- Email: [email protected]
Tonsillectomy or tonsillotomy? — a randomized study comparing postoperative pain and long-term effects
International Journal of Pediatric Otorhinolaryngology 51 (1999) 171 – 176 www.elsevier.com/locate/ijporl
Tonsillectomy or tonsillotomy? — a randomized study comparing postoperative pain and long-term effects Elisabeth Hultcrantz *, Arne Linder, Agneta Markstro¨m Department of Otorhinolaryngology, ENT Clinic, Uni6ersity Hospital, 75185 Uppsala, Sweden Received 9 June 1999; received in revised form 8 September 1999; accepted 12 September 1999
Abstract Background: tonsillectomy (TE) is currently the most common treatment for children with snoring and sleep apnea. Many of these children have not had any severe throat infections. To cure such children from their obstructive problems, without influencing the immunological function of the tonsils, tonsillotomy (TT) with CO2-laser was performed in a randomized study comparing it to regular tonsillectomy, with special attention to postoperative pain and symptom recurrence. Method: 41 children 3.5–8 years-old were included — 21 ‘TT’s’ and 20 ‘TE’s’. They were all operated under the same anesthesia and followed the same postoperative scheme for analgesia. A visual analogue scale for pain measurements with faces was used for the first 24 h. After that, each day until pain-free, the parents registered the child’s pain on a three graded scale, what the child was able to eat, and the amount of analgesic drugs used. Results: all the children were cured from their breathing obstruction. The mean time used for the surgery was the same and no postoperative bleeding was seen in either group. ‘TT children’ were pain-free after 5 days and ‘TE children’ after 8 days. Eight to ten days after surgery, the TT-children had gained weight and the TE children lost weight significantly. The TE group used twice as much analgesic drugs as the TT group during the first postoperative week. The TT group was healed with normal-looking, but small tonsils after 8 – 10 days; the TE group often still showed edema and crusts. At the one-year follow-up 2/21among the ‘TT-children’ snored, but did not require re-surgery. Conclusion: tonsillotomy is much less painful than TE and children recover more quickly. Results with respect to breathing obstruction are almost the same for both methods at 1-year follow-up. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Tonsil surgery; Laser in children; Randomized study; Postoperative pain
1. Introduction
* Corresponding author. Fax: +46-18-665360. E-mail address: [email protected] Hultcrantz)
(E.
During the last few years, a new surgical technique for tonsillar surgery has developed — a partial removal of the tonsils with use of CO2 laser [1,2]. The technique is used thus far in a few clinics in Scandinavia, where it is employed
0165-5876/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 5 8 7 6 ( 9 9 ) 0 0 2 7 4 - 8
172
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
mostly on children who have obstructive problems caused by tonsillar hypertrophy. The general impression has been that this kind of surgery seems to be gentler to the patients; they are noted to recover faster and to have less pain. Before recommending the technique to other centers, some issues must be addressed. Are there risks associated with partial removal of the tonsils? One obvious possibility is that the remaining parts of the tonsils will start to grow again and give the child the same symptoms once more. Another possible risk is that the partial resection will cause scar tissue in the tonsils. The deformed tissue may predispose to deeper infections or peritonsillitis. Older colleagues, who have witnessed how guillotine tonsillotomies were performed frequently during the first part of this century and then abandoned, draw a parallel with that technique. However, going through the literature we found that there is really no convincing evidence for such risks if the surgery was performed under safe conditions in skilful hands [3,4]. Thus, the fear of late infectious complications after a partial resection of the tonsil may well be unfounded. Furthermore, if the risks can be estimated, we also need to document the benefits. Are there any real advantages compared to regular tonsillectomy and in that case how can these advantages be weighed against the risks? The purpose of the present investigation is to compare the effect of tonsillotomy using CO2 laser (TT) with tonsillectomy using regular surgical technique (TE) with respect to postoperative pain, duration of surgery, perioperative bleeding, and need for hospitalization. The long-term effects on the obstructive symptoms after 6 months and 1 year are also evaluated. The patients were randomized to the two surgical procedures.
2. Method Out of 50 children, who were randomized to the study 41 were included, 15 girls and 26 boys 3.5–8 years old (69 1.5). All of them had been put on the waiting list for tonsillar surgery due to obstructive problems: snoring and/or sleep apnea, mouth breathing, and/or eating problems. They all had a verified tonsillar hyperplasia and none of them had had repeated streptococcal throat infections. The randomization was performed from the waiting list and the parents informed by mail. They could change the decision about the choice of method, but in that case the child was excluded from the study. The Ethics Committee at Uppsala University approved the investigation. All patients were kept on the ward 24 h postoperatively, irrespective of the procedure performed. The children were weighed on the morning of the surgery ‘on an empty stomach’. The surgery was performed in 21 cases as tonsillotomy (TT) with CO2-laser, 20 W (continuous mode). The surgical technique is described in detail in our previous paper [2]. The same surgeon (AL) performed all the tonsillotomies. Those specialists in the department, who regularly do tonsillectomies, operated the 20 control patients. Three children in each group had an adenoidectomy on the same occasion using conventional surgical technique. The total time in the operating theatre, the duration of the surgery, and the amount of blood loss were recorded. Postoperative pain was measured by means of a pediatric visual analogue scale (VAS) showing faces with increasing expressions of distress (Fig. 1). VAS measurements were obtained during the first 24 h, once every hour while the patient was awake. Pain, general condition and alertness were recorded, as well as the elapsed time before resumption of drinking and eating.
