Short-Term Weight Gain after Tonsillectomy Does Not Lead to Overweight: A Systematic Review

Different studies and systematic reviews have reported weight increase after tonsillectomy. However, the odds of a child being overweight or obese after tonsillectomy were no different than before surgery, according to a few studies. This systematic review aims to analyze the impact of adenotonsillectomy (TA) on weight gain and identify subgroups of children and adolescents at risk of experiencing weight gain. A systematic search included studies published in the last ten years. The PICO framework was used in the selection process, and evidence was assessed using the GRADE system. A total of 26 studies were included, and moderate–high level quality ones showed that children who underwent TA could present an increase in BMI z-score. However, this weight gain was significant in individuals younger than six years old and was considered catch-up growth in underweight subjects at baseline. In contrast, for normal-weight or overweight individuals, TA did not lead to overweight per se. At the same time, diet changes and overfeeding did not have a leading role in weight gain. In conclusion, TA may not be an independent risk factor for unfavorable weight gain in children; however, individuals who were underweight pre-operatively or younger than six years reported more weight gain after TA than expected.


Introduction
In the last 20 years, the prevalence of overweight and obesity among children and adolescents has increased rapidly.The principal cause is an imbalance between calorie intake and physical activity, with socioenvironmental factors playing a crucial role [1].
Tonsillectomy is a standard surgical procedure in the pediatric age, and it is indicated for treating obstructive sleep apnea (OSA) and chronic throat infections.Adenoidectomy is performed for nasal passage obstruction, OSA, recurrent otitis media, and rhinosinusitis.Obesity has been recognized as a risk factor for OSA [2], and children with obesity have more tonsillectomy/adenoidectomy procedures compared with their peers [3,4].Although tonsillectomy, with or without adenoidectomy, is a relatively safe surgery, postoperative complications have been reported in multiple studies.Among these, post-tonsillectomy hemorrhage is the most common, followed by possible infections, taste disturbances, breathing problems, and dehydration [5].
The connection between tonsillectomy and weight gain and growth has been controversial and discussed for decades [6,7].Weight increase after tonsillectomy has been reported in different studies and the most recent systematic reviews [8,9], but the odds of a child being overweight or obese after tonsillectomy in the next 12-18 months were not different than they were before surgery, according to a few studies [10,11].Other study designs, methods, baseline weight and age of the subjects, additional indications for tonsillectomy, and outcomes analyzed likely explain the heterogeneity in the literature results.
The mechanisms underlying weight gain after adenotonsillectomy (TA) surgery are still unknown [12]: (i) children with enlarged adenoids and/or tonsils could experience more frequent infections, swallowing problems, dysphagia, or odynophagia, and these disorders can lead to decreased calorie intake; when the tonsils are removed, children can consume a more significant amount of calories [13]; (ii) in children with OSA, intermittent upper airway obstruction during sleep has been thought to increase the work of breathing, and therefore increase energy expenditure at night, and probably after TA, this energy expenditure decreases [9]; (iii) in addition, reduced episodes of OSA after TA could lead to hormonal dysregulation, increasing levels of insulin-like growth factor-1 and insulin growth factor binding protein-3 [14].
The aims of this systematic review are: (1) to provide an up-to-date summary of the available evidence on the impact of TA on weight gain and eventually increased prevalence of overweight and obesity; (2) to identify subgroups of children and adolescents at risk of weight gain; (3) to elucidate mechanisms beyond increased caloric intake that underlie weight gain.

Search Strategy
We searched electronic databases (Pubmed, EMBASE, The Cochrane Library, Web of Science, Clinicaltrial.gov,International Clinical Trials Registry Platform) for studies published between 1 January 2013, and 1 August 2023.Search terms or "MESH" (Medical Subject Headings) for this systematic review included different combinations: "weight gain" or "overweight", or "obesity" or "obese", or "BMI" or "weight" or "unhealthy weight" or "fat" AND "tonsillectomy" or "adenoidectomy" or "adenotonsillectomy".
We also screened the reference lists of eligible studies to avoid missing any relevant studies.

Criteria for Study Selection
We conducted a systematic search of the literature according to the PICOS model (Population, Intervention, Comparison, Results, Study design): Randomized clinical trials (RCTs), observational studies (cohort, case-control, cross-sectional studies), exploratory studies, mix of qualitative and quantitative studies Inclusion criteria were: (i) study population: children and adolescents (aged 1-18 years) who had undergone tonsillectomy with or without adenoidectomy; (ii) study type: observational studies (cohort, case-control, cross-sectional studies), exploratory studies, a mix of qualitative and quantitative studies; (iii) review articles were excluded, but their reference lists were screened to identify potentially eligible studies; (iv) only published full papers were included, whereas abstracts only were not included; (v) data on weight gain: weight, height, BMI and relative percentiles, metabolic and hormonal changes, before and after tonsillectomy; (vi) publication date: last ten years (2013-2023) to include studies after the last systematic review reported in the literature [9].
Exclusion criteria: (i) data available only for adults ≥18 years or subjects with syndromes (Down, Prader Willi, Rohadnet, etc.); (ii) animal data; (iii) case reports; studies with <10 patients who underwent tonsillectomy; (iv) full paper not available; (v) study not yet published; (vi) studies on tonsillectomy not reporting weight variation within 24 months; (vii) languages other than English were not "a priori" exclusion criteria.

