Clinical course of community-acquired respiratory syncytial virus pneumonia in newborns hospitalized in neonatal intensive care unit

(1)

Clinical course of community-acquired respiratory syncytial virus pneumonia in newborns hospitalized in neonatal intensive care unit

Banu AYDIN¹, Ayşegül ZENCİROĞLU¹, Dilek DİLLİ¹, Nurullah OKUMUŞ¹, M. Şah İPEK¹, Mustafa AYDIN¹, Nuran UZUNALIÇ¹, Nilay HAKAN¹, Ahmet Afşin KUNDAK¹, Arzu DURSUN¹, Nilgün KARADAĞ¹, Gülay KORUKLUOĞLU²

1Dr. Sami Ulus Kadın Doğum Çocuk Sağlığı ve Hastalıkları Eğitim ve Araştırma Hastanesi, Yenidoğan Kliniği, Ankara,

2Türkiye Halk Sağlığı Kurumu, Ankara.

ÖZET

Yenidoğan yoğun bakım ünitesine yatırılan yenidoğanlarda toplumda gelişen respiratuar sinsityal virüs pnömonisinin klinik seyri

Giriş:Respiratuar sinsityal virüs (RSV), tüm dünyada süt çocuklarında pnömoninin önemli bir nedenidir. Bu çalışmada, yenidoğan yoğun bakım ünitesine yatırılan yenidoğanlarda toplumda gelişen RSV pnömonisinin klinik seyrinin değerlen- dirilmesi amaçlandı.

Materyal ve Metod:Kasım 2010-Nisan 2011 tarihleri arasında yatan ve pnömoni tanısı konulan tüm yenidoğanlar prospek- tif olarak değerlendirildi. Elli dört nazofarengeal sekresyon örneği eş zamanlı olarak hızlı antijen testi (HAT) ve multipleks polimeraz zincir reaksiyonu (PCR) ile RSV yönünden incelendi. Pnömonili olgularda hastalığın ağırlığını değerlendirmede Downes’s skoru kullanıldı.

Bulgular:PCR esas alındığında, RSV tanısında HAT’ın duyarlılığı %90, özgüllüğü %78.5 idi. RSV pnömonili hastalardan dördünde atriyal septal defekt (ASD) saptandı ve hepsi mekanik ventilasyon desteği gerektirdi. Birinci gün Downes skoru intravenöz sıvı gereksinimi (p= 0.001, r= 0.48), toplam oksijen desteği (p= 0.000, r= 0.63), tekrar beslenmeye başlama sü- resinin uzunluğu (p= 0.001, r= 0.46) ile doğru orantılıydı. Kan pH düzeyi ile Downes skoru negatif ilişkiliydi (p= 0.031, r=

0.46). İkinci gün Downes skoru ASD’li hastalarda ASD olmayanlara göre daha yüksekti (3.8 ± 2.6 vs. 2 ± 1.1, p= 0.01).

Hastanede yatış süresini etkileyen en önemli faktör ikinci gün Downes skoru idi (p= 0.02, OR: 1.9, %95 GA (1.1-3.2)]. Tüm hastalar sağlıklı olarak taburcu edildi.

Sonuç:RSV tanısında HAT duyarlı ve özgül bir testtir. Hekimler, RSV pnömonili olgularda hastalığın ağırlığını değerlendir- mede Downes skorunu kullanabilir. Bu olgularda ASD daha ağır seyretmektedir.

Anahtar Kelimeler: Respiratuar sinsityal virüs, pnömoni, yenidoğan.

Yazışma Adresi (Address for Correspondence):

Dr. Dilek DİLLİ, Dr. Sami Ulus Kadın Doğum Çocuk Sağlığı ve Hastalıkları Eğitim ve Araştırma Hastanesi, Yenidoğan Kliniği, Altındağ, ANKARA - TURKEY

e-mail: dilekdilli2@yahoo.com

(2)

INTRODUCTION

Respiratory syncytial virus (RSV) is a major cause of pneumonia in infants and young children worldwide (1- 3). The clinical differentiation of bacterial from viral (RSV) pneumonia is very difficult. Therefore, a rapid and reliable diagnosis is essential for further clinical management, e.g., isolation or cohortation measures, with- holding of antibiotic treatment. Laboratory diagnosis by cell culture and viral serology is usually necessary to identify the etiologic agent, although the final results of RSV isolation by tissue culture usually require several days. Rapid antigen test (RAT) results are available in 15 minutes (4). Multiplex real time polymerase chain reaction (RT-PCR) is a highly sensitive method for diagnosis of viral infection and has been used successfully in children with RSV. The multiplex RT-PCR has a sensitivity for RSV of 90% and a specificity of 99.2% using virus culture as the gold standard (5,6). Because of the

higher costs and longer turnaround time of the molecular tests, as compared to the antigen- detection tests, some laboratories use antigen detection to screen inco- ming clinical samples and only perform a more sensitive molecular test on the negative samples (7). Although there are many studies investigating predictive values of RAT for RSV in children and adults (8-10), there are limited data on this issue in newborns (11,12). Further- more, to our 3 knowledge there is no study using Dow- nes’ score in assessment of disease severity in newborns with RSV pneumonia. Therefore, in this study, we aimed to define predictive values of RAT for RSV in newborns. The factors affecting the clinical course of community- acquired RSV pneumonia in hospitalized newborns were also evaluated.

