Declining incidence of cerebral palsy in South Korea: Linked analysis of the National Health Insurance database and the health screening data

Healthcare delivery system in Korea

Almost all Koreans are covered by the health insurance policies of the Korea National Health Insurance (KNHI) of the Health Insurance Review and Assessment Service, except for 3% of the population that the Medical Aid Program covers. Thus, the KNHI claims database contains claims information for most Koreans except for uninsured procedures, such as plastic surgery. As a part of the National Health Insurance Corporation healthcare system, the national health-screening program for infants and children, for which children aged 4 to 80 months are eligible, began in 2007 and is composed of seven consecutive health examinations according to age (4–9, 9–18, 18–30, 30–42, 42–54, 54–66, and 66–80 months). The national health-screening program consists of history taking, physical examination, developmental screening, visual acuity, and a dental examination.

Study population

Information on pregnancy and birth was obtained by linking the KNHI claims database and national health screening program data for infants and children. Using the KNHI claims database, we identified all women who had delivered a baby between 2007 and 2015. A total of 3,778,561 women who had deliveries were identified. Excluding 56,925 deliveries of multiple births, a total of 3,278,916 single births were included in our study. Detailed information, such as medical illness, delivery mode, pregnancy complications, and single or multiple pregnancies, was confirmed using data from the KNHI claims and the national health-screening program for infants and children.

Definitions of CP

According to The International Classification of Diseases, Tenth Revision, CP was defined as a medical claim case with a CP diagnostic code. Included diagnostic codes were as follows: spastic quadriplegic CP (G80.0), spastic diplegic CP (G80.1), spastic hemiplegic CP (G80.2), dyskinetic CP (G80.3), ataxic CP (G80.4), other CP (G80.8), CP, unspecified (G80.9), flaccid hemiplegia (G81.0), spastic hemiplegia (G81.1), hemiplegia, unspecified (G81.9), flaccid paraplegia (G82.0), spastic paraplegia (G82.1), paraplegia, unspecified (G82.2), flaccid tetraplegia (G82.3), spastic tetraplegia (G82.4), tetraplegia, unspecified (G82.5), and paralytic syndromes (G83.0-83.3). Hereditary spastic paraplegia (G11.4) was excluded. The observational period was at least four years after birth, and upon diagnosis of CP, the number of people was summed based on the child's year of birth.

Statistical analysis

The Student's t-test was used to compare continuous variables between groups, and the categorical variables were compared using the χ² test. We compared temporal trends by using the χ² test. Multivariate logistic regression analysis was used to estimate the adjusted odds ratio (OR) and 95% confidence interval (CI). All P-values were two-sided and were considered statistically significant if 0.050 or less. Statistical analysis was performed using SPSS software version 12.0 (SPSS Inc., Chicago, IL, USA)

This study was approved by the institutional review board of the Korea University Guro Hospital approved the study (2020GR0468). Informed consent of the patients was waived because of its retrospective nature. All methods were carried out in accordance with relevant guidelines and regulations.

Trends in vaginal and cesarean births by year

As shown in Fig. 1, the number of single vaginal births from 2007 to 2015 appeared to be modestly declining, while the number of cesarean deliveries increased slightly, though the changes were insignificant.

Incidence of cerebral palsy

It takes approximately four years after birth for most children to be diagnosed with CP; more than 70% of children were diagnosed within two years, more than 80% within three years, and more than 90% within four years (Fig. 2). As shown in Fig. 3, the incidence of CP decreased from 4.77 to 2.52 per 1000 single live births, from 6.23 to 3.37 among cesarean deliveries, and from 3.96 to 1.96 among vaginal births (P < 0.001 for vaginal births and total live births; P = 0.0012 for cesarean births).

Comparison of the study population

According to a combined analysis of maternal and their babies' records, the frequency of advanced maternal age, hypertension and diabetes before and during pregnancy, diabetes mellitus before pregnancy, cesarean delivery, gestational hypertension, gestational diabetes mellitus requiring insulin treatment, chorioamnionitis, premature rupture of membrane, oligo- and polyhydramnios, preterm birth, low birth weight, male sex, and large for gestational age was higher in the group diagnosed with CP than in the controls (Table 1).

Risk factors for developing cerebral palsy

On multivariate analysis, maternal DM (OR = 1.16; 95% CI: 1.07–1.26), cesarean section (OR = 1.25; 95% CI: 1.20–1.31), chorioamnionitis (OR = 1.19; CI: 1.04–1.36), oligohydramnios (OR = 1.49; CI: 1.33–1.66), polyhydramnios (OR = 4.23 CI: 3.36–5.32), low birth weight (OR = 4.64; CI: 4.31–4.99), and male (OR = 1.34; 95% CI: 1.29–1.40) were associated with a higher risk of CP, but large for gestational age (OR = 0.87; 95% CI: 0.77–0.99) was associated with a lower risk of CP as shown in Table 2. Preterm birth was associated with an increase in the risk of CP, and in particular, the lower the gestational age at birth, the greater the risk of CP.

Yearly occurrence of risk factors

The number of occurrences by year of the three major risk factors is shown in Fig. 4. The number of preterm and low-birth-weight babies decreased yearly, while the number of pregnancies with polyhydramnios increased slightly (P < 0.001).

