Child behavior and quality of life before and after tonsillotomy versus tonsillectomy

International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl

Child behavior and quality of life before and after tonsillotomy versus tonsillectomy§,§§ Elisabeth Ericsson a,b,*, Inger Lundeborg c, Elisabeth Hultcrantz b a

Department of Nursing Science, School of Health Sciences, Jo¨nko¨ping University, Sweden Department of clinical and experimental medicine, Division of Oto-rhino-laryngology, Linko¨ping University, Sweden c Department of clinical and experimental medicin, Division of speech and language pathology, Linko¨ping University, Sweden b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 25 March 2009 Received in revised form 17 May 2009 Accepted 19 May 2009 Available online 17 June 2009

Objectives: Compare two techniques for pediatric tonsil surgery with respect to postoperative pain and morbidity and changes in sleep behavior, health related quality of life (HRQL) and benefits due to surgery. Methods: 67 children (4.5–5.5 years) with tonsillar hypertrophy and obstructive sleep-disordered breathing with or without recurrent tonsillitis were randomized to either regular tonsillectomy (TE) (n = 32) or intracapsular tonsillectomy/tonsillotomy (TT) (n = 35) with Radiofrequency surgical technique (ellman Int.). Before TT/TE, the parents completed a validated Quality of Life survey of pediatric obstructive sleep apnea, the OSA-18 (Obstructive Sleep Apnea-18) and a standardized assessment of their children’s behavior with the Child Behavior Checklist (CBCL). Six months after surgery, the parents repeated these measurements, and assessed the health related benefits of the surgery using the Glasgow Children’s Benefit Inventory (GCBI). Results: In the TT group, the children recorded less pain from the first day after surgery onwards, used fewer doses of painkillers and were pain-free 3 days earlier than the children in the TE group. Six months after surgery, there were no significant difference between TT and TE with regard to snoring and ENT-infections. The differences in the total scores and in all the individual domains between the initial OSA-18 and postsurgery scores were all significant (P < 0.0001). The improvement in the total problem score measured with CBCL was also significant (P < 0.01) and there was no difference between the TT and TE children. The improvements in all subscores of the GCBI indicated a significant health benefit of both TT and TE. Conclusions: TT with RF-surgery causes less pain and postoperative morbidity than regular TE and has an equal effect on snoring and recurrent infections. Pre-school children with tonsillar hypertrophy and obstructive sleep-disordered breathing all show an impact on HRQL and behavior before surgery and improve dramatically just as much after TT as after TE. Therefore TT would be considered for treatment of small children. ß 2009 Elsevier Ireland Ltd. All rights reserved.

Keywords: Tonsillotomy Tonsillectomy Postoperative pain Quality of life

1. Introduction Sleep-disordered breathing (SDB) is a common problem in children with a spectrum of upper airway disorder ranging in

§ Economical support: Financial support for this project was received from the Research Council of South East Sweden (FORSS). §§ Part of the results from the present article was presented at the 8th International Conference in Pediatric Otorhinolaryngology Budapest, Hungary, June, 2008. * Corresponding author at: Department of Nursing Science, School of Health Sciences, Jo¨nko¨ping University, Box1026, SE-551 11 Jo¨nko¨ping, Sweden. Tel.: +46 36101197; fax: +46 36101250. E-mail addresses: [email protected], [email protected] (E. Ericsson).

0165-5876/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2009.05.015

severity from primary snoring to obstructive sleep apnea (OSA). Symptoms that are included in SDB are oral breathing, long lasting snoring, sleep apnea, difficulty in breathing, restless sleep, frequent awakening, failure to thrive, abnormal bite and behavioral disturbance [1–4]. SDB even in less affected children, has been shown to cause behavioral disturbances and cognitive impairments including impaired concentration, reduced memory, lower intelligence scores and problem behavior [4–8]. The most common physical findings in children with symptoms of SDB are enlargement of the adenoid and tonsils. The indications for tonsil surgery have changed in recent decades such that obstructed breathing now rivals recurrent infection as the most common surgical indication for tonsil surgery [3,9]. In the last 10 years, tonsillotomy (partial tonsillectomy) for obstructive symptoms has reappeared in the medical literature in connection with the

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

1255

consent/assent regarding the study meant and about the type of surgery their child would receive and in case of TT the potential risks of tonsillitis and growth of the tonsil remnant. No sleep studies were performed before surgery (in accordance with Swedish praxis for ‘‘regular cases’’). As some children had been on the waiting list for six months the final decision about surgery was done after a pre-operative evaluation of the present breathing distress by the parents together with the otorhinolaryngologist. All children were snoring regularly and had hyperthrofied tonsils. Other symptoms like daytime sleepiness, oral breathing, enuresis or eating difficulties varied and were to be ‘‘objectified’’ in the study by the use of OSA-18. Participants and withdrawals are illustrated in Fig. 1. All children were evaluated by a speech and language pathologist before and six months after surgery and by an orthodontist before surgery with a planned follow-up after 2 years. The results of these evaluations will be presented elsewhere.

