Sibling risk of pediatric obstructive sleep apnea syndrome and adenotonsillar hypertrophy.

(1)

SLEEP, Vol. 32, No. 8, 2009

SLEEP DISORDERED BREATHING (SDB) IS A CONDI- TION CHARACTERIZED BY INTERMITTENT PARTIAL OR COMPLETE UPPER AIRWAY OBSTRUCTION THAT disrupts normal ventilation during sleep and normal sleep pat- terns.¹ There is no consensus on the definition of pediatric SDB, but clinically there is a spectrum of symptoms, in which the milder forms only have primary snoring and mouth-breathing, while the more severe forms have symptoms similar to the more defined entity obstructive sleep apnea syndrome (OSAS), i.e., intermittent breathing pauses (apneas), habitual snoring, snorts or gasps, disturbed sleep,¹ and daytime neurobehavio- ral problems with impaired school performance.² Complica- tions of OSAS include failure to thrive and, in severe cases, cor pulmonale and even death.¹ To diagnose OSAS, sleep studies are required; the gold standard is full-night polysomnography (PSG) with electroencephalography (EEG).¹ However, PSG is expensive, time-consuming, and not widely available; thus the term SDB is most often clinically used.

The correlation between clinical history of snoring and ap-he correlation between clinical history of snoring and apnea and the polysomnographic findings have been investigated.

One study showed that apneic episodes every night detected by parents were the most important single risk factor for OSAS.³ Further, previous adenoidectomy was a risk factor, suggesting that the epipharyngeal space does not play a central role in pedi-

atric OSAS.³ Another study showed that tonsillar hypertrophy was correlated to more severe apnea among pre-school-age children, but not among school-age children.⁴

The most important risk factor to SDB and OSAS is adenotonsillar hypertrophy.⁵ There are diverging results concerning if the severity of OSAS is proportional to the size of the tonsils and adenoids,⁶ or not.⁷ Other risk factors are obesity, craniofacial anomalies, and/or neuromuscular disorders etc.^8,9 Surgical removal of adenotonsillar hypertrophy is the first treatment op- tion in children.^8,10-12 Removal of adenoids from the nasopharynx alone may not be sufficient to treat OSAS and is mostly performed to treat primary snoring and mouth breathing.³

A recent epidemiologic review suggests that the prevalence of parent-reported symptoms of SDB on questionnaire is 4% to 11%; OSA diagnosed by varying criteria on diagnostic studies, 1% to 4%.¹³ The prevalence of SDB is higher among boys than girls, and among African Americans compared to other races.¹³ In addition, obesity and overweight showed to have different importance among older versus younger children, in whom en- larged tonsils are believed to play a major role in SDB risk.¹³ The prevalence of adenotonsillar hypertrophy in children is un- known.

Genetic studies of adults and children indicate that also hered- ity can be an important risk factor for OSA or SDB.^14-18 Genome scan of white adults in USA has been performed.¹⁴ Evidence for linkage to the apnea-hypopnea index (severity of OSA) was found in 2 chromosome regions. The Cleveland Family Study in the United States suggested that the angiotensin-converting enzyme (ACE) deletion allele may protect against hypertension in the setting of obstructive sleep apnea.¹⁵ Further shown in OSA children is that RNA derived from peripheral leuko- cytes confirms the presence of involved genes related to the

Sibling RiSk of PediatRic oSa

Sibling risk of Pediatric Obstructive Sleep Apnea Syndrome and Adenotonsillar Hypertrophy

Danielle Friberg, MD, PhD¹; Jan Sundquist, MD, PhD²; Xinjun Li, MD, PhD²; Kari Hemminki, MD, PhD^2,3; Kristina Sundquist, MD, PhD²

1Department of Otorhinolaryngology, Karolinska Institutet, Stockholm, Sweden; ²Centers for Family and Community Medicine, Karolinska Institutet, Stockholm, Sweden; ³Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany

objectives: To estimate sibling risk of hospitalization for children with sleep disordered breathing (SDB), diagnosed with (1) obstructive sleep apnea syndrome (OSAS), or (2) adenotonsillar hypertrophy in the total Swedish population.

design, Setting, and Participants: Using the MigMed database at the Karolinska Institute, we divided the population of Sweden aged 0–18 years into sibling groups based on a shared mother and father and presence of a primary hospital diagnosis of OSAS or adenotonsillar hypertrophy for each individual born between 1978 and 1986, during the follow-up period 1997–2004. Individuals with at least one affected sibling were identified and the incidence rates were computed, using standardized incidence ratios (SIRs) with 95% confidence intervals (CIs). Reference groups were boys and girls with unaffected siblings of 2 or more.

