Prevalence of craniosynostosis in Finland, 1987–2010: A population‐based study

Craniosynostosis is a prevalent craniofacial malformation in Finland; however, comprehensive population‐based epidemiological data are limited. This study aimed to estimate the total and birth prevalence of craniosynostosis in Finland from 1987 to 2010 and examine temporal trends.

significant increase in the syndrome group.In live births increase was significant only within the syndrome subgroup, primarily due to an increase in Muenke syndrome patients.The rising prevalence of syndromes necessitates further investigation.Contrasting with trends in Europe, Australia, and the USA, Finland showed no significant increase in metopic craniosynostosis.

| INTRODUCTION
Primary craniosynostosis refers to the premature fusion of one or more skull sutures during fetal development.Single-suture synostosis is estimated to occur in 3-4 cases per 10,000 live births (Lajeunie et al., 1995(Lajeunie et al., , 1996(Lajeunie et al., , 1998)), while multisutural synostoses are typically associated with a craniosynostosis syndrome with a genetic basis.These syndromes are characterized by a low prevalence rate of approximately 1.5 cases per 100,000 births for the most frequent syndromes, such as Crouzon and Apert syndromes (Cohen et al., 1992;Cohen Jr. & Krelborg, 1992;Martínez-Frías et al., 1991).They are often accompanied by other anomalies, including facial and limb deformities.Another group of multisutural craniosynostoses exists, exhibiting atypical features and lacking known associations with genetic mutations.These cases are referred to as "multisutural non-syndromic synostoses," also known as "complex craniosynostoses" (Czerwinski et al., 2011) and may have an unknown genetic etiology.
Secondary craniosynostoses arise because of another malformation or disease, or due to treatment for a separate condition, such as sagittal synostosis following hydrocephalus shunting.
Certain investigations from Europe, Australia, and the United States have documented an increase in the prevalence of craniosynostoses, particularly in relation to metopic craniosynostoses (Kweldam et al., 2011;Lee et al., 2012;Selber et al., 2008;van der Meulen et al., 2009).The European network of population-based registries for the epidemiological surveillance of congenital anomalies (EUROCAT) similarly reports an increased prevalence of craniosynostoses, attributing the trend to increased awareness and enhanced diagnostics (Morris et al., 2018).
To better understand the epidemiology of singlesuture, multisutural non-syndromic, and syndromic craniosynostosis in Finland, we conducted a nationwide population-based register study.Our analysis focused on prevalence data from 1987 to 2010, including 1,447,951 live births and a total of 1,453,780 births.

| Identification
We meticulously gathered information over a period of 24 years from 1987 to 2010 on all fetuses and infants in Finland, who were diagnosed or suspected to have craniosynostosis, craniofacial malformation, skull deformity, or craniosynostosis syndrome.For complete case ascertainment, we collected information from multiple sources to identify all affected fetuses and children with craniosynostosis.The collected data covered live births, stillbirths, and terminations of pregnancy due to fetal anomaly (TOPFA), with the diagnoses classified according to the 9th version of the International Classification of Diseases (ICD-9, Finnish modification) and the 10th version of the International Statistical Classification of Diseases and Related Health Problems (ICD-10, Finnish modification).In addition to the ICD-codes, we utilized procedure codes according to the national Classification of Surgical Procedures published by Sairaalaliitto, the Finnish Hospital Association, in 1983, as well as procedure codes according to the NOMESCO Classification of Surgical Procedures, in inpatient care from 1987 to 2013 in all hospitals in Finland, and specialist care outpatient care from 1998 to 2013, as recorded in the Hospital Discharge Register (1987)(1988)(1989)(1990)(1991)(1992)(1993) and the Care Register for T A B L E 1 ICD-10 (Finnish modification) and ICD-9 (Finnish modification) codes and procedure codes used in the data search.

