Original Articles Indian Pediatrics 2002; 39:529-537 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Respiratory Distress in Neonates with Special Reference to Pneumonia |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESPIRATORY distress is one of the commonest disorder encountered within the first 48-72 hours of life. It occurs in 0.96 to 12%(1-5) of live births and is responsible for about 20% of neonatal mortality(4). Respiratory pathology is the commonest (32-54%) autopsy finding among early neonatal deaths(6-8). The spectrum of respiratory distress in neonates includes pneumonia, transient tachypnea of the newborn, hyaline membrane disease, meconium aspiration syndrome and other miscellaneous causes. In developing countries there is a paucity of studies on causes of respiratory distress in neonates(1-3) and all respiratory distress in neonates are treated as pneumonia at the first referral unit. The clinical features of pneumonia in neonatal period are often non-specific(9). The present study was planned to: (i) find the causes of respiratory distress in neonates brought to a referral neonatal unit with symptoms suggestive of respiratory disorder; (ii) evaluate clinical signs like cough, difficulty in feeding, cyanosis, respiratory rate, chest retractions, flaring of alae nasi and adventitious sounds for diagnosis of neonatal pneumonia; (iii) determine bacterial etiology of neonatal pneumonia; and (iv) determine indicators of fatality in neonatal pneumonia. Subjects and Methods The study was conducted in the Referral Neonatal Unit, Department of Pediatrics, Maulana Azad Medical College in collaboration with the Department of Microbiology, Maulana Azad Medical College, New Delhi. The unit caters to neonates delivered at home and brought directly or delivered in smaller hospitals in Delhi and surrounding states and referred for neonatal care. One hundred and fifty consecutive admissions of neonatal respiratory distress fulfilling the inclusion criteria were enrolled in the study. All neonates presenting with respiratory symptoms characterized by any of the following were included in the study: (i) rapid, noisy or difficult breathing; (ii) respiratory rate >60/min; (iii) chest retraction; (iv) cough; and (v) grunting. Surgical problems causing respiratory distress,i.e., congenital malformations affecting respiratory tract and congenital heart disease were excluded from the study. The baby was evaluated in between feeds and in quiet state. Respiratory rate was recorded for at least 1 minute. The diagnosis of respiratory problems was based on guidelines recommended by the National Neonatology Forum (NNF)(10). Pneumonia was diagnosed in the presence of respiratory distress with: (a) positive blood culture or (b) if any two of the following were present: (i) existing or predisposing factors characterized by any one of the following: (a) maternal fever (>38º C), (b) foul smelling liquor; (c) prolonged rupture of membranes (>24 hours); (ii) clinical picture of sepsis characterized by any of the following: (a) poor feeding, (b) lethargy, (c) poor reflexes, (d) hypo or hyperthermia, (e) abdominal distention; and (iii) X-ray picture suggestive of pneumonia characterized by any of the following: nodular or coarse patchy infiltrates, diffuse haziness or granularity, air bronchogram, and lobar or sublobar consolidation(11). Transient episodes of consolidation lasting less than 48 hours due to pulmonary edema were excluded from the diagnosis of neonatal pneumonia. (iv) Positive sepsis screen. Transient tachypnea of the new born was diagnosed as respiratory distress in a term or borderline term neonate starting within 4 hours after birth, often requiring supplemental oxygen but recovering spontaneously within 3-4 days and showing characteristic X-ray changes, i.e., linear streaking at hilum and interlobar fluid. Hyline membrane disease (HMD) was diagnosed when the following three criteria were present: (a) preterm neonates; (b) respiratory distress having onset within hours of birth; and (c) skiagram of chest showing poor expansion with air bronchogram or reticulogranular pattern or ground glass opacity. Meconium aspiration syndrome was diagnosed in the presence of at least two of the following: (i) meconium staining of the liquor or staining of nails or umbilical cord or skin; (ii) respiratory distress soon after birth; and (iii) radiological evidence of aspiration pneumonitis (atelectasis or hyperinflation). Following investigations were done in the neonate at the time of inclusion into the study and sepsis screen was considered positive if at least two of these were positive: (i) peripheral smear with bandemia more than 20%; (ii) total leukocyte count interpreted as per reference value(12); (iii) micro ESR-interpreted as per criteria suggested earlier(13); and (iv) CRP done by rapid slide latex agglutination method using commerical kits (E. Merck India Limited MIDC, Taloja). All bacterial isolates were identified by conventional methods. X-ray chest was interpreted as per suggested criteria(14). Blood from radial or posterior tibial artery was taken by complete aseptic precaution in a 1ml syringe after heparinizing by a sterile concentrated heparin solution. Arterial line was used whenever available. Care was taken to ensure that the baby was not crying. The fraction of inspired oxygen (FiO2) was measured just before taking the ABG sample. Alveolar arterial oxygen gradient (PAO2- PaO2) and arterial alveolar oxygen tension ratio (PaO2/PAO2) were calculated. The