Transactions of the Royal Society of Tropical Medicine and Hygiene Advance Access published June 1, 2015
ORIGINAL ARTICLE
Trans R Soc Trop Med Hyg doi:10.1093/trstmh/trv042
Epidemiology of adenovirus respiratory infections among hospitalized children in Seremban, Malaysia Khuen Foong Nga,*, Kah Kee Tana, Boon Hong Nga,1, Pritiss Naira,2 and Wan Ying Ganb a
Pediatric Department, Tuanku Jaafar Hospital, 70300 Seremban, Malaysia; bDepartment of Nutrition and Dietetics, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43300 Serdang, Malaysia *Corresponding author: Tel: 6067684000; E-mail: [email protected] Present address: Hematology Department, Ampang Hospital, 68000 Ampang, Malaysia 2 Present address: Anaesthesiology Department, Tuanku Jaafar Hospital, 70300 Seremban, Malaysia 1
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Received 24 March 2015; revised 5 May 2015; accepted 6 May 2015 Background: There is scarcity of data regarding epidemiology and clinical aspects of human adenovirus acute respiratory infection (ARI) among children in developing countries. Methods: Retrospective data on demographics, clinical presentation, outcomes and laboratory findings of 116 children admitted into Tuanku Jaafar Hospital in Seremban, Malaysia from 2012 to 2013 with documented diagnosis of community-acquired adenovirus ARI were collected and analysed. Results: Male to female ratio was 1.70. Median age was 14 (1–107) months. The commonest symptoms were fever (94.8%; 110/116), cough (82.8%, 96), rhinorrhea (63.8%; 74), interrupted feeding (66.4%; 77), diarrhea (33.6%; 39) and conjunctivitis (21.6%; 25). Mean temperature on admission was 38.48C+0.98C. Among all 116 subjects, 20.7% (24) needed oxygen supplementation, 57.8% (67) required intravenous hydration, 11.2% (13) were admitted into the pediatric intensive care unit and 6.9% (8) required mechanical ventilation. Only 1% (1/87) had positive blood culture (Streptococcus pneumoniae) among 87 who received antibiotic treatment. Case fatality rate was 2.6% (3/116) and 1.7% (2/116) developed bronchiolitis obliterans. Median length of hospital stay was 4 (1–50) days. Conclusion: Adenovirus ARI caused significant morbidity and substantial resource utilization among hospitalized Malaysian children. It should be considered in the differential diagnosis of infants below two years presenting with ARI associated with high fever. Antibiotics should not be prescribed as secondary bacterial infections are uncommon. Keywords: Adenovirus, Children, Epidemiology, Malaysia, Respiratory viral infection
Introduction Acute respiratory infection (ARI) caused by human adenovirus is well documented among children in developed countries. There is a paucity of data among children hospitalized with adenovirus infections in developing countries due to lack of diagnostic facilities. Adenoviruses are non-enveloped double-stranded DNA viruses. To date, there are more than 60 recognized human adenovirus types. According to International Committee on Taxonomy of Viruses (ICTV), 54 types of human adenovirus are grouped into seven subgroups (A to G). The clinical presentations vary with serotypes, age, and host factors, e.g., immune status. Serotypes 40 and 41 are commonly associated with infantile gastroenteritis, serotypes 8, 19 and 37 with epidemic keratoconjunctivitis, serotypes 1–5, 7 and 21 with upper respiratory infection and
pneumonia and serotypes 11, 34 and 35 with hemorrhagic cystitis and interstitial nephritis. Adenoviruses affect mostly children, but adults are not spared.1 Risk factors for severe disease are children younger than 7 years old, individuals with underlying chronic disease and immunosuppressed host, e.g., transplant recipients, and patients with cancer and congenital immunodeficiency syndrome.2–4 Laboratory diagnosis can be achieved by viral culture, direct fluorescent antibody (DFA) test and PCR assay.5 There is a scarcity of data regarding epidemiology, clinical characteristics and outcome of adenovirus respiratory infection among children in the region of Southeast Asia and developing countries. The objectives of this study are to report epidemiology and clinical features of children admitted into Tuanku Jaafar Hospital in Seremban, Malaysia with adenovirus ARI.
# The Author 2015. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved. For permissions, please e-mail: [email protected].
1
K. Foong Ng et al.
Methods
Results There were 5043 and 5393 children admitted in 2012 and 2013 respectively for ARI. Respiratory secretion specimens were collected from 21.0% (1060/5043) patients in 2012 and 33.8% (1824/5393) patients in 2013. Of the 116 subjects positive for adenovirus ARI, 9.5% (11) were admitted in 2012 and the remaining 90.5% (105) in 2013. The seasonal variability of the 116 children admitted is shown in Figure 1. The majority of the admissions were between March and June and there was a small peak between September and December (Figure 1). Demographics, clinical, laboratory, management and outcome data are presented in Table 1. More males than females (male to female ratio 1.70) were infected. The majority were below 2 years old (80; 69.0%). The commonest clinical symptoms were fever (110; 94.8%), cough (96; 82.8%), rhinorrhea (74; 63.8%) and interrupted feeding (77; 66.4%). Diarrhea (39; 33.6%) and conjunctivitis (25; 21.6%) were present in a substantial number of patients. Mean and SD of temperature on admission was 38.48C+0.98C. Among the patients with comorbidity (18; 15.5%),
2
Figure 1. Seasonal variability of 116 children admitted to Tuanku Jaafar Hospital, Seremban, Malaysia for adenovirus acute respiratory infection, 2012–2013.
nine had asthma, six had congenital heart diseases, one had history of intussusception with hemicolectomy, one had HIV infection and one had cerebral palsy. The child with HIV infection was admitted for Cytomegalovirus pneumonitis and then acquired adenovirus ARI during her stay in the ward. A complete blood count was done for 90.5% (105) patients and 76.7% (89) of the subjects had their blood sampled for blood culture. Antibiotics were started for 87 (75.0%) but only one had a positive blood culture (Streptococcus pneumoniae). Among all positive isolates with adenovirus, there was one patient coinfected with respiratory syncytial virus and another with parainfluenza 3, influenza A and B. These two patients had an uncomplicated hospital stay and undifferentiated clinical features but a meaningful comparison with patients having single adenovirus ARI was not possible due to the small number. Of the 116 patients, 13 (11.2%) were admitted into PICU, out of which eight (61.5%) required tracheal intubation and mechanical ventilation. Case fatality rate was 2.6% (3/116). Of the three patients who developed complications, two were diagnosed to have bronchiolitis obliterans (both were caused by adenovirus serotype 7) and one had pneumococcal pneumonia with bacteremia and meningitis. The serotyping for the two patients with bronchiolitis obliterans was done in view of their chronic pulmonary sequelae. However, typing was not performed for the other cases. Length of hospital stay was associated with height of temperature, presence of tachypnea, oxygen and IVH requirement, longer duration of oxygen supplementation and IVH, antibiotic treatment, ribavirin administration, transfusion with blood products and IVIG, PICU admission and longer PICU length of stay (Tables 2 and 3). Admission to PICU was associated with younger age, presence of tachypnea, need for oxygen supplementation, IVH, transfusion with blood products and IVIG, antibiotic and ribavirin treatment. Mortality and complications are associated with PICU admission (Table 4).