surgical techniques used. Removing only the protruding parts of the tonsils seems to have the same beneficial long-term effect as regular tonsillectomy (TE) on obstructive symptom due to hypertrophic tonsils regardless of the surgical method used. Intracapsular tonsillectomy/tonsillotomy (TT) reduces postoperative complications, causes less pain and results in more rapid recovery than TE [9–13]. Pediatric OSA has been shown to have a significant impact on health related quality of life (HRQL) and interest in measuring quality of life (QoL) is growing. Children with diseased tonsils have poorer HRQL than healthy children, and the general health perception of children with tonsil and adenoid disease, is similar to that of those with asthma and juvenile rheumatoid arthritis [2,14,15]. Large improvements in HRQL, behavior, and cognitive function occur after tonsil surgery and these findings are maintained over the long-term [2,8,16,17]. However, the literature lacks controlled studies on HRQL in pediatric TE versus TT in which validated specific instruments are used to measure the impact of HRQL before TT/TE and after surgery [2,8]. In recent investigations, we have shown that children and young adults in a wide age-group from 5 to 25 years with tonsillar hypertrophy, with and without recurrent infections, benefit from the less invasive tonsil surgery technique of tonsillotomy using radiofrequency technique (RF) [16,17]. The estimated prevalence of OSA in 4-year-old children is 1–3% [1]. In Sweden during 2007, 20% of all tonsil-surgery was done in the age group 4–6 years; the main indication was obstructive breathing (85%)[18]. In this age group, the Waldeyer’s lymphatic ring is most active and hypertrophic in children, which also is a cause of obstructive symptoms [19]. It is therefore important to examine members of this age group of younger children, regarding preoperative symptoms and postoperative morbidity, HRQL, obstructive symptoms and infections in connection to tonsil surgery. The aim is to compare the effects of tonsillotomy using radiofrequency technique with regular tonsillectomy using cold steel, effects including postoperative pain and morbidity and, at six months follow-up, including health related quality of life, obstructive symptoms, infections and benefit due to surgery.

See Appendix A, details of the measurements and flow sheet (Fig. 1). Before surgery, the parents completed two questionnaires at home: the Child Behavior Checklist (CBCL) [21] which is a standardized assessment tool for specific childhood behavior and the Obstructive sleep apnea-18 (OSA-18) [22,23] which is a health-related quality of life measure of pediatric SDB and of longitudinal changes. After discharge, the pain experienced by the child and in addition the child’s eating and drinking pattern, were scored by the parents using a five point verbal scale. The parents logged the doses of analgesics given and recorded the date when the child was back to normal activity. Six months after surgery, the parents repeated the OSA-18 and CBCL, and assessed the benefits of the surgery using the Glasgow Children’s Benefit Inventory (GCBI) [24]. GCBI is designed to measure the benefit to the quality of life of a child’s day-to-day life resulting from interventions such as surgery. The parents also completed a questionnaire (Qu) that asked about general health, snoring, eating difficulties and infections [16,17].

2. Methods

2.2. Surgical technique

This study was approved by the Human Research Ethics Committee, Linko¨ping University. Three ENT clinics participated (one university clinic and two county council hospitals). The children included were 4.5–5.5 years old with tonsillar hypertrophy and sleep disordered breathing with or without recurrent tonsillitis. The children were initially randomized from the existing ordinary waiting list for tonsil surgery either to tonsillotomy using radiofrequency technique or to regular Tonsillectomy (TE), according the method of Zelen [20], and a few were later excluded due to the exclusion criteria. The criteria for exclusion were:

TT was performed with the ellman 4.0 MHz Surgitron1 Dual Radiowave unit (ellman International Inc. 3333 Royal Ave., Oceanside, NY 11572, USA). The protruding parts of the tonsils were removed parallel with the tonsillar pillars [10,11]. TE was performed with cold knife and blunt dissection. Adenoidectomy was performed using ring knife. All surgery was performed by the clinics’ ordinary otolaryngologists.

 treatment with antibiotics for throat infections during the last three months,  previously treatment for peritonsillitis,  records stating small tonsils,  other complicating disease needing special care,  children and/or caregivers not speaking Swedish. The randomization procedure was implemented using a computergenerated sequentially numbered list. An independent person drew from this list and assigned even numbers to TT and odd numbers to TE. After randomization for both surgery groups, the parents were invited to have their child participate in the study; prior to enrollment, they received information about what informed

2.1. Measures

2.3. Anesthesia and pain treatment Pre-medication, paracetamol 40 mg/kg bwt was given orally. Two of the ENT-clinics used i.v. induction with propofol and fentanyl and maintained anesthesia with sevorane/air/O2 and the third total intravenous anesthesia with remifentanil and propofol. Oral intubation was performed. During the first three postoperative days, pain treatment was given as paracetamol 100 mg/kg bwt/day, divided into four doses, and then reduced to 75 mg/kg bwt/day. Diclophenac 25 mg, was given in one (<24 kg) or two (24 kg) doses. The parents were instructed to first withdraw diclophenac and thereafter to reduce the dosage of paracetamol while keeping the pain level ‘‘below 3’’ where 1 is ‘‘no pain’’ and 5 ‘‘worst possible’’, i.e., just accepting a minor/slight pain level. The instructions for pain medication were given both verbally and in writing. The families were prepared for

1256

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

Fig. 1. Flow sheet, number of patients, dropouts and data collected.

the child to be in day-surgery which meant that the children would need to stay at least 6 h after surgery in the hospital. The day after discharge all parents were contacted by phone, and during the entire postoperative period they had a direct number to call if they experienced any problems.

in the questionnaires. Spearman’s rank-correlation coefficient was used for correlation between questions. Differences in CBCL between the population in this study and the comparative normative populations [25] were tested using Student’s t-test (2-tailed) for normally distributed continuous variables. P values < 0.05 were considered statistically significant.