Fig. 1. Visual analogue scale (VAS) used in the pain measurements during the over-night stay on the ward. The child indicates a face that appears to match the pain they are feeling at that moment. The nurse or surgeon records the ordinal number of the face (the leftmost face showing no distress is given the number 0).
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
A simplified rating was used for the subsequent seven days. At discharge 24 h after surgery, the parents were instructed to fill out a form each day recording the child’s pain on a three graded scale, and how much and at what time analgesic drugs were administered. They were also asked to record how well the child ate (less, more, normal amounts) and which kind of food (liquid, pure´e or normal). While hospitalized and afterwards, paracetamol or paracetamol/codeine were used (Supp. Citodone® 25 mg/kg bwt) for pain relief. Pethidine i.v. could be given if required, in doses of 5 mg during the first few hours. Children weighing more than 30 kg (only three), could also be given supp. diclofenac, which they could continue with at home as long as needed. However, mix. Paracetamol was the drug most children used after discharge. The parents recorded the number of doses given. All the children attended a post-operative check-up in the office 7 – 9 days after surgery. The child and the parents brought the written report to this visit and also gave their opinion orally of the procedure. The child was examined and weighed. At 6 months, and at one year after surgery, the parents were sent a questionnaire, where they reported on the child’s preoperative symptoms and how these had changed after surgery. Visual analogue scales were used. Snoring, breathing stops and eating problems were asked about, and there was also a more general question: Are you and your child satisfied with the result of the surgery?
2.1. Statistics The inter group differences in pain score and in incidence of weight gain/loss were tested with non-parametric methods (x 2 or Fisher). The other variables were compared using Student’s t-test or confidence intervals.
3. Results Twenty-one patients were operated on by TT, and 20 by TE of the 50 children who were origi-
173
nally randomized from the waiting list. Only one child in each group was excluded because the parents had a preference for the other surgical technique; in these cases the child was not operated according to the random allocation. Of the rest, four parents did not want any surgery at all, three of these were randomized to TE. The remaining three children excluded were randomized to TT, but at admission, the case history showed too many throat infections; in these cases, we preferred to perform tonsillectomy. The total perioperative blood loss and the average duration of surgery for the two groups are reported in Table 1. The postoperative pain-score was significantly lower for the first 24 h for the TT children than for the TE children, (see Fig. 2). The amount of analgesic drugs administered during the day of surgery was the same for both groups, four doses, despite the higher pain score for the TE children. Only two of the TE children had no scoring above three compared to to seven in the TT group. One of the TT children had to stay another night due to a high temperature. During the following seven days at home, the TT children received significantly fewer doses of analgesics than the TE children did. The mean difference was 8.8 doses (P= 0.0016, CI 95% 3.6– 14.1). (See Table 1). On the average, the TT children were pain-free three days earlier than the TE children (see Table 1). Likewise, TT children returned to a normal diet three days before the TE children. At the postoperative control, the TT children had gained weight and the TE children had lost weight. The mean difference was almost 800 g (P= 0.01, CI 95% 1.33–0.19) (see Table 1). The breathing obstruction symptoms were completely relieved 8–10 days postoperatively in all the children of both groups. The eating problems had also disappeared except for five TE children who still were in pain. Six months post surgery, two children in the TT group had started to snore again. These children were checked up in the office. One was found to have hypertrophied tonsils and the other was heavily overweight. The parents reported that snoring was less pronounced than pre surgery and the children no longer had apneas.
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
174
Table 1 Comparison of results after regular tonsillectomy (TE) and tonsillotomy with CO2-laser (TT)a
Time for surgery (min) Peri-operative bloodloss (ml) Doses of analgetics given during day 1–7 First pain-free day after surgery No. with weight loss by day 8–10 after surgery No. with unchanged weight 8–10 days after surgery No. with weight gain by day 8–10 after surgery Mean weight change (gram) by day 8–10 after surgery No. snoring one week after surgery No. snoring 6 months after surgery No. snoring 12 months after surgery
a
TE (n= 20)
TT (n =21)
27.59 13 39.5 9 39 1699 79 4 11
24.5 9 14 11 918 897 592 4
Mean difference Confidence inter- Level of signifival (95%) cance – 28.2 2.7 8.8 –
– 8.4–48.0 7.4–4.7 3.6–14.1 –
n.s. PB0.01 PB0.01 PB0.01 P= 0.0004 (chi2)
7
3
–
–
P =0.0004 (chi2)
2
14
–
–
P =0.0004 (chi2)
−400
+362
762
0 0
0 2
– –
– –
0
2
–
–
1331–193
P =0.01 ns P= 0.2560976 (Fisher), n.s. P= 0.2560976 (Fisher), n.s.
n.s., Not significant.
