prepared for immunouorescence staining. Data analysis Groups were compared by means of the v2 test using Minitab version 11. Results A total of 350 specimens were tested for the presence of adenovirus by SV, CC or DFA. Table I shows the number positive for adenovirus by the assays used. The number positive for adenovirus by SV, CC and DFA was 54 (15.4%). Adenovirus was more frequently isolated with CC (88.9%) than with SV (63.0%) or DFA (55.6%). The age and sex distributions of the positive and negative cases of adenovirus are shown in Table II. The children’s ages ranged from 1 month to 13 years (mean 7 months). Data showed that adenovirus was highly associated with children aged 1– , 5 years (p , 0.001). Table III shows the Žndings in children positive for adenovirus. Adenovirus infection was associated with pneumonia in 38.9% and with bronchopneumonia in 31.5%. In this study, 75.9% of hospitalized children with adenovirus had received at least one type of antibiotic. The frequency of signs and symptoms in the children positive for adenovirus is shown in Fig. 1. Cough and wheezing were the most prominent, 81.5% and 77.8%, respectively. Comparison of SV with CC and DFA for isolation of adenovirus is shown in Tables IV and V. The sensitivity and speciŽcity of SV compared with CC were 68.8% and 99.7%, respectively. 62 M. M. Meqdam et al. TABLE I. Number of adenoviruses positive by SV, CC or DFA assays in nasopharyngeal aspirates of 350 children No. of specimens positive by SV CC DFA No. of positive adenovirus (%) 54 (15.4) 34 (63.0) 48 (88.9) 30 (55.6) TABLE II. Age and sex distribution of children positive for adenovirus by SV, CC or DFA assay Age (yrs) No. of children (M:F) % No. of positive children (M:F) % ,1 1– , 5 5–13 236 (126:110) 102 (60:42) 17 (11:6) 67.4 29.1 4.9 19 (12:7) 33 (18:15)* 2 (1:1) 8.1 32.4 11.8 Total 350 54 * p , 0.001. TABLE III. Observations in 54 children positive for adenovirus by SV, CC or DFA assay Observations Clinical diagnosis Bronchiolitis Pneumonia Bronchopneumonia Other respiratory infection (bronchitis, croup, upper respiratory tract) No. (%) of positive children 7 (13.0) 21 (38.9) 17 (31.5) 9 (16.7) Duration of hospitalization (days) 1–3 4–6 7–9 23 (42.6) 22 (40.7) 3 (5.6) No. of antibiotics received 0 1 2 3 13 (24.0) 26 (48.1) 9 (16.7) 6 (11.1) Discussion This is the Žrst study in Jordan to examine the role of adenovirus as an aetiological agent of RTI in children under 13 years of age. The nasopharyngeal aspirates were collected during the winter (November 1997 to May 1998) when RTI is more prevalent in northern Jordan.17 The infection rate of adenovirus varies from region to region and from one year to another,8,10,18 and in this study it was higher than expected (15.4%). The rate of infection with adenovirus in children hospitalized during the winter months in Saudi Arabia was very low (0.7%).19 The reported infection rate has ranged from 2 to 4% in hospitalized children in various developing countries18 and a 6-year study revealed an only slightly higher infection rate of 8.2% among 1,416 sick children.20 Our Žndings differ but agree with the reported infection rate of 14.3% among 169 hospitalized children in Argentina.21 A similar study of 80 hospitalized children reported that the most important viral agent of RTI was RSV followed by adenovirus.22 In addition to these two studies, data from WHO show that 13% of a total of 135,702 cases with respiratory illnesses were associated with adenovirus9 and another study reports an adenovirus infection rate of 18% among 96 children hospitalized with respiratory illnesses.23 Our results indicate that adenovirus infection occurs mainly in children aged 1– , 5 years, in agreement with other reports.21,22 Detection of adenovirus in Jordan 63 FIG . 