MATERIALS and METHODS

Study was conducted on patients attending outpatient clinics of Department of Neonatology and hospitalized SUMMARY

Clinical course of community-acquired respiratory syncytial virus pneumonia in newborns hospitalized in neonatal intensive care unit

Banu AYDIN¹, Ayşegül ZENCİROĞLU¹, Dilek DİLLİ¹, Nurullah OKUMUŞ¹, M. Şah İPEK¹, Mustafa AYDIN¹, Nuran UZUNALIÇ¹, Nilay HAKAN¹, Ahmet Afşin KUNDAK¹, Arzu DURSUN¹, Nilgün KARADAĞ¹, Gülay KORUKLUOĞLU²

1Clinic of Neonatology, Dr. Sami Ulus Maternity and Children Health and Diseases Training and Research Hospital, Ankara, Turkey,

2Turkey Public Health Institute, Ankara, Turkey.

Introduction:Respiratory syncytial virus (RSV) is a major cause of pneumonia in infants worldwide. We aimed to evalu- ate the clinical course of community-acquired RSV pneumonia in newborns hospitalized in neonatal intensive care unit.

Materials and Methods:All the newborns diagnosed as pneumonia were prospectively evaluated for RSV infection betwe- en November 2010 and April 2011. Fifty-four specimens of nasopharyngeal secretions were tested in parallel with the RAT and the multiplex real time polymerase chain reaction (RT- PCR). Downes’ score was used to assess the disease severity in patients with pneumonia.

Results:RAT has a sensitivity of 90% and a specificity of 78.5%, as the PCR technique target assay. Four of the patients with RSV pneumonia had secundum atrial septal defect (ASD) and all of four needed mechanical ventilation support. The first day Downes’ score was positively correlated with time of intravenous fluid requirement (p= 0.001, r= 0.48), total oxygen need (p= 0.000, r= 0.63), and re-enteral feeding (p= 0.001, r= 0.46). Blood pH (p= 0.031, r= 0.46) were negatively correla- ted with Downes’ score. The second day Downes’ score was higher in patients with ASD than those of without ASD (3.8

± 2.6 vs. 2 ± 1.1, p= 0.01). The most possible risk factor for longer hospital stay was the higher second day Downes’ score (p= 0.02 OR: 1.9, CI 95% (1.1-3.2). All infants were discharged from hospital in a good health.

Conclusion:RAT is sensitive and specific in detecting RSV infections in newborns. Physicians may use Downes’ score for evaluation of disease severity in infants with RSV pneumonia. In these patients, ASD has increased the disease severity.

Key Words: Respiratory syncytial virus, pneumonia, clinical course, newborn.

Tuberk Toraks 2013; 61(3): 235-244 • doi: 10.5578/tt.4598

(3)

in Neonatal Intensive Care Unit (NICU) of Dr. Sami Ulus Maternity and Children Training and Research Hospital, Ankara, Turkey from November 2010 to Ap- ril 2011. The study was approved by Local Ethics Committee.

Inclusion and Exclusion Criteria

The newborns younger than 30 days and attended to the outpatient clinics with complaints of respiratory symptoms were assessed by the physicians and hospitalized in our NICU if required. During the study period, a total of 54 newborns hospitalized in NICU with signs and symptoms of pneumonia were included in the study. Exclusion criteria were presence of severe congenital malformations, intrauterine infections, and parents unwilling to give informed consent.

Demographic and Clinical Characteristics

Detailed history was taken from the parents of the newborns with sign and symptoms of pneumonia. Demog- raphic and clinical characteristics were recorded to previously prepared study forms. Diagnosis of pneumonia was made using standard clinical, laboratory and radiological criteria. The clinical findings were apnea, tachypnea, poor feeding, abdominal distention, ja- undice, emesis, respiratory distress, and circulatory collapse. Ventilator- dependent infants might have increased oxygen and ventilator requirements or purulent tracheal secretions. Blood white blood cell count, serum C-reactive protein (CRP) values, blood culture, Gram staining and culture of tracheal aspirates could help diagnosis. Bilateral alveolar densities with air bronchograms were defined as characteristic in pneumonia. Bronchiolitis was considered if there were mild symptoms of an upper respiratory tract infection at the beginning of the disease and then it progressed to cough, wheeze, onset of dyspnea, increased respiratory rate and retractions of the respiratory muscles, indica- ting lower respiratory tract involvement. Although there were no definite criteria, clues to a diagnosis of RSV pneumonia might be the finding of crackles, respiratory distress and dullness to percussion (13-15). At the first and second days of the admission, the severity of the respiratory distress was assessed by Downes' sco-

re which is a comprehensive scoring system and can be applied to any gestational age and condition (Table 1) (16). A score of > 6 is indicative of impending respiratory failure. Appropriate supportive treatment including oxygen, fluid, resuscitation and inotropic agents were administered. Antibiotic and/or antiviral medica- tions were started according to advice of in charge con- sultant neonatologist. Even though the use of antibiotics for all infants with bronchiolitis is not thought to be helpful, their use in a selected subset who may also have bacterial pneumonia may be indicated. A prog- ressively increasing FiO₂requirement to maintain a saturation of 90-92% in a preterm and 94-96% in a term baby is a sensitive indicator of the severity and progress of distress. We used clinical deterioration, blood culture and/or tracheal aspirate positivity in combinati- on with chest radiograph, temperature abnormality, high blood leucocyte count and serum CRP values to make decisions for treatment with antibiotics (17). Co- agulase negative staphylococcus generally were considered to be contaminants, except when two positive blood cultures and/or evidence of a true infection based on clinical course and laboratory data existed.