At 1-year follow-up, the same children reported snoring while none of the others had started to snore again. The snoring remained less pronounced than preoperatively.
4. Discussion
4.1. Pain de6elopment after surgery Tonsillectomy is a routine procedure in ENT surgery. However, recovery from tonsillectomy is a painful experience [5,6]. Almost all children got Pethidine, 5 mg i.v. in the recovery room. During the first 24 h post-operatively the TT-children showed significant lower pain judging from the VAS score, although both groups received the same amount of analgesic drugs. This indicates that the medication followed the conventional dosing schedule rather than the pain of the child. As a routine at our hospital, after discharge from the hospital the children were prescribed paracetamol or paracetamol/codeine 10 – 15 mg/ kg body weight (bwt) four times/day. The three children, who weighed more than 30 kg in addition also received supp diclofenac 25 mg, three
times/day. The parents had been told to give the child analgesics when the child was in pain. On the visit 7–9 days after operation, we could see from the report form that even if the TE children had indicated severe pain, several parents had not given the prescribed analgesics. Probably, the parents had not taken into account that refusal to eat could be an indirect indication of pain. This is in agreement with several studies that show that in general, post tonsillectomy pain is poorly managed by parents [7] and professionals [8].
Fig. 2. Postoperative pain recordings with VAS during the first 24 h after surgery. The diagram shows the distribution of all the hourly pain scores from the children of each group.
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
The TT children were pain-free after 5 days compared with the TE children who were not pain-free until day 8. This differences between the groups would probably have been less if the parents had been told to give analgesics four times/ day and in a higher dosage, whether the child claimed to be in pain or not. Andersson et al., have shown that paracetamol could be increased to 15–25 mg/bwt given as suppositories up to six times/day [9]. All variables measured in the present study consistently indicated less pain for the TT children: VAS, the amount of analgesic drugs needed, and which day normal eating was resumed. Especially, the influence on weight, during the first postoperative week for the two groups is remarkable, an increase in weight for the TT children and a loss for the TE children. The amount of weight gain and weight loss, respectively, can not be evaluated from the figures as the children preoperatively were weighed ‘on an empty stomach’ and at the postoperative visit came after lunch, some having eaten and some not. However, those conditions were the same for both the TE and TT children, and thus the comparison is valid. The decrease of weight for the TE children indicates probably both less intake of food and a catabolic metabolism due to pain. The findings in this study support the conclusion from our previous study [2] that a partial resection of the tonsil causes less pain than a complete tonsillectomy. The CO2-laser is an indispensable tool for tonsillotomy, but use of the laser does not by itself reduce the pain. In several studies [10,11] complete tonsillectomy performed with surgical laser caused a delayed pain that may be worse than the pain after conventional tonsillectomy. The difference is in the surface left to heal: the lymphoid tissue of the tonsils does not seem to be as pain sensitive as the tonsillar pillars and does not show the intense inflammation clearly visible during the recovery after tonsillectomy.
4.2. Post-operati6e bleeding There is a small but alarming risk of bleeding 7 – 10 days after tonsillectomy [6]. The intra opera-
175
tive bleeding observed in the present study was small in both groups, but was significantly smaller for the TT children. No postoperative bleeding was noted. Due to the comparatively small group, the results must be interpreted with care. However, our experience to date includes more than 100 children operated with CO2-laser: the 21 children in the present study, 33 children reported in a previous study [2] and more than fifty children operated after completion of the studies. In none of these cases we have had any postoperative bleeding. We do not expect such bleeding since the tonsils are completely epithelialized and look normal after 7–10 days.
4.3. Need for hospitalization In this study, all children were admitted as inpatients to insure uniform care for both groups and to provide similar conditions for pain registration. Only one child in the TT-group, who developed high temperature, was kept for an extra night. The positive results with respect to pain and bleeding were so convincing, that we have not hesitated to continue to do tonsillotomies with CO2-laser in day care surgery. In countries, like ours, where tonsillectomies are still performed with patients staying in the hospital for one or more nights, a change to tonsillotomy will be advantageous economically. However, with the pain better controlled after TE, nothing else in our investigation contradict the possibility to perform both kinds of surgery in day care.