2.4. Statistical analysis 3. Results Statistical analysis was performed with SPSS1 Windows version 15.0. Parametric data are expressed as number of cases and mean  standard deviation (SD). Non-parametric data are expressed in median and inter-quartile range. Non-parametric methods were employed since the variables are at an ordinal level of measurement and the data are not normally distributed (Kolmogorov–Smirnov test). The Wilcoxon signed rank test was used for comparison of preoperative and postoperative data and for changes in scores over time. The Mann– Whitney U-test was used for comparison between two subgroups

3.1. Subjects and setting The initial study population comprised 118 children. The flow sheet, number of patients, dropouts and data collected are presented in Fig. 1. Thirty percent (22TE/14TT) of the families declined participation. Other children were excluded in accordance with the exclusion criteria and due to randomization error. There were no dropouts in the study-groups at six months follow-up.

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262 Table 1 Peroperative data for tonsillotomy (TT) and tonsillectomy (TE).

Last day for Paracetamol Last day for Diclofenac First pain-free day First day of no eating problems Days at home a b

1257

Table 2 Impact on quality of life in TT and TE-group, classified by OSA-18 scores.

TTa

TEa

P-valueb

OSA-18 scores (18–126)

5 2 4 3 6

8 4 8 7 9

<0.00001 <0.00001 <0.0001 <0.0001 <0.00001

TT

TE

TT

TE

Low impact (<60)a Moderate impact (60–80)a Large impact (<80)a

17 10 8

14 13 5

33 2

31 0 1

Total scoreb

58 (49–75)

61 (46–74)

27 (22–34)

25 (23–32)

(3–6) (0–3) (3–6) (1–6) (4–7)

(6–9) (2–6) (6–10) (5–9) (7–10)

Median (inter-quartile range). Mann–Whitney U-test.

a b

The final study populations included 67 children between 4.5 and 5.5 years old who were operated on according to the randomization—35/TT and 32/TE, 39 boys and 28 girls with a mean age of 4.8 years. Twenty percent of the children had had one or more episodes of tonsillitis in addition to obstruction, and the remainder had only obstructive symptoms as indication for surgery. Eighty percent had adenoidectomy on the same occasion (28TT/25TE), and six (10%) children (5TT/1TE) had earlier had an adenoidectomy. 3.2. Postoperative morbidity in relation to surgical method Seventy percent of the children in both groups were discharged on the day of surgery. The reasons staying overnight in the hospital were nausea and vomiting, living far away from the hospital, or having had surgery late in the afternoon. All postoperative logbooks were collected after the children were pain free. There were no gender differences regarding pain, or use of analgesics. The TT group recorded less pain from the first day and less need for analgesics compared to the TE group (Table 1). The TT group was back to normal activity 3 days earlier than the TE group and was pain free 4 days earlier and returned to normal eating habits. The parents of the TT group took less time off from work compared to TE (Table 1). Fig. 2 illustrates the significance difference of pain between TT and TE from the day after surgery to day 9. There was no primary hemorrhage. Two TE-patients returned to the hospital on day 5 due to postoperative bleeding and they were observed overnight, but needed no other intervention. There was no secondary bleeding in the TT-group. One TE-child had an adenoidectomy three months after the tonsil surgery due to recurrent periodic fever (PFAFA) which had not been solved by the tonsil surgery.

Fig. 2. Median logged level of pain score within each group. Median and interquartile range (25th and 75th percentiles) TT group recorded less pain from the first day. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; *****P < 0.00001 (Mann– Whitney U-test).

Pre-surgery

Post-surgery

Number of patients. Median (inter-quartile range).

3.3. HRQL behavior There were no gender differences with respect of the answers of CBCL, OSA-18, GCBI or the follow-up questionnaire (Qu). The impact on HRQL in different surgical groups is presented in Table 2. Before surgery, 13 children had a high total score indicating a large impact on HRQL classified by OSA-18, 23 had a moderate impact, and 31 had a low score, low impact on HRQL. Six months after surgery, one child reported a large impact, two children moderate impact, and 64 had a low score and thus low impact. Preoperative responses for the OSA-18 survey and the change scores are summarized in Table 3. ‘‘Sleep disturbance’’ and ‘‘Physical suffering’’ were the highest rated domains, followed by ‘‘Emotional distress’’, ‘‘Daytime problems’’ and the ‘‘Concerns of caregivers’’. There were no significant differences between the TT and the TE groups in all domains before surgery and six months after. The change score indicated a large change six months after surgery in ‘‘Sleep disturbance’’, ‘‘Physical symptoms’’ and ‘‘Caregivers concern’’ for both groups. The impact on ‘‘Daytime function’’ was large in the TE-group and moderate in the TT- group. In both groups, the change ‘‘Emotional distress’’ was moderate (Table 3). The total OSA-18 score and each of the domain scores show large improvement (P < 0.001–P < 0.0001) for both tonsillectomy and tonsillotomy after six months. The change in the OSA-18 survey domains is illustrated graphically in Fig. 3. The parents in both surgical groups rated the child’s quality of life in the visual scale (OSA-18) and the results showed large improvements after surgery (P < 0.00001). The GCBI demonstrated a significant improvement in all GCBI-scores (Fig. 4). There were no differences between the TT and TE group either in the answers to the 24 questions or the four dimensions/subscales. A comparison with normative values on the CBCL showed that there was a difference before surgery between the study groups’ score and the normative value (Table 4). The scores on the CBCL improved in both groups at the six months follow-up, and the mean total problem score was lower (Table 5). In both groups, 50% of the parents reported that surgery had affected the child’s enjoyment of food for the better (GCBI) and that the child had a better appetite (Qu). The answers to single questions (Qu) concerning, ‘‘temper’’, ‘‘stamina’’, and ‘‘concentration’’ showed a clear improvement and a good correlation with the GCBI-subscales ‘‘Emotional’’, ‘‘Vitality’’ and ‘‘Learning’’ (rs = 0.65–0.79; P < 0.0001). Preoperatively, a fair to good correlation was seen between OSA-18, GCBI, Qu and behavior scores (CBCL), and also with the postoperative changes in those measures. There was a positive correlation between the preoperative CBCL ‘‘total problem’’ and the preoperative OSA-18 ‘‘total score’’ (rs = 0.47 P < 0.001) and the change in the CBCL ‘‘total problem’’ score with the OSA-18 ‘‘total change’’ score (rs = 0.56 P < 0.001). 3.4. Snoring, infections and general health After six months, there were no significant differences between the TT and TE-groups either in frequency or loudness of snoring