Discussion In hospitalized children with viral ARI about 5–6% are caused by adenovirus in the developing world, e.g., Malaysia, Vietnam and Madagascar.6–8 In developed countries, e.g., France and Canada,
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Adenovirus ARI was defined by the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) codes for adenovirus pneumonia (J12.0). When a single child had multiple hospital admissions and/or samples, the earliest sample was selected for this analysis. Each eligible patient contributed only one sample to the analysis. Not all pediatric patients who were admitted for ARI had their respiratory secretion sampling done for viral studies. During the 2-year period (2012 to 2013), 116 patients admitted to Tuanku Jaafar Hospital, Seremban with positive nasopharyngeal or tracheal aspirates for adenovirus, detected by DFA method were included in our study population. The same DFA test also detects influenza A and B, respiratory syncytial virus, and parainfluenza 1, 2 and 3. Molecular serotyping was not routinely done. Clinical data were extracted from medical records using a structured and standardized questionnaire form. Information collected was demographic data (age, gender, race, date of specimen collection), type of presenting symptoms (fever, cough, rhinorrhea, sore throat, tachypnea, apnea, poor feeding, conjunctivitis, diarrhea) and their duration, temperature on admission, presence of comorbidity, asthma, history of prematurity, birth weight, gestational age, mode of delivery, laboratory findings on admission (coinfection with other respiratory viruses, blood culture, white cell, lymphocyte and neutrophil counts, hemoglobin and platelet), oxygen and intravenous hydration (IVH) requirement and duration, antibiotic and ribavirin treatment, transfusion of blood product including intravenous immunoglobulin (IVIG), pediatric intensive care unit (PICU) admission and duration of stay, invasive and noninvasive ventilation and their duration, total length of hospital stay (LOHS), mortality and type of complication. Statistical analysis was performed using SPSS version 22.0 software (IBM Corp., Armonk, NY, USA). Descriptive statistics were used to describe the demographic and clinical data. The x2 or Fisher exact test was employed for categorical variables and the independent-samples t-Test or Mann-Whitney U test for continuous variables. Non-parametric correlation between two continuous data was examined using Spearman rank order correlation analysis. A p-value of ,0.05 was considered significant.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 1. Demographic data, clinical features, laboratory findings, treatments and outcomes in children admitted for adenovirus acute respiratory infection %
Median
Range
IQR
73 43 NA
62.9 37.1 NA
NA NA 14
NA NA 1–107
NA NA 17
80 24 12
69.0 20.7 10.3
NA NA NA
NA NA NA
NA NA NA
110 96 74 2 34 4 77 25 39 18 9 16 NA NA
94.8 82.8 63.8 1.7 29.3 3.4 66.4 21.6 33.6 15.5 9.5 13.8 NA NA
5 5 5 NA 1 NA 2 3 2 NA NA NA 37 3000
1–30 1–21 1–21 NA 1–6 NA 1–11 1–14 1–15 NA NA NA 28–40 1170–4500
4 4 4 NA 2 NA 4 3 2 NA NA NA 3 550
29 79
26.9 73.1
NA NA
NA NA
NA NA
NA 4 31 NA NA NA NA 9 14
NA 3.8 29.5 NA NA NA NA 8.6 13.3
12.0 NA NA 5.9 4.2 11.6 271 NA NA
3.4–36 NA NA 0.6–20.5 0.6–13.1 6.4–16.5 44–604 NA NA
9.0 NA NA 6.0 3.2 1.8 147 NA NA
1 88 2
1.1 98.9 1.7
NA NA NA
NA NA NA
NA NA NA
24 67 13 8 8 87 23 11 8
20.7 57.8 11.2 6.9 6.9 75 19.8 9.5 6.9
7 2 6 4 4 NA NA NA NA
1–35 1–35 1–36 2–14 1–7 NA NA NA NA
10 3 5 3 4 NA NA NA NA
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Gender, n¼116 Male Female Age (month), n¼116 Age group, n¼116 0–2 years old 2 to 5 years old .5 years old Clinical features, n¼116 Fever (day) Cough (day) Rhinorrhoea (day) Apnoea Tachypnoea (day) Sorethroat Poor feeding (day) Conjunctivitis (day) Diarrhoea (day) Comorbidity Asthma Prematurity Gestational age (week), n¼38 Birth weight (g), n¼82 Mode of delivery, n¼108 C-section Vaginal Complete blood count Leukocyte (109/L), n¼105 Leukopeniaa Leukocytosisb Neutrophil (109/L), n¼102 Lymphocyte (109/L), n¼102 Hemoglobin (g/dL), n¼104 Platelet (109/L), n¼105 Thrombocytopeniac Thrombocytosisd Blood culture, n¼89 Positive Negative Coinfection with other respiratory viruses Treatment, n¼116 Oxygen (day) IVH (day) PICU (day) Ventilation (day) NIV (day) Antibiotic Ribavirin IVIG Transfusion
Frequency
Continued
3
K. Foong Ng et al.
Table 1. Continued
Length of hospital stay (day), n¼116 Outcome, n¼116 Mortality Complication
Frequency
%
Median
Range
IQR
NA
NA
4
1–50
3
3 3
2.6 2.6
NA NA
NA NA
NA NA
the proportion is around 4%9,10; however, the figure is higher in China (13%).11 These discrepancies can be explained by different research methodology and/or diagnostic tests. We experienced an increase of adenovirus ARI in 2013 when there were 9.5 times more admissions in comparison with 2012 (11 in 2012 and 105 in 2013). This could possibly be due to an ongoing outbreak of adenovirus transmission in 2013. The majority of the admissions were between March and June with a small peak between September and December. The within-year seasonality pattern of our study is similar to an earlier study done in Malaysia and this may represent the seasonal variability of adenovirus circulation in a tropical country such as Malaysia.6 In general, rainfall over the west coast of Peninsular Malaysia where Seremban is situated has two peaks, maximal in October to November and April to May. This may explain the seasonality of adenovirus circulation seen in our study. The predominance of male children among our cohort of patients (male to female ratio 1.70) is not surprising because it was also commonly reported in literature, and this is also true for other causes of viral respiratory infections.6 In this study, the commonest clinical features of adenovirus ARI are fever, cough, rhinorrhea and interrupted feeding. It is not uncommon for adenovirus respiratory infection to present with diarrhea and conjunctivitis. These findings are almost similar to other case series in Taipei, Taiwan and Manitoba, Canada.12,13 Some of these symptoms can be prolonged, as documented in our study: fever up to 30 days, cough and rhinorrhea up to 21 days, poor feeding up to 11 days, conjunctivitis and diarrhea up to 2 weeks. However, these symptoms are not unique to adenovirus ARI only and it will be clinically difficult to distinguish adenovirus ARI from other viral agents, and acute bacterial pneumonia. Only 3.4% (4/116) of our patients complained of sore throat and this can be explained by a large proportion of our patients below 2 years old (80/116; 69%) who might not be able to verbalize their symptoms well. A child with HIV infection was admitted for Cytomegalovirus pneumonitis and developed healthcare-associated adenovirus infection. This highlights the potential of adenovirus to cause healthcare-associated infection: because its non-enveloped properties render lipid disinfectant, alcohol and ether ineffective, adenovirus is known to survive on environmental surfaces for a long time and patients who had previously been infected by this
4
virus can continue to shed it in urine, stool and throat intermittently for months.14 There have been reports documenting outbreak of healthcare-associated adenovirus infection causing epidemic keratoconjunctivitis and pneumonia.15–17 Among the patients who had complete blood counts done on admission, the median leukocyte and platelet counts were nonspecific, their ranges were wide and the majority of patients had normal readings, suggesting such investigation provides no discriminating values. Antibiotics were prescribed for 75.0% (87/ 116) of the subjects but only 1% (1/87) had a positive blood culture (Streptococcus pneumoniae). This finding indicates that coinfection with pathogenic bacteria upon admission to the hospital is uncommon among patients with adenovirus ARI and antibiotics should be prescribed judiciously because the presence of fever and/or leukocytosis may not be indicative of concurrent bacterial infection. PCR is highly sensitive and specific in detecting adenovirus DNA from respiratory specimens while viral antigen detection assay techniques such as DFA is specific but less sensitive. PCR is especially useful among the immunocompromised population. However, laboratory diagnosis of adenovirus ARI by PCR is costly and is not available in resource limited countries. The DFA test is relatively inexpensive and able to provide rapid results within a few hours, making it an attractive alternative for laboratory diagnosis. We recommend utilization of viral antigen detection assays such as DFA in hospitals with scarce resources because it is inexpensive, offers early diagnosis, limits usage of antibiotics, prevents distress associated with blood sampling and may result in potential cost savings by reducing antimicrobial usage. Cidofovir is the drug of choice for adenovirus infection because it is more active against adenovirus in vitro and in vivo with demonstration of viral load reduction by real-time PCR assay and improved survival among hematopoeitic stem cell and lung transplant patients but it is associated with nephrotoxicity (up to 25% of cases), proteinuria (50% of cases) and Fanconi syndrome (1% of cases).18–20 Ganciclovir, vidarabine and ribavirin have inconsistent in vitro activity against adenovirus. However ribavirin and IVIG were used in our cohort of patients with severe adenovirus ARI because cidofovir is not available in Malaysia and there are case reports of successful treatment with ribavirin and pooled IVIG, although data suggests efficacy only to serogroup C adenoviruses.21,22