4.4. Gender aspects Almost twice as many boys as girls taken from the ordinary waiting list, were operated in this study. Is tonsillar hypertrophy more commons in boys? A recent epidemiological study of snoring and other obstructive problems done in the same geographic area among 4 and 6-year-old children did not find sex differences [12]. An Icelandic study has earlier shown that the prevalence of sleep apnea for boys is highest at 2.5 years and for girls at 5 years [13]. Based on those data, we would have expected more girls to be treated. Could the reason be that parents (and the doc-
176
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
tors?) are more hesitant to have surgery performed on girls than on boys — or are the symptoms more noticeable in boys? These questions cannot be fully addressed here, but will be followed up in a later study.
the children operated with CO2-laser. A technique, which reduces post-operative pain, must be taken seriously. However, treatment of children who experience pain while still in the hospital and later at home still needs to be improved.
4.5. Long term impro6ement of obstruction References All the children had been put on the waiting list for operation due to obstructive symptoms: snoring and/or sleep apneas, mouth breathing, and/or eating problems. These symptoms were completely relived in both groups at 7 – 9 days after operation. Six months later, two children in the TT group had started to snore again. Only one child showed a recurrence of tonsillar hypertrophy. At the 1-year follow up, this child was still snoring, but according to the parents less than before surgery with no apneas, why we decided further expectancy. The symptoms progressed during the second year and the boy is now scheduled for TE. The second child, who snored after 6 months, was considerably overweight which could be the cause of snoring since no hypertrophy of the tonsils could be noted. This child was still snoring after 1 year but to a lesser degree and both the child and the parents were satisfied with the situation. An increase of weight the first year after tonsillectomy has been reported earlier [14]. It is probably due to both increased appetite and that the child develops a more anabolic situation after relief of the stressful, obstructive problems.
5. Conclusion The present study shows that it may be time to reconsider the operation technique for tonsillhypertrophy. The CO2-laser technique should not be confused with older methods of tonsillotomy [3,4]. The results of the present study indicate that tonsillotomy with CO2-laser are safe to perform and efficient in cases where obstruction and not streptococcal infections motivate the surgery. The improvement of symptoms and signs remained one year after surgery. The pain was much less for .
[1] A. Eliasson, D. Andesson, O. Densert, et al., Tonsillotomi hos barn med tonsillhyperplasi, Abstr Svenska La¨karsa¨llskapets Rikssta¨mma (Swe) (1996) 332. [2] A. Linder, A. Markstro¨m, E. Hultcrantz, Using the carbondioxide laser for tonsillotomy in children, Int. J. Pediatr. Otolaryngol. 50 (1999) 31 – 36. [3] N.G. McGuire, A method of guillotine tonsillectomy with an historical review, J. Laryngol. Otol. 81 (1966) 187 – 195. [4] M. Wake, P. Glossop, Guillotine and dissection tonsillectomy compared, J. Laryngol. Otol. 103 (1989) 588 – 591. [5] A.G. Toma, J. Blanshard, Eynon-Lewis, M.W. Bridger, Post-tonsillectomy pain: the first ten days, J. Laryngol. Otol. 109 (1995) 963 – 964. [6] B.E. Linden, C.W. Gross, T.E. Long, R.H. Lazar, Morbidity in pediatric tonsillectomy, Laryngoscope 100 (1990) 120 – 124. [7] L.H. Kotiniemi, P.T. Ryhanen, J. Valanne, R. Jokela, A. Mustonen, E. Poukkula, Postoperative symtoms at home following day-case surgery in children: a multicentre survey of 551 children, Anaesthesia 52 (1997) 963 – 969. [8] F.F. Warnock, J. Lander, Pain progression, intensity and outcomes following tonsillectomy, Pain 75 (1998) 37 – 45. [9] B.J. Anderson, N.H. Holford, Rectal paracetamol dosing regimens: determination by computer simulation, Paediatr. Anaest. 7 (1997) 451 – 455. [10] S.A. Martinez, D.P. Akin, Laser tonsillectomy and adenoidectomy, Otolaryngol. Clin. North Am. 20 (1987) 371 – 376. [11] R.E. Oas, J.P. Bartels, KTP-532 laser tonsillectomy: a comparison with standard technique, Laryngoscope 100 (1990) 385 – 388. [12] B. Lo¨fstrand-Tidestro¨m, B. Thilander, J. Ahlqvist-Rastad, O. Jakobson, E. Hultcrantz, Breathing obstruction in relation to craniofacial and dental arch morphology in four year-old children, J. Eur. Orthodon. 21 (1999) 323 – 332. [13] T. Gislason, B. Benediktsdottir, Snoring, apneic episodes, and nocturnal hypoxemia among children 6 months to 6 years old. An epidemiological study of lower limit of prevalence, Chest 107 (1995) 963 – 966. [14] J. Ahlquist-Rastad, E. Hultcrantz, H. Melander, H. Svanholm, Body growth in relation to tonsillar enlargement and tonsillectomy, Int. J. Pediatr. Otolaryngol. 24 (1992) 55 – 61.
