Differences in Possible Risk Factors, Treatment Strategies, and Outcomes of Neonatal Pneumothorax in Preterm and Term Infants

Abstract

Objective:

The study aimed to compare the risk factors, treatment strategies, and early outcomes of symptomatic neonatal pneumothorax (NP) between preterm and term newborns.

Materials and Methods:

This retrospective cross-sectional study was conducted in a neonatal intensive care unit between 2015 and 2022, consisting of hospitalized neonates with symptomatic NP. The cases were divided into three groups according to their gestational ages: <34^0/7 (group 1), 34^0/7-36^6/7 (group 2), and ≥37^0/7 weeks (group 3). Risk factors, treatment strategies, and mortality rates of the study groups were compared using Kruskal–Wallis analysis.

Results:

Fifty-nine infants with a diagnosis of symptomatic NP were included in the study. The number of participants was as follows: 25 (42.3%) in group 1, 18 (30.5%) in group 2, and 16 (27.1%) in group 3. The need of delivery room (DR) resuscitation was significantly higher in group 1 (40%, P= .003). The surfactant administration rate was significantly higher in group 1 when compared to group 2 and group 3 (68% vs. 22% and 19%, respectively), P< .001. Similarly, the invasive mechanical ventilation percentage was significantly higher in group 1 than group 2 and group 3, P = .014. However, compared to group 3 (63%), the percentage of chest drain insertion (CDI) need was significantly higher in group 1 (96%) and group 2 (89%) (P = .014).

Conclusion:

Exposure to DR resuscitation and the need for surfactant are the most common risk factors for NP in preterm infants. Although oxygen and/or needle aspiration treatments are less invasive in symptomatic NP, the improvement rate without CDI is very low in preterm infants born before 34 weeks of gestational age.

What is already known on this topic?

• It is known that the major risk factors for neonatal pneumothorax are respiratory distress syndrom, meconium aspiration syndrome, congenital malformations, infections, transient tachypnea of the newborn, immaturity, and specific respiratory procedures in all newborns.

What this study adds on this topic?

• We compared the groups of newborns as term, late preterm, and early preterm. Although oxygen and/or needle aspiration treatments are less invasive in symptomatic neonatal pneumothorax, we found that preterm infants born before 34 weeks of gestation have a very low rate of recovery without chest drain insertion.

Introduction

Neonatal pneumothorax (NP) is defined as the abnormal accumulation of air between the visceral and parietal pleura of the lungs in infants who have not completed the first 28 days after birth or 44-week corrected gestational age (GA).¹ The incidence of symptomatic NP is between 0.8 and 1.9 per 1000 live births.^2,3 Pneumothorax is more common in newborns, particularly in the first 3 days of life, compared to all other age groups.⁴ It is more common in preterm infants, at as much as 5%-7% in newborns weighing less than 1500 g.³

Pneumothorax is classified as primary or secondary pneumothorax according to its etiology.⁵ Primary pneumothorax was defined as presenting spontaneously soon after birth and without significant underlying lung disease or the presence of predisposing factors such as prematurity, invasive or noninvasive ventilation.^5-7 On the other hand, secondary pneumothorax is defined as the pneumothorax that occurs due to traumatic injury to the pleura or risk factors such as meconium aspiration syndrome (MAS), respiratory distress syndrome (RDS), congenital pneumonia, transient tachypnea of the newborn (TTN), surfactant treatment, endotracheal suctioning, pulmonary hypoplasia, or congenital lung malformations, and the use of positive pressure ventilation (PPV) during resuscitation.^7-9 Respiratory distress syndrome was found to be the most common possible underlying lung disease of NP for preterm infants and TTN for term infants.³

In the clinical evaluation, chest wall asymmetry, cardiac impulse shift, and decreased breath sounds may also be observed in tension pneumothorax, although the symptoms seen in NP are mostly nonspecific, such as tachypnea, grunting, nasal flaring, retractions, cyanosis, apnea, and bradycardia.⁶ Infants receiving respiratory support in the neonatal intensive care unit (NICU) due to respiratory distress are evaluated for NP if there is a sudden increase in the need for respiratory support. Unless the clinical condition requires an immediate chest drain insertion (CDI), a chest x-ray is definitely performed in the NICU to confirm the diagnosis of NP.³