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
IVH: intravenous hydration; IVIG: intravenous immunoglobulin; NA: not applicable; NIV: noninvasive ventilation; PICU: pediatric intensive care unit. a leukocyte count ,5×109/L. b leukocyte count .15×109/L. c platelet count ,150×109/L. d platelet count .400×109/L.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 2. Difference in length of hospital stay according to gender, clinical characteristics, treatments and pediatric intensive care unit admission n
p-value
73 43
4 (1–43) 5 (1–50)
NS
110 6 96 20 74 42 34 82 77 39 25 91 39 77 18 98 9 107 29 79 92 16
4 (1–50) 2.5 (1–9) 4 (1–50) 4 (2–17) 4 (1–50) 4 (1–43) 5 (1–50) 4 (1–14) 4 (1–50) 4 (1–35) 4 (1–43) 4 (1–50) 4 (1–11) 4 (1–50) 5 (1–35) 4 (1–50) 3 (1–8) 4 (1–50) 4 (2–13) 4 (1–50) 4 (1–50) 4 (1–11)
NS
24 92 67 49 87 29 23 93 11 105 8 108 13 103
10 (2–50) 4 (1–14) 5 (1–50) 3 (1–9) 5 (1–50) 3 (1–9) 11 (5–50) 4 (1–8) 13 (7–50) 4 (1–19) 13.5 (5–50) 4 (1–35) 12 (7–50) 4 (1–14)
NS NS 0.030 NS NS NS NS NS
Demographic data Age Clinical features Temperature Fever duration Cough duration Rhinorrhoea duration Tachypnoea duration Conjunctivitis duration Diarrhoea duration Gestational age Birth weight Laboratory findings Leukocyte Neutrophil Lymphocyte Hemoglobin Platelet Treatment Oxygen duration IVH duration PICU length of stay
n
rho
p-value
116
20.008
NS
116 110 95 72 33 24 39 38 82
0.306 20.014 0.136 0.127 0.035 20.192 0.111 0.224 20.149
0.001 NS NS NS NS NS NS NS NS
105 102 102 104 105
0.064 0.088 20.029 20.134 0.037
NS NS NS NS NS
24 67 13
0.917 0.709 0.756
,0.001 ,0.001 0.003
NS NS
IVH: intravenous hydration; NS: not significant; PICU: pediatric intensive care unit.
,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001
IVH: intravenous hydration; IVIG: intravenous immunoglobulin; NS: not significant; PICU: pediatric intensive care unit.
Our PICU admission rate (13/116; 11.2%) due to adenovirus ARI is almost similar to cohorts of patients in high-income countries: Kansas, USA (8/79; 10%)23 and Taipei (31/202; 15.3%).12 Our percentage of mechanically ventilated patients (8/116; 6.9%) is
almost comparable with other studies done in other parts of the world: 10.9% (21/193) in Manitoba13 and 10% in Taipei.12 This could partly be explained by the availability of noninvasive ventilation in our center and differences in management across centers. Despite the lack of extracorporeal membrane oxygenation service in our facility, our case fatality rate (3/116; 2.6%) is considered low; it is higher than what was observed in Manitoba (0/193; 0%)13 but lower than what was reported in a study done in Taipei (7/202; 3.5%).12 The rate of chronic lung sequelae among our patients (2/116; 1.7%) is almost similar to the studies done in Manitoba (8/193; 4.1%)13 and Taoyuan, Taiwan (2/76; 2.6%).24 Chronic lung disease in two patients was caused by serotype 7. The adenovirus serotypes commonly associated with severe disease are 3 and 7, both are associated with higher case fatality rate, severe acute respiratory failure, hypotension and chronic pulmonary sequelae, e.g., bronchiolitis obliterans and Swyer James syndrome.25,26 The findings of our study that may predict the severity of adenovirus ARI are young age, height of temperature, presence of tachypnea, need for oxygen supplementation and IVH. In the Taiwan study clinical features of severe adenovirus ARI were duration of fever, peak temperature, dyspnea, exudative tonsillitis, abnormal lung auscultation, presence of underlying disease such as prematurity, neurologic, hematologic and metabolic
5
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Demographic data Gender Male Female Clinical features Fever Yes No Cough Yes No Rhinorrhoea Yes No Tachypnoea Yes No Poor feeding Yes No Conjunctivitis Yes No Diarrhoea Yes No Comorbidity Yes No Asthma Yes No Mode of delivery C-section Vaginal Term/preterm Term Preterm Treatment Oxygen Yes No IVH Yes No Antibiotic Yes No Ribavirin Yes No IVIG Yes No Transfusion Yes No PICU admission Yes No
Median LOHS (day)
Table 3. Correlation analysis of length of hospital stay with age, clinical features, laboratory findings and treatments in children admitted for adenovirus acute respiratory infection
K. Foong Ng et al.
Table 4. Pediatric intensive care unit admission among children with adenovirus acute respiratory infection by demographic data, clinical features, laboratory findings, treatments and outcomes PICU admission No, n¼103
p-value
9 (69.2%) 10 (2–23)
64 (62.1%) 15 (1–107)
NS 0.047
13 (100%) 38.7+1.0 12 (92.3%) 8 (61.5%) 12 (92.3% 10 (76.9%) 3 (23.1%) 2 (15.4%) 3 (23.1%) 0 1 (9.1%) 3000 1 (8.3%) 38.5 (36–40)
97 (94.2%) 38.4+0.9 84 (81.6%) 66 (64.1%) 22 (21.4%) 67 (65.0%) 22 (21.4%) 37 (35.9%) 15 (14.6%) 9 (8.7%) 28 (28.9%) 3000 15 (15.6%) 37 (28–40)
NS NSa NS NSb ,0.001 NS NS NS NS NS NS NS NS NS
12.9 (4.7–17.6) 7.4 (2.4–11.2) 4.4 (0.6–9.5) 11.0 (9.1–16.5) 250 (73–580)
11.2 (3.4–36) 5.7 (0.6–20.5) 4.1 (1.2–13.1) 11.6 (6.4–14.1) 284 (44–604)
NS NS NS NS NS
13 (100%) 13 (100%) 13 (100%) 13 (100%) 9 (69.2%) 7 (53.8%)
11 (10.7%) 54 (52.4%) 74 (71.8%) 10 (9.7%) 2 (1.9%) 1 (1.0%)
,0.001 ,0.001 0.036 ,0.001 ,0.001 ,0.001
3 (23.1%) 3 (23.1%)
0 0
0.001 0.001
IVH: intravenous hydration; IVIG: intravenous immunoglobulin; NS: not significant; PICU: pediatric intensive care unit. independent-sample t-test. b 2 x test. a
disorders.12 In their study, the abnormal chest radiograph findings, anemia, thrombocytopenia, leukopenia, hyponatremia, transaminitis and raised lactate dehydrogenase were all associated with severe adenovirus disease. However, our study showed that complete blood count indices were not helpful in determining the possibility of severe adenovirus ARI. Median LOHS of our study patients was 4 days (range 1–50) while the study in Manitoba reported a median LOHS of 5 days (range 1–189).13 Another study carried out in Taoyuan revealed a mean LOHS of 9.89+9.55 days.24 In a study performed in
6
Massachusetts, USA, median LOHS was 3.5 days (range 1–292).27 This showed that the average LOHS due to adenovirus ARI is similar across geographical regions. There are several limitations in our study. Mild cases with adenovirus associated ARI might not be hospitalized. Nasopharyngeal aspirates for adenovirus DFA are not routinely done for all patients, contributing to potential underdiagnosis. Molecular methods were not available and may lead to underdiagnosis, as PCR methods are more sensitive than DFA.28 Serotyping of adenovirus by PCR method is not available in our hospital laboratory to document temporal
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Demographic data Male Median age in month (range) Clinical features Fever Mean temperature (8C) Cough Rhinorrhoea Tachypnoea Poor feeding Conjunctivitis Diarrhoea Comorbidity Asthma C-section Birth weight (g) Prematurity Gestational age (week) Laboratory findings Leukocyte (109/L) Neutrophil (109/L) Lymphocyte (109/L) Hemoglobin (g/dL) Platelet (109/L) Treatment Oxygen IVH Antibiotic Ribavirin IVIG Transfusion Outcome Mortality Complication
Yes, n¼13
Transactions of the Royal Society of Tropical Medicine and Hygiene
trends and severity of illness. This study adds to the limited data on epidemiology and outcome of adenovirus ARI in developing countries. In conclusion, adenovirus ARI caused variable duration and severity of symptoms among hospitalized Malaysian children. It should be recognized as a significant cause of childhood morbidity in developing countries, with potential impact on hospitalization and resource utilization. Laboratory confirmation by DFA method is inexpensive, offers early diagnosis, limits usage of antibiotics, and potentially reduce cost of hospitalization.