Data Extraction and Management
Using the search strategy, two independent investigators (AN, LO) screened the titles and abstracts of the identified studies for inclusion.Any discrepancies were resolved by consensus or consultation with a third investigator (RF).After abstract selection, four investigators conducted a complete paper analysis (EMA, LL, LF, LO).
The following characteristics were evaluated for each study in the full paper: (i) reference details: authorship(s); published or unpublished; year of publication; period in which the study was conducted; other relevant cited papers; (ii) study characteristics: study design, topic, treatment period, follow-up duration, region; (iii) population characteristics: number of participants who underwent tonsillectomy with or without adenoidectomy, number with OSA, age and demographic data, baseline weight, BMI, fat, waist circumference and other related parameters; comparator characteristics; (iv) methodology: assessment of growth measurements, metabolic and hormone changes, caloric intake, physical exercise; (v) main results: growth pattern, body composition parameters, factors influencing post-operative growth.

Assessment of the Certainty of the Evidence
We used the GRADE approach (Grading of Recommendations Assessment, Development, and Evaluation) to rank the quality of evidence (www.gradeworkinggroup.org,accessed on 4 October 2023) for the included studies.Two authors (EM and RF) independently assessed the certainty of the evidence for each of the outcomes, and MM resolved discrepancies.In case of risk bias in the study design, imprecision of estimates, inconsistency across studies, indirectness of the evidence, and publication bias, the recommended option of decreasing the level of certainty by one or two levels according to the GRADE guidelines was applied [15].The GRADE approach results in an assessment of the certainty of a body of evidence and allocation to one of four grades:

Results
After duplicates were removed, 278 studies were identified following the literature review.After reviewing titles and abstracts, 229 additional records were excluded: 39 review articles, two guidelines, nine studies including only participants with peculiar syndromes (Prader Willi, Down syndrome), 168 studies reporting outcomes different from those of interest, eight studies not available as full papers, three studies with a publication period before 2013, and one study with less than 10 patients.
A total of 49 full-text manuscripts were assessed for eligibility.After a full-text examination, 23 studies were excluded, and a final number of 26 studies were included in this systematic review.The PRISMA flow diagram (Figure 1) summarizes the study screening process.A summary of the studies included in this systematic review and the grading of evidence is reported in Table 1 .
this systematic review.The PRISMA flow diagram (Figure 1) summarizes the study screening process.A summary of the studies included in this systematic review and the grading of evidence is reported in Table 1 .The intervention group had a higher baseline waist circumference z-score (p = 0.063).The intervention group had a significantly greater increase in WT +3.3 ± 2.1 vs. +2.2± 1.5 kg, p = 0.014.The BMI z-score remained similar at baseline and follow-up for both groups.The waist circumference z-score increased significantly in the control group (p < 0.05).
Studies with a sample size of <50 patients or with a short follow-up (<6 months) have been considered at risk of bias in the study design and/or at imprecision for estimating the analyzed outcome.Therefore, they were assigned a low level of evidence.In the same way, other studies, retrospective in design or prospective without a control group, were evaluated in the entire design, considering sample size, follow-up range (a wide range makes comparisons less reliable), and outcome results.In some cases, the level of certainty was considered "low".Below, we report the results on growth outcomes based on the 14 studies, graded as a moderate-high quality level (Table 2).None of these studies evaluated body composition or hormonal changes.Therefore, we report available evidence of low-quality grades.

Changes in Growth Pattern after Tonsillectomy
Among growth pattern outcomes, all the studies included in this review evaluated BMI z-score pre-and post-tonsillectomy (T) or TA; five studies analyzed the impact on height and/or height z-score, one on waist circumference [40], and in two studies, changes in the percentage of subjects with overweight/obesity were reported (Table 2).Below, we write a summary of the evidence.
BMI z-score did not change significantly compared to the control group in the two prospective studies in which participants were normal weight at baseline [17,37] in 1 RCT [40], and the 2 RCTs referred to the re-analysis of data of the CHAT cohort [35,36] (Table 2).In one large population-based cohort study, shortly after TA, the BMI z-score increased significantly in the operation group (0.41 ± 0.02) vs. the control group (0.18 ± 0.01; p < 0.001) [23].In four studies that included subjects Uw, Nw, Ow, and Ob [22,24,27,32], BMI z-score increased across all cohorts, but the determinant for this result was the data of the Uw group, which was not the most prevalent (3.5%, 9.2%, 14.9%, 1.3%, respectively).Three studies, mainly involving patients Ow or Ob at baseline, reported an increased BMI z-score after TA [16,25,33].A recent RCT in children with OSA concluded that there is an increase in BMI z-score between 0 and 12 months but not from 12-24 months, and the mean BMI z-score included catch-up of Uw subjects, while only eight of the TA group had undesirable weight gain [39].
Given the differences in ethnic origin, growth chart used for percentiles, and follow-up timing recorded in the various papers, a meta-analysis using BMI z-score as the outcome was not conducted.At the same time, we performed an analysis to search for determinants of BMI z-score.

