Evaluation of Premature Infants Hospitalized in Neonatal Intensive Care Unit between 2010-2012

Objective: With continuing developments in the field of neonatology, survival rates of low birth weight and small for gestational age infants have increased, which in turn has brought important prematurity-related problems. The aim of this study was to evaluate retrospectively the prematurity problems that are the significant causes of morbidity and mortality. Materials and Methods: 613 premature infants hospitalized in the neonatal intensive care unit of Ataturk University Medical Faculty Hospital between January 2010 and January 2012 were included in this study. Infants were divided into groups according to their birth weight and gestational age. Results: 323 infants were male (52.6%) and 290 were female (47.4%). 63.9% of infants weighed ≥1500 grams, and 58.5% had a gestational age of ≥33 weeks. Respiratory distress syndrome (RDS) was detected in 249 (40.6%), bronchopulmonary dysplasia (BPD) in 124 (20.2%), necrotizing enterocolitis (NEC) in 41 (6.6%), retinopathy of prematurity (ROP) in 202 (32.9%), and intracranial hemorrhage (ICH) in 15 (2.4%). RDS, BPD, NEC, ROP, and ICH rates were inversely proportional to decreases in gestational age and birth weight, and were found to be statistically significant. Conclusion: Mortality and morbidity rates were similar to the other data published from our country, but the rates were above those reported in developed countries. We believe that our morbidity and mortality rates can reach levels comparable to those of developed countries with improved antenatal care, regular follow-up of pregnancy and increased numbers of physicians and health care personnel per patient.


Introduction
Preterm delivery is among the leading causes of neonatal mortality and morbidity globally, including in developed countries.Recently, thanks to the scientific and technological advances in the field of neonatology, leading to an increased quality of neonatal intensive care unit care, perinatal mortality among premature babies has been reduced, particularly in developed countries [1].
Based on the recent statistics issued in the United States, the premature delivery rate was reported to be 12% in 2007 [2].The Turkish Population and Health Survey results reveal a rate of 11% for babies with a low delivery weight in Turkey [3].
However, despite all the advances in neonatal and perinatal care, particularly the morbidity among preterm babies remains a significant issue.A portion of these babies die during or after delivery, with surviving babies potentially experiencing major complications [4,5].Primary complications include infections, immune and genitourinary system disorders, respiratory distress syndrome (RDS), bronchopulmonary dysplasia (BPD), intracranial hemorrhage (ICH), patent ductus arteriosus (PDA), retinopathy of prematurity (ROP), and necrotizing enterocolitis (NEC) [4].
While data on these premature infant complications are retained better in developed countries with a high number of relevant publications, there are potential deficiencies related to data recording and follow-up in our country [6].
Mortality and morbidity statistics related to preterm infants are important indicators of the quality of the perinatal health care system on an institutional or national basis and are necessary for the identification of problems and implementation of preventive measures.However, the perinatal mortality and morbidity statistics on an institutional or national basis in Turkey are inadequate.In this study, we aimed to analyze the mortality and morbidity among preterm infants in our tertiary care university hospital and to compare these results with previous reports.We also aimed to document the causes of mortality and morbidity in preterm infants to facilitate an evaluation of the preventive care of these babies.This is the first report to show the mortality and morbidity rates in a high-altitude region of Turkey.

Materials and Methods
The trial was conducted in Ataturk University Medical Faculty Hospital, Department of Pediatric Health and Diseases, Neonatal Intensive Care Unit.The study was approved by the Ataturk University Medical Faculty Research Ethics Committee (Date: 18.08.2011No: 7/15).Six hundred and thirteen premature infants treated and monitored in the Unit were included in the trial.
The medical records of the babies included in the trial were reviewed according to gestational week and birth weight.The "New Ballard Score" assessment was performed in those babies for whom the gestational week could not be determined clearly by the last menstruation date and/or antenatal ultrasonography (USG).The babies were grouped into three categories according to gestational week as: <25 weeks, 26-28 weeks, 29-32 weeks, and 33-37 weeks.Based on the birth weight, the babies were classified as <750 grams, 750-1000 grams, 1001-1499 grams, and ≥1500 grams.
The diagnosis of RDS was made based on the clinical and pulmonary radiography results.The "New BPD" definition was used for BPD identification [7].NEC was described according to the "Modified Bell" criteria [8].ICH was diagnosed via transfontanellar USG.
The ophthalmological examinations of the babies for the diagnosis of PR were performed by the Ataturk University Medical Faculty Hospital, Ocular Diseases Department.

Statistical Analysis
The statistical analyses were conducted using the Statistical Package for the Social Sciences (SPSS) 20.0 software.In assessing the study data, for quantitative data comparison, the one-way ANOVA test, as well as the descriptive statistical methods (mean, standard deviation, and frequency), was employed for comparing the parameters with a normal distribution between the groups.For comparing the qualitative data, the chi-square test was used.The results were evaluated at the significance level of p<0.05 in the 95% confidence interval.