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

1258

Table 3 Preoperative responses for the OSA-18 and change scores in TT and TE.

a

Sleep disturbance Physical symptomsa Emotional distressa Daytime functiona Caregiver concernsa

TT, n = 35

Change scoreb at 6 mo

TE, n = 32

Change scoreb at 6 mo

P-valuec TT/TE Change scores

4.2  1.3 3.9  1.3 3.2  1.7 3.1  1.4 3.0  1.6

2.7  1.5 1.9  1.7 1.0  1.4 1.2  1.5 1.6  1.5

3.9  1.4 3.8  1.4 3.1  1.4 3.3  1.3 2.8  1.4

2.6  1.5 2.1  1.6 1.1  1.5 1.7  1.2 1.5  1.3

NS NS NS NS NS

a

Mean  standard deviation (SD). Change score = The follow-up mean survey score and the individual domain scores subtracting from the baseline mean and individual domain mean scores; Change score < 0.5 = trivial change; 0.5–0.9 = small change; 1.0–1.4 = moderate change; and 1.5 = large change. c Mann–Whitney U-test. b

Fig. 3. Preoperative and postoperative median OSA-18 domain scores in TT and TE and the significant improvement in both group, six months after surgery.

Fig. 4. Scores of the GCBI-subscales in both TT and TE group.

(Qu). When comparing the two of the five Likert-items ‘‘never snore’’ and ‘‘more seldom’’, snoring was still present in a higher percentage of the TT children (P < 0.05) (Fig. 5). Two parents in each group scored the snoring of the child to be ‘‘about the same’’

as before surgery (Qu). About 90% of the parents in both groups rated the child’s sleep ‘‘much better’’/‘‘better’’ (GCBI). The domain with the greatest change in mean score was ‘‘sleep disturbance’’ (OSA-18), to the same degree in both surgical groups. The answers to the questions about infections showed postoperative improvement. There was no difference between the TT and TE-groups regarding ENT infections (upper respiratory infection including sore throats) (Fig. 6) or the use of antibiotics. In the TE group, two children had been treated with antibiotics for strep throat, and in the TT group, four children had had tonsillitis and had been treated with antibiotics. One TT and six TE children had been treated with antibiotics for otitis media (Qu). About 60% in both groups rated ‘‘much better’’/‘‘better’’ improvement on the effect of surgery on the child’s proneness to catch colds, need to visit a doctor, and to take medication (GCBI). The questions and items about general health all showed good correlation between the OSA-18 ‘‘Physical change’’ score and GCBI-subscale ‘‘Physical health’’ (rs = 0.52 P < 0.001) and the question about health (Qu) and GCBI ‘‘Physical Health’’ (rs = 0.77 P < 0.0001).

Table 4 Child Behavior Checklist before TE versus TT and compared with normal range. TT (n = 35)

TE (n = 32)

P-value TT CBCL/Normal rangea,c

P-value TE CBCL/Normal rangea,c

P-value TE/TT CBCL

Before surgery Internalizationb Externalizationb Total problemsb

5.8  4.6 9.8  7.0 25.6  19.1

4.2  3.6 7.8  6.1 20.9  12.4

<0.01 NS <0.001

NS NS <0.01

NS NS NS

Six months after surgery Internalizationb Externalizationb Total problemsb

3.7  5.4 7.6  6.7 19.5  18.4

2.4  2.7 6.3  4.5 13.5  9.8

NS <0.05 <0.05

<0.05 NS NS

NS NS NS

NS = not significant. a Normal range. Swedish population [25]. b Mean  SD. c Mann–Whitney U-test.

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

1259

Table 5 Impact of tonsillotomy and tonsillectomy on Child Behavior Checklist scores. TT, n = 35) Preop Internalization Withdrawn Somatic complaints Anxious or depressed Externalization Attention problems Delinquent behavior Aggressive behavior Social problems Thought problems Total problems

a

5.8  4.6 1.9  2.1 1.7  1.4 2.2  2.7 9.8  8.6 2.5  2.9 1.5  2 8.3  7. 1.5  2.2 0.5  1 25.6  19.1

TE (n = 32) Postop

a

3.7  5.4 1.4  1.8 0.7  1 1.7  3.6 7.6  6.7 2.0  2.9 0.9  1.1 6.7  5.9 1.5  2.3 0.5  1.1 19.5  18.4

Change

a

2  5.1 0.54  2.1 1  1.4 0.5  2.7 2.2  5.6 0.5  2.4 0.6  1.7 1.6  4.5 0.0  1.4 0.1  0.9 6.1  14.