There are mainly 3 options for the management of NP: (1) conservative treatment by giving only oxygen; (2) drainage with needle aspiration (NA) by temporary insertion of a needle into the pleural space; or (3) definitive drainage of the air by insertion of a chest drainage catheter (intercostal) into the pleural space.^9,10 The oxygen therapy is affected by nitrogen washout. Nitrogen flushing is thought to help resolve pneumothorax by increasing the nitrogen absorption gradient from the extrapulmonary space. However, there is worrying evidence in the literature that oxygen therapy has harmful effects secondary to free radical damage.¹¹ At the drainage with NA, the needle is inserted into the second or third intercostal space at the midclavicular line, passing just above the rib to reduce the risk of rupturing the intercostal artery. Air flow into the syringe confirms that NP has been reached by the needle, which should not be inserted further to avoid lung damage.¹² Chest drain insertion is usually performed by inserting a chest tube into the pleural space. In the sixth intercostal space, a small incision is made through the skin in the midaxillary line, subcutaneous tissue is dissected, and a subcutaneous route is made into the intercostal space. The chest tube is passed into the pleural opening, rotated anteriorly and directed to the location of the NP, and then connected to an underwater seal with continuous suction at a pressure of 10-20 cm H₂O.¹²

Although there are numerous studies in the literature on NP risk factors, especially in extremely preterm and term infants, studies comparing NP treatment strategies between preterm and term infants are very limited.^9,10,13

The objective of the study was to compare possible NP risk factors, treatment strategies, and early outcomes among preterm, late preterm, and term newborns with pneumothorax and to reveal differences. Our main rationale is to provide data to develop individualized treatment strategies for preterm, late preterm, and term infants.

Materials and Methods

This retrospective cross-sectional study was conducted between January 2015 and December 2022 at the tertiary NICU in Kırıkkale University Hospital, regarding the principles of the Declaration of Helsinki. And the study protocol was approved by the Ethics Committee of Kırıkkale University (meeting date: January 11, 2023; meeting number: 2023/01; decision number: 2023.01.07). Verbal informed consent was obtained from the patients who agreed to take part in the study.

Over the course of the study, all newborns admitted to the NICU with the diagnosis of symptomatic NP or those who developed symptomatic NP during their NICU stay were involved in the study. Neonates who were admitted to our unit after birth from an outside center, or neonates with major congenital anomalies, and those who required thoracic tube insertion due to cardiac or thoracic surgery were excluded.

Diagnosis and Treatment Strategies of Neonatal Pneumothorax in our Neonatal Intensive Care Unit

Neonatal pneumothorax was suspected in infants hospitalized in the NICU when they exhibited nonspecific respiratory symptoms, such as respiratory distress, tachypnea, increased retraction, increased oxygen or pressure demand, and increased anterior–posterior chest diameter in infants on mechanical ventilators. Besides, in the delivery room, NP was considered, especially in infants who did not respond to resuscitation. These infants were imaged with a chest x-ray.

On the condition that there was no tension pneumothorax and that the amount of air leakage observed in chest radiography was less than one-third of the lung, oxygen therapy that kept NP neonates’ oxygen saturation within normal limits was preferred as the first step of the treatment. When supplemental oxygen therapy failed, NA was performed as the second treatment choice. However, CDI was applied if there was a tension pneumothorax or if the NP persisted despite the treatment choices mentioned above. Local anesthesia and systemic analgesic agents were used in the CDI procedure. Chest distress insertion administration was discontinued after the pneumothorax air was completely gone and/or the patient's need for pressure on the mechanical ventilator decreased.

Surfactant was applied to study infants with RDS by the intubated-surfactant-extubation (INSURE) procedure. Surfactant was administered to premature infants with signs of respiratory distress, an oxygen requirement (above 40%), and signs of RDS on chest radiography in the study. Some infants who underwent the INSURE procedure could not be extubated immediately afterwards and were followed up intubated.

Study Groups and Data Collection

The newborns with NP were divided into 3 groups based on their GA as <34^0/7 (group 1), 34^0/7-36^6/7 (group 2), and ≥37^0/7 weeks (group 3).

These 3 groups’ gender, birth weight (BW), GA, the first and fifth minutes Apgar scores, underlying diagnoses, need for resuscitation in the delivery room (DR) or at NICU, postnatal day of pneumothorax, lung lateralization of pneumothorax, pneumothorax treatment strategies (only oxygen supplementation, NA or CDI), chest tube length of stay, invasive or noninvasive mechanical ventilation administration and durations, surfactant treatment, NICU stay, and mortality rate were recorded from the hospital electronic database.