Results: After accounting for socioeconomic status, age, and geo- graphic region, boys with at least one sibling with OSAS had an in-

creased risk of having OSAS (SIR, 33.2; 95% CI, 16.5–64.8), and in girls the SIR was 40.5 (19.4–81.4). For hypertrophy of the tonsils or hypertrophy of the adenoids and tonsils the corresponding SIRs were 4.53 (3.0–6.8) for boys and 4.94 (3.3–7.4) for girls.

conclusions: The study indicate an increased sibling risk of sleep dis- ordered breathing in children, which may be due to heritable genes and/or shared environment such as increased awareness among family members or referring doctors. Caregivers should ask �arents if sib�Caregivers should ask �arents if siblings have similar symptoms, and thus offer them early treatment.

keywords: Heritability, population-based, sibling risk, sleep apnea, children

citation: Friberg D; Sundquist J; Li X; Hemminki K; Sundquist K. Sib- ling risk of pediatric obstructive sleep apnea syndrome and adenoton- sillar hypertrophy. SLEEP 2009;32(8):1077-1083.

Submitted for publication october, 2008 Submitted in final revised form April, 2009 Accepted for publication May, 2009

Address correspondence to: Dr. Danielle Friberg, ENT Department, Karo- linska University Hos�ital, Huddinge, Karolinska Institutet, CLINTEC, Stockholm, Sweden; Tel: +46-8-58581408; Fax: +46-8-7467551; E-mail:

danielle.friberg@karolinska.se

Sibling Risk of OSAS—Friberg et al

Downloaded from https://academic.oup.com/sleep/article-abstract/32/8/1077/2679665 by Uppsala Universitetsbibliotek user on 03 April 2019

(2)

SLEEP, Vol. 32, No. 8, 2009

inflammatory response.¹⁶ Apolipoprotein E (ApoE) epsilon4 allele was associated with not only increased odds of having SDB, but also with an increased risk for neurocognitive dysfunction,¹⁷ and polymorphisms involving more than one locus in the ApoE gene and its regulatory region were associated with OSA in Caucasian children.¹⁸

In Sweden, there is a special hospital diagnostic code for OSAS. However, it is seldom used for children as they only oc- casionally undergo sleep studies, i.e., polysomnography at hospitals, with measurements of breathing, pulse oximetry, and/or transcutaneous carbon dioxide, can be performed. SDB has no special diagnostic code and therefore, children with such symptoms and suspected OSAS get the code adenotonsillar hypertrophy instead. If a child previously has undergone adenoidectomy but still has symptoms, the code of hypertrophy of tonsils is used.

The construction of a large population-based patient register has led to the rapid development of genetic-epidemiological studies in Sweden.^19,20 In addition, a recently published large- scale population-based study from our group has found that, after accounting for socioeconomic status, age, geographic region, and period of diagnosis, adults with at least one sibling who had a hospital diagnostic code of OSAS had a 3-fold increased risk of getting diagnosed with the same code during 8 years.²¹

In the present study we included hospital data on all children in Sweden during the study period between 1997 and 2004, i.e., a total of 2.7 million individuals. The individual study population was siblings born in 1978 and onward. The use of hospital register data eliminated recall bias. Recall bias is a potential problem when conducting case-control studies of familial risk because patients in the case group are prone to report a positive family history of the disease or other health problem being studied.

The main aim of this study was to define the familial risk of hospitalization for primary diagnostic codes of (1) OSAS and/

or (2) suspected OSAS, with clinical signs of SDB and adenotonsillar hypertrophy, in Swedish children less than 19 years of age. A further aim was to determine whether there were any differences by age and gender.

MetHodS

MigMed Research database

Data used in this study were retrieved from the MigMed database, located at the Center for Family and Community Medicine at the Karolinska Institute in Stockholm. MigMed is a single, comprehensive database that has been constructed using several national Swedish data registers, including, but not limited to, the Total Population Register, the Multigeneration Register, and the Swedish Hospital Discharge Register (1986–

2004).^22-24 Information from the various registers in the database is linked at the individual level via the national 10-digit civic registration number assigned to each person in Sweden for his or her lifetime. Prior to inclusion in the MigMed database, civic registration numbers were replaced by serial numbers to ensure the anonymity of all individuals.