9 Berciaud S, Rayne F, Kassab S et al. Adenovirus infections in Bordeaux University Hospital 2008–2010: clinical and virological features. J Clin Virol 2012;54:302–7. 10 Banerji A, Greenberg D, White LF et al. Risk factors and viruses associated with hospitalization due to lower respiratory tract infections in Canadian Inuit children: a case-control study. Pediatr Infect Dis J 2009;28:697–701. 11 Zhang C, Zhu N, Xie Z et al. Viral etiology and clinical profiles of children with severe acute respiratory infections in China. PLoS One 2013;8:e72606. 12 Tsou TP, Tan BF, Chang HY et al. Community outbreak of adenovirus, Taiwan, 2011. Emerg Infect Dis 2012;18:1825–32.
Acknowledgements: We thank Nurul Zamil Mohd Muzzamil for assisting the authors to obtain research registration from National Medical Research Registry of Malaysia. We also appreciate non-financial support from Director General of Health, Malaysia and Clinical Research Centre (CRC), Tuanku Jaafar Hospital. Funding: None. Competing interests: None declared. Ethical approval: This research was approved by Medical Research and Ethics Committee of Malaysia.
References 1 Ampuero JS, Vıctor Ocana V, Gomez J et al. Adenovirus respiratory tract infections in Peru. PLoS One 2012;7:e46898. 2 Kojaoghlanian T, Flomenberg P, Horwitz MS. The impact of adenovirus infection on the immunocompromised host. Rev Med Virol 2003; 13:155–71. 3 Ison MG. Adenovirus infections in transplant recipients. Clin Infect Dis 2006;43:331–9. 4 Gray GC, McCarthy T, Lebeck MG et al. Genotype prevalence and risk factors for severe clinical adenovirus infection, United States 2004– 2006. Clin Infect Dis 2007;45:1120–31. 5 Tabain I, Ljubin-Sternak S, Cepin-Bogovic´ J et al. Adenovirus respiratory infections in hospitalized children: clinical findings in relation to species and serotypes. Pediatr Infect Dis J 2012;31:680–4. 6 Khor CS, Sam IC, Hooi PS et al. Epidemiology and seasonality of respiratory viral infections in hospitalized children in Kuala Lumpur, Malaysia: a retrospective study of 27 years. BMC Pediatr 2012;12:32.
14 Buehler JW, Finton RJ, Goodman RA et al. Epidemic keratoconjunctivitis: report of an outbreak in an ophthalmology practice and recommendations for prevention. Infect Control 1984;5:390–4. 15 Jernigan JA, Lowry BS, Hayden FG et al. Adenovirus type 8 epidemic keratoconjunctivitis in an eye clinic: risk factors and control. J Infect Dis 1993;167:1307–13. 16 Finn A, Anday E, Talbot GH. An epidemic of adenovirus 7a infection in a neonatal nursery: course, morbidity, and management. Infect Control Hosp Epidemiol 1988;9:398–404. 17 Faden H, Wynn RJ, Campagna L, Ryan RM. Outbreak of adenovirus type 30 in a neonatal intensive care unit. J Pediatr 2005;146:523–7. 18 Leruez-Ville M, Minard V, Lacaille F et al. Real-time blood plasma polymerase chain reaction for management of disseminated adenovirus infection. Clin Infect Dis 2004;38:45–52. 19 Neofytos D, Ojha A, Mookerjee B et al. Treatment of adenovirus disease in stem cell transplant recipients with cidofovir. Biol Blood Marrow Transplant 2007;13:74–81. 20 Shah DP, Ghantoji SS, Mulanovich VE et al. Management of respiratory viral infections in hematopoietic cell transplant recipients. Am J Blood Res 2012;2:203–18. 21 Sabroe I, McHale J, Tait DR et al. Treatment of adenoviral pneumonitis with intravenous ribavirin and immunoglobulin. Thorax 1995;50: 1219–20. 22 Emovon OE, Lin A, Howell DN et al. Refractory adenovirus infection after simultaneous kidney-pancreas transplantation: successful treatment with intravenous ribavirin and pooled human intravenous immunoglobulin. Nephrol Dial Transplant 2003;18:2436–8. 23 Selvaraju SB, Kovac M, Dickson LM et al. Molecular epidemiology and clinical presentation of human adenovirus infections in Kansas City children. J Clin Virol 2011;51:126–31. 24 Chen SP, Huang YC, Chiu CH. Clinical features of radiologically confirmed pneumonia due to adenovirus in children. J Clin Virol 2013;56:7–12. 25 Lai CY, Lee CJ, Lu CY et al. Adenovirus serotype 3 and 7 infection with acute respiratory failure in children in Taiwan, 2010–2011. PLoS One 2013;8:e53614. 26 Callaway Z, Kim SH, Kim JY et al. Adenovirus infection with serious pulmonary sequelae in Korean children. Clin Respir J 2011;5:92–8.
7 Do AH, van Doorn HR, Nghiem MN et al. Viral etiologies of acute respiratory infections among hospitalized Vietnamese children in Ho Chi Minh City, 2004–2008. PLoS One 2011;6:e18176.
27 Gupta P, Sunali Goyal S, Tobias JD et al. Risk factors associated with hospital admission among healthy children with adenovirus infection. Turk J Pediatr 2011;53:597–603.
8 Hoffmann J, Rabezanahary H, Randriamarotia M et al. Viral and atypical bacterial etiology of acute respiratory infections in children under 5 years old living in a rural tropical area of Madagascar. PLoS One. 2012;7:e43666.
28 Stroparo E, Cruz CR, Debur Mdo C et al. Adenovirus respiratory infection: significant increase in diagnosis using PCR comparing with antigen detection and culture methods. Rev Inst Med Trop Sao Paulo 2010; 52:317–21.
7
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Authors’ contributions: KKT conceived the study; KFN designed the study protocol; BHN and PN reviewed the case notes and entered the data; KFN and WYG carried out the data analysis and interpretation of these data. KFN drafted the manuscript; KKT and KFN critically revised the manuscript for intellectual content. All authors read and approved the final manuscript. Guarantor of this paper is KKT.
13 Alharbi S, Caeseele PV, Consunji-Araneta R et al. Epidemiology of severe pediatric adenovirus lower respiratory tract infections in Manitoba, Canada, 1991–2005. BMC Infect Dis 2012;12:55.