Tonsillectomy or tonsillotomy? — a randomized study comparing postoperative pain and long-term effects Elisabeth Hultcrantz *, Arne Linder, Agneta Markstro¨m Department of Otorhinolaryngology, ENT Clinic, Uni6ersity Hospital, 75185 Uppsala, Sweden Received 9 June 1999; received in revised form 8 September 1999; accepted 12 September 1999
Abstract Background: tonsillectomy (TE) is currently the most common treatment for children with snoring and sleep apnea. Many of these children have not had any severe throat infections. To cure such children from their obstructive problems, without influencing the immunological function of the tonsils, tonsillotomy (TT) with CO2-laser was performed in a randomized study comparing it to regular tonsillectomy, with special attention to postoperative pain and symptom recurrence. Method: 41 children 3.5–8 years-old were included — 21 ‘TT’s’ and 20 ‘TE’s’. They were all operated under the same anesthesia and followed the same postoperative scheme for analgesia. A visual analogue scale for pain measurements with faces was used for the first 24 h. After that, each day until pain-free, the parents registered the child’s pain on a three graded scale, what the child was able to eat, and the amount of analgesic drugs used. Results: all the children were cured from their breathing obstruction. The mean time used for the surgery was the same and no postoperative bleeding was seen in either group. ‘TT children’ were pain-free after 5 days and ‘TE children’ after 8 days. Eight to ten days after surgery, the TT-children had gained weight and the TE children lost weight significantly. The TE group used twice as much analgesic drugs as the TT group during the first postoperative week. The TT group was healed with normal-looking, but small tonsils after 8 – 10 days; the TE group often still showed edema and crusts. At the one-year follow-up 2/21among the ‘TT-children’ snored, but did not require re-surgery. Conclusion: tonsillotomy is much less painful than TE and children recover more quickly. Results with respect to breathing obstruction are almost the same for both methods at 1-year follow-up. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Tonsil surgery; Laser in children; Randomized study; Postoperative pain
1. Introduction
* Corresponding author. Fax: +46-18-665360. E-mail address: [email protected] Hultcrantz)
(E.
During the last few years, a new surgical technique for tonsillar surgery has developed — a partial removal of the tonsils with use of CO2 laser [1,2]. The technique is used thus far in a few clinics in Scandinavia, where it is employed
0165-5876/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 5 8 7 6 ( 9 9 ) 0 0 2 7 4 - 8
172
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
mostly on children who have obstructive problems caused by tonsillar hypertrophy. The general impression has been that this kind of surgery seems to be gentler to the patients; they are noted to recover faster and to have less pain. Before recommending the technique to other centers, some issues must be addressed. Are there risks associated with partial removal of the tonsils? One obvious possibility is that the remaining parts of the tonsils will start to grow again and give the child the same symptoms once more. Another possible risk is that the partial resection will cause scar tissue in the tonsils. The deformed tissue may predispose to deeper infections or peritonsillitis. Older colleagues, who have witnessed how guillotine tonsillotomies were performed frequently during the first part of this century and then abandoned, draw a parallel with that technique. However, going through the literature we found that there is really no convincing evidence for such risks if the surgery was performed under safe conditions in skilful hands [3,4]. Thus, the fear of late infectious complications after a partial resection of the tonsil may well be unfounded. Furthermore, if the risks can be estimated, we also need to document the benefits. Are there any real advantages compared to regular tonsillectomy and in that case how can these advantages be weighed against the risks? The purpose of the present investigation is to compare the effect of tonsillotomy using CO2 laser (TT) with tonsillectomy using regular surgical technique (TE) with respect to postoperative pain, duration of surgery, perioperative bleeding, and need for hospitalization. The long-term effects on the obstructive symptoms after 6 months and 1 year are also evaluated. The patients were randomized to the two surgical procedures.
2. Method Out of 50 children, who were randomized to the study 41 were included, 15 girls and 26 boys 3.5–8 years old (69 1.5). All of them had been put on the waiting list for tonsillar surgery due to obstructive problems: snoring and/or sleep apnea, mouth breathing, and/or eating problems. They all had a verified tonsillar hyperplasia and none of them had had repeated streptococcal throat infections. The randomization was performed from the waiting list and the parents informed by mail. They could change the decision about the choice of method, but in that case the child was excluded from the study. The Ethics Committee at Uppsala University approved the investigation. All patients were kept on the ward 24 h postoperatively, irrespective of the procedure performed. The children were weighed on the morning of the surgery ‘on an empty stomach’. The surgery was performed in 21 cases as tonsillotomy (TT) with CO2-laser, 20 W (continuous mode). The surgical technique is described in detail in our previous paper [2]. The same surgeon (AL) performed all the tonsillotomies. Those specialists in the department, who regularly do tonsillectomies, operated the 20 control patients. Three children in each group had an adenoidectomy on the same occasion using conventional surgical technique. The total time in the operating theatre, the duration of the surgery, and the amount of blood loss were recorded. Postoperative pain was measured by means of a pediatric visual analogue scale (VAS) showing faces with increasing expressions of distress (Fig. 1). VAS measurements were obtained during the first 24 h, once every hour while the patient was awake. Pain, general condition and alertness were recorded, as well as the elapsed time before resumption of drinking and eating.