Results
Three hundred and twenty-three of the premature babies were males (52.6%) and 290 were females (47.4%).The mean gestational week of the premature babies included in the trial was 32.5±3.5 weeks, and the mean birth weight was 1732±665 grams.The distribution of the patient demographics by birth weight and gestational week is presented in Table 1 and Table 2, respectively.
One hundred and sixty-one babies were born via normal spontaneous vaginal delivery (NSVD) (26.2%), while 452 were born via cesarean section (C/S) (73.8%).The rate of those born via C/S was 2.8-fold higher compared to those born via NSVD.
The complications occurring during the monitoring of the babies according to birth weight and gestational week are presented in Table 3 and Table 4, respectively.As the gestational week and birth weight decreased, the rates of RDS, BPD, ROP, NEC, and ICH were observed to be statistically increased (p<0.001).However, no statistically significant correlation was detected between the birth weight and gestational week and pneumothorax (p=0.06,p=0.053).
Patent ductus arteriosus was detected in 137 babies (22.3%).In 93 (67.8%) babies with hemodynamically significant PDA, the ductus was observed to close following 1-3 courses of ibuprofen treatment.Nineteen (13.8%) babies underwent surgery since the ductus did not close despite 3 courses of ibuprofen treatment.Ten (7.3%) babies who received no ibuprofen treatment died.Ten of the 15 babies (7.3%) with no hemodynamically significant PDA detected and no treatment administered experienced spontaneous Eurasian J Med 2015; 47: 13-20 closure of the ductus during follow-up, while in the remaining 5 (3.6%) babies, post-discharge outpatient follow-up was recommended.
The causes of mortality among premature babies included in the study are presented in Table 5.The most common cause was RDS, which was detected in 34 babies (34/87, 39.1%).Fifty percent of these babies were premature babies of <25 weeks.
Based on the VON 2005 data, the mortality rates among babies between 500 and 1500 grams were reported to be between 10% and 18%.These rates are from the investigative sites equipped with the most developed facilities (such as nitric oxide, extracorporeal membrane oxygenation) and qualified staff [11].
In our trial, the mortality rate was 14.1% among 613 premature infants (87 babies).The mortality rates were 57.6%, 27.5%, 15.6%, and 8.1%, respectively, for babies with a birth weight of <750 grams, 750-1000 grams, 1001-1499 grams, and ≥1500 grams.The mortality rates were 85%, 34.9%, 11.2%, and 6.6%, respectively, for babies with a gestational week <25 weeks, 26-28 weeks, 29-32 weeks, and 33-37 weeks.Compared to the developed sites, we observed high mortality rates for babies with a birth weight ≤1500 grams.However, comparing these results to the data from the Turkish Neonatology Association, no marked differences in mortality rates were detected.In line with the literature, the mortality rate was detected to increase as the gestational week and birth weight decreased (p<0.001).Among the exitus babies, 39% were detected to have advanced prematurity and RDS, 16% and 11.4% were detected to have sepsis and congenital heart disease, respectively, and 9.1% had asphyxia.
Respiratory distress syndrome is a common cause of morbidity among premature babies.The most significant risk factors are low gestational age and low birth weight, while late premature delivery (35-36 gestational weeks) and spontaneous elective cesarean delivery may also be considered as risk factors.Based on the National Institute of Child Health and Human Development (NICHD) data, the RDS incidences among babies according to birth weight are 44%, 71%, 55%, 37%, and 23%, respectively, for babies with birth weights of 501-1500 grams, 501-750 grams, 751-1000 grams, 1001-1250 grams, and 1251-1500 grams [10].
In the current trial, 40.6% of the 613 premature babies were detected to have RDS; as the gestational week and birth weight decreased, the incidence of RDS increased (p<0.001).Compared to the data from developed countries, RDS rates were observed to be higher in our trial.A factor that reduces the risk of RDS is antenatal steroid intake [4,5].Based on the NICHD data [10], the antenatal steroid intake was detected to be between 47% and 90%, while this figure was relatively low (33.6%) for our trial.In our district, most pregnant women can reach the hospital only at the moment of delivery due to their low socioeconomic level and the closure of roads, particularly during winter.This leads to a low level of antenatal care and steroid use in relation to the lack of follow-up in pregnant women.
While the rates of administering surfactant to babies with a very low birth weight has varied between 21% and 40% in the trials conducted in Turkey, this figure was 52% based on NICHD data [5,6,[12][13][14].The rate of surfactant administration was detected to be 38% in our trial (n=236), with 38% of these 236 babies (n=90, 14.6% of all patients) being ≤28 weeks.