b

a

TT/TE a

P value pre//post

Preop

Postop

<0.01 NS <0.01 NS <0.05 NS <0.05 <0.05 NS NS <0.01

4.2  3.6 1.4  1.6 1.3  1.6 1.5  1.6 7.8  6.1 2.3  2.1 0.9  1.2 6.8  5.3 1.3  1.3 0.1  0.3 20.9  12.3

2.4  2.7 0.9  1.1 0-6  1.0 0.9  1.5 6.3  4.5 1.3  1.5 0.8  1.1 5.5  3.9 0.8  1.1 00 13.5  9.8

Change

b

1.8  2.3 0.44  1.2 0.69  1.3 0.7  1.1 1.5  4.8 1.1  1.5 0.2  1 1.3  4.2 0.5  1.3 0.1  0.3 7.4  8.6

b

P value pre//post

P-value

<0.001 NS <0.01 <0.01 NS <0.001 NS NS <0.05 NS <0.0001

NS NS NS NS NS NS NS NS NS NS NS

c

Change score

CBCL = Child Behavior Checklist; NS = not significant. a Mean  SD. b Wilcoxon signed rank test. c Mann–Whitney U-test.

This investigation can clearly demonstrates that pre-school children on an ordinary waiting list for adenotonsil surgery all have a low health related quality of life as measured with the validated instrument OSA-18. They also have behavioral problems more than ‘‘normal’s’’ measured with CBCL. Both TT and TE yielded large improvement in HRQL, behavioral problems as well as for snoring and infections, at the six months follow-up. What is more, the study demonstrates less postoperative morbidity both with respect to pain and time for recovery for the children operated with TT compared to them who had received TE. The study had a relatively large dropout of 30% of the invited families before enrolment in the study, which certainly was caused by the randomization method according to Zelen [20]; with which the patients are randomized before being contacted about surgery. A few ‘‘declined’’ to participate without explanation, the same

frequency for the TE and TT groups, but several in the TE-group did not participate because they wanted their child to be operated with TT. The opposite was seen in our previous study 5 years ago [11] which shows that tonsillotomy now is recognized as a surgical method with less postoperative morbidity. Although tonsillectomy may be performed safely in young children, the risks of bleeding, dehydration, and poor oral intake are more critical in younger patients. Tonsillectomy is associated with considerable pain and negative behavioral outcomes especially in young children (nightmares, new fears, sleep and eating problems, bed wetting, separation anxiety and regression) [26,27]. Also, in the present investigation, the pain level for some TEchildren was too high in spite of the prescribed scheme for high postoperative analgesic treatment. However, the tonsillotomy achieved the goal with less pain than the standard tonsillectomy. The TT-group was pain free and also went back to normal activity earlier than the TE-group. This should not only be of benefit for the TT-children, but may also reduce the parental distress in many ways. One was that the TT parents’ had to take less time off from work compared to the TE-parents. The children in the present investigation were similar in the two surgical groups with respect to the results from the quality of life surveys before and six months after surgery; both groups demonstrated comparable and very significant improvements. Our results are consistent with the findings by Colen et al. [28], who also demonstrates a significant improvement in the overall OSA-18 score after powered intracapsular tonsillectomy at three months and at 1 year. The results of the present study also agree with several studies after adenotonsillectomy (ATE) with six

Fig. 5. Parents’ rating of how often children snore six months after surgery in comparison with snoring before operation (Qu).

Fig. 6. Parents’ assessment of how prone to ENT infections children have been during the first six months (Qu).

Both groups expressed overall satisfaction with the surgery regardless of the method used. One parent in each group rated being less satisfied with the surgery. Their children snored about the same after as before surgery. Both these children had adenoidectomy at the same occasion as tonsil surgery. The TT child’s parent’s rated moderate impact of OSA-18 pre- and post surgery. This child had had one episode of tonsillitis after surgery. The TE-child’s parents rated large impact of OSA18 pre and post surgery. This boy had swollen nasal mucosa and was slightly overweight and is now being treated with nasal steroids. 4. Discussion

1260

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

months follow-up OSA-18 have been used to evaluate HRQL in children with OSA [15,29–31]. The children in the present investigation displayed behavior problems (CBCL) before surgery, and both TT and TE demonstrated significant improvements in behavior and quality of life after surgery. This is consistent with several studies after ATE [6,7,29], which suggest that SDB may be associated with behavioral problems regardless of the relative severity of the sleep disorder. The extent of improvement after ATE is not related to whether a child has mild SDB or OSA [6,7]. Emotional disturbance (OSA-18) showed the smallest, but strongly significant improvement (P < 0.001), which also has been earlier reported [15]. Factors related to emotional disturbance may take longer than six months to show improvement. Our study (with a young age group) did not report improvement in all individual syndrome scales (CBCL) after six months which also can be a too short period for some problem behaiviors, but a significant improvement was found in the total problem score classification after both TT and TE. This is in accordance with comparative studies which used CBCL [7,15,16,29–31]. This study and others can report a good correlation between HRQL (OSA-18) and behavior (CBCL) in children with SDB [29,30]. Two children in the TT-group had CBCL scores which were ‘‘borderline’’ before surgery. Their total problem scores were lower after surgery, but still not normal, which lead to a medical consultation referral to a child psychiatrist for possible diagnosis of attention deficit hyperactivity disorder (ADHD). This is in accordance with results from Li et al. [32] who found that behavioral problems could not be completely resolved by surgery in every child with primary or SDBrelated ADHD. Schwenter et al. [33] have shown similar improvement in GCBI after TE as was demonstrated in the present study for TT and TE after six months as well as in our previous investigation up to 3 years after surgery [16]. Complementing OSA-18 with GCBI to evaluate HRQL was done in order to strengthen the evaluation of the effects of surgery. GCBI is a specific instrument for outcome analysis in children after various surgical interventions. The GCBI mean score is not influenced by any ‘‘critical life event’’ or related chronic disease. This inventory is thus a powerful and durable estimate of the individual outcome after surgery. There was a significant relation between GCBI scores and the postoperative improvement of the scores in OSA-18. The present investigation confirms several studies reporting positive effects in follow-ups after TT with respect to snoring and other symptoms of obstruction [12,16,17,34,35]. Two parents in each group reported unchanged snoring after surgery. All other children had stopped ‘‘snoring regularly’’ although a few more in the TT group still snored ‘‘now and then’’, This means that more questions are needed to evaluate the postoperative effects. The more thorough analysis with OSA-18 and GCBI in the present study, make it clear that there are no significant differences between the surgical methods, but long-term follow-ups are needed and planned. The outcome of surgery was evaluated using questionnaires and clinical examination without any sleep studies. This is praxis in Sweden as well as several other countries for ‘‘regular cases’’ when the clinical findings are consistent with the case history of obstructive problems during sleep [1,3]. However, in a recent study [34], the PSG data after TT (using CO2 laser), also demonstrated that TT is highly effective in the treatment of pediatric OSA. Several studies describe only moderate association between PSG findings and QoL, The PSG data (objective measure) and QoL and behavior data (subjective measure) are related and complementary and assess different but associated constructs [6,36]. Twenty percent of the children had had one or more episodes of tonsillitis in addition to experiencing obstruction. Six months after surgery; some children in both groups had been treated post-