ICD-10 ICD-9
Procedure codes NOMESCO (1996) Health Care (1994Care ( -2013) ) (Table 1).We collected the data from national registers up to year 2013 to find all those born in later years of our study period.
In addition, we sourced data from the Register of Congenital Malformations in Finland (RCM), and three other national health registers: the Medical Birth Register, the Register on Induced Abortions, and the Register on Small Premature Infants, all maintained by the Finnish Institute for Health and Welfare (Gissler & Haukka, 2004).The RCM is a comprehensive database that captures information on all live births, stillbirths, and fetuses from spontaneous abortions and TOPFAs, each collected case having at least one major congenital anomaly.The RCM collects data also from cytogenetic laboratories, the cause of death data maintained by Statistics Finland, and from the National Supervisory Authority for Welfare and Health, the latter of which nationally grants permission to all terminations of pregnancy due to major fetal anomalies or fetal and maternal diseases.
Furthermore, for complete ascertainment of identification of all cases, we also cross-checked all data from the national registers with the registers of Helsinki Regional Hospital Aurora (treating pediatric and craniosynostosis patients 1987-1994) and Helsinki University Hospital, which is one of the two university hospitals that provide surgical care for children with these malformations in Finland (Figure 1).The final study data were pseudonymized and all analyses were performed without unique identifiers.

| Screening
We combined the case data collected from various sources using the mother's and child's national personal identity codes.A craniofacial surgeon (PV) and a clinical geneticist of RCM (AR) reviewed the register data for all patients, including individual medical records and autopsy records in stillbirths, terminations of pregnancy, and infant deaths, to confirm and provide an accurate diagnosis.The surgeon and geneticist initially reviewed all the documents separately and then together, discussing and searching for more information if needed.The RCM has the right to gather information via individual medical records from all hospitals in Finland.Throughout the study period, it was recommended in Finland, that all patients with a suspected or confirmed diagnosis of craniosynostosis, including those with sagittal synostosis, undergo a genetic consultation with a clinical geneticist.Genetic testing was advised as necessary, though it was not routinely recommended for cases of sagittal synostosis alone.
The diagnostic criteria for craniosynostosis were based on clinical diagnosis and confirmation via radiologic examination, findings in surgery, autopsy examination, and/or cytogenetic examination.All the codes for co-occurring birth defects and diseases were reviewed, too.We included relatives of index patients only if they were treated and examined in the same manner as patients.We excluded patients with secondary craniosynostosis, deformational plagiocephaly, metopic ridge, mild skull deformities, or cases with other craniofacial malformations.We excluded also spontaneous abortions, as well as cases with insufficient information.Finally, we cross-linked the data with the Medical Birth Register and included children born in Finland from 1st of January 1987 to 31st of December 2010 in the study.Children born abroad were excluded, even if they were treated in Finland (Figure 1).The Hospital Discharge Register and the Care Register for Health Care found all the patients that we had identified directly for cross-checking from the registers of Helsinki University Hospital and Helsinki Regional Hospital Aurora, and whose diagnoses we had confirmed from the original documents.

| Statistical analysis
We used tests of relative proportions, chi-square tests, and Fisher's exact tests where appropriate, to analyze the data.Trend figures were calculated by using the least squares method in linear regression as illustrated in Figure 3.To represent statistical uncertainty, we calculated 95% confidence intervals with Microsoft Excel using the Poisson distribution (Uhari, 1992) and checking the results with the EUROCAT guide for calculation of prevalence and their 95% confidence intervals (https://eu-rd-platform.jrc.ec.europa.eu/sites/default/files).Depending on the rarity of the anomaly, we calculated the rates per 10,000 or 100,000 live births/live births + stillbirths + TOPFAs.

| Ethics approval
Helsinki University Hospital and the Finnish Institute for Health and Welfare approved the protocol of this retrospective study.In addition, Helsinki City (the data from Helsinki Regional Aurora Hospital) and Statistics Finland granted permission to use their sensitive health register data in this study.All work was conducted in accordance with the Declaration of Helsinki.

| RESULTS
We identified 877 craniosynostoses after reviewing the medical records of a total of 1878 patients.These 877 craniosynostoses included 864 live births, five stillbirths, and eight TOPFAs (Figure 2).