The variables which may affect the clinical course were recorded.

Laboratory Evaluation

Whole blood count, arterial blood gases, CRP, and blood cultures were taken from all subjects. Pulmonary X- ray was performed. All the patients were evaluated for congenital heart diseases by echocardiography. Speci- men collection A total of 54 nasopharyngeal aspirates from newborns with clinical signs of pneumonia were taken. The specimens were taken within the first 72 hours after admission. Nasopharyngeal aspirates from both nostrils were combined in a standardized manner in NaCl 0.9% without any additives by using a sterile suction trap. The secretions were immediately trans- ported to the laboratory and either analyzed immediately (RAT, during working hours) or stored at +4°C for a maximum of 24 h until further processing (RAT, after working hours). According to the manufacturer’s instructions, specimens (nasal washes and aspirates) can

Table 1. Downes’ score.

Score Respiratory rate Cyanosis Air entry Grunt Retraction

0 < 60/minutes Nil Normal None Nil

1 60-80/minutes In room air Mild? Ausc with stethoscope Mild

2 > 80/minutes In ≥ 40% Marked? Audible with naked ear Moderate

(4)

be stored up to 24 hours at +2-8°C. Samples for PCR testing were stored at +4°C and stored for up to 5 days.

If a longer storage time before sending was foreseeab- le, samples were frozen at -70°C. Specimens were tested by a RAT test for RSV in our NICU and in parallel by multiplex RT-PCR in Refik Saydam National Hygiene Center. PCR technique was used as the target assay.

Rapid Antigen Test

The RAT was performed following the manufacturer’s instructions [Respi Strip (Coris BioConcept, Gembloux, Belgium)]. A nasopharyngeal swab was obtained and placed in a tube containing saline solution and agita- ted. An aliquot was then mixed with extraction buffer and stirred thoroughly. The strips were then inserted in- to the tube and incubated for 15 minutes before re- ading. The presence of a positive control line with a positive test line was considered a positive result. All tests were performed by the primary investigator (B.A).

Each box of Respi-Strip was checked with a positive control included in the box before being used in patients. The rapid test was evaluated before knowing the PCR result.

Multiplex RT-PCR ELISA

Respiratory system viruses including RSV A and B, rhinovirus, influenza A and B, coronavirus, H1N1, human metapneumovirus, parainfluenza, and adenovirus were investigated by multiplex RT-PCR. For multiplex RT- PCR, the oligonucleotide primers were designed to amplify the nucleoprotein (N) and phosphoprotein (P) genes of RSV, because they are highly conserved and are regions of the RSV genome which allow subgro- uping of RSV strains into A and B types (14). The RT- PCR was performed with “in-house” RT-PCR protocol provided by Center for Disease Control (CDC, USA), using 2x buffer and SuperScript III Platinum^®One-Step Quantitative RT-PCR System (Invitrogen; CA, USA) with Stratagene Mx3005P (Strategene; California, USA) RT-PCR machine. The 25 mL PCR 22 mixture contained 5 mL of extracted RNA, 1 mL each of for- ward and reverse primers, 1 mL probe, 0.5 mL Su-

perScript III RT/Platinium Taq mix, 12.5 mL of 2X Mas- ter mix, and 4 mL nuclease-free water. RT-PCR ampli- fication conditions were as follows; reverse transcription at 50°C for 30 minutes, Taq inhibitor activation 95°C for 2 minutes and 45 cycles at 95°C for 15 seconds, 55°C for 30 seconds.

Statistical Analyses

SPSS 16.0 (SPSS, Chicago ILL, USA) was used for statistical analysis. Kolmogorov Smirnov test was used to analyze the distribution of data. Data were expressed as the arithmetic mean ± standard deviation. Differen- ces between two groups were tested using Mann Whit- ney U test. Chi-square tests were performed for cate- gorical variables. Spearman test was used to analyze correlation between variables. Sensitivity, specificity, and positive and negative predictive values of RAT for RSV was calculated. Positive and negative likelihood ratios were also given. Factors affecting the longer hospital stay in RSV pneumonia were investigated by multivariate logistic regression analysis. The level of signi- ficance was set at 5% in all comparisons.

RESULTS

A total of 44 newborns with community-acquired pneumonia were included the study. All the 15 patients were healthy previously.