operatively for tonsillitis or ‘‘sore throat’’ which is normal for children of this age group. In our previous TT-RF studies with the same inclusion criteria but older children, in fact both TT and TE, demonstrated a large improvement with less infection problems at both the 1 and 3 years follow-up after surgery [16,17] which is similar to other studies where TT has been compared with TE after six [12] and 10–14 years [35]. Both groups expressed overall satisfaction with the surgery regardless of method used. One parent of a TT-child was less satisfied with the surgery because the child snored about the same after as before surgery, and also has had one incident of tonsillitis after surgery. After TT, a regrowth of the tonsils might be expected, but the risk seems to be low [9]. The rate of tonsillar regrowth for different methods for tonsillotomy is estimated to be between 0 and 3.5% for up to 3 years follow-up [9,12,13,16,35,37]. In a consecutive study [38] of a TT-group with RF, regrowth was reported in 7 out of 42 children who all were younger than 4 years. Since the age is low in the present study, it is worth noting that young age might be a risk factor for tonsillar re-growth which will be checked for in the planned 2-year follow-up. We do not consider it to be ethical to perform more randomized pain studies of TT with comparison to full tonsillectomy as instead the few cases of symptom-giving regrowth after TT very well can receive re-TT and not full TE (EHz). Issues regarding the biologic function of tonsillar tissue and its preservation especially in very young children must also be taken into consideration [19]. 5. Conclusion TT with RF-surgery is a method that causes less pain and postoperative morbidity than regular TE and has a similar beneficial effect on snoring and recurrent infections in 4–6year-old children with tonsillar hypertrophy and sleep-disordered breathing who, all show an impact on HRQL and behavior before surgery which improve dramatically just as much after TT as after TE. Therefore TT would be considered for treatment of small children. Acknowledgements Financial support for this project was provided by the Research Council of South East Sweden (FORSS). Special appreciation is extended to the parents and children who willingly participated; Jonas Graf MD, Otolaryngologist, and Anita McAllister, PhD, SLP, for their time in the clinical and science work and generous support. Appendix A The CBCL is a standardized 113-item survey of specific child behaviors that was completed by the parents both at the time of surgery and at the 1-year follow-up. The CBCL is scored to obtain a total problem score, as well as scores for ‘‘Internalizing behavior’’ (subscores: Withdrawn, Somatic Complaints, Anxious/Depressed) and ‘‘Externalizing behavior’’ (subscores: Delinquent Behavior and Aggressive Behavior) [21]. This instrument is well known, and normative data for the Swedish population are available from 1991 for the version of the instrument used. Each item is scored 0, ‘‘not true’’; 1, ‘‘somewhat’’ or ‘‘sometimes true’’; or 2, ‘‘very true’’ or ‘‘often true’’. The scores from the present study were compared to those normative data [25]. The CBCL consists of two parts: social competence and behavior/emotional problems. In this study, only the items from the behavioral/emotional part was used.