| Prevalence
During the period 1987-2010, the combined number of live and stillbirths was 1,453,780 in Finland.The total (LB + SB + TOPFA) prevalence of craniosynostosis from 1987 to 2010 was 6.0 per 10,000 (95% CI 5.6-6.5),ranging from 5.0 in 1987 to 8.0 in 2010.During the later years of F I G U R E 3 Total prevalence of all craniosynostoses per 10,000 including live births (LB), stillbirths (SB), and terminations of pregnancy for fetal anomaly (TOPFA) in Finland from 1987 to 2010.The data are presented with 95% confidence intervals (CIs) for each year.The trend line was calculated by using the least squares method in linear regression.The Y-axis represents the prevalence per 10,000 (LB this period (2005)(2006)(2007)(2008)(2009)(2010) the prevalence increased to 7.3 per 10,000 (95% CI 6.4-8.2) (Figure 3).The increase was statistically significant.All TOPFAs and stillbirths occurred in the syndromic and multisutural nonsyndromic groups and six out of eight TOPFAs and one stillbirth during the later years (2005)(2006)(2007)(2008)(2009)(2010).
The number of live births in the period 1987-2010 was 1,447,958 in Finland.The prevalence of craniosynostoses in live births from 1987 to 2010 was 6.0 per 10,000 (95% CI 5.6-6.4)ranging from 5.0 in 1987 (one case/2002 live births) to 7.5 per 10,000 in 2010 (one case/1334 live births).During the later years of this period (2005-2010), the prevalence was 7.1 per 10,000 live births (95% CI 6.3-8.0).

| Single-suture synostoses
The single-suture synostoses group comprised all the non-syndromic, single-suture synostosis patients, including patients with other unrelated minor or major anomalies.In this context, non-syndromic means that the patient did not have a diagnosed craniosynostosis syndrome.All patients with single-suture synostosis were born alive.Single-suture synostosis group comprised 83% (727/877) of all craniosynostosis patients, with sagittal synostosis being the most common in this group (568 patients, 65% of all cases).This was followed by metopic synostosis (80 patients, 9% of all cases), unicoronal synostosis (62 patients, 7% of all cases), and unilambdoid synostosis (17 patients, 2% of all cases) (Table 2).The prevalences of single-suture synostoses per year from 1987 to 2010 are presented in Figure 4.

| Craniosynostosis syndromes
Only craniosynostosis syndromes of genetic cause were included in the syndromic craniosynostosis group.Not all patients had been genetically tested, but all had typical clinical and radiological findings characteristic of certain syndrome types.Thanatophoric dysplasia (lethal skeletal dwarfism) (TD) is caused by a mutation in the FGF3 gene and some of them have a "cloverleaf" type of multisutural craniosynostosis (Langer et al., 1987).Those patients with TD syndrome and cloverleaf malformation were included in the syndromic craniosynostosis subgroup.Crouzon syndrome and Pfeiffer syndrome were classified in distinct subgroups.The diagnosis of Pfeiffer syndrome encompassed patients exhibiting typical clinical features in addition to craniosynostosis including more severe hand and foot anomalies than those observed in Crouzon syndrome, multiple associated congenital anomalies, and/or mutation in FGFR1 or typical FGFR2 mutation.Notably, there may be clinical overlap between Crouzon and Pfeiffer syndromes, and we could have combined these groups.

| Trend analysis
The trend analysis was conducted for all data and additionally conducted separately for live births.The time period 1987-2010 was divided into three parts for trend analysis (Table 5).A trend analysis of total prevalence (LB + SB + TOPFA) was conducted between the first period (1987)(1988)(1989)(1990)(1991)(1992) and the last period (2005)(2006)(2007)(2008)(2009)(2010).This showed a statistically significant increase in the prevalence of all craniosynostoses ( p = .003),and, separately, metopic craniosynostosis ( p = .043)and syndromic craniosynostosis ( p = .017).However, when evaluating all years, we observed no statistically significant change in metopic craniosynostosis, but there was a significant increase in all craniosynostoses (p = .042)and in the syndromic group ( p = .002).When evaluating live births across all periods, the only statistically significant rise was in the syndromic group (Table 6).
There was statistically significant rise in overall prevalence including live births, stillbirths, and TOPFAs across all years.This rise is attributed to increase in all craniosynostosis subgroups, albeit these individual increases were not statistically significant.Specifically, the prevalence of sagittal synostosis increased from 3.63 to 4.39 per 10,000 between the first and last period, metopic synostosis from 0.44 to 0.81 per 10,000, unicoronal from 0.39 to 0.45 per 10,000, unilambdoid from 0.08 to 0.17, multisutural non-syndromic from 0.44 to 0.53 per 10,000 and craniosynostosis syndrome from 0.55 to 1.0.Notably, the increase in craniosynostosis syndrome prevalence was independently statistically significant.
The same trend was observed in live births.The prevalence was increasing in all groups of craniosynostosis (live birth columns in Table 5.) but was individually significant only in the craniosynostosis syndrome group.
T A B L E 5 Absolute numbers and prevalence (per 10,000) of craniosynostoses in Finland for the whole period (1987-2010) and split into three shorter time periods : 1987-1992, 1993-2004, and 2005-2010. 1987-2010 1987-1992 1993-2004 2005 Of the 16 patients in the dataset with Muenke syndrome (Muenke et al., 1997), 12 were born alive during the last reviewed period (2005)(2006)(2007)(2008)(2009)(2010).This was a substantial increase in the proportion of this syndrome among all syndromic cases compared to the previous years (p = .005,Fisher's exact test) and explains the increase of the prevalence in the syndromic craniosynostosis group in live births.Patients with Muenke syndrome were all assessed for mutation in the FGFR 3 gene, confirming the diagnosis.