Diagnostic Results for RAT and Multiplex RT-PCR Forty-four specimens were examined by both RAT for RSV and multiplex RT-PCR. Thirty (68.2%) samples were positive for RAT, 27 (90%, 27/30) samples of which were positive for PCR testing. By RAT for RSV 14 (31.8%) samples were negative. Eleven (78.6%, 11/14) of these specimens were negative for multiplex RT-PCR. According to the study results, RAT has a sensitivity of 90% and a specificity of 78.5%. Positive and negative predictivity were calculated to be 90% and 78.6%, respectively. Positive likelihood ratio was 4.09 with a low negative likelihood ratio (0.12) (Table 2).

During the study period, nosocomial pneumonia developed in five of the patients who were hospitalized in NI-

Table 2. RAT and multiplex RT-PCR results of the patients.

Multiplex RT-PCR positive Multiplex RT-PCR negative Total

RSV A/B positive 27 (RSV A/B + influenza A + rhinovirus; 3 30

n= 2 and RSV A/B + rhinovirus positive; n= 1)

RSV A/B negative 3 (RSV A/B + influenza A + rhinovirus; 11 (rhinovirus, n= 1, influenza A; 14 n= 1, RSV A/B + adenovirus; n= 1) n= 1, and SWH1N1; n= 2)

Total 30 14

(5)

CU. They were evaluated by RAT and test was positive in two of them. These patients were not included in analysis. Influenza A and B, rhinovirus, adenovirus, SWH1N1 were the other agents detected by multiplex RT-PCR (Table 2). Regarding the results the highest RSV incidence (n= 18) was on January.

Comparison of Multiplex RT-PCR RSV Positive Patients According to Downes’ Score

Further analyses were performed on 30 patients who- se RSV positivity was confirmed by multiplex RT-PCR and their data were compared according to Downes’

score. There was no significant difference between the two groups according to demographic characteristics except gestational age (Table 3). Demographic and clinical characteristics of the multiplex RT-PCR positive patients according to Downes’ score are seen on Table 3 and 4. Gestational age was significantly lower among patients with high Downes’ score. All patients had cough and rales. The rates of retraction, roncus, and cardiac murmur were higher in patients with high Downes’

score. On admission, leucocyte (9500 ± 2769/mm³vs.

8116 ± 1544) and CRP values (5.7 ± 8.8 mg/L) 17 vs.

6.3 ± 8.8 mg/L) were similar in infants with low and high Downes’s score, respectively 18 (p> 0.05 and p>

0.05). In all RSV positive patients, the common radiological findings were hyperinflation (n= 16, 53.3%), infiltration (n= 12, 40%), and atelectasis (n= 2, 6.7%). Co- agulase negative staphylococci was grew in 4 (13.3%) patients, but it was considered to be contamination as there were no additional data supporting blood stream infection, Cupriavidus pauculus (n= 1, 3.3 %) and Esc-

herichia coli (n= 1, 3.3%) grew in blood cultures of RSV positive infants. Streptococcus viridians grew in the tracheal aspirate of 1 (3.3%) patient with high Dow- nes’s score.

Among RSV positive patients, antibiotic treatment was started in 18 with severe clinical findings, high CRP levels, and positive blood cultures. Only supportive treatment was given to twelve of the patients. Oseltamivir was started in four patients, who needed mechanical ventilation, until multiplex RT-PCR results were obtained.

Clinical Course of Multiplex RT-PCR RSV Positive Patients According to Downes’ Score

There were significant differences between the infants with low and high Downes’ scores according to intravenous fluid requirement, time of tachpnea, time to re- enteral feeding, and total day of oxygen support.

Length of hospital stay was also longer in infants with high Downes’ score, although not significantly (Table 5). The first and second day Downes’ scores were higher in infants requiring mechanical ventilation 5.5 ± 1.2 vs. 2.8 ± 1.5, p= 0.007 and 5.0 ± 2.4 vs. 2.0 ± 1.3, p=

0.01, respectively). The duration of hospital stay was negatively correlated with gestational age (p= 0.03, r=

0.32). However, duration of tachypnea (p= 0.02, r=

0.34), IV fluid requirement (p= 0.001, r= 0.47), 14 total oxygen support (p= 0.01, r= 0.38), and second day Downes’ score (p= 0.03, r= 0.31) were 15 positively correlated with hospital stay. Possible factors affecting longer hospital stay (> 10 days) including gestational

Table 3. Demographic and clinical characteristics of the multiplex RT-PCR RSV positive patients according to Downes’ score.

Downes’ score ≤ 6 Downes’ score > 6

(n= 24) (n= 6) p

Gender (male) (n, %) 12 (50) 3 (50) 1.0

Birthweight (mean ± SD) 3720 ± 519 3248 ± 822 0.74

Gestational age (week) (mean ± SD) 38.7 ± 1.1 37.1 ± 1.4 0.02

Prematurity (37 week of gestation) (n, %) 3 (12.5) 2 (33.3) 0.25

Delivery type (cesarean) (n, %) 11 (45.8) 2 (33.3) 0.67

History of upper respiratory system infection in-family (n, %) 20 (83.3) 4 (66.7) 0.57

Passive smoking (n, %) 9 (37.5) 2 (33.3) 0.85

Feeding type (n, %)

Formula 7 (29.2) 2 (33.3) 0.84

Breastfeeding 17 (70.8) 4 (66.7)

Family size (number of persons at home) (mean ± SD) 5 ± 2 5 ± 1 0.78

Age at admission (day, mean ± SD) 19.2 ± 6.7 20.0 ± 5.3 0.11

(6)

week, present of ASD, second day Downes’ score were evaluated by multivariate analysis. It was found that longer hospital stay was related to higher second day

Downes’ score (p= 0.02, OR: 1.9, CI 95% (1.1-3.2).