Fig. 1. Visual analogue scale (VAS) used in the pain measurements during the over-night stay on the ward. The child indicates a face that appears to match the pain they are feeling at that moment. The nurse or surgeon records the ordinal number of the face (the leftmost face showing no distress is given the number 0).
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
A simplified rating was used for the subsequent seven days. At discharge 24 h after surgery, the parents were instructed to fill out a form each day recording the child’s pain on a three graded scale, and how much and at what time analgesic drugs were administered. They were also asked to record how well the child ate (less, more, normal amounts) and which kind of food (liquid, pure´e or normal). While hospitalized and afterwards, paracetamol or paracetamol/codeine were used (Supp. Citodone® 25 mg/kg bwt) for pain relief. Pethidine i.v. could be given if required, in doses of 5 mg during the first few hours. Children weighing more than 30 kg (only three), could also be given supp. diclofenac, which they could continue with at home as long as needed. However, mix. Paracetamol was the drug most children used after discharge. The parents recorded the number of doses given. All the children attended a post-operative check-up in the office 7 – 9 days after surgery. The child and the parents brought the written report to this visit and also gave their opinion orally of the procedure. The child was examined and weighed. At 6 months, and at one year after surgery, the parents were sent a questionnaire, where they reported on the child’s preoperative symptoms and how these had changed after surgery. Visual analogue scales were used. Snoring, breathing stops and eating problems were asked about, and there was also a more general question: Are you and your child satisfied with the result of the surgery?
2.1. Statistics The inter group differences in pain score and in incidence of weight gain/loss were tested with non-parametric methods (x 2 or Fisher). The other variables were compared using Student’s t-test or confidence intervals.
3. Results Twenty-one patients were operated on by TT, and 20 by TE of the 50 children who were origi-
173
nally randomized from the waiting list. Only one child in each group was excluded because the parents had a preference for the other surgical technique; in these cases the child was not operated according to the random allocation. Of the rest, four parents did not want any surgery at all, three of these were randomized to TE. The remaining three children excluded were randomized to TT, but at admission, the case history showed too many throat infections; in these cases, we preferred to perform tonsillectomy. The total perioperative blood loss and the average duration of surgery for the two groups are reported in Table 1. The postoperative pain-score was significantly lower for the first 24 h for the TT children than for the TE children, (see Fig. 2). The amount of analgesic drugs administered during the day of surgery was the same for both groups, four doses, despite the higher pain score for the TE children. Only two of the TE children had no scoring above three compared to to seven in the TT group. One of the TT children had to stay another night due to a high temperature. During the following seven days at home, the TT children received significantly fewer doses of analgesics than the TE children did. The mean difference was 8.8 doses (P= 0.0016, CI 95% 3.6– 14.1). (See Table 1). On the average, the TT children were pain-free three days earlier than the TE children (see Table 1). Likewise, TT children returned to a normal diet three days before the TE children. At the postoperative control, the TT children had gained weight and the TE children had lost weight. The mean difference was almost 800 g (P= 0.01, CI 95% 1.33–0.19) (see Table 1). The breathing obstruction symptoms were completely relieved 8–10 days postoperatively in all the children of both groups. The eating problems had also disappeared except for five TE children who still were in pain. Six months post surgery, two children in the TT group had started to snore again. These children were checked up in the office. One was found to have hypertrophied tonsils and the other was heavily overweight. The parents reported that snoring was less pronounced than pre surgery and the children no longer had apneas.
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
174
Table 1 Comparison of results after regular tonsillectomy (TE) and tonsillotomy with CO2-laser (TT)a
Time for surgery (min) Peri-operative bloodloss (ml) Doses of analgetics given during day 1–7 First pain-free day after surgery No. with weight loss by day 8–10 after surgery No. with unchanged weight 8–10 days after surgery No. with weight gain by day 8–10 after surgery Mean weight change (gram) by day 8–10 after surgery No. snoring one week after surgery No. snoring 6 months after surgery No. snoring 12 months after surgery
a
TE (n= 20)
TT (n =21)
27.59 13 39.5 9 39 1699 79 4 11
24.5 9 14 11 918 897 592 4
Mean difference Confidence inter- Level of signifival (95%) cance – 28.2 2.7 8.8 –
– 8.4–48.0 7.4–4.7 3.6–14.1 –
n.s. PB0.01 PB0.01 PB0.01 P= 0.0004 (chi2)
7
3
–
–
P =0.0004 (chi2)
2
14
–
–
P =0.0004 (chi2)
−400
+362
762
0 0
0 2
– –
– –
0
2
–
–
1331–193
P =0.01 ns P= 0.2560976 (Fisher), n.s. P= 0.2560976 (Fisher), n.s.
n.s., Not significant.
