Long-term effects of adenotonsillectomy in children with attention deficit hyperactivity disorder

Abstract

Study Objectives:

Adenotonsillectomy (AT) improves short-term symptoms of attention deficit hyperactivity disorder (ADHD) in children; however, its long-term effects remain unclear. We aimed to verify the therapeutic long-term effects of AT in children with ADHD.

Methods:

This retrospective control study included children ages < 18 years who were diagnosed with ADHD and receiving ADHD medications. Participants were divided into groups depending on whether AT was performed (AT [+] or AT [–] groups) and matched 1:1 for age, sex, and year and month of diagnosis using randomized nonreplacement selection.

Results:

Among patients with ADHD (n = 171,112), 3,615 underwent AT. In both groups, the number of drugs taken gradually increased before and decreased after the AT date (ATD). There was no difference in the number of drugs used before (P = .88) and after ATD (P = .06). Before ATD, the average number of outpatient visits (nOV) did not change in both groups (AT [+]: P = .12; AT [–]: P = .71). After ATD, the average number of outpatient visits decreased only in the AT (+) group (P = .001). However, there was no difference in the average number of outpatient visits between the two groups before (P = .47) and after ATD (P = .17). Before ATD, methylphenidate doses between the groups were not different (P = .06); however, a significant increase was noted after ATD in the AT (+) group (P < .001).

Conclusions:

AT does not result in significant long-term therapeutic effects in terms of medication use and health care utilization in children with ADHD.

Citation:

Lee J, Choi A, Kim S, Kim K. Long-term effects of adenotonsillectomy in children with attention deficit hyperactivity disorder. J Clin Sleep Med. 2024;20(5):727–733.

BRIEF SUMMARY

Current Knowledge/Study Rationale: To confirm the long-term therapeutic effects of adenotonsillectomy, this retrospective study included children with attention deficit hyperactivity disorder and compared those who had undergone adenotonsillectomy with those who had not. Unlike previous studies, changes in medication dosage and health care utilization were identified to confirm the therapeutic effects of adenotonsillectomy.

Study Impact: No significant long-term differences between the two groups were found in medication dosage or health care utilization after adenotonsillectomy in children with attention deficit hyperactivity disorder. This study represents a unique endeavor to investigate the long-term effects of adenotonsillectomy in pediatric patients with attention deficit hyperactivity disorder through an extensive cohort, yielding differing results from prior research that focused on the short-term effects of adenotonsillectomy.

INTRODUCTION

Attention deficit hyperactivity disorder (ADHD), characterized by inattention, hyperactivity, and impulsivity, is a relatively common neurodevelopmental disorder in childhood, with symptoms beginning mainly before the age of 12. The prevalence of ADHD in children varies among studies; however, it is generally reported to be approximately 5%–8% in children, which is more than twice as that in adults.^1–3 In Korea, the prevalence of ADHD in children and adolescents increased from 533.3 per 100,000 in 2008 to 783.4 per 100,000 in 2018.⁴

ADHD has been associated with sleep-disordered breathing (SDB),⁵ which occurs at a high rate in patients with ADHD. Similarly, behavioral problems such as inattention and hyperactivity, which correspond to the diagnostic criteria of ADHD, were often observed in patients with SDB.^6,7

Adenotonsillectomy (AT) is the most effective treatment for SDB.⁸ Several studies have examined the association between AT and changes in ADHD symptoms in patients with SDB and ADHD.^8–13 In these studies, patients showed improvement in ADHD-related behavioral issues as well as sleep problems after AT. The results of these studies suggest that AT may help improve behavioral symptoms in patients with ADHD and sleep problems.

However, previous studies included small sample sizes,^9–16 short observation periods,^{8,10,11,13–15,17} and self-reported assessments of the therapeutic effect of AT on ADHD through surveys or questionnaires.^8,10–17 To overcome these limitations, we used long-term health insurance claims data to validate the positive effects of AT on ADHD symptoms.

METHODS

Data source

We used the health insurance claims data of the Health Insurance Review and Assessment Service in Korea that were collected from January 1, 2007, to December 31, 2020.