Because the database contains information from the Multi- generation Register, it is possible to link more than 10 million

index persons (persons born in or after 1932 and registered in Sweden any time since 1961) with their biological parents. The latest version of the Multigeneration Register, which has been incorporated in the MigMed database, includes supplementary data from church records on index persons domiciled in Sweden between 1947 and 1961, including information about biological parents, children, siblings, and adoptions. This population includes both first and second generation immigrants. We have no data on ethnicity or country of birth.

outcome Variables

The 10th revision of the International Classification of Dis- eases (ICD-10) was used to identify all first hospital admissions for the outcome variables: (1) OSAS, (2) hypertrophy of tonsils, (3) hypertrophy of adenoids and tonsils, or (4) only hypertrophy of adenoids, during the study period (1997–2004) in individuals aged 0–18 years. ICD-10 code G47.3 was used to define OSAS, J 35.1 to define hypertrophy of tonsils, and J 35.3 to define hypertrophy of both adenoids and tonsils. Also, J 35.2 to define hypertrophy of adenoids was used, but as such children often have milder forms of SDB and seldom have suspected OSAS, the results are only showed in figures, not in tables. Children with primary diagnosis of upper airway infections (acute tonsillitis, pharyngitis) were excluded in the present study.

Diagnostic codes at the individual level were retrieved from the Swedish Hospital Discharge Register in the MigMed database. Sweden has a social welfare system that includes public primary and hospital health care for all individuals. During the study period, most children with suspected OSAS would have been referred to an otorhinolaryngologic clinic by a primary health care or hospital physician. The patients in the present study were mainly on hospital wards because of symptoms of SDB and suspected OSAS; for surgical removal of adenotonsillar hypertrophy, or, in rare cases, hospitalized for sleep studies. There are only a few private hospitals in Sweden, and the Swedish Hospital Discharge Register includes data from these hospitals, as well as from the public hospitals. However, some older children undergo ambulant sleep studies and therefore are not included in the present study. In addition, the routines for the overnight studies vary between different regions and hospitals in Sweden, which we have partly accounted for in the present analysis (see below).

explanatory Variables

Explanatory variables included gender, age at first hospital diagnosis of the outcome variable in question, socioeconomic status (defined as family income), and geographic region of residence (i.e., in most cases geographic region of hospitalization). Family income was divided into 3 categories based on the income level registered by the taxation authorities.

Geographic region was divided into large cities (cities with a population of more than 200,000, i.e., Stockholm, Gothen- burg, and Malmö), Southern Sweden, and Northern Sweden.

Geographic region was included as an explanatory variable to adjust for possible differences between geographic regions in Sweden with regard to hospital admissions for the 3 outcome variables.

(3)

SLEEP, Vol. 32, No. 8, 2009 Statistical analysis

Using the individual-level data in the MigMed database, the entire pediatric population of Sweden was sorted into sibling groups (families) based on a shared mother and father. The database was then used to determine the presence or absence of a primary hospital diagnosis of pediatric OSAS or hypertrophy of the tonsils, or hypertrophy of the adenoids and tonsils, in each individual during the follow-up period. Next, individuals were categorized as positive or negative for sibling OSAS, or hypertrophy of the tonsils, or hypertrophy of the adenoids and tonsils, based on the presence of the disorder in at least one of their siblings. Individuals without siblings were excluded from the analysis. The individual serial numbers described in the section on the MigMed research database were used to check that those with hospital diagnoses of pediatric OSAS, or hypertrophy of the tonsils, or hypertrophy of the adenoids and tonsils appeared only once in the data set, i.e., for their first hospital diagnosis during the study period.

Person-years were calculated from the start of follow-up on January 1, 1997, to hospitalization for OSAS, or hypertrophy of the tonsils, or hypertrophy of the adenoids and tonsils, death, emigration, or the end of the study on December 31, 2004.

Age-specific incidence rates (defined as first hospitalization rates during the study period) were calculated for the whole follow-up period. The results are shown as standardized incidence ratios (SIRs) with 95% confidence intervals (CIs). SIRs were calculated as the ratio of the observed to the expected number of cases for age, gender, time period, region, and socioeconomic status. Sibling risks were calculated separately for boys and girls categorized as positive for the disease in question in at least one of their siblings compared with boys or girls characterized as negative for sibling OSAS, or hypertrophy of the tonsils, or hypertrophy of the adenoids and tonsils, using the cohort methods as described in a study by Hemminki et al.²⁵ Briefly, we defined a cohort of individuals with at least one affected sibling and computed the incidence rates in this cohort over the study period. In a family with two or more affected siblings, each affected individual is included in the cohort (as the sibling of an affected individual).