ORIGINAL ARTICLE
Trans R Soc Trop Med Hyg doi:10.1093/trstmh/trv042
Epidemiology of adenovirus respiratory infections among hospitalized children in Seremban, Malaysia Khuen Foong Nga,*, Kah Kee Tana, Boon Hong Nga,1, Pritiss Naira,2 and Wan Ying Ganb a
Pediatric Department, Tuanku Jaafar Hospital, 70300 Seremban, Malaysia; bDepartment of Nutrition and Dietetics, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43300 Serdang, Malaysia *Corresponding author: Tel: 6067684000; E-mail: [email protected] Present address: Hematology Department, Ampang Hospital, 68000 Ampang, Malaysia 2 Present address: Anaesthesiology Department, Tuanku Jaafar Hospital, 70300 Seremban, Malaysia 1
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Received 24 March 2015; revised 5 May 2015; accepted 6 May 2015 Background: There is scarcity of data regarding epidemiology and clinical aspects of human adenovirus acute respiratory infection (ARI) among children in developing countries. Methods: Retrospective data on demographics, clinical presentation, outcomes and laboratory findings of 116 children admitted into Tuanku Jaafar Hospital in Seremban, Malaysia from 2012 to 2013 with documented diagnosis of community-acquired adenovirus ARI were collected and analysed. Results: Male to female ratio was 1.70. Median age was 14 (1–107) months. The commonest symptoms were fever (94.8%; 110/116), cough (82.8%, 96), rhinorrhea (63.8%; 74), interrupted feeding (66.4%; 77), diarrhea (33.6%; 39) and conjunctivitis (21.6%; 25). Mean temperature on admission was 38.48C+0.98C. Among all 116 subjects, 20.7% (24) needed oxygen supplementation, 57.8% (67) required intravenous hydration, 11.2% (13) were admitted into the pediatric intensive care unit and 6.9% (8) required mechanical ventilation. Only 1% (1/87) had positive blood culture (Streptococcus pneumoniae) among 87 who received antibiotic treatment. Case fatality rate was 2.6% (3/116) and 1.7% (2/116) developed bronchiolitis obliterans. Median length of hospital stay was 4 (1–50) days. Conclusion: Adenovirus ARI caused significant morbidity and substantial resource utilization among hospitalized Malaysian children. It should be considered in the differential diagnosis of infants below two years presenting with ARI associated with high fever. Antibiotics should not be prescribed as secondary bacterial infections are uncommon. Keywords: Adenovirus, Children, Epidemiology, Malaysia, Respiratory viral infection
Introduction Acute respiratory infection (ARI) caused by human adenovirus is well documented among children in developed countries. There is a paucity of data among children hospitalized with adenovirus infections in developing countries due to lack of diagnostic facilities. Adenoviruses are non-enveloped double-stranded DNA viruses. To date, there are more than 60 recognized human adenovirus types. According to International Committee on Taxonomy of Viruses (ICTV), 54 types of human adenovirus are grouped into seven subgroups (A to G). The clinical presentations vary with serotypes, age, and host factors, e.g., immune status. Serotypes 40 and 41 are commonly associated with infantile gastroenteritis, serotypes 8, 19 and 37 with epidemic keratoconjunctivitis, serotypes 1–5, 7 and 21 with upper respiratory infection and
pneumonia and serotypes 11, 34 and 35 with hemorrhagic cystitis and interstitial nephritis. Adenoviruses affect mostly children, but adults are not spared.1 Risk factors for severe disease are children younger than 7 years old, individuals with underlying chronic disease and immunosuppressed host, e.g., transplant recipients, and patients with cancer and congenital immunodeficiency syndrome.2–4 Laboratory diagnosis can be achieved by viral culture, direct fluorescent antibody (DFA) test and PCR assay.5 There is a scarcity of data regarding epidemiology, clinical characteristics and outcome of adenovirus respiratory infection among children in the region of Southeast Asia and developing countries. The objectives of this study are to report epidemiology and clinical features of children admitted into Tuanku Jaafar Hospital in Seremban, Malaysia with adenovirus ARI.
# The Author 2015. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved. For permissions, please e-mail: [email protected].
1
K. Foong Ng et al.
Methods
Results There were 5043 and 5393 children admitted in 2012 and 2013 respectively for ARI. Respiratory secretion specimens were collected from 21.0% (1060/5043) patients in 2012 and 33.8% (1824/5393) patients in 2013. Of the 116 subjects positive for adenovirus ARI, 9.5% (11) were admitted in 2012 and the remaining 90.5% (105) in 2013. The seasonal variability of the 116 children admitted is shown in Figure 1. The majority of the admissions were between March and June and there was a small peak between September and December (Figure 1). Demographics, clinical, laboratory, management and outcome data are presented in Table 1. More males than females (male to female ratio 1.70) were infected. The majority were below 2 years old (80; 69.0%). The commonest clinical symptoms were fever (110; 94.8%), cough (96; 82.8%), rhinorrhea (74; 63.8%) and interrupted feeding (77; 66.4%). Diarrhea (39; 33.6%) and conjunctivitis (25; 21.6%) were present in a substantial number of patients. Mean and SD of temperature on admission was 38.48C+0.98C. Among the patients with comorbidity (18; 15.5%),
2
Figure 1. Seasonal variability of 116 children admitted to Tuanku Jaafar Hospital, Seremban, Malaysia for adenovirus acute respiratory infection, 2012–2013.
nine had asthma, six had congenital heart diseases, one had history of intussusception with hemicolectomy, one had HIV infection and one had cerebral palsy. The child with HIV infection was admitted for Cytomegalovirus pneumonitis and then acquired adenovirus ARI during her stay in the ward. A complete blood count was done for 90.5% (105) patients and 76.7% (89) of the subjects had their blood sampled for blood culture. Antibiotics were started for 87 (75.0%) but only one had a positive blood culture (Streptococcus pneumoniae). Among all positive isolates with adenovirus, there was one patient coinfected with respiratory syncytial virus and another with parainfluenza 3, influenza A and B. These two patients had an uncomplicated hospital stay and undifferentiated clinical features but a meaningful comparison with patients having single adenovirus ARI was not possible due to the small number. Of the 116 patients, 13 (11.2%) were admitted into PICU, out of which eight (61.5%) required tracheal intubation and mechanical ventilation. Case fatality rate was 2.6% (3/116). Of the three patients who developed complications, two were diagnosed to have bronchiolitis obliterans (both were caused by adenovirus serotype 7) and one had pneumococcal pneumonia with bacteremia and meningitis. The serotyping for the two patients with bronchiolitis obliterans was done in view of their chronic pulmonary sequelae. However, typing was not performed for the other cases. Length of hospital stay was associated with height of temperature, presence of tachypnea, oxygen and IVH requirement, longer duration of oxygen supplementation and IVH, antibiotic treatment, ribavirin administration, transfusion with blood products and IVIG, PICU admission and longer PICU length of stay (Tables 2 and 3). Admission to PICU was associated with younger age, presence of tachypnea, need for oxygen supplementation, IVH, transfusion with blood products and IVIG, antibiotic and ribavirin treatment. Mortality and complications are associated with PICU admission (Table 4).