1. The frequency of signs and symptoms in 54 children with adenovirus infection. TABLE IV. Comparison of SV, CC and DFA for diagnosing adenovirus in nasopharyngeal aspirates of 350 children Possible result 1 /1 1 /2 2/1 2/2 SV/CC SV/DFA CC/DFA 33 1 15 301 24 7 11 308 27 3 21 299 Adenovirus is responsible for a signiŽcant amount of clinical respiratory illness. Approximately 10% of childhood pneumonia is probably caused by adenovirus. Other lower respiratory illnesses related to adenoviruses include bronchiolitis and bronchitis.9,10,24–26 A signiŽcant association was found between adenovirus and pneumonia in children older than 1 year. 21 Our study supports these Žndings in that adenovirus was identiŽed in 38.9% children with pneumonia and in 31.5% with bronchopneumonia. This study shows that 41 (76%) of our children received at least one type of antibiotic after admission. This exceeds the proportion reported in the literature. In one study, 61% of children were given antibiotics despite the fact that 35% of them had a conŽrmed viral aetiology.22 The frequent use of antibiotics in our patients might reect the fact that the Princess Rahma Hospital has no facility to test for viral infection and doctors might give antibiotics to avoid secondary bacterial infection. Wide variation in the sensitivity of SV for TABLE V. Percentage of sensitivity, speciŽcity, predictive value and agreement of SV, CC and DFA in nasopharyngeal aspirates of 350 children Predictive value Assays compared SV/CC SV/DFA CC/DFA Sensitivity SpeciŽcity 1 2 Agreement 68.8 68.6 56.3 99.7 97.8 99.9 97.1 77.4 90.0 95.3 96.6 93.4 95.4 94.9 93.1 64 M. M. Meqdam et al. detecting adenovirus has been reported. One study found that SV was approximately 80% sensitive for detecting adenovirus26 while in another study SV assay using Hep-2 cell lines was only 52% sensitive at 24 hours for specimens previously known to be positive.22 Furthermore, only 10 of 21 adenovirus isolates were detected by SV assay after 4 days incubation.27 In one study comparing SV with traditional tube cultures using A-549 cell lines for detecting adenovirus, 77% of specimens were positive in SV at 2 days and 100% at 5 days post-inoculation, and most specimens were eye swabs.28 It has been reported that SV works well when specimens contain relatively large amounts of the virus but are insensitive when specimens contain small amounts of the virus.13 This might explain the high rate of detection of adenovirus in the earlier study.28 Furthermore, the average time to detect adenovirus in specimens positive by SV was 4 days compared with 8 days for specimens negative in shell vial and positive only in CC.13 In the present study, 33 (61 %) of 54 adenovirus isolates were detected by both SV and CC. However, 15 (28%) samples were positive by CC and negative by SV. Although SV assay is highly sensitive for detecting viruses, its sensitivity is only fairly good (69%) for detecting adenovirus. Based on our Žndings, it might be necessary to extend the incubation period of SV for isolating adenovirus or simply to use the CC assay. The variation in SV sensitivity for adenovirus could be owing to the use of different types of specimen; for example, throat or eye swabs yield higher numbers of adenovirus than nasopharyngeal specimens.1,23 Low titres of viruses in specimens, especially adenovirus, make detection difŽcult.1 Our Žndings explain the importance of selecting the right laboratory assay to diagnose viral infection when managing children hospitalized with RTI. Positive laboratory diagnosis of respiratory viruses might persuade physicians to stop unnecessary antibiotic therapy because of its cost and side-effects and to shorten the duration of hospitalization; it might also reduce the need for other diagnostic tests. Acknowledgment Financial support for this research was provided by Jordan University of Science and Technology, grant no. 63/97. References 1 Stout C, Murphy MD, Lawrence S, Julian S. Evaluation of a monoclonal antibody pool for rapid diagnosis of respiratory viral infections. J Clin Microbiol 1989; 27:448–52. 2 World Health Organization. Basic Principles for Control of Acute Respiratory Infections in Children in Developing Countries. A joint WHO/UNICEF Statement. Geneva: WHO, 1986; 116–26. 