Relative weights used to calculate the age-standardized incidence rates were based on the European standard population.

We judged this standard population to be a precise choice because a study from 2002found that the 1960 standard popula-that the 1960 standard population (the Segi standard) was as precise as more recent standard populations in the calculation of age-standardized rates.²⁶

The test statistic χ² was used to calculate the probability (P value) of the SIR ratio between gender and age.

ethical considerations

This study was approved by the Ethics Committee of the Karolinska Institute, Stockholm, Sweden.

ReSUltS

A total of 854 boys and 627 girls aged 0–18 years had a first hospital diagnosis of pediatric OSAS during the study period (Table 1). The first hospitalization rates in the entire popula-

tion were 10.3 per 100,000 person-years for boys and 8.0 per 100,000 person-years for girls (significant gender difference, P = 0.014). The incidence rate of boys with a sibling history of OSAS was 544.8 and for girls 462.8 (no significant gender difference, P = 0.503).

There were 23 children (8%) diagnosed with pediatric OSAS who were also diagnosed with hypertrophy of the tonsils, or adenoids and tonsils; data not shown.

A total of 13,656 boys and 11,648 girls aged 0–18 years had a first hospital diagnosis of hypertrophy of the tonsils, or hypertrophy of adenoids and tonsils. The first hospitalization rates in the entire population were 165.0 per 100,000 person-years for boys and 148.8 per 100,000 person-years for girls (significant gender difference, P < 0.001). The incidence rate for boys with a sibling history of hypertrophy of the tonsils or hypertrophy of the adenoids and tonsils was 861.6 and, for girls, 812.6 (no significant gender difference, P = 0.185).

In Table 2, the SIRs for the first hospitalization for OSAS, hypertrophy of the tonsils, or hypertrophy of adenoids and tonsils, among children with ≥ 1 sibling with the disorders are shown by age groups. The models are adjusted for all the ex- planatory variables simultaneously. The overall SIR of OSAS among those who had ≥ 1 affected sibling was 33.23 (95% CI, 16.54–64.84) for boys and 40.49 (95% CI, 19.44–81.35) for girls, compared with the reference group of those without a sibling history of the disorder, no significant gender differences.

The overall SIR of hypertrophy of the tonsils, or hypertrophy of adenoids and tonsils, among those who had at least one affected sibling was 4.53 (3.01–6.79) for boys and 4.94 (3.27–7.44) for girls, compared with the reference group of those without a sibling history of these disorders, a significant gender difference, P = 0.049. Children with only hypertrophy of adenoids, had a SIR of 17.12 (95% CI, 8.33–34.04) in boys, and a SIR of 8.88 (95% CI, 3.33–21.5) in girls, not shown in the table.

Table 3 shows SIRs and the observed number with pediatric OSAS who had at least one sibling with pediatric OSAS by gender of the affected siblings. When a male sibling had pediatric OSAS, the risk (SIR) that a male sibling would also have the disorder was 30.80 (95% CI, 13.28–67.38) and that a female sibling would have the disorder was 36.65 (95% CI, 15.06–83.15). When a female sibling was affected the risk (SIR) that a male sibling would have OSAS was 38.83 (95% CI, 15.96–88.11) and that a female sibling would have the disorder was 42.70 (95% CI, 16.45–101.58).

In children with hypertrophy of the tonsils or hypertrophy of the adenoids and tonsils who had at least one sibling with hypertrophy of the tonsils or hypertrophy of the adenoids and tonsils, the results were: when a male sibling had the disorder, the risk (SIR) that a male sibling would also have it was 4.38 (95% CI, 2.86–6.72) and that a female sibling would have the disorder was 4.71 (95% CI, 3.04–7.28). When a female sibling was affected the risk (SIR) that a male sibling would have the disorder was 4.75 (95% CI, 3.06–7.33) and that a female sibling would have it was 5.15 (95% CI, 3.32–7.96).