In the DR, PPV is routinely performed with a T-piece resuscitator in all infants in need of resuscitation. Infants requiring invasive respiratory support were ventilated in pressure-controlled synchronized intermittent mandatory ventilation (SIMV) or patient trigger ventilation (PTV) modes. Volume guarantee modes were not used during the study period in the NICU. High-frequency oscillatory ventilation mode was used as a rescue mode throughout the study. High-frequency oscillatory ventilation was initiated in patients who did not respond to conventional ventilation.

Statistical Analysis

The data were analyzed using Statistical Package for the Social Sciences Statistics for Windows version 28.0 (IBM Corp.; Armonk, NY, USA) and tested for normality with the Shapiro–Wilk test. They were expressed as the mean (SD) and median (minimum–maximum) as appropriate. The data were found to follow a nonnormal distribution; thus, median (minimum–maximum) was used according to the distribution of the demographic data. Besides, group differences were assessed using the Kruskal–Wallis test. Also, the chi-square test or Fisher’s exact test (when chi-square test assumptions do not hold due to low expected cell counts), where appropriate, was used to compare these proportions in different groups. In the analysis, a P-value less than 0.05 was accepted as statistically significant.

Results

Demographic Data

During the study period, a total of 7650 infants were born at the obstetric clinic in the hospital. A total of 3050 of these infants were admitted in the NICU. Fifty-nine newborns with a diagnosis of symptomatic NP were included in the study. The incidence of symptomatic NP was 0.07% and 1.9% for live-born babies and for hospitalized newborns, respectively.

The majority of the study group consisted of premature infants who were born before 37 weeks of gestation, 72.8% (43/59). The number of participants was as follows: 25 (42.3%) in group 1, 18 (30.5%) in group 2, and 16 (27.1%) in group 3 (Figure 1).

Figure 1.

Study groups.

Cesarean section rate, sex distribution, and side of the lung with pneumothorax were not significantly different among the groups (Table 1). The medians of Apgar scores at the first and fifth minutes of life were significantly lower in group 1 than in group 2 and group 3, P = .02 and P = .034 respectively (Table 1). The median days (minimum–maximum) of NP in group 1, group 2, and group 3 were 2 (1-9), 2 (1-4), and 1 (1-3), respectively (P = .058).

Table 1.

Demographic and Clinic Data of the Study Infants

	Group 1 (n = 25)	Group 2 (n = 18)	Group 3 (n = 16)	P
GA** (weeks), median (minimum–maximum)	31 (24-33)	35 (34-36)	38 (37-41)	< .0001*
BW** (g), mean ± SD	1680 (570-2450)	2693 (2025-3540)	3050 (2300-4400)	<.001*
Apgar first minute**, median (minimum–maximum)	7 (0-9)	8 (8-9)	8 (6-9)	.02*
Apgar fifth minute**, median (minimum–maximum)	8 (0-10)	9 (6-10)	9 (8-10)	.034*
CS, n (%)	23 (92)	16 (88.9)	15 (93.8)	1.000†
Male, n (%)	10 (40)	11 (61.1)	10 (62.5)	.259†
Right-sided pneumothorax, n (%)	10 (40)	8 (44.4)	8 (50)	.846†
Left-sided pneumothorax, n (%)	9 (36)	8 (44.4)	5 (31.3)
Bilaterally pneumothorax, n (%)	6 (24)	2 (11.2)	3 (18.7)

P < .0001 was statistically valuable which was given in bold. BW, birth weight; CS, cesarean section; GA, gestational age.

^*Kruskal–Wallis Test.

^**Median (minimum–maximum).

^†Fisher’s exact test.

Possible Risk Factors for Neonatal Pneumothorax

All infants included in the study were diagnosed with secondary pneumothorax due to the underlying diseases (RDS, TTN, neonatal pneumonia, and MAS). The most common underlying cause of the NP was found to be RDS (47.5%, 28/59), followed by neonatal pneumonia (27.1%, 16/59) and TTN (23.7%, 14/59). When analyzing group 1, RDS (n = 18, 72%) was detected to be the main reason behind NP. However, the number of infants diagnosed with TTN and neonatal pneumonia was 4 (16%) and 3 (12%), respectively. Considering group 2, RDS (n = 8, 45%), neonatal pneumonia (n = 6, 33%), and TTN (n = 4, 22%) were the underlying causes of NP, respectively. Finally, when examining group 3, neonatal pneumonia (n = 7, 44%) took the lead and was followed by TTN (n = 6, 38%), RDS (n = 2, 12%), and MAS (n = 1, 6%).