The first day Downes’ score was positively correlated with time of intravenous fluid requirement (p= 0.001, Table 5. Clinical course of the multiplex RT-PCR RSV positive patients according to Downes’ score.

(n= 24) (n= 6) p

IV fluid requirement (day) (mean ± SD) 1.4 ± 1.0 5.8 ± 4.5 0.01

Tachypnea (day) 1.6 ± 1.5 6.1 ± 4.1 0.001

Time of re-enteral feeding (day) 1.8 ± 0.7 6 ± 4.3 0.004

Hospital stay (day) 9.6 ± 3.8 13.0 ± 5.0 0.11

Need of oxygene treatment

Free oxygen (n, %) 14 (58.3) 6 (100) 0.07

Nasal CPAP (n, %) 0 1 (33.3) 0.03

Mechanical ventilation (n, %) 0 3 (50.0) 0.001

Total day of oxygen support (mean ± SD) 1.8 ± 2.0 6.1 ± 3.6 0.003

Table 4. Clinical findings, vital parameters, physical findings of the multiplex RT-PCR RSV positive patients according to Downes’s score.

(n= 24) (n= 6) p

Clinical findings (n, %)

Cough 24 (100) 6 (100) NA

Nasal congestion 23 (95.8) 5 (83.3) 0.36

Grunting 18 (75.0) 6 (100) 0.30

Vomiting 11 (45.8) 4 (66.7) 0.65

Nasal discharge 11 (45.8) 3 (50.0) 0.10

Cyanosis 10 (41.7) 3 (50.0) 0.72

Respiratory distress 6 (25.0) 3 (20.0) 0.32

Fever 5 (20.8) 2 (33.3) 0.60

Apnea 5 (20.8) 2 (33.3) 0.60

Diarrhea 5 (20.8) 0 NA

Vital parameters

Fever (°C) (mean ± SD) 36.7 ± 0.5 36.8 ± 0.5 0.70

Heart rate (per min) 157 ± 17 155 ± 22 0.86

Respiratory rate (per min) 63 ± 5.5 68 ± 7.4 0.23

Tension arterial (systolic-mmHg) 86 ± 12 92 ± 21.4 0.70

Oxygen saturation (SpO₂, %) 92.1 ± 4.6 91 ± 5.4 0.98

Physical findings

Retraction 8 (33.3) 5 (83.3) 0.02

Rales 24 (100) 6 (100) NA

Roncus 3 (12.5) 2 (33.3) 0.03

Cardiac murmur 3 (12.5) 5 (83.3) 0.002

Hepatomegaly 2 (8.3) 1 (16.7) 0.54

(7)

r= 0.48), total oxygen need (p= 0.000, r= 0.63), and re-enteral feeding (p= 0.001, r= 0.46). Blood pH (p=

0.031, r= 0.46) were negatively correlated with Dow- nes’ score. There was no correlation between gestational age and Downes’ score.

Echocardiography Results of Multiplex RT-PCR RSV Positive Patients

Patent ductus arteriosus (PDA) (n= 1, 3.3%), patent fo- ramen ovale (PFO) (n= 13, 43.3%), secundum atrial septal defect (ASD) (n= 7, 23.3), and pulmonary hypertension (PH) (n= 1, 3.3%) were noted. Four of the patients with secundum ASD (13.3%) needed mechanical ventilation support. It was noted that clinical course of RSV pneumonia was worse in patients with ASD; with much more need of oxygen support (4.7 ± 4.6 days vs. 1.9 ± 2.0, p= 0.02), intravenous fluid treatment (4.8 ± 4.7 days vs. 1.5 ± 1 days), p= 0.004, and longer hospital stay (14 ± 4.6 days vs. 9.2 ± 3.4, p=

0.007). The second day Downes’ score was higher in patients with ASD than those of without ASD (3.8 ± 2.6 vs. 2 ± 1.1, p= 0.01).

Complications of Multiplex RT-PCR RSV Positive Patients

When evaluated for encountered complications during follow-up in NICU, seconder bacterial pneumonia developed in 2 (6.6%) patients. Sinus tachycardia was detected in another 1 (4.3%) patient at the 5^thof admission; there was no fever or finding supporting myocarditis. Apneic convulsion was observed in 1 (3.3%) infant; any metabolic abnormalities were not detected and cranial imaging studies were normal. Diarrhea developed in 3 (10.0%) infants; Rotavirus was shown in 2 (6.7%) of them. Conjunctivitis was observed in 2 (6.7%) patients; conjunctival cultures were negative.

All the infants discharged from the hospital with a good health. No problems were observed in any of the patients at 6^th month visit. The patients with ASD have been followed-up by department of pediatric cardi- ology.

DISCUSSION

We found that RAT showed a high sensitivity and thus proved to be satisfactory to exclude RSV etiology in a newborn with pneumonia confirms a RSV diagnosis.