There were no families with > 2 siblings affected with OSAS. However, there were families with > 2 siblings affected with hypertrophy of the tonsils or hypertrophy of the adenoids and tonsils. Boys and girls with no affected sibling were used as a reference. For 64 boys, the SIR was 96.63 (52.61–174.59) Sibling Risk of OSAS—Friberg et al

(4)

SLEEP, Vol. 32, No. 8, 2009

showed similar increases, as well as the ratios for boys and girls with one, two, or more affected siblings. There were significant gender differences in the first hospitalization rates in the entire study population with a higher rate for OSAS in boys than girls, as well as for hypertrophy of lymphoid tissue in boys compared to girls. This is in accordance with previous studies, which have shown that boys have a higher frequency of SDB than girls.¹³ There was no significant gender difference in the incidence rate of OSAS among those with a sibling history of OSAS, which was in contrast to adenotonsillar hypertrophy, in which a statis-adenotonsillar hypertrophy, in which a statis- tically significant gender difference was found; the SIR in girls was 4.9 compared to 4.5 in boys. However, this small gender difference is probably of minor clinical importance.

The results of the current study, with sibling aggregation of pediatric SDB are consistent with those of previous studies, i.e., the Cleveland Family Study investigated predictors of sleep disordered breathing in 577 children with a mean age of 10 years.²⁷ Ten percent of the children had at a mean of 5 years previously undergone tonsillectomy and/or adenoidectomy. An independent risk factor for SDB in these children was, besides obesity and black race, a high familial risk (related to an affected proband). It suggested that there are both shared and unshared genetic factors underlying susceptibility to OSA and obesity and that the genetic determinants of obesity may be modulated by the severity of apnea.

A recently published study by our group on adult OSAS was conducted using the same methods as in the present study.²³ We showed an increased risk in males with at least one sibling who had OSAS with a SIR of 3.4 in males and 3.2 in females for having OSAS. To the best of our knowledge, these are the only for having the disorder and, for 47 girls, it was 77.92 (40.47–

146.63); there was no significant gender difference, P = 0.26;

data not shown in table.

We also performed an additional analysis, to calculate age differences, in order to investigate the effect of environmental factors on sibling risk. Large age differences indicate less shared environment and vice versa. The children were divided in 2 groups: < 4 years old and ≥ 4 years old. There was no significant age difference in pediatric OSAS between 25 younger boys, SIR of 41.18 (18.82–86.08), and 12 older boys, SIR of 23.71 (8.62–58.75), P = 0.11. For the few number of girls with OSAS, there was a significant age difference between 23 younger girls, SIR of 56.33 (25.22–119.71), and 8 older girls, SIR of 22.38 (6.76–62.68), P = 0.02. However, in children with hypertrophy of the tonsils or hypertrophy of the adenoids and tonsils, there were only small, non-significant age differences;

data not shown in table.

diScUSSion

The sibling risk of pediatric OSAS was extremely high in the present study, with an incidence ratio of 33.2 in boys and 40.5 in girls in children who had at least one sibling with an OSAS diagnosis. We also included children with SDB and suspected OSAS with adenotonsillar hypertrophy, which is the main cause of pediatric OSAS.⁵ The familial risk of OSAS was much higher than in the group with adenotonsillar hypertrophy, the incidence ratio being 4.5 in boys and 4.9 in girls. However, the increase was highly significant and the numbers of children were much larger than in OSAS. The gender-specific ratio

Table 1—Age-Specific Incidence Rates of Obstructive Sleep Apnea Syndrome, Hypertrophy of Tonsils and Hypertrophy of Adenoids and Tonsils Per 100,000 Person-Years in Boys and Girls Aged 0–18 Years

Boys Girls

Entire population With sibling

history of OSAS Entire population With sibling

history of OSAS Age at

diagnosis (y) No. Incidence

rates No. Incidence

rates OSAS

0–6 601 21.3 28 941.2 418 15.6 22 747.5

7–12 187 6.5 8 324.4 151 5.5 6 245.2

13–18 66 2.6 1 74.1 58 2.4 3 229.4

All 854 10.3 37 544.8 627 8.0 31 462.8

Hypertrophy of

tonsils Entire population With sibling history of hypertrophy of tonsils or hypertrophy of adenoids

and tonsils

Entire population With sibling history of hypertrophy of tonsils

or hypertrophy of adenoids and tonsils or hypertrophy

of adenoids and tonsils

No. Incidence

rates No. Incidence

rates

0–6 8896 315.2 797 1526.5 6186 231.2 638 1299.3

7–12 3745 130.0 254 533.8 3626 132.9 263 578.8

13–18 1015 39.5 36 136.5 1836 75.8 69 277.9

All 13656 165.0 1087 861.6 11648 148.8 970 812.6

(5)

SLEEP, Vol. 32, No. 8, 2009

the indication of tonsillectomy mainly was because of recurrent and/or chronic tonsillitis. We know from the diagnostic codes that the children in the present study underwent tonsillectomy primarily because of OSAS and adenotonsillar hypertrophy, and not tonsillitis. However, we cannot rule out the possibility that the children secondarily also had an acute infection, which may have worsened their SDB symptoms. The heritability for recurrent and/or chronic tonsillitis and hypertrophy of lymphoid tissue is probably different, which makes it difficult to compare the results. Unfortunately, we did not have enough statistical power to perform a twin study of the Swedish children. How- ever, recent genetic studies of both adult and pediatric patients studies that have investigated family risks in siblings with a his-

tory of medically verified OSAS or in children with adenotonsillar hypertrophy.