| DISCUSSION
Craniosynostosis encompasses various distinct conditions that result from the premature fusion of one or more skull sutures.In our study, we observed an increasing total prevalence (including live births, stillbirths, and TOPFAs) of craniosynostosis in Finland during the study period, with a rise from 5.0 cases per 10,000 in 1987 to 8.0 per 10,000 in 2010.This increase was statistically significant considering the total prevalence of all craniosynostoses and in the syndromic group alone.However, when looking at only live births, a significant increase was observed only in the syndromic group, caused by an increased number of Muenke patients in more recent years.Single-suture synostosis was the most common variant accounting 83% of all cases, with sagittal synostosis being as the most prevalent type of craniosynostoses (65% of all cases) followed in decreasing order, metopic (9%), unicoronal (7%), and lambdoid synostoses (2%).The prevalence of all single-suture synostosis subgroups showed an increase, although these increases were not statistically significant.The prevalence of unicoronal synostosis remained relatively stable across the study period, and we observed no statistically significant increase in the prevalence of metopic synostosis either.Craniosynostosis syndromes with known genetic causes comprised 10% of the cases, and multisutural, nonsyndromic synostoses accounted for 7%.
Finland is a high-income country with 5.5 million inhabitants.The public sector provides free or heavily subsidized health services and medical care, and all craniosynostoses are diagnosed and treated in public healthcare.Since 2006, surgical treatment for craniosynostosis has been centralized in two university hospitals-one in Helsinki and one in Oulu.Since the 1960s, Finland has had a national population registration system with personal identification numbers connecting a wide range of government-maintained national registers and even individual-level medical information.Public organizations govern these, and data can be obtained based on detailed applications, as done in our study.Several studies have found the National Register of Congenital Malformations in Finland to be of high quality regarding both its sensitivity and specificity (Koskimies et al., 2011;Pakkasjärvi et al., 2006;Syvänen et al., 2014).
Our study is notable for its inclusion of stillbirths, terminations of pregnancy for fetal anomalies, and patients who died before receiving treatment at a craniofacial center.It also benefits from national coverage via the use of different registers, data collection from treating hospitals to enable cross-checking of collected records, and confirmation of each individual diagnosis by a clinical geneticist and a craniofacial surgeon.Thus, having combined, confirmed and cross-checked data from multiple sources in the current study, we likely obtained reliable data for Finnish patients for the given study period.
Nevertheless, one limitation of our retrospective study is that some patients with mild craniosynostosis may have remained undiagnosed and, thus, may have never come to the attention of hospitals or have been recorded in registers.One difficulty of such a retrospective register study is the inaccuracy of the ICD-10 code system and, even more so, the ICD-9 code system for craniosynostosis; however, this was mitigated by confirming each diagnosis from the relevant individuals' medical records (Gonzalez et al., 2019).
We compared our study to four other studies focusing on live birth patients.Three of these studies were Birth Defect Register-based: two from Western Australia (Junaid et al., 2022;Singer et al., 1999) and the third from Atlanta, USA (Boulet et al., 2008).The fourth study was based on live birth patients born in the Netherlands and collected from the registry of the Dutch Association for Cleft Palate and Craniofacial Anomalies, with data cross- referenced with information from treating hospitals (Cornelissen et al., 2016).
Our findings of a prevalence of 5.5 per 10,000 live births (95% CI 4.8-6.3)during the first study period, from 1987 to 1992, and 5.7 (95% 5.1-6.2) during the second period, 1993-2004, align with the prevalence of 5.1 per 10,000 births in Western Australia from 1980 to 1994 and 4.3 per 10,000 live births from Atlanta, USA, from 1989 to 2003.Notably, both the Atlanta and Western Australia studies reported a high proportion of lambdoid synostoses in their data, which could be attributed to possible misclassification of infants with deformational plagiocephaly as having lambdoid synostosis (Boulet et al., 2008).In a recent study on craniofacial malformations utilizing the same Birth Defect Register in Western Australia for records from 1980 to 2010 (Junaid et al., 2022), the overall prevalence of all craniosynostoses was 5.3 per 10,000 births (including stillbirths), but likely still included cases of misclassified lambdoid synostoses.Interestingly, the prevalence of lambdoid synostoses increased from 0.8 to 1.5 per 10,000 births between 1990 and 1999 but then substantially decreased to 0.03 in the following decade; at the same time, the overall prevalence rose from 4.6 to 6.3 and then returned to 4.9.This suggests that diagnostic improvements may have led to a more accurate differentiation between true lambdoid synostosis and deformational plagiocephaly.