At 1-year follow-up, the same children reported snoring while none of the others had started to snore again. The snoring remained less pronounced than preoperatively.
4. Discussion
4.1. Pain de6elopment after surgery Tonsillectomy is a routine procedure in ENT surgery. However, recovery from tonsillectomy is a painful experience [5,6]. Almost all children got Pethidine, 5 mg i.v. in the recovery room. During the first 24 h post-operatively the TT-children showed significant lower pain judging from the VAS score, although both groups received the same amount of analgesic drugs. This indicates that the medication followed the conventional dosing schedule rather than the pain of the child. As a routine at our hospital, after discharge from the hospital the children were prescribed paracetamol or paracetamol/codeine 10 – 15 mg/ kg body weight (bwt) four times/day. The three children, who weighed more than 30 kg in addition also received supp diclofenac 25 mg, three
times/day. The parents had been told to give the child analgesics when the child was in pain. On the visit 7–9 days after operation, we could see from the report form that even if the TE children had indicated severe pain, several parents had not given the prescribed analgesics. Probably, the parents had not taken into account that refusal to eat could be an indirect indication of pain. This is in agreement with several studies that show that in general, post tonsillectomy pain is poorly managed by parents [7] and professionals [8].
Fig. 2. Postoperative pain recordings with VAS during the first 24 h after surgery. The diagram shows the distribution of all the hourly pain scores from the children of each group.
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
The TT children were pain-free after 5 days compared with the TE children who were not pain-free until day 8. This differences between the groups would probably have been less if the parents had been told to give analgesics four times/ day and in a higher dosage, whether the child claimed to be in pain or not. Andersson et al., have shown that paracetamol could be increased to 15–25 mg/bwt given as suppositories up to six times/day [9]. All variables measured in the present study consistently indicated less pain for the TT children: VAS, the amount of analgesic drugs needed, and which day normal eating was resumed. Especially, the influence on weight, during the first postoperative week for the two groups is remarkable, an increase in weight for the TT children and a loss for the TE children. The amount of weight gain and weight loss, respectively, can not be evaluated from the figures as the children preoperatively were weighed ‘on an empty stomach’ and at the postoperative visit came after lunch, some having eaten and some not. However, those conditions were the same for both the TE and TT children, and thus the comparison is valid. The decrease of weight for the TE children indicates probably both less intake of food and a catabolic metabolism due to pain. The findings in this study support the conclusion from our previous study [2] that a partial resection of the tonsil causes less pain than a complete tonsillectomy. The CO2-laser is an indispensable tool for tonsillotomy, but use of the laser does not by itself reduce the pain. In several studies [10,11] complete tonsillectomy performed with surgical laser caused a delayed pain that may be worse than the pain after conventional tonsillectomy. The difference is in the surface left to heal: the lymphoid tissue of the tonsils does not seem to be as pain sensitive as the tonsillar pillars and does not show the intense inflammation clearly visible during the recovery after tonsillectomy.
4.2. Post-operati6e bleeding There is a small but alarming risk of bleeding 7 – 10 days after tonsillectomy [6]. The intra opera-
175
tive bleeding observed in the present study was small in both groups, but was significantly smaller for the TT children. No postoperative bleeding was noted. Due to the comparatively small group, the results must be interpreted with care. However, our experience to date includes more than 100 children operated with CO2-laser: the 21 children in the present study, 33 children reported in a previous study [2] and more than fifty children operated after completion of the studies. In none of these cases we have had any postoperative bleeding. We do not expect such bleeding since the tonsils are completely epithelialized and look normal after 7–10 days.
4.3. Need for hospitalization In this study, all children were admitted as inpatients to insure uniform care for both groups and to provide similar conditions for pain registration. Only one child in the TT-group, who developed high temperature, was kept for an extra night. The positive results with respect to pain and bleeding were so convincing, that we have not hesitated to continue to do tonsillotomies with CO2-laser in day care surgery. In countries, like ours, where tonsillectomies are still performed with patients staying in the hospital for one or more nights, a change to tonsillotomy will be advantageous economically. However, with the pain better controlled after TE, nothing else in our investigation contradict the possibility to perform both kinds of surgery in day care.