The increased risks for OSAS and also the main causes of pediatric OSAS, adenotonsillar hypertrophy, could be an expression of genetic or shared environmental mechanisms. However, the results from the present study cannot differ between these mechanisms. The influence of heritability has been investigated in a twin-study from 1991, using questionnaires.²⁸ The persons had undergone childhood tonsillectomy from 1939 to 1964 and the heritability factor varied between 29% to 82%. Their study has a recall bias and was performed during a time era when

Table 2—Standardized Incidence Ratios (SIRs) and Observed Number of Cases of Obstructive Sleep Apnea Syndrome (OSAS), Hypertrophy of Tonsils and Hypertrophy of Adenoids and Tonsils in Boys and Girls Aged 0–18 Years with at Least One Sibling with the Disease in Question

Sibling history Age at Boys Girls Gender

diagnosis difference

(years) O SIR 95% CI O SIR 95% CI P value

OSAS All 37 33.64 15.79 68.83 31 40.49 19.44 81.35 0.42

0–6 28 32.85 9.92 92.00 22 39.90 17.65 85.56

7–12 8 26.74 0.01 216.8 6 36.52 9.29 113.15

13–18 1 33.23 16.54 64.84 3 60.00 8.00 251.18

Hypertrophy of tonsils All 337 4.38 2.78 6.89 296 4.36 2.74 6.91 0.95

0–6 221 4.97 3.06 8.01 171 5.77 3.49 9.48

7–12 92 3.65 2.08 6.33 85 3.47 1.96 6.07

13–18 24 3.33 1.51 7.02 40 2.90 1.46 5.59

Hypertrophy of

adenoids and tonsils All 750 4.59 3.02 6.98 674 5.24 3.43 8.00 0.013

0–6 576 4.92 3.20 7.55 467 5.73 3.69 8.88

7–12 162 3.87 2.33 6.39 178 4.49 2.73 7.36

13–18 12 2.78 1.01 6.89 29 3.88 1.84 7.90

Hypertrophy of tonsils or hypertrophy of adenoids

and tonsils All 1087 4.53 3.01 6.79 970 4.94 3.27 7.44 0.049

0–6 797 4.93 3.25 7.48 638 5.74 3.75 8.78

7–12 254 3.79 2.36 6.06 263 4.10 2.56 6.55

13–18 36 3.12 1.55 6.12 69 3.25 1.79 5.81

O = observed number of cases; SIR = standardized incidence ratio; CI = confidence interval. Bold type: 95% CI does not include 1.00.

Table 3—Gender-Specific Standardized Incidence Ratios and Observed Number of Cases of Obstructive Sleep Apnea Syndrome and Hyper- trophy of Tonsils or Hypertrophy of Adenoids and Tonsils

OSAS Male sibling with OSAS Female sibling with OSAS

O SIR 95% CI O SIR 95% CI

Boys 20 30.80 13.28 67.38 17 38.83 15.96 88.11

Girls 17 36.65 15.06 83.15 14 42.70 16.45 101.58

Hypertrophy of tonsils or Male sibling with hypertrophy of tonsils Female sibling with hypertrophy of tonsils hypertrophy of adenoids and tonsils or hypertrophy of adenoids and tonsils or hypertrophy of adenoids and tonsils

O SIR 95% CI O SIR 95% CI

Boys Hypertrophy of tonsils 188 4.35 2.65 7.09 155 4.42 2.65 7.31

Hypertrophy of tonsils and adenoids 410 4.40 2.82 6.86 335 4.91 3.11 7.73

All 598 4.38 2.86 6.72 490 4.75 3.06 7.33

Girls

Hypertrophy of tonsils 149 4.43 2.65 7.35 141 4.30 2.56 7.17

Hypertrophy of tonsils and adenoids 340 4.85 3.08 7.63 339 5.61 3.56 8.83

All 489 4.71 3.04 7.28 480 5.15 3.32 7.96

O = observed number of cases; SIR = standardized incidence ratio; CI = confidence interval