Study participants

The working definition of a patient with ADHD in our study was a patient under the age of 18 who was newly diagnosed with ADHD and had been prescribed ADHD medication between January 1, 2008, and December 31, 2018. The diagnosis of ADHD was confirmed using the F90.0 code of the Korean Standard Classification of Diseases, which is a modification of the International Statistical Classification of Diseases and Related Health Provisions to suit the situation in Korea. The ADHD medications included central nervous system stimulants (methylphenidate) or nonstimulants (atomoxetine and clonidine). The prescription period, prescription dose, action code corresponding to AT, and date of AT were included in the study data.

Patients newly diagnosed with ADHD were defined as those diagnosed between 2008 and 2018 and with no diagnosis or no medication prescribed for ADHD in the 1-year lookback period. The patients newly diagnosed with ADHD with no prescription of any medications for ADHD were excluded (n = 62,779). Patients who underwent AT before an ADHD diagnosis were also excluded (n = 19,393). Finally, patients were divided into an intervention group that underwent AT and a control group that did not; these groups were matched at a 1:1 ratio for age, sex, year, and month of ADHD diagnosis using randomized nonreplacement selection.¹⁸

From January 1, 2008, to December 31, 2018, 171,112 patients under the age of 18 years were newly diagnosed with ADHD and prescribed medication at least once. Among them, 3,621 patients underwent AT, and 3,615 patients were matched in a 1:1 ratio with the control group. An equal number of patients from the 161,582 patients who did not undergo AT were included as a control group (Figure 1).

Figure 1. Study population.

Children newly diagnosed with ADHD and on medication from 2008 to 2018 were classified into two groups, and the patients were matched 1:1 based on age, sex, and the year and month of ADHD diagnosis. *Any ADHD medication, either central nervous system stimulants (methylphenidate) or nonstimulants (atomoxetine or clonidine). ADHD = attention deficit hyperactivity disorder.

Study outcomes

The outcomes included changes in the average number of medications for ADHD, the average dose of methylphenidate, and the average number of outpatient visits before and after AT. Variables were aggregated by month. The date of AT (ATD) of a participant in the intervention group was presumed to be the ATD of a matched counterpart in the control group.

Statistical analysis

We used Pearson’s chi-squared test and Fisher’s exact test to analyze categorical values and Student’s t test to investigate quantitative data. We performed an interrupted time series analysis¹⁹ to determine the effects of AT. SAS Enterprise Guide 7.1^® (SAS Institute Inc., Cary, NC, USA) and R version 4.0.0 (R Foundation for Statistical Computing, Vienna, Austria) were used for all statistical analyses.

Ethics statement

This study was approved by the Institutional Review Board of Eunpyeong Saint Mary’s Hospital (PC22ZISI0099). The study data were obtained from the Health Insurance Review and Assessment Service database (M20210818443). This study was granted an exemption from the requirement for informed consent considering its retrospective nature.

RESULTS

Characteristics of the study participants

Patient characteristics are summarized in Table 1. The median age of the patients was 8 years (interquartile range, 6.0–11.0), and 79.3% were male. The median age of patients undergoing AT was 11 (interquartile range, 8.0–15.0) years. Methylphenidate was the most prescribed medication in both groups. There were no differences in the duration of medication use or follow-up period between the two groups. Only diagnoses associated with AT (adenotonsillar hypertrophy, chronic tonsillitis/adenitis, sleep apnea, and snoring) were significantly more frequent in the intervention group.

Table 1.

Clinical characteristics of the study participants.