The prevalence of craniosynostosis in the Netherlands from 2008 to 2013 was 7.2 per 10,000 live births, which is similar to our prevalence of 7.1 per 10,000 live births from 2005 to 2010 (95% CI 6.3-8.0).However, while 11.5% of our patients had metopic synostosis during that period, the percentage in the Netherlands was 32% (Cornelissen et al., 2016).They found no statistically significant increase in prevalence of craniosynostoses during the study period from 2008 to 2013.In a previous study from the Netherlands, Kweldam et al. (2011) noted a significant increase in prevalence from 2.6 in 1997 to 6.4 in 2007 per 10,000 live births (Kweldam et al., 2011).
In Sweden, the treatment of craniosynostosis is centralized to Uppsala and Gothenburg University Hospitals, where 707 patients treated from 2012 to 2019 were included in a study (Tarnow et al., 2022).The prevalence in Sweden was 7.7 children per 10,000 live births, which is comparable to our study.
The prevalence of craniosynostosis syndromes in Finland from 1987 to 2010, at 0.53 per 10,000 live births, is comparable to the prevalence in the Netherlands (as reported by Cornelissen et al., 2016) and in Sweden (as reported by Tarnow et al., 2022), which were 0.5 and 0.6 per 10,000 live births, respectively.However, during the last period of our study, from 2005 to 2010, the prevalence in Finland in live births was notably higher at 0.84 per 10,000 live births due primarily to increase of Muenke patients.
In Norway, 386 patients treated at Oslo University Hospital from 2003 to 2017 exhibited an increasing prevalence of craniosynostoses from 3.3 to 5.5 per 10,000 live births (Tønne et al., 2020).However, these numbers are lower compared to the figures in Finland for the years 2005 to 2010.Neusel et al. (2018) collected 142 craniosynostosis patient records from treating hospitals from the state of Saxony-Anhalt in Germany and from the Malformation Monitoring Center of Saxony-Anhalt (including stillbirths and terminations of pregnancy for fetal anomaly) from 2000 to 2017.The live birth prevalence was 4.2 per 10,000 live births and total prevalence 4.8 per 10,000 births and TOPFAs, which are lower values than in our study during the periods 1993-2004 (total prevalence 5.7, 95% CI 5.1-6.2) and 2005-2010 (total prevalence 7.3, 95% CI 6.4-8.2) (Neusel et al., 2018).Lajeunie et al. (1998) estimated the prevalence of metopic synostosis at 0.7 per 10,000 live births in France.Kweldam et al. (2011) found that the prevalence of metopic craniosynostosis increased from 0.6 to 1.9 per 10,000 live births in the Netherlands from 1997 to 2007.Several European craniofacial centers, including those in the Netherlands, France, the UK, and Spain, have observed an increase in the prevalence of metopic synostosis in recent years (Di Rocco et al., 2009;van der Meulen et al., 2009).Specifically, in the multicenter study by van der Meulen et al. (2009), a total of 3240 patients collected from treating hospitals in Europe, were diagnosed with craniosynostosis; 41% of them had sagittal suture synostosis, while 23% had metopic synostosis.In this study, the increase in the prevalence of metopic synostosis was statistically significant, both in absolute terms and as a percentage.Among all craniosynostoses, 20.1% versus 25.5% in children born in 1997-2000 versus 2001-2005, respectively, were metopic synostoses.Meanwhile, there was a slight but statistically nonsignificant increase in the prevalence of sagittal synostoses in their study.The authors postulated that the increased prevalence might be explained by greater public awareness of craniosynostoses in general.Another possible factor behind this increase could be a trend toward higher paternal age; in the Netherlands, the proportion of fathers over 40 rose from 9% in 1990 to 11% in 2000.Maternal age was not discussed in their study.
The study from the USA, utilizing the Texas Birth Defect Registry, concentrating to risk factors and temporal trends on prevalence of isolated craniosynostosis from 1999 to 2014, observed an increase of birth prevalence for metopic synostosis: from 0.57 to 1.10 per 10,000 live births (Schraw et al., 2021).Birth prevalence of sagittal synostosis increased from 1.35 to 2.03 per 10,000 live births during the same time period.The prevalence of coronal synostosis increased from 1999 to 2007 and subsequently decreased from 2007 to 2014.However, there was no significant change in the prevalence of coronal synostosis over the entire study period in their study.We observed the same in our study in Finland from 1987 to 2010: no significant change in the prevalence of unicoronal synostosis.In live births of metopic synostosis in Finland from 1987 to 2010, there was an increase from 0.17 (95% CI 0.00-0.93) to 0.82 (95% CI 0.27-1.91)per 10,000 live births and in sagittal synostosis the prevalence rose from 3.35 (95% CI 2.05-5.17) to 4.41 (95% CI 2.91-6.42).