The comparison of the possible NP risk factors among the study groups is demonstrated in Table 2. The percentages of surfactant requirements in group 1 were significantly higher than those in group 2 and group 3 (P < .001). The percentage of resuscitation requirement (40%) was found to be significantly higher at the DR in group 1, P = .003. The percentages of invasive mechanical ventilation (IMV) administrations before developing NP were significantly higher in group 1 (80%) and group 2 (83%) compared to group 3 (63%) (P = .014) (Table 2). The median days of IMV duration, which were found to be significantly higher in group 1 than those in group 2 and group 3, were as follows: 5 (0-46), 3.5 (0-12), and 2 days (0-8) (P = .004).

Table 2.

Risk Factors for Neonatal Pneumothorax

	Group 1 (n = 25)	Group 2 (n = 18)	Group 3 (n = 16)	P
DR resuscitation, n (%)	10 (40)	0	3 (18.8)	.003**
NICU resuscitation, n (%)	5 (20)	0	4 (25)	.062**
Surfactant administration before NP, n (%)	17 (68)	4 (22)	3 (19)	<.001**
Noninvasive MV exposure before NP, n (%)	20 (80)	15 (83)	10 (63)	.340**
IMV exposure before NP, n (%)	13 (52)	2 (11)	4 (25)	.014*
HFOV exposure before NP, n/n (%)	5/24# (21)	1/18 (6)	2/16 (13)	.351**

P < .0001 was statistically valuable which was given in bold. DR, delivery room; HFOV, high-frequency oscillatory ventilation; IMV, invasive mechanical ventilation; MV, mechanical ventilation; NICU, neonatal intensive care unit; NP, neonatal pneumothorax.

^*Chi-square test.

^**Fisher’s exact test.

^#There was one missing data in group 1 for HFOV exposure.

Treatments Strategies and Early Outcomes

The percentages of infants with NP improved with conservative oxygen therapy alone were 31% in group 3 and 11% in group 2. However, no infant recovered with conservative oxygen therapy alone in group 1, n = 0 (Table 3). The percentage of NA needed was significantly higher in group 3 than in group 1 and group 2 (P = .037) (Table 3). The percentage of CDI need was significantly higher in group 1 and group 2 than in group 3: 96% (n = 24), 89% (n = 16), and 63% (n = 10), respectively (P = .014) (Table 3). There were no cases requiring pleurodesis for persistent NP.

Table 3.

Comparison of Treatment Strategies of the Infants with Neonatal Pneumothorax Among Study Groups

	Group 1 (n = 25)	Group 2 (n = 18)	Group 3 (n = 16)	P *
Only oxygen supplementation, n (%)	0	2 (11)	5 (31)	.007
Drainage with NA, n (%)	1 (4)	2 (11)	5 (31)	.037
CDI, n (%)	24 (96)	16 (89)	10 (63)	.014
Re-thoracostomy need after the thorax tube was removed, n (%)	3 (12)	3 (16.7)	1 (6.3)	.785

P < .007 was statistically valuable which was given in bold.CDI, chest drain insertion; NA, needle aspiration.

^*Fisher’s exact test.

The median length of stay of the thorax tube was significantly lower in Group-3 than in group 1 and group 2: 1 day (0-10), 5 days (1-17), and 4 days (0-15), respectively (P = .022). There was no significant difference in mortality rates among the study groups. A total of 12% of infants with NP died (7/59). Five infants under 34 weeks of gestation and 2 patients over 37 weeks of gestation also died (P = .095).

Discussion

The incidence of symptomatic NP was found to be 0.07% among live-born babies and 1.9% among hospitalized newborns in the present study. The incidence of pneumothorax has been reported between 0.08% and 0.14% among newborns in the literature.^7,14 Furthermore, the incidence of NP among neonates hospitalized in NICUs was 1%-2% and was shown to exceed 40% in the presence of RDS.¹⁵

There were no cases of primary (idiopathic) NP due to the fact that all of the infants included in the present study had lung diseases such as RDS, TTN, neonatal pneumonia, and MAS. Although the risk of developing NP is high in premature infants who need resuscitation or who are exposed to IMV due to underlying respiratory problems, the incidence of primary NP in premature infants is also high.^9,16 In newborn infants, prematurity and RDS are the most common possible risk factors for NP.⁵ According to the study conducted by Joshi et al,⁷ 50.1% of their study population consisted of preterm infants. The most common underlying cause in infants with NP was RDS (35.5%). In the present study, the majority of the study group consisted of premature infants who were born before 37 weeks of gestation (72.8%). In addition, similar to the study mentioned above, RDS was the most common underlying cause for not only the whole study population but also for group 1 and group 2, formed by preterm and late preterm newborns. However, considering group 3 in which term infants were involved, neonatal pneumonia was the most common one.

