Is partial tonsillectomy equivalent to total tonsillectomy for obstructive symptoms?

Abstract

Partial tonsillectomy (“tonsillotomy,” “intracapsular tonsillectomy,” or “subtotal tonsillectomy”) was commonplace until the 1930s, when the procedure was largely replaced by complete tonsillectomy for a variety of reasons, including concerns about possible regrowth and effects of retained tonsil tissue. Since the late 1980s, there has been renewed interest in partial tonsillectomy, particularly in the treatment of children with obstructive symptoms. Numerous studies have compared surgical techniques, and analyzed perioperative morbidity and recovery after partial versus total tonsillectomy. The evidence consistently shows that partial tonsillectomy—whatever technique is used—has lower morbidity and equivalent or easier recovery than total tonsillectomy. However, the question remains: is partial tonsillectomy as effective as total tonsillectomy for patients with airway obstruction? That is the aim of this brief review. One of the earliest randomized prospective studies to address efficacy was published by Densert et al. in 20011 (Table I). In this study, 43 children with obstructive symptoms were randomized to receive either total or partial tonsillectomy. Snoring, apnea, and well-being were evaluated by parents using a visual analog scale before surgery, and at 3 months and 2 years after surgery. The study found that both groups improved after surgery, and that there was no statistically significant difference in clinical symptoms between groups. Chan et al.2 reported a multisite, prospective study of 55 children with obstructive symptoms, randomized to receive either partial or total tonsillectomy. Efficacy was assessed by parents at 3 and 12 months after surgery using a questionnaire rating frequency of 13 obstructive symptoms (nonvalidated). Children also had postoperative clinical examinations. At 3 and 12 months, improvement in obstructive symptoms did not differ statistically between treatment groups. Like several other early studies, this was a small study with no power analysis reported, leaving open the possibility that it was underpowered to detect a significant difference between groups. In 2006, Ericsson et al.3 published a prospective randomized study of 92 Swedish children undergoing either partial or total tonsillectomy, comparing frequency of relapse in snoring or infections and possible long-term behavioral changes. Parents completed the Child Behavior Checklist (validated and widely used instrument) and a nonvalidated clinical questionnaire preoperatively, and 1 year and 3 years after surgery. Also, at 3 years, parents completed the Glasgow Children's Benefit Inventory by mail. There were significant improvements in health status, behavior, snoring (frequency and loudness), concentration, temper, and stamina at 1 year and 3 years after surgery in both groups, and there were no differences between groups in those outcomes. At 3 years, 2 of 49 (4%) of children in the partial tonsillectomy group had subsequently undergone total tonsillectomy, one for persistent tonsillitis and another for increased snoring. In addition to subjective and health-related quality of life outcomes, some have reported more objective outcome data, such as Apnea Hypopnea Index (AHI). In general, these studies tend to be small, and without control groups. Friedman et al.4 published the largest such study to date. This was a 5-year retrospective uncontrolled review of 159 patients who underwent partial tonsillectomy with adenoidectomy. Mean preoperative AHI was 17.8 and mean postoperative AHI was 3.3 (P < .05). A postoperative AHI <1 was achieved in 54.7% of patients. The authors compared this to pooled studies of AHI after total tonsillectomy, which found similar results: 61% to 66% of children achieved a postoperative AHI ranging from 1 to 5 in total tonsillectomy studies. The authors concluded that although complete resolution was not seen in all cases, partial tonsillectomy and adenoidectomy significantly improved AHI at a rate similar to total tonsillectomy and adenoidectomy. Direct comparisons between studies are difficult, however, because of different entry criteria and nonstandardized outcomes. The theoretic risk of tonsillar regrowth following partial tonsillectomy is often alluded to, although rarely directly addressed. A long-term study by Eviatar et al.5 in Israel evaluated several outcomes, including tonsil regrowth, 10–14 years after surgery. They attempted to contact 50 patients who had undergone partial tonsillectomy and a selected (nonrandomized) control group of 21 total tonsillectomy patients. All completed a telephone survey (nonvalidated), and any patients identifying a problem were advised to come in for physical examination. Response rate was 33 of 50 (66%) for partial tonsillectomy and 16 of 21 (76%) for total tonsillectomy. The authors found no statistically significant difference between groups with respect to reported clinical outcomes such as snoring or recurrent infection, or tonsil regrowth. Based upon the available evidence, partial tonsillectomy is equivalent to total tonsillectomy for the treatment of obstructive symptoms in children; however, the outcomes are not standardized and many studies are limited. There are several level 1 studies, but more level 4 studies. Some of the level 1 studies are likely underpowered, but nevertheless there is adequate level 1 evidence comparing partial tonsillectomy to total tonsillectomy for subjective outcome of obstructive symptoms.