In our study, the increase in the prevalence of metopic craniosynostosis was not statistically significant.Instead, we identified a weak, albeit statistically nonsignificant trend when comparing data from the first and last study periods.Furthermore, paternal age is not recorded in the RCM, but both maternal age (Klemetti et al., 2016) and paternal age (Paavilainen et al., 2016) have been on the rise in Finland.It is possible that our diagnostic criteria for metopic craniosynostosis have been stricter than in other studies, and this may have resulted in some milder cases in our data being categorized as mild trigonocephalies within the normal variation range in Finland.
Interestingly, the prevalence of craniosynostoses in live births reported in several other studies from Australia, the USA, and Europe falls within the confidence intervals of our study.However, the distribution of craniosynostosis types in Finland differs from the findings in these studies, with a lower proportion of metopic synostoses (9%) and a higher proportion of sagittal synostoses (66%).Our study concluded in 2010, and the situation in Finland may have evolved since then (Table 7).
The reasons for the increased prevalence in live births in our study are obvious only in the craniosynostosis syndrome group due to a higher number of patients with Muenke syndrome in later years of our study period.This could be due to a combination of factors, including more routine genetic testing and potential demographic changes within the population.It is important to note that not all patients in our study underwent genetic assessment, particularly in the earlier years.Hence, it is possible that the Muenke syndrome would have been even more prevalent if all the unicoronal and bicoronal non-syndromic patients had been genetically assessed.Cornelissen et al. (2016) grouped multisutural non-syndromic patients with craniosynostosis syndromes in their study in the Netherlands.As research into the genetic basis of craniosynostoses progresses, and genetic testing becomes even more routine, this transition between patient groups is likely to increase in frequence.
Increased general awareness among both the population and healthcare professionals, along with advancements in diagnostic methods, may have contributed to the increasing prevalence live births observed in our study, albeit without statistical significance.Magnetic resonance imaging, three-dimensional computed tomography, and advanced genetic testing have become routine diagnostic tools during our study period enabling more precise diagnostics and identification of milder cases that may have previously gone undiagnosed.The centralization of diagnostic and treatment services for craniosynostosis in Finland may have had a similar effect.Additionally, the improved efficiency and automation of data collection from hospitals to national databases in Finland may have had a role in increasing prevalence in Finland.
Race and ethnicity can influence the prevalence of craniosynostosis (Sacks et al., 2019).While the Finnish population was historically homogenous and Caucasian, immigration and increased population heterogeneity over our study period may have had an impact.The number of people with foreign backgrounds living in Finland rose from 37,618 in 1990 to 237,066 in 2010 and births with foreign background rose from 117 in 1990 to 2908 in 2010 (Statistics Finland).Although race or ethnicity was not recorded in the RCM, this information was documented in the medical records of the patients included in our study.
We hypothesize that changes in maternal factors, such as increasing maternal age, may have influenced the total prevalence in our study.Maternal age has been steadily rising in Finland (Klemetti et al., 2016).Advanced maternal age is well established as a risk factor for chromosomal abnormalities, but research on its relationship with major congenital malformations in the absence of aneuploidy is limited (Goetzinger et al., 2016).However, in our current study, we did not specifically investigate the impact of advanced maternal age on the increased prevalence of craniosynostosis.Nevertheless, we have initiated a study utilizing this verified data to explore the demographic and risk factors associated with craniosynostosis, including ethnicity, maternal age, socioeconomic status, fertility treatments, smoking, and medication during pregnancy as well as regional and geographic variations in craniosynostosis.

