Pulmonary surfactant (PS) replacement therapy has been established as a safe and effective treatment for immaturity-related surfactant deficiency since the early 1990s.1 Systematic reviews have reported that surfactant administration in preterm infants with established respiratory distress syndrome (RDS) decrease the mortality rate, incidence of pulmonary air leak, and risk of chronic lung disease or death at 28 days of age.23 However, recent studies have suggested that in addition to PS deficiency, there are other causes leading to RDS, especially in near-term and term infants.4

In Korea, surfactant replacement therapy has been available for neonatal RDS since 1991,5 and several products for artificial PS replacement therapy are currently being marketed (e.g., Surfacten^®, Exosurf^®, Newfactan^®, Curosurf^®, and Infasurf^®). Insurance coverage for single-dose surfactant replacement was initiated in 1992 and for multiple doses in 2007. In 2011, the insurance system also extended coverage for early prophylactic therapy with a single administration within 2 hours after birth for premature infants with a birth weight < 1,250 g or a gestational age < 30 weeks.6 Since then, the survival rates of very-low-birth-weight infants (VLBWIs) and extremely-low-birth-weight infants have significantly improved.78 Moreover, minimally invasive PS administration has been applied with insurance coverage since December 2018. The survival rate has been encouraging and is similar to that achieved in advanced countries in Europe and North America.9

PS has traditionally been administered through an endotracheal tube either as bolus, in smaller aliquots, or by infusion. However, the optimal method of PS administration in preterm infants has not been elucidated to date owing to the availability of limited and conflicting data. Repeated PS doses for predetermined indications have decreased the mortality rate and morbidity.3 However, there is insufficient evidence to recommend the optimal number of fractional doses of surfactant or the optimal body position during administration.1 With the development of new methods of minimally invasive PS administration, prophylactic PS replacement therapy may be advantageous compared to continuous positive airway pressure therapy with rescue administration of PS with less invasive surfactant administration.10

Several studies have reported on the PS preparations in Korea; however, these studies were published before 2010.111213 The latest multicenter collaborative study in 2010 showed increasing trends of PS use. The study reported that 3,160 patients in 72 hospitals were treated with 8,364 vials of surfactant, which cost 4,993 million Korean won (KRW).11 Among them, 318 infants died, yielding a mortality rate of 10.1%.11 The use of PS therapy may have further increased with the provision of extended insurance coverage since 2011; however, nationwide data are not available regarding this. National epidemiologic data can be useful for estimating the medical resources required for disease control. Thus, this study aimed to investigate the epidemiologic data of PS replacement therapy for RDS in Korea and analyze the complications associated with RDS. To the best of our best knowledge, this is the first nationwide study on surfactant replacement therapy using the national insurance health data in Korea.

A total number of 26,392 PS replacement therapy were performed for 20,671 infants between 2014 and 2018. Five products were used in Korea, the proportion of them showed 45%, 40%, 11%, 3%, and 1%. The indication for PS replacement therapy was RDS in 19,442 of the 20,671 infants (94%). Multiple doses of PSs were administered to 1,688 infants (8.7%), as follows: 2 doses in 929 infants (4.8%), 3 doses in 438 infants (2.3%), 4 doses in 191 infants (1.0%), 5 doses in 84 infants (0.4%), 6 doses in 32 infants (0.2%), 7 doses in 9 infants (0.1%), 8 doses in 4 infants (0.02%), and 9 doses in 1 infant (0.01%). The incidence of RDS was 0.99%, and the annual incidence rate showed an increasing trend from 0.84% in 2014 to 1.16% in 2018. The total number of term infants with RDS and incidence of RDS in term infants markedly increased over 5 years, respectively. Incidence of RDS in preterm infants also showed an increasing trend. In addition, RDS showed 0.25% of infants with a birth weight of 2,500 g or more, 5.1% of infants with a birth weight of 1,500 to 2,500 g. Incidence of RDS in VLBWIs was 48.1%, showing an increasing trend by 9.5% during 5 years (Table 1). The proportion of SGA among infants with RDS was 3.3% (646 infants) ranging from 1.9% to 4.2% during 5 years. The average total amount of admission payment was 21,115,115 ± 17,031,077 KRW, the payment by patients was 42,686 ± 184,351 KRW and the payment by insurance was 21,072,428 ± 1,003,284 KRW. The distribution of RDS incidence by province is shown in Fig. 1; ranging from 0.13% to 2.38%.