Overall specificity, however, was nearly high, making it acceptable to confirm a RSV diagnosis. In a recent study performed on 3691 newborn subjects, it was de- monstrated that RAT has high sensitivity (90.3%) and specificity (88.2%) for the diagnosis of RSV respiratory tract disease (11). In this study, we used the rapid IC test and showed that the sensitivity and specificity of

RAT were 90% and 78.5%, respectively. We think that rapid confirmation or elimination of RSV would facilita- te prompt and adequate measures to restrict transmis- sion of the virus in a NICU containing high-risk infants.

It was suggested that because of the limited antibody response and reduced concentration of passively acquired maternal antibodies in neonates, and having smal- ler airways preterm neonates were highly susceptible to RSV (18). During the study period, prematurity rate was 16.6% among infants with community-acquired RSV pneumonia. It is not clear how breast-feeding re- duces the risk of RSV infection. Although recent immu- nological studies have tried to define the mucosal and/or systemic mechanisms of protection against respiratory infections during the early months of life, more studies are required to identify which elements mo- dify the evolution of disease (19). Interestingly, 70% of our patients with RSV pneumonia were breastfed. Co- ugh, rhinorrhea and fever were reported to be the most common symptoms, and respiratory distress and apnea were serious complications frequently seen in young infants caused by RSV bronchiolitis and pneumonia (20). In our study, cough, nasal congestion, grunting, vomiting, nasal discharge and cyanosis were the major clinical findings, while hyperinflation, atelectasis, and infiltration were the most common radiological findings observed in the RSV infected infants, consistent with previous data. It was reported that lower respiratory distress with wheezing caused by RSV was com- monly found in children less than two years old (21).

Rales were heard in all of our study patients with RSV pneumonia. It can be said that RSV infection should be suspected in the presence of such symptoms during the RSV season. Premature infants in their first six months, children with underlying cardiac or pulmonary disease in their first two years, immune-compromised children (particularly transplant patients), and healthy infants younger than six weeks of age are at the highest risk of severe RSV infection. These children are likely to have prolonged hospital stays, and are more likely to require admission to the NICU and to need mechanical ventilation. The absence of a vaccine to prevent RSV infection narrows preventive measures to a combinati- on of public health advice and the use of passive im- munization with palivizumab, the only licensed product available for prevention of RSV lower respiratory tract disease in high-risk infants and children during the RSV season (22). In our practice, palivizumab prophylaxis has been performed on high-risk infants during the winter season. According to recommendations of Ame- rican Academy of Pediatrics (AAP), congenital heart diseases requiring palivizumab prophylaxis include

(8)

cyanotic heart diseases, moderate to severe pulmonary hypertension, and congestive heart failure requiring medication (19). In the current study, all the patients were evaluated by echocardiography and secundum ASD was shown in seven of them. The four had RSV infection and required mechanical ventilation. Pul- monary hypertension and cardio-pulmonary insufficiency developed in these infants. We think that ASD might negatively affect the course of the disease with increased viral load in lungs due to increased pulmonary blood flow. Although it may be difficult to say palivizumab prophylaxis in infants with ASD because to small number of the cases, but it might be recommen- ded that the patients with severe RSV infection should be evaluated by echocardiography. Previous studies have shown that bacteriemia or serious bacterial infection was present in some children admitted with RSV infection (23). Blood cultures were considered to be positive in 6.6% of our patients with RSV infection. Ext- rapulmonary manifestations are common in children with severe RSV infection, suggesting RSV may infect organs other than the lung. Life-threatening extrapulmonary manifestations of RSV infection include central apneas, seizures, status epilepticus, ventricular tachycardia and fibrillation, heart block and pericardial tam- ponade and can be detected by adequate monitoring (24). Although underlying metabolic abnormalities such as hypoglycemia, and/or hyponatremia may cause seizure, the pathophysiological mechanism of RSV- related seizures remains unclear (25). Detection of RSV in the cerebrospinal fluid has also supported a direct in- vasion of the central nervous system in RSV disease.

Apnea developed in 23.3% of our patients. One had apneic convulsion and she had no any metabolic or int- racranial abnormalities. Presumably, the arrhythmias may be secondary to RSV pericarditis, myocarditis, or some other infection of the heart. The possibility that the cardiac arrhythmias may have been caused by one of the cardiotropic viruses infecting the child at the sa- me time as RSV has also been reported (26). In our study, sinus tachycardia not related to myocarditis, cardiac insufficiency or anemia developed in two patients. The other extra pulmonary manifestations of RSV infection include hyponatremia, hepatitis hypothermia, rash, thrombocytopenia and conjunctivitis. Conjuncti- vitis was found in two of our patients and there was no bacterial grew in conjunctival cultures. Severe RSV infections resulting in acute respiratory distress syndrome with a mortality rate of 40-70% have been reported in the literature (27). The statistical significant associ- ation between RSV infection and days of hospitalization and oxygen therapy confirms that this infection is cause of severity illness and that prompt recognition of

the diagnosis that include early treatment with oxygen therapy helped to improve the clinical state of the children. The severity of respiratory distress may be assessed by Silverman-Anderson Score and Downes’

