The main findings in the present randomized double-blinded study are in line with previous comparative studies, which have shown the advantages of TT compared with TE concerning postoperative pain and bleeding.
Pain
The results suggest that ATT is associated with less postoperative pain than ATE, but the differences were relatively small. Only two of the six pain-related outcomes showed significant differences between the groups: The first day when the child reported being pain free were day median (IQR) 8 (5–10) after ATE and day 5 (3–8) after ATT, and the first day caregiver reported reduction of the child’s pain to VAS ≤ 5 was day median (IQR) 1(1–4.5) for ATE and day 0 (1–2) for ATT. Other pain-related variables showed no significant differences between the groups in this small sample.
Comparable results with our study were found in a review by Walton et al. [
14], evaluating TT versus TE in pediatric populations, concluding that TT was equivalent or superior to TE regarding recovery-related outcomes. Only some of the studies in this review were blinded, and none of the blinded studies compared cold steel tonsillectomy with coblation TT, as in the present study. However, one study [
17] from the review [
14] showed similar results as ours with a randomized, double-blinded study. They compared three different surgical techniques: electrocautery TE, coblation TT, and microdebrider TT, which showed better recovery after TT (coblation and microdebrider were comparable) than after TE in terms of days with pain and return to a normal diet, but with no differences in average pain scores [
17].
Moreover, Lister et al. performed a randomized blinded study in which the tonsil on one side was removed with electrosurgical TE, and the other side with microdebrider TT in the same individual, with significantly less pain on the TT side until postoperative day 10 [
16]. A proposed explanation to the reduced pain after TT is the preservation of tonsil tissue and the tonsillar capsule, with preserved protection of underlying vessels and nerves. Further, a correlation of the inflammatory response and the extent of the surgical intervention has been suggested, however, a randomized trial evaluating the inflammatory response to surgery after TT vs TE in children did not show any significant differences [
19].
The surgical depth of TT tends to vary in previous studies, TTs were often performed all the way to the tonsillar capsule, which is deeper than in our study where TTs were less invasive, stopping at the level of the anterior and posterior pillars. Possibly, the depth of TT could be another factor correlating to the level of postoperative pain. Also, the technique for TT and TE varies in previous studies, making direct comparisons to our results difficult. For example, a randomized study compared cold technique TT (scalpel and scissors) with cold steel TE in a pediatric population, showing less postoperative analgesic use in the TT group, but no differences in pain scoring [
20].
Our results demonstrate smaller group differences in pain-related outcomes than some previous non-blinded randomized studies [
6,
21]. This discrepancy to our double-blinded study could suggest a possible factor of expectation: the patient’s and parent’s expectation that the pain might be less after ATT than after ATE, depending on the preoperative information given.
Another possible explanation of the smaller group differences is that all the TEs in our study were performed with cold steel technique. A large Swedish study of > 18,000 children 1–18 years undergoing TE, reported less postoperative pain if TE was performed with cold technique than with hot technique [
16]. The study also reported a correlation to age, with more pain for children of older ages [
22] and all children in our study were very young (mean age of 47 ± 15 months). However, a recent review comparing TE vs TT concluded that neither the extent of surgery or surgical technique seemed to affect the recovery outcomes [
23]. The study included 16 RCTs showing only moderate advantages for TT. Nonetheless, we consider our conservative method for TT safe, but as always, the risk of re-growth of the tonsils must be taken into consideration. This risk is highest for the youngest children, especially those under 3 years [
24].
Bleeding
A significant difference in perioperative blood loss was seen, with higher values in the ATE group, but this can be considered of little clinical relevance since the bleeding volumes were relatively small in both groups; 55.1 ± 33.9 ml for TE and 28.6 ± 15.6 ml for TT.
Two cases (5%) of postoperative hemorrhage occurred in the ATE group, one (2.5%) primary (within 24 h) with a return to theatre and one secondary (7 days after surgery). No case of postoperative bleeding was reported in the ATT group.
In comparison, another recent study compared TT and TE in children with OSA reported a postoperative bleeding frequency of 0.2% after TT and 2.9% after TE, of which 1.6% needed a return to theatre [
25].
Moreover, the aforementioned Swedish study by Elinder et al. of TE in > 18,000 children, reported a primary bleeding frequency of 2.0% for children undergoing tonsillectomy on OSA indication, with increased risk at older ages [
22].
The present study showed no significant difference between the groups regarding postoperative bleeding, which may be due to the small sample size.
Strengths and limitations
The primary strength of the present study is the randomized, double-blinded design, minimizing the risk of selection bias as well as interpreter- and expectation bias. Moreover, the young age of the children in the study, 2–6 years, is another strength since these are the age groups where tonsil surgery incidence peaks and, therefore, of high clinical interest. Only a few previous studies have included children as young as 2 years.
However, the young age of the included children is also a weakness of the study, as there is no validated tool for assessing pain in children younger than four years, which was the case for 65% of the study population. We used the FPS-R, validated for children 4 years and older [
13], and this was chosen since it is easy to use for both parents and children and a recommended tool to assess acute and postoperative pain in children [
14]. There are many sources of bias in children’s self-report of pain, e.g., children under five years have a tendency to use only the extremes of scales [
26]. It cannot be ruled out that this could have affected the results. On the other hand, returning to normal diet is probably independent of the child’s age, and this parameter did not show significant differences between the groups.
Another weakness is that TT can be performed with several techniques and this study analyzed only a conservative ATT performed with coblation and, therefore, the generalizability can be limited. We chose coblation since it is the most common method for TT in Sweden [
27]. Further, the relatively small study population and that no power analysis was performed fort these secondary outcomes. Thus, there is a risk that the study was underpowered to investigate smaller differences between groups. In summary, we consider our results, with the minor differences between techniques, to be of uncertain clinical significance.