Characteristics	Intervention Group (n = 3,615)	Control Group (n = 3,615)	P
Age of diagnosis			–
0–5 (preschool)	489 (13.5)	489 (13.5)
6–12 (school)	2,235 (61.8)	2,235 (61.8)
13–18	891 (24.6)	891 (24.6)
Sex			–
Male	2,868 (79.3)	2,868 (79.3)
Medication
Methylphenidate	3,391 (93.8)	3,411 (94.4)	.344
Atomoxetine	1,101 (30.5)	1,013 (28.0)	.024
Clonidine	125 (3.5)	83 (2.3)	.004
Use of medication duration (years)	0.9 [0.2–2.5]	0.9 [0.2–2.7]	.352
Follow-up period (years)	2.6 [0.5–5.9]	2.5 [0.5–5.7]	.511
Diagnostic code-related SDB
Adenotonsillar hypertrophy	2,827 (78.2)	168 (4.6)	<.001
Chronic tonsillitis/adenitis	2,054 (56.8)	428 (11.8)	<.001
Sleep apnea	202 (5.6)	10 (0.3)	<.001
Snoring	216 (6.0)	14 (0.4)	<.001

Data are presented as number (%) or median [interquartile range]. SDB = sleep-disordered breathing.

Effectiveness of AT in patients with ADHD

In the intervention group, the average number of prescribed medications gradually increased significantly before ATD and decreased after ATD (before, 0.007, 95% confidence interval [CI] 0.006 to 0.008; P < .001; after, –0.009, 95% CI –0.011 to –0.008; P < .001); this pattern was the same in the control group (before, 0.007, 95% CI 0.006 to 0.008; P < .001; after, –0.011, 95% CI –0.012 to –0.01; P < .001) (Figure 2A). However, there was no significant difference between the two groups (before ATD: –0.0001, 95% CI –0.0016 to 0.0014; P = .88; after ATD: –0.0015, 95% CI –0.003 to 0.001; P = .06) (Figure 2B).

Figure 2. Changes in drug use in patients with ADHD.

(A) Monthly number of prescribed ADHD medications. (B) Difference in the monthly number of prescribed ADHD medications between the AT (+) and AT (–) groups. T₀ is the date of adenotonsillectomy, indicated by the dotted line. ADHD = attention deficit hyperactivity disorder; AT = adenotonsillectomy.

The changes in the average number of outpatient visits before and after ATD are shown in Figure 3A. Before ATD, there was no change in either group (intervention group, 0.007, 95% CI –0.002 to 0.015; P = .12; control group, –0.002, 95% CI –0.013 to 0.009; P = .71). After ATD, the average number of outpatient visits decreased in the intervention group (–0.017, 95% CI –0.026 to –0.008; P = .001); however, there was no significant change in the control group (–0.007, 95% CI –0.018 to 0.004; P = .22). In addition, there was no significant difference in the average number of outpatient visits between the two groups before (0.002, 95% CI –0.003 to 0.007; P = .47) and after ATD (–0.004, 95% CI –0.009 to 0.002; P = .17) (Figure 3B).

Figure 3. Changes in the number of outpatient visits in patients with ADHD.

(A) Number of outpatient visits per month in the AT (+) and AT (–) groups. (B) Difference in the number of outpatient visits between the AT (+) and AT (–) groups. T₀ is the date of adenotonsillectomy, indicated by the dotted line. ADHD = attention deficit hyperactivity disorder; AT = adenotonsillectomy.

The numbers of patients who were prescribed methylphenidate at least once between 2008 and 2018 were 3,391 (93.8%) and 3,411 (94.4%) in the intervention and control groups, respectively. As shown in Figure 4A, the number of patients prescribed methylphenidate gradually increased in both groups before ATD. In the intervention group, the dose did not show a significant change before ATD (–3.54 mg, 95% CI –9.17 to 2.09; P = .21), and although the number of patients prescribed methylphenidate gradually decreased after ATD, the dose gradually increased (17.19 mg, 95% CI 11.26 to 23.12; P < .001). In the control group, no significant changes in dose of methylphenidate were observed (before ATD, 4.10 mg, 95% CI –9.75 to 17.95; P = .55; after ATD, 5.28 mg, 95% CI –9.31 to 19.87; P = .47). As shown in Figure 4B, there was no significant difference between the two groups before ATD (–6.44 mg, 95% CI –13.11 to 0.23; P = .06); however, the difference in dose of methylphenidate increased significantly after ATD (12.82 mg, 95% CI 5.80 to 19.85; P < .001).

Figure 4. Changes in the use of methylphenidate.