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

The OSA-18 consists of 18 items grouped into 5 domains: sleep disturbance, physical symptoms [23], Emotional distress, daytime function, and caregivers concerns. Items are scored on a 7-point ordinal scale that assesses the frequency of specific symptoms, scored as follows: 1, ‘‘none of the time’’; 2, ‘‘hardly any of the time’’; 3, ‘‘a little of the time’’; 4, ‘‘some of the time’’; 5, ‘‘a good bit of the time’’; 6, ‘‘most of the time’’; and 7, ‘‘all of the time’’. Item responses are summed to produce a total score ranging from 18 (best QoL) to 126 (poorest QoL). A total score less than 60 suggests a small impact on disease-specific QoL, scores of 60–80 suggest a moderate impact, and scores greater than 80 suggest a large impact. A mean survive score and individual domain mean scores are also calculated. The OSA-18 change scores are calculated by subtracting the follow-up mean survey score and the individual domain mean scores from the baseline mean and individual domain mean scores. Positive values indicate clinical improvement and negative values indicate deterioration. A change score of less than 0.5 indicates trivial change; 0.5–0.9, small change; 1.0–1.4 moderate change; and 1.5 or greater, large change. The OSA-18 also provides a direct global rating of SDB-related HRQL via 10-point visual analogue scale with specific semantic anchors. The Glasgow Children’s Benefit Inventory (GCBI) is a health-related quality-of-life instrument assessing the child’s condition after intervention/surgery [24]. GCBI is comprised of 24 questions on the consequences of a specified intervention on various aspects of the child’s day-to-day-life. The GCBI is constructed to evaluate general symptoms and it is applicable to children of any age. The questionnaire has a high internal consistency. The GCBI shows 4 dimensions related to ‘‘emotion’’, ‘‘physical health’’, ‘‘learning’’ and ‘‘vitality’’ and it is suitable for use in pediatric otolaryngology. A GCBI summary score can be calculated. The GCBI scores are scaled from S100 to 100 with positive score implying an improvement in HRQL. Questionnaire (Qu) [12,16,17] included 11 questions compared to the time before surgery about general health a), temper a), stamina/ energy a), concentration a), snoring often b) and loudly c), appetite, ENT-infection d), antibiotic infections e) and satisfaction f). This questionnaire have been used in two other tonsil-studies [16,17]. The questions were assessed with a Likert scale a) ‘‘Much better’’, ‘‘Some what better’’, ‘‘About the some’’, ‘‘Somewhat worse’’, ‘‘Much worse’’; b) ‘‘Never snores’’ ‘‘Mores seldom’’ ‘‘About the some’’, ‘‘Some what more’’, ‘‘Much more often’’; c) ‘‘Does not snores’’, ‘‘Less loudly’’, ‘‘About the same’’, ‘‘Somewhat more loudly’’, ‘‘Much more loudly’’; d) ‘‘No infections’’, ‘‘More seldom’’, ‘‘About the some’’, ‘‘Some what more often’’, ‘‘Much more often’’; e) numbers f) ‘‘Very satisfied’’, ‘‘Satisfied’’, ‘‘Some what satisfied’’, ‘‘Less than satisfied’’, ‘‘Not satisfied at all’’. The original English versions of the GCBI and OSA-18 were, before use, carefully translated into Swedish independently by two qualified users of the English language, and independently back-translated into English to ensure accuracy. The versions from translators were compared and one target version was drafted and discussed with people having knowledge of pediatric otolaryngology.

References [1] N.J. Ali, D.J. Pitson, J.R. Stradling, Snoring, sleep disturbance, and behaviour in 4–5 year olds, Arch. Dis. Child 68 (1993) 360–366. [2] C.M. Baldassari, R.B. Mitchell, C. Schubert, E.F. Rudnick, Pediatric obstructive sleep apnea and quality of life: a meta-analysis, Otolaryngol. Head Neck Surg. 138 (2008) 265–273. [3] R.A. Weatherly, E.F. Mai, D.L. Ruzicka, R.D. Chervin, Identification and evaluation of obstructive sleep apnea prior to adenotonsillectomy in children: a survey of practice patterns, Sleep Med. 4 (2003) 297–307.