Score (16,28). While the Silverman Anderson Retracti- on Score is more suited for preterms with RDS, the Downes’ score is more comprehensive and can be applied to any gestational age and condition. Scoring should be done at half hourly intervals and a chart mainta- ined to determine progress. In a study, Downes’ score

> 7 or FiO₂> 50% after 15-20 minutes of CPAP were found to be predictive of CPAP failure (29). Pillai et al.

defined persistent/worsening respiratory distress if Sil- verman score ≥ 4 addition to other criteria (30). In our study, objective assessment of disease severity was made using Downes’ score. Downes’ scores of the patients were higher among patients requiring mechanical ventilation. The length of hospital stay was longer in infants with higher Downes’ score and having longer oxygen support. We speculate that Downes’ score is prognostic in evaluation and follow-up of patients with RSV pneumonia. The treatment and care for RSV de- pend on the assessment of the severity of respiratory compromise and of respiratory exhaustion with decre- ased respiratory effort. In addition, the consequences of respiratory compromise, in particular the inability to feed and drink, determine the management of these infants. Additional risk factors and preexisting illness (e.g., chronic lung disease of prematurity or history of reactive-airway disease) also influence disease management. Among RSV positive patients, antibiotic treatment was started in 18 with severe clinical findings, high CRP levels, and positive blood cultures. Ribavirin is an antiviral drug that is very effective against RSV in vitro and is licensed for use by inhalation for severe RSV bronchiolitis. However, due to teratogenic side effects, ribavirin cannot be administered as an aerosol in the presence of pregnant women (e.g., medical staff).

A further problem with antiviral drugs is the timing of application; for example, anti-influenza virus drugs need to be applied during the first 48 hours of illness to be effective (31). A Cochrane review found that published reports of trials of ribavirin lack the power to provide reliable estimates of its effects but suggested that ribavirin may reduce the duration of mechanical ventilation and hospitalization (32). RSV is spread by infected respiratory secretions (33). To control the spread of RSV in NICU during a community outbreak of RSV infection isolation of all infants with respiratory disease is generally impossible. Utilizing six infection control procedures (isolation of infected infants, handwashing between infants, change of gowns between caring for infants, cohorting of staff to infants, isolation of high risk in-

(9)

fants, limitations of visitors) would reduce nosocomial infection rate. During the RSV season, we implemented control procedures including isolation of the positive cases, contact precautions, limitation of visitors and closure to elective admissions. These measures were effective and an outbreak did not develop in our NICU.

RSV was detected in only two patients among hospitalized infants.

To our knowledge, this is the first study using Downes’

score in evaluating newborns with RSV pneumonia. Ho- wever, our study has a limitation; this study reports the results of only one tertiary care medical center in Tur- key. A comprehensive study including local clinics, regional hospitals and medical centers would provide more details of RSV infections in Turkey. As conclusion, RAT is sensitive and specific in detecting RSV infections in newborns. Physicians may use Downes’ score for evaluation of disease severity in infants with RSV pneumonia and subsequently target preventive and monitoring strategies toward newborns at high risk. ASD may increase the disease severity in RSV pneumonia.

CONFLICT of INTEREST None declared.

REFERENCES

1. Simoes EA. Respiratory syncytial virus infection. Lancet 1999;

354: 847-52.

2. Hall CB, Walsh EE, Long CE, Schnabel KC. Immunity to and frequency of reinfection with respiratory syncytial virus. J In- fect Dis 1991; 163: 693-8.

3. Hall CB. Respiratory syncytial virus and parainfluenza virus.

N Engl J Med 2001; 344: 1917-28.

4. Kelly JH. Advances in the laboratory diagnosis of viral respi- ratory disease. Pediatr Infect Dis J 2004; 23: 6-10.

5. Weinberg GA, Erdman DD, Edwards KM, Hall CB, Walker FJ, Griffin MR, et al. Superiority of reverse-transcription polymera- se chain reaction to conventional viral culture in the diagnosis of acute respiratory tract infections in children. J Infect Dis 2004; 189: 706-10.

6. Puppe W, Weigl JAI, Aron G, Gröndahl B, Schmitt HJ, Niesters HGM, et al. Evaluation of a multiplex reverse transcriptase PCR ELISA for the detection of nine respiratory tract patho- gens. J Clin Virol 2004; 30: 165-74.

7. Gruteke P, Glas A, Dierdorp M, Vreede W, Pilon JW, Bruisten S.

Practical implementation of a multiplex PCR for acute respira- tory tract infections in children. J Clin Microbiol 2004; 42:

5596-603.

8. Kawai N, Ikematsu H, Iwaki N, Kawashima T, Maeda T, Hirot- su N, et al. Detection of respiratory syncytial virus with nes-

ted RT-PCR and a new rapid detection test kit in patients with influenza-like illness, including elderly adults. Kansenshoga- ku Zasshi 2008; 82: 1-5.

9. Ginocchio CC, Swierkosz E, McAdam AJ, Marcon M, Storch GA, Valsamakis A, et al. Multicenter study of clinical perfor- mance of the 3M Rapid Detection RSV test. J Clin Microbiol 2010; 48: 2337-43.