The incidence rate of cerebral palsy

A previous study in South Korea found that the incidence of CP increased from 2.2 to 3.2 per 1000 children born between 1999 and 2003.[3] The incidence of CP in Korea decreased from 4.77 to 2.52 per 1000 babies between 2007 and 2015, close to the CP incidence in developed countries. According to a meta-analysis analyzing 19 previous reports using live births as the denominator, the overall prevalence of CP is 2.11 per 1000 births.[4] The included studies were from the USA, Canada, Australia, or Europe, except one study from China. Nationwide population studies in other Asian countries have recently reported a prevalence of CP of 1.09 per 1000 children under 20 years of age in Taiwan,[5] 1.19 in children aged 0–6 years in China,[6] and 2.26 in children aged 0–4 years in Japan.[7]

Risk factors for cerebral palsy

Medical advances may be a possible explanation for why the incidence of CP has decreased. Among them, early detection and prevention of preterm labor, magnesium sulfate treatment for preterm birth, and post-brain injury management, such as brain cooling, are at the forefront of medical progress. Our study showed that low birth weight (< 2500 g), preterm birth, and polyhydramnios are still major risk factors for developing CP in Korea.

1) Preterm birth

Aside from our study results, preterm birth is already well-known as a risk factor for CP.[8, 9] The immature brain cannot effectively maintain a blood supply because there are fewer collateral vessels or anastomoses around the peripheral blood vessels and immature walls. The vessels cannot compensate for hypoxic-ischemic damage to limited vasodilation capacity.[10] Thus, various attempts have been made to prevent preterm birth; 17α-hydroxyprogesterone caproate may help prevent recurrent preterm birth.[11–13] Furthermore, vaginal progesterone prevents premature birth of mothers with short cervixes and improves neonatal outcomes.[14–16] Overall, progesterone administration lowered the rate of preterm births by approximately 50%.[15, 17] Cervical pessary use or cervical cerclage prevents repeated preterm births in high-risk women with a short cervix.[18] Despite medical advances, the rate of preterm births increased from 2.9% (1997–1999) to 4.5% (2011–2013)[19] and from 3.31% (1997–1998) to 6.44% (2013–2014)[20] in Korea. The increase in twin births is thought to have contributed to the rise in preterm births. As shown in Fig. 4, singleton's incidence of preterm birth gradually decreased between 2007 and 2015 (P < 0.001) in our study.

2) Low birth weight

The association between the development of CP and fetal growth restriction is well known. This association is not related to the gestational age at delivery.[21] Fetal growth restriction has been reported as a more critical risk factor for CP than fetal inflammation and birth asphyxia combined.[22] Intrauterine infection, one of the major causes of intrauterine growth retardation, can cause fetal inflammatory response syndrome: fetal growth restriction, damage to blood vessels in the fetal brain, placental vascular disorders associated with congenital disabilities, and white matter damage. Among them, white matter damage is a typical pathologic hallmark of CP. Our study also showed that the risk of developing CP was 4.6 times higher in babies weighing less than 2500 g. Fortunately, between 2007 and 2015, the number of low birth weight babies decreased (Fig. 4, P < 0.001).

3) Polyhydramnios

Polyhydramnios was found to quadruple the risk of CP in our study. Polyhydramnios has also been associated with increased perinatal morbidity and mortality risk, such as preterm birth, aneuploidy, cesarean section, fetal anomalies, and perinatal and postnatal mortality.[23, 24] Even when the results of a detailed ultrasound examination of the fetus were normal, polyhydramnios doubled the risk of genetic syndromes, neurologic disorders, and fetal malformations diagnosed after birth.[25] The incidence of polyhydramnios slightly increased over the study period (Fig. 4, P < 0.001). Older maternal age could be one of the underlying reasons for this change. In addition, considering that efforts to reduce the occurrence of CP have focused on preventing perinatal asphyxia and preterm birth and developing treatments to minimize brain damage and not the prevention of polyhydramnios, it is considered a natural result.

4) Direct efforts to reduce the chances of neonatal brain injury

There have been various advances in medical technology that have minimized neonatal brain injury and mortality. First, magnesium sulfate (MgSO₄) stabilizes blood pressure, reduces vasoconstriction in the cerebral arteries, and restores circulation in preterm neonates.[26, 27] Treatment with MgSO₄ in preterm labor may lower the risk of CP.[28–30] The proportion of moderate to severe CP decreased significantly in babies born in women with preterm birth who were treated with magnesium (relative risk, 0.55; 95% CI 0.32–0.95)[29]; several meta-analyses also support this result.[31–33] Second, the prenatal administration of corticosteroids for fetal lung maturity may reduce the occurrence of CP.[34] Finally, brain or whole-body cooling has become standard management for neuroprotection in newborns with birth asphyxia.[35, 36]

With all these efforts, Korea's infant mortality rate decreased from 4.7 in 2004 to 3.0 in 2014.[37] Our study found that the incidence of CP also continues to decline in Korea.

Limitations

Our data do not include perinatal or postnatal neonatal risks, such as birth asphyxia, neonatal sepsis, or respiratory distress syndrome. We did not include early infantile risks, such as encephalitis or head trauma, which may also cause CP. A child can be diagnosed with CP when a precipitating event occurs before the affected function has developed.[1]