Resuscitation is one of the known possible risk factors for NP.^1,7 In the study of Jiang et al,¹⁷ the rate of NP was found to be high in late-premature newborns who needed resuscitation in the DR. The rate of DR resuscitation was found to be higher in group 1 in the current study, proving that premature infants undergoing DR resuscitation are more vulnerable to NP. However, there was no significant difference between preterm and term infants with NP in terms of resuscitation rate during the NICU stay.

Although NP was observed in group-1 and group 2, which included preterm infants, in the later postnatal days, no statistically significant difference was found when compared with group 3. The later development of pneumothorax in preterm infants could be related to lung immaturity. In the study of Basheer et al,¹⁸ the median age at the diagnosis time for the preterm neonates was found to be higher than the term neonates, 31 hours vs. 7 hours, respectively. In addition, NP was diagnosed later in newborns born below 2500 g compared to those born above 2500 g.¹⁹

García-Muñoz Rodrigo et al²⁰ described that surfactant administration was the independent risk factor associated with NP. Acute changes in lung mechanics after surfactant administration are considered to be a risk factor of NP.²⁰ Moreover, underlying causes of lung damage like RDS may also have increased the rate of NP.²⁰ On the other hand, in a Korean study, an increase in pneumothorax was observed, particularly in term infants after surfactant therapy.²¹ The rate of surfactant administration before NP was significantly higher in group 1 in the present study.

The need for IMV has been reported as a crucial risk factor that may affect the short-term prognosis of NP.^17,22,23 The rate and duration of IMV were found to be significantly higher in infants <34 weeks of gestation. The need for intubation and IMV is inversely proportional to the gestational week in newborns. Therefore, premature infants are more vulnerable to developing pneumothorax than term infants. In addition, when IMV is administered to patients with NP, the recovery time of NP may be relatively prolonged.²²

In the treatment of NP, conservative oxygen support and/or NA methods are recommended to avoid complications of both intubation and CDI.¹⁰ In the present study, although oxygen and/or NA strategies showed successful results in the treatment of NP for term infants significantly, they failed in preterm infants. One of the reasons why neonates in group 1 did not respond to oxygen therapy alone may be due to their relative larger pneumothorax compared to term neonates because they had a lower birth weight and a lower lung capacity. According to the best of our knowledge, the number of studies in the literature comparing differences between preterm and term infants based on treatment strategies remains limited.

The CDI application rate for NP treatment was found significantly higher in infants below 34 weeks of gestation in the present study. Halibullah et al²⁴ reported that CDI was required in preterm infants in particular and that older newborns were treated with NA. In a meta-analysis that included 2 different studies, the first study found that 45% of newborns with a 31-week median GA and with NP treated with NA did not require CDI.¹² In the other study, the newborns with a 36.4-week mean GA were treated with only NA. But in that study, the needle used in NP was kept in the newborn for an average of 27.1 hours after the procedure.¹²

The mortality rate of NP has been reported to vary between 13% and 65% in the literature.^3,10,25-27 In our study, the mortality rate was 12%. We suggest that this big difference in mortality rates is related to the socioeconomic status of the country where the study was conducted and the proportion of premature infants included in the study.

The inclusion of neonates with NP at all gestational ages is a strength of this study. The limitations of this study are that it was a single-center study, and a relatively low number of cases and complications of preterm infants such as intraventricular hemorrhage, necrotizing enterocolitis, bronchopulmonary dysplasia, and retinopathy of prematurity could not be obtained. In addition, all neonates born during the study period were not screened for NP due to the retrospective nature of the study, so cases of spontaneous pneumothorax may have been missed.

Symptomatic NP is more common in premature infants. Exposure to DR resuscitation and the need for surfactant are the most common risk factors for preterm infants. In term infants, neonatal pneumonia and TTN are the most common underlying causes of NP. Although oxygen and/or NA treatments are less invasive in symptomatic NP, the improvement rate without CDI is very low, especially in premature infants under 34 weeks of age.

As a result of this study, NP treatment with oxygen and NA should be considered in neonates over 34 weeks of age, and it should be known that CDI is usually not necessary. However, it should be kept in mind that CDI, although invasive, is often necessary in premature infants (<34 weeks). In this way, oxygen toxicity and additional complications of NA can be avoided in preterm infants.

Funding Statement

This study received no funding. This study was presented as an oral presentation at Turkish National Neonatology Congress on April 26-30, 2023.