3 Leowski J. Mortality from acute respiratory infection in children under 5 years of age: global estimate. World Health Stat Q 1990; 39:138–44. 4 StanŽeld SK. Acute respiratory infection in the developing word: strategies for prevention, treatment and control. Pediatr Infect Dis J 1987; 6:622– 9. 5 Zaman K, Baqui AH, Yunus M, et al. Acute respiratory infections in children: a community-based longitudinal study in rural Bangladesh. J Trop Pediatr 1997; 43:133–7. 6 International Conference on Acute Respiratory Infections. The Forgotten Pandemic. Canberra, Australia: The Australian National University, 1997. 7 Narain JP. Epidemiology of acute respiratory infections. Indian J Pediatr 1989; 54:153–60. 8 Collier L, Oxford J. Human Virology. Oxford: Oxford University Press, 1993; 111–22. 9 Schmitz H, Wigand R, Heinrich W. Worldwide epidemiology of human adenovirus infections. Am J Epidemiol 1983; 117:455–66. 10 Zuckerman AJ, Banatavala JH, Pattison JR. Principles and Practice of Clinical Virology. Chichester, UK: John Wiley, 1994; 214–303. 11 Balows A. Manual of Clinical Microbiology. Washington, DC: American Society of Microbiology, 1991. 12 Krisher KK, Menegus MA. Evaluation of three types of cell culture for recovery of adenovirus from clinical specimens. J Clin Microbiol 1987; 25:1323–4. 13 Olsen MA, Shuck KM, Sambol AR, Flor SA, O’Brien J, Cabrera BJ. Isolation of seven respiratory viruses in shell vials: a practical and highly sensitive method. J Clin Microbiol 1993; 31:422–5. 14 Espy MJ, Hierholzer JC, Smith TF. The effect of centrifugation on the rapid detection of adenovirus in shell vials. Am J Clin Pathol 1987; 88:358–60. 15 Meziere A, Mollat C, Lapied R, Billaudel S, Courtieu AL. Detection of respiratory syncytial Detection of adenovirus in Jordan 16 17 18 19 20 21 22 virus antigen after seventy-two hours of culture. J Med Virol 1990; 31:241–4. Isenberg HD. Clinical Microbiology Procedure Handbook. Washington, DC: American Society of Microbiology, 1992. Meqdam MMM, Rawashdeh MO, Shurman AA, Abuharfeil N. Respiratory syncytial virus infection in infants hospitalized with respiratory illness in northern Jordan. J Trop Pediatr 1997; 43:92–5. Berman S. Epidemiology of acute respiratory infection in children of developing countries. Rev Infect Dis 1991; 13:454–62. Bakir TM, Halawani M, Ramia S. Viral etiology and epidemiology of acute respiratory infections in hospitalized Saudi children. J Trop Pediatr 1998; 44:100–3. Uduman SA, Ijaz MK, Kochiyil J, Mathew T, Hossam MK. Respiratory syncytial virus infections among hospitalized young children with acute lower respiratory illnesses in Al Ain, UAE. J Commun Dis 1996; 28:245–52. Tecu C, Gherghina I, Constantinescu C, et al. Viral etiology of acute respiratory infections in infants and young children. Roum Arch Microbiol Immunol 1996; 55:333–9. Edwards KM, Thompson J, Paolini J, Wright PR. Adenovirus infections in young children. Pediatrics 1985; 76:420–4. 65 23 Takimoto S, Grandien M, Pereira MS, et al. Comparison of enzyme-linked immunosorbent assay, direct immunouorescence assay, and virus isolation for detection of respiratory viruses in nasopharyngeal secretions. J Clin Microbiol 1991; 29:470–4. 24 Hijazi Z, Pacsa A, El-Gharbawy F, et al. Acute lower respiratory tract infections in children in Kuwait. Ann Trop Paediatr 1997; 17:127–34. 25 Misra PK, Chaudhary RS, Jain A, Pande A, Mathar A, Chatravedi UC. Viral aetiology of acute respiratory infections in children in north India. J Trop Pediatr 1990; 36:24–7. 26 Kinney J, Robertsen CM, Johnson KM, et al. Seasonal respiratory viral infection. Impact on infants with chronic lung disease following discharge from neonatal intensive care unit. Pediatr Adolesc Med 1995; 149:81–8. 27 Mahafzah AM, Landry ML. Evaluation of immunouorescent reagents, centrifugation, and conventional cultures for the diagnosis of adenovirus infections. Diagn Microbiol Infect Dis 1989; 12.407–11. 28 Nwankwo MU, Okuonghae HO, Currier G, Schuit KE. Respiratory syncytial infections in malnourished children. Ann Trop Paediatr 1994; 14:125– 30.