(6)

SLEEP, Vol. 32, No. 8, 2009

lar signs or symptoms in siblings. If the answers are positive, these children should be sent to an ear nose throat specialist for further examination and treatment.

abbReViationS

OSAS - obstructive sleep apnea syndrome SDB - sleep disordered breathing

SIR - standardized incidence ratio CI - confidence interval

acknowledgMentS

Financial support: This study was supported by grants from the Swedish Research Council (K2005-27X-15428-01A, the Swedish Council for Working Life and Social Research (2005- 0039, 2006-0386 and 2007-1754) and The Swedish Research Council Formas (2006-4255-6596-99 and 2007-1352).

diScloSURe StateMent

This was not an industry supported study. The authors have indicated no financial conflicts of interest.

RefeRenceS

1. American Academy of Pediatrics. Clinical practice guidelines:

diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics 2002;109:704-12.

2. Kaemingk KL, Pasvogel AE, Goodwin JL, et al. Learning in children and sleep disordered breathing: findings of the Tucson Chil- dren’s Assessment of Sleep Apnea (tuCASA) prospective cohort study. J Int Neuropsychol Soc 2003;9:1016-26.

3. Nieminen P Tolonen U, Löppönen H, et al. Snoring, children: factors predicting sleep apnea. Acta Otolaryngol Suppl 1997;529:190-4

4. Valera FCP, Avelino MAG, Pettermann MB, et al. Otolaryngol Head Neck Surg 2005; 132:268-72.

5. Davidson W, Sally L, Marcus CL. Obstructive sleep apnea in in- fants and young children. J Clin Neurophysiol 1996;13:198-207.

6. Arens R, McDonough JM, Costarino AT, et al. Magnetic reso- nance imaging of the upper airway structure of children with obstructive sleep apnea syndrome. Am J Respir Crit Care Med 2001;164:698-703.

7. Laurikainen E, Erkinjuntti M, Alihanka J, et al. Radiological parameters of the bony nasopharynx and the adenotonsillar size compared with sleep apnea episodes in children. Int J Pediatr Otorhinolaryngol 1987;12:303-10

8. Shine NP, Coates HL, Lannigan FJ. Obstructive sleep apnea, mor- bid obesity, and adenotonsillar surgery: A review of the literature.

Int J Pediatr Otorhinolaryngol 2005;69:1475-82.

9. Erler T, Paditz E. Obstructive sleep apnea syndrome in children:

a state-of-the-art review. Treat Respir Med 2004;3:107-22.

10. Zucconi M, Strambi LF, Pestalozza G, et al. Habitual snoring and obstructive sleep apnea syndrome in children: effects of early tonsil surgery. Int J Pediatr Otorhinolaryngol 1993; 26:235-43 11. Tarasiuk A, Simon T, Tal A, et al. Adenotonsillectomy in children

with obstructive sleep apnea syndrome reduces health care utili- zation. Pediatrics 2004;113:351-6

12. Mitchell RB, Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and postoperative polysomnography. Laryngoscope 2007;117:1844-54

with OSAS indicate that genetic mechanisms do play an important role.^14-18 Future studies will probably clarify this.

Possible environmental factors are the increased medical awareness of SDB over time, both among parents and doctors.

It is likely that parents who have one child with symptoms of SDB are more prone to seek evaluation for a sibling, even if the symptoms in the sibling are milder. Also the referring doc-Also the referring doctors may influence the sibling incidence, as the awareness and personality differs between doctors. There were only small age differences in the present study which indicate that the children had more shared environment.

There are weaknesses in the present study. Firstly, there are no strict definitions of adenotonsillar hypertrophy and SDB. In addition, few children undergo polysomnography or other sleep studies in Sweden, which make the diagnosis more uncertain.

On the other hand, the use of only hospitalization data (leading either to sleep studies or adenotonsillectomy), suggest that we have included children with more severe symptoms of SDB, and suspected OSAS. Children with only hypertrophy of adenoids often have milder symptoms of SDB, and therefore did we only present results from those in figures and not in the tables. On the other hand, the increased sibling risks were similar in all investigated groups. Secondly, children with milder forms of SDB more often undergo surgery on both private and public ambulatory wards without hospitalization. The fact that we have only calculated the SIR of siblings with hospital diagnosis and compared it with a group of siblings without hospital diagnosis means that we can have missed children with milder forms of non-hospitalized OSAS and/or adenotonsillar hypertrophy. However, this is a non-differential bias, as it most likely affected both investigated groups equally. Thirdly, we had no data on individual risk factors for pediatric OSAS, such as obesity or craniofacial abnormalities.