In our trial, the BPD incidence among 613 babies was 20.2%.The incidence rates were 42.3%, 66.2%, 35.6%, and 4.8%, respectively, for babies with birth weights of <75 grams, 750-1000 grams, 1001-1499 grams, and ≥1500 grams.By gestational week, the incidence rates were 15%, 61.4%, 35%, and 4.7% for babies <25 weeks, 26-28 weeks, 29-32 weeks, and 33-37 weeks, respectively.In our trial, the BPD incidence was generally low among babies with a birth weight <750 grams.However, the high mortality observed below this weight may have yielded this result.As the gestational week and birth weight decrease, the BPD incidence increases.In the literature, male gender is reported to be correlated with BPD [15].However, there was no significant gender correlation among babies with BPD in our trial (p=0.112).Despite all the advances in the neonatal intensive care units, prematurity results in significant emergent gastrointestinal complications.The most important among these is NEC.In a trial by Tayman et al. [16], the NEC incidence was reported to be 11.4% among 532 premature infants <32 gestational weeks.Based on the data of VON 2005 from babies with a very low birth weight, the NEC incidence ranged between 3% and 9% [11].The most recent rates as reported by NICHD are as follows: 11%, 9%, 5%, 3%, and 7%, respectively, for babies with a birth weight of 501-750 grams, 751-1000 grams, 1001-1250 grams, 1251-1500 grams, and >1500 grams [10].
Premature retinopathy is among the primary causes of vision loss in premature infants.However, it can be avoided through appropriate and timely treatment regimens.The ROP rates obtained in previous trials are given in Table 7.
A multi-center randomized trial by the Cryotherapy for Retinopathy of Prematurity (CRYO-ROP) group reported a ROP (at any stage) incidence of 65.8% for 4099 neonates with a birth weight <1251 grams.PR of several grades was detected in 90%, 78% and 47% of babies with a birth weight of <750 grams, 750-999 grams and 1000-1250 grams, respectively [19].The Early Treatment for Retinopathy of Prematurity (ETROP) trial detected a PR incidence of 68% for 6998 neonates <1251 grams from 26 investigative sites [20].
The ROP incidence of 32.9% detected in our trial is consistent with the results from the studies reviewed in our trial.By gestational week, the ROP incidence was detected to be 15%, 57.8%, 65.5%, and 14.4% for babies <25 weeks, 26-28 weeks, 29-32 weeks, and 33-37 weeks, respectively.By birth weight, the incidence rates were detected to be 42.3%, 62.5%, 59.1%, and 18.6% for babies with birth weights of <750 grams, 750-1000 grams, 1001-1499 grams, and ≥1500 grams, respectively.Taking babies <1500 grams of weight into consideration, the rates of ROP at any stage appear to be above 60%.We believe that the low rates being observed below 750 grams result from the high mortality in this group.In line with the literature, the ROP occurrence rate in our trial increased as the gestational week and birth weight decreased.
Another major morbidity is PDA.The 2002 NICHD [10] data revealed a PDA incidence of 29% with no significant change The risk of germinal matrix-intraventricular hemorrhage is increased in preterm babies.Most of the cerebral injury among premature babies occurs via intraventricular hemorrhage.The ICH incidence increases as the birth weight and gestational week decrease.While the germinal matrix hemorrhage incidence was 40-45% before 1980, it had regressed to 12-30% in the 1990s [21].Based on the VON 2008 data from babies with a very low birth weight, the severe ICH incidence was between 5.6% and 12.5% [11].
Tavosnanska et al. [22] reported an ICH incidence of 10.1% in their trial among 1169 babies with birth weights of 500-1499 grams, monitored between 2008 and 2010.In our trial, the ICH incidence was 4.97% for babies with a birth weight <1500 grams and 2.4% for babies overall.We attributed this low incidence to the fact that 23 of our exitus patients died within the first 3 days before transfontanellar USG was performed.
The altitude above sea level in our geographic region should also be considered while assessing our data because perinatal complications are reported to increase even between 2000 and 2999 meters, a range defined as moderately high altitude (Erzurum central region: 1900-2200 meters) [30].While the pulmonary pressure in babies born at sea level immediately reaches sea level and adapts to outer circumstances more quickly, such adaptation is reported to become slower, thereby leading to a higher pulmonary pressure at high altitude [31].We believe this could affect the morbidity and mortality rates.However, further studies need to be performed in this regard.
Some limitations of the study should be mentioned.The first relates to the retrospective nature of the study.As patient data were collected retrospectively, some information, such as days and amount of oxygen or ventilation support, was not available for inclusion in the study.Thus, we were unable to discuss the effect of the high altitude in the preterm infants.Second, because of the high mortality rate in preterm infants below 750 g and 25 gestational weeks, the information collected regarding morbidity was insufficient, and this affected our statistical analysis results in these groups.
In conclusion, although the mortality rate has decreased among babies with a low gestational age, we still observed a high rate of morbidity.We believe improving the prenatal care conditions and regular pregnancy monitoring would reduce the incidence of premature babies.We also think that increasing the number of nurses and physicians per patient during the postnatal period would reduce mortality and morbidity rates, leading us to the levels observed in the developed countries.

Table 1 . Demographics of the infants (by birth weight)
M: male; F: female; NSVD: normal spontaneous vaginal delivery; C/S: cesarean section

Table 2 . Demographics of the infants (by gestational week)
M: male; F: female; NSVD: normal spontaneous vaginal delivery; C/S: cesarean section