10. Liao RS, Tomalty LL, Majury A, Zoutman DE. Comparison of viral isolation and multiplex real-time reverse transcription- PCR for confirmation of respiratory syncytial virus and influ- enza virus detection by antigen immunoassays. J Clin Micro- biol 2009; 47: 527-32.

11. Yen AB, Harrison G. Rapid antigen testing to detect respira- tory syncytial virus performs well in neonates. Pediatr Infect Dis J 2011; 30: 234-37.

12. Savic N, Jankovic B, Minic P, Vasiljevic Z, Sovtic A, Pejic K, et al. Clinical characteristics of respiratory syncytial virus infec- tion in neonates and young infants. Vojnosanit Pregl 2011; 68:

220-4.

13. Katayama Y, Minami H, Enomoto M, Takano T, Hayashi S, Lee YK. Usefulness of Gram 5 staining of tracheal aspirates in ini- tial therapy for ventilator-associated pneumonia in 6 extre- mely preterm neonates. J Perinatol 2010; 30: 270.

14. Cleveland RH. A radiologic update on medical diseases of the newborn chest. Pediatr Radiol 1995; 25: 631.

15. Fete TJ, Noyes B. Common (but not always considered) viral infections of the lower 10 respiratory tract. Pediatr Ann 1996;

25: 577-84.

16. Wood DW, Downes’ JJ, Locks HI. A clinical score for the diag- nosis of respiratory failure. Am J Dis Child 1972; 123: 227-9.

17. Spurling GK, Fonseka K, Doust J, Del Mar C. Antibiotics for bronchiolitis in children. Cochrane Database Syst Rev 2007;

(1): CD005189.

18. de Sierra TM, Kumar ML, Wasser TE, Murphy BR, Subbarao EK. Respiratory syncytial virus-specific immunoglobulins in preterm infants. J Pediatr 1993; 122: 787-91.

19. Bryan DL, Hart PH, Forsyth KD, Gibson RA. Immunomodula- tory constituents of human milk change in response to infant bronchiolitis. Pediatr Allergy Immunol 2007; 18: 495-502.

20. Forster J, Schumacher RF. The clinical picture presented by premature neonates infected with the respiratory syncytial vi- rus. Eur J Pediatr 1995; 154: 901-5.

21. Rattanadilok NA, Bhuket T, Sunakorn P, Suwanjutha S, Na- wanoparatkul S, Teeyapaiboonsilpa P. Wheezing-associated lower respiratory infections in under 5-year-old children:

study in Takhli District Hospital. J Med Assoc Thai 2002; 85:

1247-51.

22. Bocchini JA Jr, Bernstein HH, Bradley JS, Brady MT, Bying- ton CL, Fisher MC, et al. From the American Academy of Pedi- atrics: policy statements-modified recommendations for use of palivizumab for prevention of respiratory syncytial virus infec- tions. Pediatrics 2009; 124: 1694-701.

23. Levine DA, Platt SL, Dayan PS, Macias CG, Zorc JJ, Krief W, et al. Risk of serious bacterial infection in young febrile infants with respiratory syncytial virus infections. Pediatrics 2004;

113: 1728.

(10)

24. Eisenhut M. Extrapulmonary manifestations of severe respira- tory syncytial virus infection a systematic review. Crit Care 2006; 10: R107.

25. Sweetman LL, Ng YT, Butler IJ, Bodensteiner JB. Neurologic complications associated with respiratory syncytial virus. Pe- diatr Neurol 2005; 32: 307-10.

26. Huang M, Bigos D, Levine M. Ventricular arrhythmia associ- ated with respiratory syncytial viral infection. Pediatr Cardiol 1998; 19: 498-500.

27. Hammer J, Numa A, Newth CJ. Acute respiratory distress syndrome caused by respiratory syncytial virus. Pediatr Pul- monol 1997; 23: 176-83.

28. Silverman WC, Anderson DH. Controlled clinical trial on ef- fects of water mist on obstructive respiratory signs, death rate and necropsy findings among premature infants. Pediatrics 1956; 17: 1-4.

29. Koti J, Murki S, Gaddam P, Reddy A, Reddy MD. Bubble CPAP for respiratory distress 20 syndrome in preterm infants.

Indian Pediatr 2010; 47: 139-43.

30. Pillai MS, Sankar MJ, Mani K, Agarwal R, Paul VK, Deorari AK. Clinical prediction 22 score for nasal CPAP failure in pre- term VLBW neonates with early onset respiratory 23 distress.

J Trop Pediatr 2011; 57: 274-9.

31. Zenciroglu A, Kundak AA, Aydin M, Okumus N, Dursun A, Ipek MS, et al. Swine influenza A (H1N1) virus infection in in- fants. Eur J Pediatr 2011; 170: 333-8.

32. Ventre K, Randolph AG. Ribavirin for respiratory syncytial vi- rus infection of the lower respiratory tract in infants and yo- ung children. Cochrane Database Syst Rev 2007: CD000181.

33. Bracht M, Heffer M, O’Brien K. Development, implementation, and evaluation of a community- and hospital-based respira- tory syncytial virus prophylaxis program. Adv Neonatal Care 2005; 5: 39-49.