Discussion In hospitalized children with viral ARI about 5–6% are caused by adenovirus in the developing world, e.g., Malaysia, Vietnam and Madagascar.6–8 In developed countries, e.g., France and Canada,
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Adenovirus ARI was defined by the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) codes for adenovirus pneumonia (J12.0). When a single child had multiple hospital admissions and/or samples, the earliest sample was selected for this analysis. Each eligible patient contributed only one sample to the analysis. Not all pediatric patients who were admitted for ARI had their respiratory secretion sampling done for viral studies. During the 2-year period (2012 to 2013), 116 patients admitted to Tuanku Jaafar Hospital, Seremban with positive nasopharyngeal or tracheal aspirates for adenovirus, detected by DFA method were included in our study population. The same DFA test also detects influenza A and B, respiratory syncytial virus, and parainfluenza 1, 2 and 3. Molecular serotyping was not routinely done. Clinical data were extracted from medical records using a structured and standardized questionnaire form. Information collected was demographic data (age, gender, race, date of specimen collection), type of presenting symptoms (fever, cough, rhinorrhea, sore throat, tachypnea, apnea, poor feeding, conjunctivitis, diarrhea) and their duration, temperature on admission, presence of comorbidity, asthma, history of prematurity, birth weight, gestational age, mode of delivery, laboratory findings on admission (coinfection with other respiratory viruses, blood culture, white cell, lymphocyte and neutrophil counts, hemoglobin and platelet), oxygen and intravenous hydration (IVH) requirement and duration, antibiotic and ribavirin treatment, transfusion of blood product including intravenous immunoglobulin (IVIG), pediatric intensive care unit (PICU) admission and duration of stay, invasive and noninvasive ventilation and their duration, total length of hospital stay (LOHS), mortality and type of complication. Statistical analysis was performed using SPSS version 22.0 software (IBM Corp., Armonk, NY, USA). Descriptive statistics were used to describe the demographic and clinical data. The x2 or Fisher exact test was employed for categorical variables and the independent-samples t-Test or Mann-Whitney U test for continuous variables. Non-parametric correlation between two continuous data was examined using Spearman rank order correlation analysis. A p-value of ,0.05 was considered significant.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 1. Demographic data, clinical features, laboratory findings, treatments and outcomes in children admitted for adenovirus acute respiratory infection %
Median
Range
IQR
73 43 NA
62.9 37.1 NA
NA NA 14
NA NA 1–107
NA NA 17
80 24 12
69.0 20.7 10.3
NA NA NA
NA NA NA
NA NA NA
110 96 74 2 34 4 77 25 39 18 9 16 NA NA
94.8 82.8 63.8 1.7 29.3 3.4 66.4 21.6 33.6 15.5 9.5 13.8 NA NA
5 5 5 NA 1 NA 2 3 2 NA NA NA 37 3000
1–30 1–21 1–21 NA 1–6 NA 1–11 1–14 1–15 NA NA NA 28–40 1170–4500
4 4 4 NA 2 NA 4 3 2 NA NA NA 3 550
29 79
26.9 73.1
NA NA
NA NA
NA NA
NA 4 31 NA NA NA NA 9 14
NA 3.8 29.5 NA NA NA NA 8.6 13.3
12.0 NA NA 5.9 4.2 11.6 271 NA NA
3.4–36 NA NA 0.6–20.5 0.6–13.1 6.4–16.5 44–604 NA NA
9.0 NA NA 6.0 3.2 1.8 147 NA NA
1 88 2
1.1 98.9 1.7
NA NA NA
NA NA NA
NA NA NA
24 67 13 8 8 87 23 11 8
20.7 57.8 11.2 6.9 6.9 75 19.8 9.5 6.9
7 2 6 4 4 NA NA NA NA
1–35 1–35 1–36 2–14 1–7 NA NA NA NA
10 3 5 3 4 NA NA NA NA
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Gender, n¼116 Male Female Age (month), n¼116 Age group, n¼116 0–2 years old 2 to 5 years old .5 years old Clinical features, n¼116 Fever (day) Cough (day) Rhinorrhoea (day) Apnoea Tachypnoea (day) Sorethroat Poor feeding (day) Conjunctivitis (day) Diarrhoea (day) Comorbidity Asthma Prematurity Gestational age (week), n¼38 Birth weight (g), n¼82 Mode of delivery, n¼108 C-section Vaginal Complete blood count Leukocyte (109/L), n¼105 Leukopeniaa Leukocytosisb Neutrophil (109/L), n¼102 Lymphocyte (109/L), n¼102 Hemoglobin (g/dL), n¼104 Platelet (109/L), n¼105 Thrombocytopeniac Thrombocytosisd Blood culture, n¼89 Positive Negative Coinfection with other respiratory viruses Treatment, n¼116 Oxygen (day) IVH (day) PICU (day) Ventilation (day) NIV (day) Antibiotic Ribavirin IVIG Transfusion
Frequency
Continued
3
K. Foong Ng et al.
Table 1. Continued
Length of hospital stay (day), n¼116 Outcome, n¼116 Mortality Complication
Frequency
%
Median
Range
IQR
NA
NA
4
1–50
3
3 3
2.6 2.6
NA NA
NA NA
NA NA
the proportion is around 4%9,10; however, the figure is higher in China (13%).11 These discrepancies can be explained by different research methodology and/or diagnostic tests. We experienced an increase of adenovirus ARI in 2013 when there were 9.5 times more admissions in comparison with 2012 (11 in 2012 and 105 in 2013). This could possibly be due to an ongoing outbreak of adenovirus transmission in 2013. The majority of the admissions were between March and June with a small peak between September and December. The within-year seasonality pattern of our study is similar to an earlier study done in Malaysia and this may represent the seasonal variability of adenovirus circulation in a tropical country such as Malaysia.6 In general, rainfall over the west coast of Peninsular Malaysia where Seremban is situated has two peaks, maximal in October to November and April to May. This may explain the seasonality of adenovirus circulation seen in our study. The predominance of male children among our cohort of patients (male to female ratio 1.70) is not surprising because it was also commonly reported in literature, and this is also true for other causes of viral respiratory infections.6 In this study, the commonest clinical features of adenovirus ARI are fever, cough, rhinorrhea and interrupted feeding. It is not uncommon for adenovirus respiratory infection to present with diarrhea and conjunctivitis. These findings are almost similar to other case series in Taipei, Taiwan and Manitoba, Canada.12,13 Some of these symptoms can be prolonged, as documented in our study: fever up to 30 days, cough and rhinorrhea up to 21 days, poor feeding up to 11 days, conjunctivitis and diarrhea up to 2 weeks. However, these symptoms are not unique to adenovirus ARI only and it will be clinically difficult to distinguish adenovirus ARI from other viral agents, and acute bacterial pneumonia. Only 3.4% (4/116) of our patients complained of sore throat and this can be explained by a large proportion of our patients below 2 years old (80/116; 69%) who might not be able to verbalize their symptoms well. A child with HIV infection was admitted for Cytomegalovirus pneumonitis and developed healthcare-associated adenovirus infection. This highlights the potential of adenovirus to cause healthcare-associated infection: because its non-enveloped properties render lipid disinfectant, alcohol and ether ineffective, adenovirus is known to survive on environmental surfaces for a long time and patients who had previously been infected by this
4
virus can continue to shed it in urine, stool and throat intermittently for months.14 There have been reports documenting outbreak of healthcare-associated adenovirus infection causing epidemic keratoconjunctivitis and pneumonia.15–17 Among the patients who had complete blood counts done on admission, the median leukocyte and platelet counts were nonspecific, their ranges were wide and the majority of patients had normal readings, suggesting such investigation provides no discriminating values. Antibiotics were prescribed for 75.0% (87/ 116) of the subjects but only 1% (1/87) had a positive blood culture (Streptococcus pneumoniae). This finding indicates that coinfection with pathogenic bacteria upon admission to the hospital is uncommon among patients with adenovirus ARI and antibiotics should be prescribed judiciously because the presence of fever and/or leukocytosis may not be indicative of concurrent bacterial infection. PCR is highly sensitive and specific in detecting adenovirus DNA from respiratory specimens while viral antigen detection assay techniques such as DFA is specific but less sensitive. PCR is especially useful among the immunocompromised population. However, laboratory diagnosis of adenovirus ARI by PCR is costly and is not available in resource limited countries. The DFA test is relatively inexpensive and able to provide rapid results within a few hours, making it an attractive alternative for laboratory diagnosis. We recommend utilization of viral antigen detection assays such as DFA in hospitals with scarce resources because it is inexpensive, offers early diagnosis, limits usage of antibiotics, prevents distress associated with blood sampling and may result in potential cost savings by reducing antimicrobial usage. Cidofovir is the drug of choice for adenovirus infection because it is more active against adenovirus in vitro and in vivo with demonstration of viral load reduction by real-time PCR assay and improved survival among hematopoeitic stem cell and lung transplant patients but it is associated with nephrotoxicity (up to 25% of cases), proteinuria (50% of cases) and Fanconi syndrome (1% of cases).18–20 Ganciclovir, vidarabine and ribavirin have inconsistent in vitro activity against adenovirus. However ribavirin and IVIG were used in our cohort of patients with severe adenovirus ARI because cidofovir is not available in Malaysia and there are case reports of successful treatment with ribavirin and pooled IVIG, although data suggests efficacy only to serogroup C adenoviruses.21,22