(A) Yellow/blue dots: monthly dose of methylphenidate in the intervention and control groups. Pink/blue bars: number of patients prescribed methylphenidate in the intervention and control groups. (B) Difference in the monthly dose of methylphenidate between two groups. T₀ is the date of adenotonsillectomy, indicated by the dotted line.

DISCUSSION

The significance of this large-scale study using national data spanning over 10 years lies in its ability to compare treatment effects over time by matching patients who underwent AT with those who did not. In addition, it was possible to differentiate the results from those of other studies since we attempted to observe the effectiveness of AT in terms of health care utilization.

In this study, the number of prescribed medications gradually increased before and decreased after ATD in both the intervention and control groups. In both groups, the number of outpatient visits did not change before ATD; however, after ATD, the number of outpatient visits significantly decreased in the intervention group. However, there was no significant difference in the number of prescribed drugs and outpatient visits between the two groups before and after ATD. Meanwhile, the dose of methylphenidate increased significantly after ATD only in the intervention group.

Before discussing our findings, we need to explain the methodological aspects we used in our research. We wanted to extract as many pediatric patients with ADHD who underwent AT due to SDB as possible to conduct this study. However, it was impossible to identify the diagnostic study for each patient in this claimed insurance data study. Therefore, among children with ADHD, patients who underwent AT were estimated to have symptoms related to SDB and defined as an intervention group, which was based on several studies. First, most children who underwent AT did not perform diagnostic polysomnography to confirm SDB.^20–22 Previously, a large proportion of patients underwent AT due to repeated infections, but recently, AT is mostly performed to treat sleep problems.^23,24 In our study, only 19% of patients claimed insurance solely under the diagnostic code related to infections (chronic tonsillitis or adenitis), while 78% and 11.6% claimed insurance under adenotonsillar hypertrophy and sleep problems, respectively. Finally, OSA-related research was conducted in a way that was quite similar to ours using claimed insurance data.²⁵ Based on these studies, we judged that it was reasonable to set the intervention group by estimating that ADHD patients with AT had sleep problems.

Childhood/adolescent ADHD can impact academic and occupational functioning, social relationships, and emotional control. The long-term prognosis of childhood ADHD varies with the severity of the disease, accompanying diseases, and treatment responses.^26–28 Some pediatric patients continue to have symptoms related to ADHD until adulthood; however, the symptoms usually improve, and the patients do not exhibit significant emotional or behavioral problems in the long run.^29,30 Patients taking medications can expect greater improvement in symptoms than those not taking medications.^31–34 All the patients in our study were prescribed medications. However, only three-quarters (73.2%) of all patients diagnosed with ADHD during the same period in Korea were prescribed medications for ADHD treatment.

Several studies have discussed the relationship between SDB and ADHD.^6,35–40 According to these studies, the symptoms of SDB are similar to the main symptoms of ADHD, including inattention and hyperactivity. In addition, SDB occurs at a high rate in patients with ADHD, and children with concurrent ADHD and SDB have more severe ADHD symptoms. Therefore, we anticipated that the intervention group would use more medications and at a higher dose and the frequency of outpatient visits would be higher than those of the control group before ATD. However, in our study, there was no difference in the pattern of medication use and outpatient visits between two groups before ATD. Several studies have been conducted based on the assumption that an improvement in SDB will help alleviate ADHD symptoms.^8–14,41 AT, which is as the most effective treatment for childhood SDB,^8,17 was performed to alleviate ADHD symptoms, and most studies reported that both ADHD and SDB symptoms improved after AT. Most of these previous studies confirmed that ADHD-related symptoms improved after AT by using the following rating scales for diagnosing ADHD or questionnaires for patient’s caregivers: Turgay DSM-IV Based Child and Adolescent Behavior Disorders Screening and Rating Scale; Korean Attention Deficit Hyperactivity Disorder Rating Scale; Conners Parent Rating Scale-Revised Short Questionnaire; Pediatric Quality of Life Inventory, parent version; and Child Behavior Checklist.^8,11–17 However, the evaluation of the effectiveness of AT using these methods has several limitations.⁴² Depending on the respondents’ motivation, honesty, memory, and ability to respond, the results may be subjectively different. Additionally, only patients who responded to the survey were included in the study, which may have resulted in selection bias. Therefore, we attempted to determine the effectiveness of AT through other indicators by identifying changes in medication or medical use, such as outpatient visits, in the real world.