Among patients with PS treatment, the number of preterm infants was 3.3 times higher than that of term infants (76.8% vs. 23.2%). The mortality rate associated with RDS was 6.3%, which showed a decreasing trend according to the year (Table 2). Among the 18,673 (96% of the total cohort) infants with available data for gestational age, the mortality rate was significantly higher in the lower gestational age group (Tables 3 and 4). Morbidity, including pulmonary hemorrhage, BPD, PDA, and IVH was also higher in the lower gestational age group. However, pneumothorax most frequently occurred in the term infant group (Tables 3 and 4). The complications associated with RDS according to the gestational age group are shown in Tables 3 and 4. There were 16,243 (84%) patients for whom data were available for subgroup analysis according to the birth weight. The incidence of most RDS complications was higher in the lower birth weight group; however, the incidence of pneumothorax was higher for the infants with birth weight ≥ 2,500 g (Tables 3 and 4).

Pulmonary hemorrhage and pneumothorax in deceased infants with RDS (n = 1,229) showed higher incidence compared to those in whole population in this cohort (13.4% vs. 2.9%, 14.2% vs. 5.5%, respectively) (Tables 2 and 5). Pulmonary hemorrhage, pneumothorax, persistent pulmonary hypertension, BPD and IVH were shown as an independent risk factor of the mortality after adjusting gestational age and RDS-related complications (OR, 3.94, 3.05, 2.21, 1.28, and 1.64, respectively). Gestational age was independently associated with the mortality of infants with RDS, which showed significantly high OR 4.70 in the infants < 28 weeks compared to the term infants, and lower odds in the infants with 28–36 weeks than term infants (Table 6).

In this study, we used national epidemiologic data and found an incidence of 1% of RDS with PS replacement therapy and relating mortality and morbidity. Our results showed that the incidence of RDS is consistently increasing in term infants despite decreasing trends in the birth rates. During the 5-year study period, the number of PS-treated term infants increased by 250, while the birth rate decreased by 100,000. RDS was prevalent in 0.34% of term infants in 2018 compared to 0.19% in 2014. This indicates either a possibility of increasing RDS incidence or a trend of active PS replacement in term infants.

In Korea, after formal import of the PS preparations, the outcomes of PS replacement therapy were reported in multicenter studies, including 60 patients in 16 hospitals in 1993,16 1,066 patients in 64 hospitals in 1997,17 1,596 patients in 62 hospitals in 2004,12 1,921 patients in 57 hospitals in 2009,13 and 3,160 patients in 72 hospitals in 2011.11 The mortality rate associated with RDS significantly decreased and was reported to be 40.0% in 1990/1991, 28.7% in 1996, 18.7% in 2002, 14.3% in 2007, and 10.1% in 2010.1112131617 However, these data are limited to before 2010, when the extended insurance coverage was not applicable. As insurance coverage was extended to include PS replacement therapy in 2011, our results can represent all patients treated with PSs in Korea, as this study is based on the national health insurance data.

With respect to the complications associated with RDS, the prevalence of pneumothorax and pulmonary hemorrhage was not significantly changed over 5 years among all infants with RDS treated with PS. However, among the term infants, the incidence of pneumothorax notably increased from 6.9% up to 11.7%. On the other hand, the incidence of pneumothorax in neonates who admitted to NICU was reported as 0.5% to 1.3% in previous single center studies in Korea.1819 This can be interpreted as increased surfactant-related complications or underlying severe pulmonary pathology in term infants. In addition, the ideal treatment of RDS is often delayed in term infants delivered in local hospital, which can contribute the higher incidence of pneumothorax than that in preterm infants who are mostly delivered in general or university hospital. Careful observation for pneumothorax and timely consideration for PS therapy can be needed.