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
IVH: intravenous hydration; IVIG: intravenous immunoglobulin; NA: not applicable; NIV: noninvasive ventilation; PICU: pediatric intensive care unit. a leukocyte count ,5×109/L. b leukocyte count .15×109/L. c platelet count ,150×109/L. d platelet count .400×109/L.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 2. Difference in length of hospital stay according to gender, clinical characteristics, treatments and pediatric intensive care unit admission n
p-value
73 43
4 (1–43) 5 (1–50)
NS
110 6 96 20 74 42 34 82 77 39 25 91 39 77 18 98 9 107 29 79 92 16
4 (1–50) 2.5 (1–9) 4 (1–50) 4 (2–17) 4 (1–50) 4 (1–43) 5 (1–50) 4 (1–14) 4 (1–50) 4 (1–35) 4 (1–43) 4 (1–50) 4 (1–11) 4 (1–50) 5 (1–35) 4 (1–50) 3 (1–8) 4 (1–50) 4 (2–13) 4 (1–50) 4 (1–50) 4 (1–11)
NS
24 92 67 49 87 29 23 93 11 105 8 108 13 103
10 (2–50) 4 (1–14) 5 (1–50) 3 (1–9) 5 (1–50) 3 (1–9) 11 (5–50) 4 (1–8) 13 (7–50) 4 (1–19) 13.5 (5–50) 4 (1–35) 12 (7–50) 4 (1–14)
NS NS 0.030 NS NS NS NS NS
Demographic data Age Clinical features Temperature Fever duration Cough duration Rhinorrhoea duration Tachypnoea duration Conjunctivitis duration Diarrhoea duration Gestational age Birth weight Laboratory findings Leukocyte Neutrophil Lymphocyte Hemoglobin Platelet Treatment Oxygen duration IVH duration PICU length of stay
n
rho
p-value
116
20.008
NS
116 110 95 72 33 24 39 38 82
0.306 20.014 0.136 0.127 0.035 20.192 0.111 0.224 20.149
0.001 NS NS NS NS NS NS NS NS
105 102 102 104 105
0.064 0.088 20.029 20.134 0.037
NS NS NS NS NS
24 67 13
0.917 0.709 0.756
,0.001 ,0.001 0.003
NS NS
IVH: intravenous hydration; NS: not significant; PICU: pediatric intensive care unit.
,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001
IVH: intravenous hydration; IVIG: intravenous immunoglobulin; NS: not significant; PICU: pediatric intensive care unit.
Our PICU admission rate (13/116; 11.2%) due to adenovirus ARI is almost similar to cohorts of patients in high-income countries: Kansas, USA (8/79; 10%)23 and Taipei (31/202; 15.3%).12 Our percentage of mechanically ventilated patients (8/116; 6.9%) is
almost comparable with other studies done in other parts of the world: 10.9% (21/193) in Manitoba13 and 10% in Taipei.12 This could partly be explained by the availability of noninvasive ventilation in our center and differences in management across centers. Despite the lack of extracorporeal membrane oxygenation service in our facility, our case fatality rate (3/116; 2.6%) is considered low; it is higher than what was observed in Manitoba (0/193; 0%)13 but lower than what was reported in a study done in Taipei (7/202; 3.5%).12 The rate of chronic lung sequelae among our patients (2/116; 1.7%) is almost similar to the studies done in Manitoba (8/193; 4.1%)13 and Taoyuan, Taiwan (2/76; 2.6%).24 Chronic lung disease in two patients was caused by serotype 7. The adenovirus serotypes commonly associated with severe disease are 3 and 7, both are associated with higher case fatality rate, severe acute respiratory failure, hypotension and chronic pulmonary sequelae, e.g., bronchiolitis obliterans and Swyer James syndrome.25,26 The findings of our study that may predict the severity of adenovirus ARI are young age, height of temperature, presence of tachypnea, need for oxygen supplementation and IVH. In the Taiwan study clinical features of severe adenovirus ARI were duration of fever, peak temperature, dyspnea, exudative tonsillitis, abnormal lung auscultation, presence of underlying disease such as prematurity, neurologic, hematologic and metabolic
5
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Demographic data Gender Male Female Clinical features Fever Yes No Cough Yes No Rhinorrhoea Yes No Tachypnoea Yes No Poor feeding Yes No Conjunctivitis Yes No Diarrhoea Yes No Comorbidity Yes No Asthma Yes No Mode of delivery C-section Vaginal Term/preterm Term Preterm Treatment Oxygen Yes No IVH Yes No Antibiotic Yes No Ribavirin Yes No IVIG Yes No Transfusion Yes No PICU admission Yes No
Median LOHS (day)
Table 3. Correlation analysis of length of hospital stay with age, clinical features, laboratory findings and treatments in children admitted for adenovirus acute respiratory infection
K. Foong Ng et al.
Table 4. Pediatric intensive care unit admission among children with adenovirus acute respiratory infection by demographic data, clinical features, laboratory findings, treatments and outcomes PICU admission No, n¼103
p-value
9 (69.2%) 10 (2–23)
64 (62.1%) 15 (1–107)
NS 0.047
13 (100%) 38.7+1.0 12 (92.3%) 8 (61.5%) 12 (92.3% 10 (76.9%) 3 (23.1%) 2 (15.4%) 3 (23.1%) 0 1 (9.1%) 3000 1 (8.3%) 38.5 (36–40)
97 (94.2%) 38.4+0.9 84 (81.6%) 66 (64.1%) 22 (21.4%) 67 (65.0%) 22 (21.4%) 37 (35.9%) 15 (14.6%) 9 (8.7%) 28 (28.9%) 3000 15 (15.6%) 37 (28–40)
NS NSa NS NSb ,0.001 NS NS NS NS NS NS NS NS NS
12.9 (4.7–17.6) 7.4 (2.4–11.2) 4.4 (0.6–9.5) 11.0 (9.1–16.5) 250 (73–580)
11.2 (3.4–36) 5.7 (0.6–20.5) 4.1 (1.2–13.1) 11.6 (6.4–14.1) 284 (44–604)
NS NS NS NS NS
13 (100%) 13 (100%) 13 (100%) 13 (100%) 9 (69.2%) 7 (53.8%)
11 (10.7%) 54 (52.4%) 74 (71.8%) 10 (9.7%) 2 (1.9%) 1 (1.0%)
,0.001 ,0.001 0.036 ,0.001 ,0.001 ,0.001
3 (23.1%) 3 (23.1%)
0 0
0.001 0.001
IVH: intravenous hydration; IVIG: intravenous immunoglobulin; NS: not significant; PICU: pediatric intensive care unit. independent-sample t-test. b 2 x test. a
disorders.12 In their study, the abnormal chest radiograph findings, anemia, thrombocytopenia, leukopenia, hyponatremia, transaminitis and raised lactate dehydrogenase were all associated with severe adenovirus disease. However, our study showed that complete blood count indices were not helpful in determining the possibility of severe adenovirus ARI. Median LOHS of our study patients was 4 days (range 1–50) while the study in Manitoba reported a median LOHS of 5 days (range 1–189).13 Another study carried out in Taoyuan revealed a mean LOHS of 9.89+9.55 days.24 In a study performed in
6
Massachusetts, USA, median LOHS was 3.5 days (range 1–292).27 This showed that the average LOHS due to adenovirus ARI is similar across geographical regions. There are several limitations in our study. Mild cases with adenovirus associated ARI might not be hospitalized. Nasopharyngeal aspirates for adenovirus DFA are not routinely done for all patients, contributing to potential underdiagnosis. Molecular methods were not available and may lead to underdiagnosis, as PCR methods are more sensitive than DFA.28 Serotyping of adenovirus by PCR method is not available in our hospital laboratory to document temporal
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Demographic data Male Median age in month (range) Clinical features Fever Mean temperature (8C) Cough Rhinorrhoea Tachypnoea Poor feeding Conjunctivitis Diarrhoea Comorbidity Asthma C-section Birth weight (g) Prematurity Gestational age (week) Laboratory findings Leukocyte (109/L) Neutrophil (109/L) Lymphocyte (109/L) Hemoglobin (g/dL) Platelet (109/L) Treatment Oxygen IVH Antibiotic Ribavirin IVIG Transfusion Outcome Mortality Complication
Yes, n¼13
Transactions of the Royal Society of Tropical Medicine and Hygiene
trends and severity of illness. This study adds to the limited data on epidemiology and outcome of adenovirus ARI in developing countries. In conclusion, adenovirus ARI caused variable duration and severity of symptoms among hospitalized Malaysian children. It should be recognized as a significant cause of childhood morbidity in developing countries, with potential impact on hospitalization and resource utilization. Laboratory confirmation by DFA method is inexpensive, offers early diagnosis, limits usage of antibiotics, and potentially reduce cost of hospitalization.