Most previous studies have evaluated the therapeutic effects on patients with ADHD 3–6 months after AT,^{8,10,11,13–15} and few studies have evaluated the effects more than 1 year after AT. Previous studies have shown that AT is effective in the short-term treatment of ADHD⁵; however, its long-term effects remain uncertain.⁴³ Therefore, we attempted to confirm the long-term effects of AT on the improvement of behavioral problems by assessing the treatment progress for up to 2 years after AT, revealing that there was no difference in the number of prescribed drugs or outpatient visits after ATD for 2 years. Therefore, the long-term effects of AT on improving symptoms of ADHD were not clearly demonstrated. In a rare study that sought to evaluate the long-term effects of AT, there was no difference in the ADHD index before and after AT. When assessing the effects of time, there were no significant effects on the ADHD index, cognitive problems/inattention, or oppositional behavior.¹⁶ In that study, the evaluation of effectiveness was conducted through surveys, and only a small number of patients were evaluated; however, the results were similar as those of our study. Contrarily, another study that analyzed the long-term effects of AT through a survey found that behavioral problems improved. However, it is difficult to generalize this result because it was drawn from a small number of patients using surveys.⁴⁴ A few prospective studies have reported the short-term effects of AT.^45,46 To date, reliable large-scale prospective studies are lacking.

Previous studies have reported that there is a linear relationship between methylphenidate dose and response.⁴⁷ As symptoms worsen, the dosage of medication increases, and while there may be individual variations, in most cases, higher doses provide better control of more significant symptoms.⁴⁸ In our study, prior to ATD, there was no difference in the dosage of methylphenidate between the two groups. However, after ATD, patients in the intervention group who were still receiving medication showed an increase in the prescribed drug dosage, and there was a significant difference in the prescribed dosage between the two groups. The reason for this increased drug dosage after ATD in the intervention group may be that even if sleep problems were improved by AT, it did not have sufficient effects to improve ADHD-related symptoms. Regarding this, even if AT was performed in patients with sleep problems, the long-term prognosis of ADHD treatment may be indirectly considered less favorable than those without sleep problems.

This study had several limitations. (1) It was conducted using medical insurance claims data and was limited to identifying each patient’s symptoms and test results. (2) The severity of ADHD-related symptoms or that of sleep problems in each patient before AT could not be assessed, and the differences in the effects of AT according to the severity of ADHD symptoms could not be confirmed. (3) This study compared the treatment outcomes between patients with ADHD who underwent AT and those who did not. Therefore, it was not possible to assess the treatment prognosis of ADHD patients with SDB symptoms who did not undergo AT nor was it possible to confirm whether or not patients with ADHD who underwent AT could have a worse prognosis had they not undergone AT. (4) ADHD medications are known to have low adherence in children and adolescents,^49,50 and this may have affected the results of our study.

CONCLUSIONS

This is a large-scale study with long-term observations of the therapeutic effects of AT in pediatric patients with ADHD. The therapeutic effects of AT were evaluated through clinical indicators such as medications or outpatient visits, and it was not confirmed whether AT had significant therapeutic effects on patients with ADHD in the long run. These results differ from those of previous studies that evaluated the short-term therapeutic effects of AT. Further long-term prospective studies are warranted to confirm these results.

DISCLOSURE STATEMENT

All authors have seen and approved the submitted manuscript. This research was funded by the Catholic University of Korea, Eunpyeong St. Mary’s Hospital, Research Institute of Medical Science. The authors report no conflicts of interest.

ABBREVIATIONS

ADHD

attention deficit hyperactivity disorder

AT

adenotonsillectomy

ATD

date of adenotonsillectomy

CI

confidence interval

SDB

sleep-disordered breathing