In a register that includes an entire population, it is not feasible to include individual data on weight, height, ethnicity and other individual risk factors for OSAS. However, the children were primarily hospitalized for OSAS, or adenotonsillar hypertrophy, the main etiological factors to OSAS, which increases the possibility that the diagnoses are valid. Fourthly, our study was not able to take account of early environmental factors, such as nutritional status early in life.

The present study has several strengths: the study population included a well-defined open cohort, the entire pediatric population of Sweden. Additionally, the data in the Swedish Hospital Discharge Register are very complete. In 2001, the main diag-In 2001, the main diagnosis was missing in 0.9% and the national civic registration number in 0.4% of hospitalizations.²² Finally, the quality of the multigenerational part of the MigMed database is very high, and it includes information about parents, children, siblings, and adoptions for index persons born in 1932 and onward and domiciled in Sweden at any time between 1947 and 2004.

The experience and knowledge of pediatric OSAS varies markedly in society at large, among both the population and caregivers. It is a serious disorder, with impairments both in the quality of life and in cognitive functions, as well as a significant risk of cardiovascular complications.¹ Environmental or heredi- tary risk factors may explain the observed familial clustering seen in the present study.

Caregivers who investigate children with clinical signs or symptoms of SDB should ask their family members about simi-

(7)

SLEEP, Vol. 32, No. 8, 2009

22. Rosen M, Hakulinen T. Use of disease registers. In: Ahrens W, Pi- geot I, eds. Handbook of epidemiology. Berlin: Springer-Verlag, 23. Statistics Sweden. The Swedish MultiGeneration Register (1960-2005

1990): http://www.scb.se/default_2154.asp (In Swedish: Regis- tret över totalbefolkningen/RTB), 2005

24. The National Board of Health and Welfare. The Swedish Hospital Discharge Register and the Cause of Death Register (1961-2001).

http://www.socialstyrelsen.se/en/.

25. Hemminki K, Vaittinen P, Dong C, et al. Sibling risks in cancer: clues to recessive or X-linked genes? Br J Cancer 2001;84:388-391

26. Bray F, Guilloux A, Sankila R, et al. Practical implications of im- posing a new world standard population. Cancer Causes Control 2002;13:175-182

27. Morton S, Rosen C, Tishler P, Aylor J, Redline S. Predictors of sleep-disordered breathing in children with a history of tonsillectomy and/or adenoidectomy. Sleep 2001; 24:823-9.

28. Martin NG, Kehren U, Battistutta D, Mathews JD. Iatrogenic influences on the heritability of childhood tonsillectomy: cohort differences in twin concordance. Acta Genet Med Gemellol 1991;40:165-72.

13. Lumeng JC, Chervin RD. Epidemiologigy of Pediatric Obstruc- tive Sleep Apnea. Proc Am Thorac Soc 2008;5: 242-52.

14. Palmer LJ, Buxbaum SG, Larkin EK, et al. A whole-genome scan for obstructive sleep apnea and obesity. Am J Hum Genet 2003;72:340-50.

15. Patel SR, Larkin EK, Mignot E, et al. The association of angiotensin converting enzyme (ACE) polymorphisms with sleep apnea and hypertension. Sleep 2007;30:531-3.

16. Khalyfa A, Capdevila OS, Buazza MO, et al. Genome-wide gene expression profiling in children with non-obese obstructive sleep apnea. Sleep Med 2008;10:75-86

17. Gozal D, Capdevila OS, Kheirandish-Gozal L, et al. APOE epsilon 4 allele, cognitive dysfunction, and obstructive sleep apnea in children. Neurology 2007;69:243-9

18. Kalra M, Pal P, Kaushal R, et al. Association of ApoE genetic variants with obstructive sleep apnea in children. Sleep Med 2008;3:260-5.

19. Hemminki K, Li X, Johansson SE, et al. Familial risks for mi- graine and other headaches among siblings based on hospitalizations in Sweden. Neurogenetics 2005;6:217-24.

20. Li X, Hemminki K. Cancer risks in men who had children with different partners from the Swedish Family-Cancer Database. Eur J Cancer Prev 2003;12:355-8

21. Sundquist J, Li X, Friberg D, Hemminki K, Sundquist K. Ob- structive sleep apnea syndrome in siblings: An 8-year Swedish follow-up study. Sleep 2008:31:817-23.