4.4. Gender aspects Almost twice as many boys as girls taken from the ordinary waiting list, were operated in this study. Is tonsillar hypertrophy more commons in boys? A recent epidemiological study of snoring and other obstructive problems done in the same geographic area among 4 and 6-year-old children did not find sex differences [12]. An Icelandic study has earlier shown that the prevalence of sleep apnea for boys is highest at 2.5 years and for girls at 5 years [13]. Based on those data, we would have expected more girls to be treated. Could the reason be that parents (and the doc-
176
E. Hultcrantz et al. / Int. J. Pediatr. Otorhinolaryngol. 51 (1999) 171–176
tors?) are more hesitant to have surgery performed on girls than on boys — or are the symptoms more noticeable in boys? These questions cannot be fully addressed here, but will be followed up in a later study.
the children operated with CO2-laser. A technique, which reduces post-operative pain, must be taken seriously. However, treatment of children who experience pain while still in the hospital and later at home still needs to be improved.
4.5. Long term impro6ement of obstruction References All the children had been put on the waiting list for operation due to obstructive symptoms: snoring and/or sleep apneas, mouth breathing, and/or eating problems. These symptoms were completely relived in both groups at 7 – 9 days after operation. Six months later, two children in the TT group had started to snore again. Only one child showed a recurrence of tonsillar hypertrophy. At the 1-year follow up, this child was still snoring, but according to the parents less than before surgery with no apneas, why we decided further expectancy. The symptoms progressed during the second year and the boy is now scheduled for TE. The second child, who snored after 6 months, was considerably overweight which could be the cause of snoring since no hypertrophy of the tonsils could be noted. This child was still snoring after 1 year but to a lesser degree and both the child and the parents were satisfied with the situation. An increase of weight the first year after tonsillectomy has been reported earlier [14]. It is probably due to both increased appetite and that the child develops a more anabolic situation after relief of the stressful, obstructive problems.
5. Conclusion The present study shows that it may be time to reconsider the operation technique for tonsillhypertrophy. The CO2-laser technique should not be confused with older methods of tonsillotomy [3,4]. The results of the present study indicate that tonsillotomy with CO2-laser are safe to perform and efficient in cases where obstruction and not streptococcal infections motivate the surgery. The improvement of symptoms and signs remained one year after surgery. The pain was much less for .
[1] A. Eliasson, D. Andesson, O. Densert, et al., Tonsillotomi hos barn med tonsillhyperplasi, Abstr Svenska La¨karsa¨llskapets Rikssta¨mma (Swe) (1996) 332. [2] A. Linder, A. Markstro¨m, E. Hultcrantz, Using the carbondioxide laser for tonsillotomy in children, Int. J. Pediatr. Otolaryngol. 50 (1999) 31 – 36. [3] N.G. McGuire, A method of guillotine tonsillectomy with an historical review, J. Laryngol. Otol. 81 (1966) 187 – 195. [4] M. Wake, P. Glossop, Guillotine and dissection tonsillectomy compared, J. Laryngol. Otol. 103 (1989) 588 – 591. [5] A.G. Toma, J. Blanshard, Eynon-Lewis, M.W. Bridger, Post-tonsillectomy pain: the first ten days, J. Laryngol. Otol. 109 (1995) 963 – 964. [6] B.E. Linden, C.W. Gross, T.E. Long, R.H. Lazar, Morbidity in pediatric tonsillectomy, Laryngoscope 100 (1990) 120 – 124. [7] L.H. Kotiniemi, P.T. Ryhanen, J. Valanne, R. Jokela, A. Mustonen, E. Poukkula, Postoperative symtoms at home following day-case surgery in children: a multicentre survey of 551 children, Anaesthesia 52 (1997) 963 – 969. [8] F.F. Warnock, J. Lander, Pain progression, intensity and outcomes following tonsillectomy, Pain 75 (1998) 37 – 45. [9] B.J. Anderson, N.H. Holford, Rectal paracetamol dosing regimens: determination by computer simulation, Paediatr. Anaest. 7 (1997) 451 – 455. [10] S.A. Martinez, D.P. Akin, Laser tonsillectomy and adenoidectomy, Otolaryngol. Clin. North Am. 20 (1987) 371 – 376. [11] R.E. Oas, J.P. Bartels, KTP-532 laser tonsillectomy: a comparison with standard technique, Laryngoscope 100 (1990) 385 – 388. [12] B. Lo¨fstrand-Tidestro¨m, B. Thilander, J. Ahlqvist-Rastad, O. Jakobson, E. Hultcrantz, Breathing obstruction in relation to craniofacial and dental arch morphology in four year-old children, J. Eur. Orthodon. 21 (1999) 323 – 332. [13] T. Gislason, B. Benediktsdottir, Snoring, apneic episodes, and nocturnal hypoxemia among children 6 months to 6 years old. An epidemiological study of lower limit of prevalence, Chest 107 (1995) 963 – 966. [14] J. Ahlquist-Rastad, E. Hultcrantz, H. Melander, H. Svanholm, Body growth in relation to tonsillar enlargement and tonsillectomy, Int. J. Pediatr. Otolaryngol. 24 (1992) 55 – 61.