The incidence of RDS decreases with advancing gestational age, from approximately 60%–80% in infants born at 26–28 weeks of gestation to approximately 15%–30% among those born at 32–36 weeks of gestation.2021 In our study, we found that RDS was prevalent in 10.6% of the preterm infants born at less than 37 weeks of gestation, 48.1% of infants below 1,500 g. From the annual report of Korean Neonatal Network (KNN), the incidence of RDS among VLBWIs was 75.9% in 2018, which showed higher than our result based on HIRA database and Statistics Korea.22 We also found that the incidence of RDS treated with PS consistently increased in the VLBW patients from 43.2% to 52.6%. However, according to the annual KNN reports, PS administration showed 3.7% of decrease during 2014–2018.2223 The difference may occur because the number of VLBWIs registered to KNN was smaller than that registered to Statistics Korea and HIRA. Also, this discrepancy in the incidence of RDS with PS may be due to omission of the detail information about the gestational age (4%), birth weight (16%), or due to the use of non-reimbursable PSs, which can lead to the underestimation of the total number of VLBWIs and/or the rates of PS administration.

The incidence of RDS at 32 weeks of gestation was approximately 30% in the US in the 1990s.24 And Hibbard et al.25 reported that the incidence of RDS was 10.5% in infants at 34 weeks of gestation and 2.8% in infants at 36 weeks of gestation. The Danish Medical Birth Registry Cohort showed that the incidence of RDS in infants born at 32–36 weeks of gestation was approximately 5.7%.26 In the same study, the incidence of intracranial or IVH was 2.0% and of PDA was 3.4% in the RDS group. In our study, the incidence of IVH or PDA is higher than previous study, it is necessary to take careful consideration for preventing complication among the late preterm infants.

The burden of RDS in the near-term and term infants has been recently highlighted, and an increasing number of studies have suggested that the clinical presentation in the near-term and term infants may be different from that observed in the very preterm infants.127 Recent studies have suggested that in addition to PS deficiency, there are other dominant causes leading to RDS, especially in the near-term and term infants; such as elective cesarean section, maternal diabetes or hypertension.42829 Tsakalidis et al.30 reported that RDS in premature infants aged < 35 weeks of gestation can be adequately treated with a single dose of surfactant, but the near-term and term infants with RDS require multiple doses of surfactant. In another study, 48 per 1,986 (2.42%) neonates developed RDS, of which 7 (14.6%) weighed ≥ 2,500 g.31 In our study, PS was used in 0.25% of the term infants, and the incidence of accompanying pneumothorax was higher than that in the preterm infants, indicating a different pathophysiology of RDS and/or complicated conditions such as delivery room care, times at treatment, and transport to another hospitals in the term infants.

The diagnosis of RDS is based on the decision by neonatologists according to the clinical manifestations and chest X-ray findings. The dose and infant position during PS administration depend on the manufacturer's instructions, and no universal guidelines are available to date. There can be some variance of practice within hospitals and among doctors.

There are some limitations of this study. We did not include data on the severity of conditions and health behavior of the recipients. As the information, including gestational age or birth weight was obtained from diagnosis codes inputted by the hospital, some data on RDS may have been omitted or limited information about the diseases of the patients may have been available. There is also a discrepancy between the diagnoses entered in the data and the actual disease experienced by the patient has. And there was a limit to clarify the clinical significance of RDS complications such as IVH or PDA. Further, as the claims data are generated to reimburse the healthcare services eligible for coverage, healthcare services not covered by insurance were not assessed. The information about the residence of beneficiaries may not be reliable because the HIRA data are collected based on the location of the providers. The area where the beneficiary has received the healthcare service may be different from the area where the beneficiary resides. However, despite these limitations, we believe that our study is valuable because the use of national health insurance data makes our findings reflective of the latest national epidemiology on RDS and PS replacement therapy. Accordingly, the findings can be used for developing related public health policies and intervention for further RDS control.

In conclusion, the advances in neonatal care and extended insurance coverage have increased the use of surfactant replacement therapy for RDS, and this, in turn, has decreased the neonatal mortality rate. The incidence of complications, such as pneumothorax, PDA, IVH, and BPD, showed a decreasing trend in the term infants, except for pneumothorax. Term and near-term RDS should be differentiated, and the appropriate therapeutic strategy for RDS should be decided according to the gestational age and lung pathology.