1261

[4] S. Blunden, K. Lushington, D. Kennedy, J. Martin, D. Dawson, Behavior and neurocognitive performance in children aged 5–10 years who snore compared to controls, J. Clin. Exp. Neuropsychol. 22 (2000) 554–568. [5] N.J. Ali, D. Pitson, J.R. Stradling, Sleep disordered breathing: effects of adenotonsillectomy on behaviour and psychological functioning, Eur. J. Pediatr. 155 (1996) 56–62. [6] R.D. Chervin, D.L. Ruzicka, B.J. Giordani, et al., Sleep-disordered breathing, behavior, and cognition in children before and after adenotonsillectomy, Pediatrics 117 (2006) e769–e778. [7] R.B. Mitchell, J. Kelly, Behavioral changes in children with mild sleep-disordered breathing or obstructive sleep apnea after adenotonsillectomy, Laryngoscope 117 (2007) 1685–1688. [8] S.L. Garetz, Behavior, cognition, and quality of life after adenotonsillectomy for pediatric sleep-disordered breathing: summary of the literature, Otolaryngol. Head Neck Surg. 138 (2008) S19–S26. [9] J.A. Koempel, C.A. Solares, P.J. Koltai, The evolution of tonsil surgery and rethinking the surgical approach to obstructive sleep-disordered breathing in children, J. Laryngol. Otol. 120 (2006) 993–1000. [10] E. Ericsson, E. Hultcrantz, Tonsil surgery in youths: good results with a less invasive method, Laryngoscope 117 (2007) 654–661. [11] E. Hultcrantz, E. Ericsson, Pediatric tonsillotomy with the radiofrequency technique: less morbidity and pain, Laryngoscope 114 (2004) 871–877. [12] E. Hultcrantz, A. Linder, A. Markstrom, Long-term effects of intracapsular partial tonsillectomy (tonsillotomy) compared with full tonsillectomy, Int. J. Pediatr. Otorhinolaryngol. 69 (2005) 463–469. [13] I.M. Vlastos, K. Parpounas, J. Economides, G. Helmis, E. Koudoumnakis, M. Houlakis, Tonsillectomy versus tonsillotomy performed with scissors in children with tonsillar hypertrophy, Int. J. Pediatr. Otorhinolaryngol. 72 (2008) 857–863. [14] M.G. Stewart, E.M. Friedman, M. Sulek, et al., Quality of life and health status in pediatric tonsil and adenoid disease, Arch. Otolaryngol. Head Neck Surg. 126 (2000) 45–48. [15] V.A. Flanary, Long-term effect of adenotonsillectomy on quality of life in pediatric patients, Laryngoscope 113 (2003) 1639–1644. [16] E. Ericsson, J. Graf, E. Hultcrantz, Pediatric tonsillotomy with radiofrequency technique: long-term follow-up, Laryngoscope 116 (2006) 1851–1857. [17] E. Ericsson, T. Ledin, E. Hultcrantz, Long-term improvement of quality of life as a result of tonsillotomy (with radiofrequency technique) and tonsillectomy in youths, Laryngoscope 117 (2007) 1272–1279. [18] Socialstyrelsens statistikdatabaser. www.socialstyrelsen.se/Statistik/statistikdatabas/ Assess date: Mars 20 2009 and http://kvalitet.onh.nu/ Assess date: Mars 20 2009. [19] P. Brandtzaeg, Immunology of tonsils and adenoids: everything the ENT surgeon needs to know, Int. J. Pediatr. Otorhinolaryngol. 67 (Suppl. 1) (2003) S69–S76. [20] M. Zelen, Alternatives to classic randomized trials, Surg. Clin. North Am. 61 (1981) 1425–1432. [21] T.M. Achenbach, C.S. Edelbrock, Behavioral problems and competencies reported by parents of normal and disturbed children aged four through sixteen, Monogr. Soc. Res. Child Dev. 46 (1981) 1–82. [22] R.A. Franco Jr., R.M. Rosenfeld, M. Rao, First place–resident clinical science award 1999. Quality of life for children with obstructive sleep apnea, Otolaryngol. Head Neck Surg. 123 (2000) 9–16. [23] H. Sohn, R.M. Rosenfeld, Evaluation of sleep-disordered breathing in children, Otolaryngol. Head Neck Surg. 128 (2003) 344–352. [24] H. Kubba, I.R. Swan, S. Gatehouse, The Glasgow Children’s Benefit Inventory: a new instrument for assessing health-related benefit after an intervention, Ann. Otol. Rhinol. Laryngol. 113 (2004) 980–986. [25] B. Larsson, M. Frisk, Social competence and emotional/behaviour problems in 6– 16 year-old Swedish school children, Eur. Child Adolesc. Psychiatry 8 (1999) 24– 33. [26] M. Karling, H. Stenlund, B. Hagglof, Child behaviour after anaesthesia: associated risk factors, Acta Paediatr. 96 (2007) 740–747. [27] E. Ericsson, M. Wadsby, E. Hultcrantz, Pre-surgical child behavior ratings and pain management after two different techniques of tonsil surgery, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 1749–1758. [28] T.Y. Colen, C. Seidman, J. Weedon, N.A. Goldstein, Effect of intracapsular tonsillectomy on quality of life for children with obstructive sleep-disordered breathing, Arch. Otolaryngol. Head Neck Surg. 134 (2008) 124–127. [29] N.A. Goldstein, M. Fatima, T.F. Campbell, R.M. Rosenfeld, Child behavior and quality of life before and after tonsillectomy and adenoidectomy, Arch. Otolaryngol. Head Neck Surg. 128 (2002) 770–775. [30] K.D. Tran, C.D. Nguyen, J. Weedon, N.A. Goldstein, Child behavior and quality of life in pediatric obstructive sleep apnea, Arch. Otolaryngol. Head Neck Surg. 131 (2005) 52–57. [31] R.B. Mitchell, J. Kelly, E. Call, N. Yao, Quality of life after adenotonsillectomy for obstructive sleep apnea in children, Arch. Otolaryngol. Head Neck Surg. 130 (2004) 190–194. [32] H.Y. Li, Y.S. Huang, N.H. Chen, T.J. Fang, L.A. Lee, Impact of adenotonsillectomy on behavior in children with sleep-disordered breathing, Laryngoscope 116 (2006) 1142–1147. [33] I. Schwentner, J. Schmutzhard, C. Schwentner, I. Abraham, S. Hofer, G.M. Sprinzl, The impact of adenotonsillectomy on children’s quality of life, Clin. Otolaryngol. 33 (2008) 56–59. [34] R. de la Chaux, C. Klemens, M. Patscheider, O. Reichel, A. Dreher, Tonsillotomy in the treatment of obstructive sleep apnea syndrome in children: polysomnographic results, Int. J. Pediatr. Otorhinolaryngol. 72 (2008) 1411–1417.

1262

E. Ericsson et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1254–1262

[35] E. Eviatar, A. Kessler, N. Shlamkovitch, M. Vaiman, D. Zilber, H. Gavriel, Tonsillectomy vs. partial tonsillectomy for OSAS in children—10 years post-surgery follow-up, Int. J. Pediatr. Otorhinolaryngol. (2009). [36] M.G. Stewart, D.G. Glaze, E.M. Friedman, E.O. Smith, M. Bautista, Quality of life and sleep study findings after adenotonsillectomy in children with obstructive sleep apnea, Arch. Otolaryngol. Head Neck Surg. 131 (2005) 308–314.

[37] C.A. Solares, J.A. Koempel, K. Hirose, et al., Safety and efficacy of powered intracapsular tonsillectomy in children: a multi-center retrospective case series, Int. J. Pediatr. Otorhinolaryngol. 69 (2005) 21–26. [38] F. Celenk, Y.A. Bayazit, M. Yilmaz, et al., Tonsillar regrowth following partial tonsillectomy with radiofrequency, Int. J. Pediatr. Otorhinolaryngol. 72 (2008) 19–22.