9 Berciaud S, Rayne F, Kassab S et al. Adenovirus infections in Bordeaux University Hospital 2008–2010: clinical and virological features. J Clin Virol 2012;54:302–7. 10 Banerji A, Greenberg D, White LF et al. Risk factors and viruses associated with hospitalization due to lower respiratory tract infections in Canadian Inuit children: a case-control study. Pediatr Infect Dis J 2009;28:697–701. 11 Zhang C, Zhu N, Xie Z et al. Viral etiology and clinical profiles of children with severe acute respiratory infections in China. PLoS One 2013;8:e72606. 12 Tsou TP, Tan BF, Chang HY et al. Community outbreak of adenovirus, Taiwan, 2011. Emerg Infect Dis 2012;18:1825–32.
Acknowledgements: We thank Nurul Zamil Mohd Muzzamil for assisting the authors to obtain research registration from National Medical Research Registry of Malaysia. We also appreciate non-financial support from Director General of Health, Malaysia and Clinical Research Centre (CRC), Tuanku Jaafar Hospital. Funding: None. Competing interests: None declared. Ethical approval: This research was approved by Medical Research and Ethics Committee of Malaysia.
References 1 Ampuero JS, Vıctor Ocana V, Gomez J et al. Adenovirus respiratory tract infections in Peru. PLoS One 2012;7:e46898. 2 Kojaoghlanian T, Flomenberg P, Horwitz MS. The impact of adenovirus infection on the immunocompromised host. Rev Med Virol 2003; 13:155–71. 3 Ison MG. Adenovirus infections in transplant recipients. Clin Infect Dis 2006;43:331–9. 4 Gray GC, McCarthy T, Lebeck MG et al. Genotype prevalence and risk factors for severe clinical adenovirus infection, United States 2004– 2006. Clin Infect Dis 2007;45:1120–31. 5 Tabain I, Ljubin-Sternak S, Cepin-Bogovic´ J et al. Adenovirus respiratory infections in hospitalized children: clinical findings in relation to species and serotypes. Pediatr Infect Dis J 2012;31:680–4. 6 Khor CS, Sam IC, Hooi PS et al. Epidemiology and seasonality of respiratory viral infections in hospitalized children in Kuala Lumpur, Malaysia: a retrospective study of 27 years. BMC Pediatr 2012;12:32.
14 Buehler JW, Finton RJ, Goodman RA et al. Epidemic keratoconjunctivitis: report of an outbreak in an ophthalmology practice and recommendations for prevention. Infect Control 1984;5:390–4. 15 Jernigan JA, Lowry BS, Hayden FG et al. Adenovirus type 8 epidemic keratoconjunctivitis in an eye clinic: risk factors and control. J Infect Dis 1993;167:1307–13. 16 Finn A, Anday E, Talbot GH. An epidemic of adenovirus 7a infection in a neonatal nursery: course, morbidity, and management. Infect Control Hosp Epidemiol 1988;9:398–404. 17 Faden H, Wynn RJ, Campagna L, Ryan RM. Outbreak of adenovirus type 30 in a neonatal intensive care unit. J Pediatr 2005;146:523–7. 18 Leruez-Ville M, Minard V, Lacaille F et al. Real-time blood plasma polymerase chain reaction for management of disseminated adenovirus infection. Clin Infect Dis 2004;38:45–52. 19 Neofytos D, Ojha A, Mookerjee B et al. Treatment of adenovirus disease in stem cell transplant recipients with cidofovir. Biol Blood Marrow Transplant 2007;13:74–81. 20 Shah DP, Ghantoji SS, Mulanovich VE et al. Management of respiratory viral infections in hematopoietic cell transplant recipients. Am J Blood Res 2012;2:203–18. 21 Sabroe I, McHale J, Tait DR et al. Treatment of adenoviral pneumonitis with intravenous ribavirin and immunoglobulin. Thorax 1995;50: 1219–20. 22 Emovon OE, Lin A, Howell DN et al. Refractory adenovirus infection after simultaneous kidney-pancreas transplantation: successful treatment with intravenous ribavirin and pooled human intravenous immunoglobulin. Nephrol Dial Transplant 2003;18:2436–8. 23 Selvaraju SB, Kovac M, Dickson LM et al. Molecular epidemiology and clinical presentation of human adenovirus infections in Kansas City children. J Clin Virol 2011;51:126–31. 24 Chen SP, Huang YC, Chiu CH. Clinical features of radiologically confirmed pneumonia due to adenovirus in children. J Clin Virol 2013;56:7–12. 25 Lai CY, Lee CJ, Lu CY et al. Adenovirus serotype 3 and 7 infection with acute respiratory failure in children in Taiwan, 2010–2011. PLoS One 2013;8:e53614. 26 Callaway Z, Kim SH, Kim JY et al. Adenovirus infection with serious pulmonary sequelae in Korean children. Clin Respir J 2011;5:92–8.
7 Do AH, van Doorn HR, Nghiem MN et al. Viral etiologies of acute respiratory infections among hospitalized Vietnamese children in Ho Chi Minh City, 2004–2008. PLoS One 2011;6:e18176.
27 Gupta P, Sunali Goyal S, Tobias JD et al. Risk factors associated with hospital admission among healthy children with adenovirus infection. Turk J Pediatr 2011;53:597–603.
8 Hoffmann J, Rabezanahary H, Randriamarotia M et al. Viral and atypical bacterial etiology of acute respiratory infections in children under 5 years old living in a rural tropical area of Madagascar. PLoS One. 2012;7:e43666.
28 Stroparo E, Cruz CR, Debur Mdo C et al. Adenovirus respiratory infection: significant increase in diagnosis using PCR comparing with antigen detection and culture methods. Rev Inst Med Trop Sao Paulo 2010; 52:317–21.
7
Downloaded from http://trstmh.oxfordjournals.org/ at Karolinska Institutet on June 2, 2015
Authors’ contributions: KKT conceived the study; KFN designed the study protocol; BHN and PN reviewed the case notes and entered the data; KFN and WYG carried out the data analysis and interpretation of these data. KFN drafted the manuscript; KKT and KFN critically revised the manuscript for intellectual content. All authors read and approved the final manuscript. Guarantor of this paper is KKT.
13 Alharbi S, Caeseele PV, Consunji-Araneta R et al. Epidemiology of severe pediatric adenovirus lower respiratory tract infections in Manitoba, Canada, 1991–2005. BMC Infect Dis 2012;12:55.