| CONCLUSIONS
The prevalence of total craniosynostosis (including live births, stillbirths, and TOPFAs) has been increasing in Finland during the study period from 1987 to 2010.This rise is attributed to increase in all craniosynostosis subgroups, albeit these individual increases were not statistically significant except in subgroup of craniosynostosis syndromes.However, a significant increase in live births was seen only in the syndromic group, caused by an increased number of Muenke patients in more recent years.The prevalence of craniosynostosis in live births is in line with reported rates in other parts of Europe and aligns well with studies from Sweden and the Netherlands.Our study did not detect a similar rise in the prevalence of metopic synostosis in Finland as observed in the rest of Europe, Australia, and the USA.Additionally, while the proportion of sagittal synostosis is higher in Finland, the prevalence of metopic craniosynostosis is lower compared to other European countries.The reasons behind these differences are currently unknown and require further research.
This study is one of the few existing population-based studies published that includes data on all cases of craniosynostosis, including live births, stillbirths, pregnancy terminations for fetal reasons, and it is the only one, to our knowledge, to include data on deaths of persons having craniosynostosis diagnoses.The individual patient data were confirmed from medical records, ensuring the accuracy of our findings.
Our study adds precise knowledge of the prevalence of craniosynostoses in Finland to the existing literature, which is crucial for treatment planning, resource allocation, and treatment centralization in the country.Furthermore, this confirmed data can be used in future studies to identify risk factors, inheritance patterns, and associated anomalies, leading to improved treatment protocols and guidelines, and, thus, ultimately better outcomes for patients with craniosynostosis.

AUTHOR CONTRIBUTIONS
Data extraction was performed by PV and AR, and data analysis was performed by PV, AR, and MG.The manuscript was drafted by PV, NP, MG, AH, and ET.All authors read and approved the last version of the manuscript.

FUNDING INFORMATION
PV received funding for this study and its publication from the Department of Musculoskeletal and Plastic Surgery, Helsinki University Hospital and University of Helsinki, Helsinki, Finland.
U R E 1 Study flowchart.This diagram demonstrates the study's methodology and progression including the steps of identifying study data, screening the data, and determining the outcome, which is the absolute number of children and fetuses diagnosed with various types of craniosynostosis in Finland from 1987 to 2010.Abbreviations: termination of pregnancy for fetal anomaly (TOPFA), bilateral lambdoid synostosis with sagittal synostosis (BLSS).
Absolute numbers of fetuses and children diagnosed with craniosynostosis in Finland from 1987 to 2010.The data include terminations of pregnancy for fetal anomaly (TOPFA), stillbirths (SB), and live births (LB).The Y-axis represents the annual total numbers of fetuses and children diagnosed with craniosynostosis.
Results of chi square tests for an increase in prevalence for craniosynostosis during the period 1987-2010 in Finland.
T A B L E 6