continuous variables were analyzed using Student ‘t’ test and proportions by Chi-square test or Fischer test. Probability of 5% was considered significant. Mann Whitney U test statistic was applied to find out the p value in variables where standard deviation was more than the mean. Multivariate unweighted logistic regression was done to find out the indicators of fatality in neonatal pneumonia for those variables which were significantly associated with outcome on univariate analysis. Results The frequency of respiratory distress was 29.2% of all admissions (n = 569). The overall mortality in neonates with respiratory distress was 32%. Of the 150 cases of respiratory distress, pneumonia was the cause of respiratory distress in 68.7% cases (62.8% in term and 76.6% in preterms). Of these, 16.6% cases of pneumonia were associated with other conditions (hypoxic ischemic encephalopathy, acute gastroenteritis, hyaline membrane disease, pulmonary hemorrhage). Mortality was high in HMD, birth asphyxia and MAS (Table I). The mortality in neonates with pneumonia as compared to neonates with respiratory distress due to other causes was not significantly different (p = 0.717). Table I- Spectrum of Causes of Respiratory Distress in Neonates and Fatality
Fifty one per cent of all neonates with respiratory distress were delivered at home. The mean age at presentation was 164 hours in neonates with pneumonia and 77 hours when respiratory distress was due to causes other than pneumonia. Fifty two per cent neonates with pneumonia were full term. Forty two mothers of neonates with respiratory distress had predisposing factors for pneumonia. These predisposing factors have high specificity and positive predictive value, but low sensitivity and negative predictive value (Table II). Table II- Sensitivity, Specificity and Predictive Value of Antenatal History in Diagnosing Pneumonia in Neonates with Respiratory Distress
No significant difference was found in respiratory rate in neonates with pneumonia and respiratory distress due to other causes (p = 0.22). Presence of cough was significantly different in neonates with pneumonia as compared to neonates with respiratory distress due to other causes. Cough, adventitious sounds and flaring of alae nasi had high specificity while chest retractions, difficulty in feeding, and RR >60/min had high sensitivity for diagnosis of pneumonia in neonates (Table III). Tweleve neonates with pneumonia had respiratory rate <60/minute and 5 of them expired; 7 of them had gestational age of <34 weeks and weighed <1800 g. A respiratory rate of <50 per minute was present in 7 neonates and 5 of them were pre-term. Four of these neonates who expired had associated pulmonary hemorrhage and 3 also had co-existing shock. Five neonates with pneumonia had respiratory rate between 50-60 per minute. Three of these were pre-terms. One neonate had coexisting shock and expired. Table III- Sensitivity, Specificity and Predictive Value of Clinical Signs and Symptoms in Diagnosing Neonatal Pneumonia
Out of 103 cases of pneumonia, blood culture was positive in 49 cases. The isolates included Klebsiella in 28, Staphylococcus aureus in 7 and, coagulase negative staphylococcus in 7. Chest X-ray in neonates with pneumonia was found to have alveolar infiltrates in 44.6%, sublobar consolidation in 17.4%, lobar consolidation in 9.7%, diffuse haziness in 11.6%, opacity with reticulogranular pattern in 1.9% and clear lungs in 14.5%. Both blood culture and chest X-ray were positive in 37.9% cases of pneumonia. In 4.8% only sepsis screen was positive (Table IV). Table IV- Investigations Used to Diagnose Pneumonia in 103 Neonates
Sepsis screen was considered positive if two or more of individual parameters of sepsis screen (total leukocyte count, peripheral smear for bandemia, C-reactive protein, micro ESR) were positive. Sepsis screen was positive in 58 neonates with pneumonia. In addition, 2 neonates with positive sepsis screen but no other features of sepsis, normal X-ray chest and sterile blood culture had diarrhea with metabolic acidosis. Though all the sepsis screen parameters had specificity and positive predictive value above 90% in neonates with pneumonia, the sensitivity was low. It was 32% for micro ESR, 33.9% for total leukocyte count, 52.4% for bandemia and 54.3% for CRP. The mean alveolar arterial oxygen gradient (AaDO2) in neonates with pneumonia was 184.06 ± 144.6 and in neonates with respiratory distress due to other causes it was 143 ± 136.5 mmHg (p = 0.11). The mean arterial alveolar tension ratio a/A in neonates with pneumonia was 0.36 ± 0.19 and in neonate with respiratory distress due to other causes it was 0.43 ± 0.20 (p = 0.00). Factors significantly associated with fatality on univariate analysis are listed in Table V. After multivariate analysis, only alveolar arterial oxygen gradient (AaDO2) >250 mmHg was found to be independent predictor of fatality in neonatal pneumonia. Discussion In the present study, 29% of all the admissions to the referral neonatal unit during the study period were found to have respiratory distress. Fifty one per cent of these neonates were delivered at home while the remaining were referred from smaller hospitals from Delhi and surrounding states. To the best of our knowledge, there is no study estimating the proportion of neonates with respiratory distress which would be referred to the First Referral Unit and district hospital once primary neonatal care gets established at the national level(15). Hence, there is no estimate of the additional infrastructure required at the district hospital for providing the respiratory therapy to neonates with respiratory distress. This estimate is of paramount importance as essential newborn care becomes a part of a National Child Survival and Safe Motherhood Program/Reproductive and Child Health Program. The strategies for the essential newborn care in these national programs include provision of basic facilities for sick and low birth weight newborns in the first referral units and District Hospitals. The spectrum of causes of respiratory distress in neonates includes several clinical conditions. In the present study neonatal penumonia was the most common cause of respiratory distress in the referred neonates (68.7%). The frequency of neonatal pneumonia was higher (68.7%) in the present study as compared to earlier reports (18.7% to 51.4%) (1-3,5,16). The difference may have been due to the fact that half of the neonates in present study were delivered at home in unhygienic settings by untrained persons or transported from smaller hospitals with limited trained personnel and inadequate hygienic precautions. Earlier studies of neonates with respiratory distress(1-3,5) comprised neonates which were exclusively inborn, delivered in hospital by trained personnel and provided continuous care since birth. The importance of respiratory distress in the neonates can be realized from the fact that the neonates with respiratory distress are 2-4 times more likely to die than those without respiratory distress(1). The knowledge of the causes of respiratory distress is important to plan facilities. The overall mortality rate in the present study was 32% which is lower than that reported from other centers(1,3). The presenting complaints in neonates with pneumonia included rapid breathing (83.4%), poor feeding (81%) and difficult breathing (79.1%). These findings are similar to an earlier series(17). In primary neonatal care rapid breathing, poor feeding and difficult breathing are useful symptoms suggestive of respiratory distress. Table V- Unweighted Logistic Regression Model for Predictors of Mortality in Neonatal Pneumonia
In the present study, there was no significant difference in respiratory rate in neonates with pneumonia compared to neonates with respiratory distress due to other causes (p = 0.22). Of the 103 cases with pneumonia, 11.6% neonates had respiratory rate less than 60/min and would have been missed by the WHO definition of pneumonia (respiratory rate greater than 60/min)(18). The fatality in 7 neonates with pneumonia with respiratory rate of less than 50/min was 57.1% which was higher than in those neonates with pneumonia having respiratory rate greater than 50/min. Earlier studies have also found high mortality (27.7% and 58.3%, respectively) in preterm neonates in whom respiratory rate was less than 50/min(9,10). In the presenty study, bacterial etiology of pneumonia was established in 49 neonates (47.5%) by blood culture. This was slightly higher than that reported earlier(17,19). The bacterial isolates are consistent with the earlier studies which suggest increasing incidence of Klebsiella(19-22). Earlier studies on neonatal pneumonia(9) have included neonates with only radiological findings(3) and have not considered blood culture positivity in diagnosis of neonatal pneumonia. Webber et al.(23), however, had classified neonatal pneumonia as ‘definitive pneumonia’ if respiratory pathogen was isolated from the blood and ‘probable pneumonia’ if blood culture failed to show a pathogen, in presence of a positive chest X-ray. In the present study the percentage of ‘definitive pneumonia’ (37.9%) and ‘probable pneumonia’ (47.6%) were similar to that observed by Webber et al(23). No earlier study has stated detailed diagnostic criteria for pneumonia in neonates utilizing blood culture or sepsis screen positivity. The National Neonatalogy Forum(10) has included them in the diagnostic criteria but have not done any study evaluating utility of these criteria separately for diagnosis of pneumonia in neonates. Fatality in neonates with pneumonia having alveolar-arterial oxygen gradient (AaDO2) >250 mm Hg was significantly more as compared to those with AaDO2 <250. AaDO2 >250 was found to be predictor of fatality in neonatal pneumonia, after multivariate analysis in our study. According to Harris et al.(24) value of AaDO2 >250 mm Hg is an indicator of severe respiratory failure. In conclusion, it may be stated that pneumonia is the most common cause of respiratory distress in neonates. Clinical features and X-ray chest would miss the diagnosis of pneumonia in neonates in 15% cases and these have to be corroborated with sepsis screen and blood culture, for a definitive diagnosis. The diagnosis of pneumonia based on respiratory rate more than 60 per minute as suggested by WHO would miss the condition in 11.6% cases and the fatality in the missed cases is higher. Alveolar-arterial oxygen gradient (AaDO2) greater than 250 mm Hg is an incdependent predictor of fatality in neonatal pneumonia and can be used as a selection criteria for targeting intensive neonatal care. Contributors: NBM conceived and designed the study, interpreted the data, drafted the article and will act as a guarantor for the paper. KG drafted the article and analyzed the data. SK analyzed the data and contributed to the drafting. Funding: None. Competing interests: None stated.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
References | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|