Determinants of BMI z-Score after TA
Baseline BMI z-score: A correlation between baseline BMI z-score and BMI z-score after TA was found in seven studies.Reduced BMI z-score at baseline (Uw population) was correlated to an increase in BMI z-score after TA in five studies [22,24,27,32,35], but not in the other two that vice versa reported an increased BMI z-score after TA in Ow-Ob subjects at baseline [25,33].
Age: Younger age (age < 4-6 years) was correlated to an increase in BMI z-score after TA in three studies [16,24,32] but not in the other two [22,33]; weight gain duration after TA: among the studies that reported an increase in BMI z-score, a rise between 0 and 12 months but not from 12-24 months was reported in one RCT study [39].The other two studies reported increased BMI z-scores up to 24 months [22,33].Most weight gain has been reported in the first six months after TA and only plateau one year after surgery in another study [32].
Surgery indication: Obstruction or OSA as an indication for TA, compared to recurrent upper airway infections, was correlated to an increase in BMI z-score after TA in one study [24] but not in the other three that evaluated this risk factor [16,22,33].
Gender: No significant difference was noted in BMI z-score between boys and girls in either the aggregate or subgroup analysis in the studies that analyzed this parameter [16,22,32,33,35].
Ethnicity: African-American children were noted to have significantly more weight gain increase than white or Hispanic and Asian children in one study [32], but these data were not confirmed in the other two studies [16,33].
Caloric intake: Increased caloric intake after TA was reported only in one study, classified as a low level of evidence [13], and none investigated physical activity levels.HT z-score increased significantly more than 12 months after TA, compared to the control group in three studies [23, 24,33] but not in the other two after 18-24 months [32,39] (Table 2).

Waist Circumference
Waist circumference was evaluated in only one study and did not change significantly after TA during the follow-up period of about ten months [40].

Percentage of Overweight/Obesity
In most studies, an increase in the prevalence of Ow or Ob was not reported, as the increase in BMI z-score was minimal or absent during the follow-up period.In one study, children with Ow/Ob decreased in BMI z-score [22], whereas, in another, the Ow group had a 27% (6/22) shift to Ob [27] (Table 2).

Changes in Body Composition after Tonsillectomy
Only one study was selected for this systematic review and classified as low evidence.It showed a significant increase in body muscle mass after TA [18].No study reported data on fat mass.

Hormonal Changes Influencing Post-Operative Growth after Tonsillectomy
Only three studies, classified as low level of evidence, reported data on hormonal changes, and below, we report a summary of the results [20,29,30] (Table 3).IGF-1: The three studies measured IGF-1 levels pre-op and post-operation.IGF-1 levels increased significantly post-op vs. pre-op (p < 0.001) in two studies [29,30], whereas in the other study, only the obstructive group showed an increase in IGF-1 (p < 0.0001), but not the infection group (p = 0.0883) [20].
Leptin: Pre-op leptin levels were significantly higher in patients than in controls (p < 0.001) and increased significantly post-op (p = 0.036), as reported in only one study [29].

Discussion
This systematic review selected 26 studies, considering 14 with moderately highquality evidence.This allowed us to conclude that children who underwent TA could present an increase in BMI z-score.However, this weight gain is considered catch-up growth in subjects that are Uw at baseline, while normal weight or overweight TA did not lead to overweight per se.These data must be contextualized in an obesogenic environment where overweight subjects in the short term could gain more weight percentiles if they do not change their lifestyle [16,25,33,41,42], and studies with randomized controlled design consented to the separation of treatment effect from natural history [35,36,39,40].
The most recent systematic review on TA as a risk factor for childhood obesity was published in 2016 by Van M et al. and included six studies in children with OSA [9]; among these studies, four showed a significant weight gain after TA, and the others did not.Most of the included studies were observational and involved relatively few patients.One RCT, conducted by Katz et al., obtained in that review the greatest level of evidence [34], but in 2021 it was revised in the statistical analysis leading to the opposite conclusion [35,36].At that time, Van M et al. concluded an association between TA and weight gain in patients with OSA in the short term [9].This conclusion was the same reported in 2011 in a previous systematic review by Jeyakumar et al. who included nine studies and suggested a correlation between TA and weight gain in normal weight and overweight children, with a mean BMI increase of 7% [8].They suggested large RCTs not available then to answer some of their questions [8].
In our systematic review, we attributed the highest level of evidence to the RCTs, even if GRADE classification is outcome-specific and RCTs do not need to have a high quality of evidence for the particular outcome we decided to analyze; however, the RCTs that we selected were well designed to answer our questions, and according to us, they add quality to the results of this systematic review.Notably, the children in the no-intervention group in these trials are not healthy ones but rather those who have OSA, a condition that appears to restrict their growth [9].One RCT, conducted by Kevat et al., reported increased BMI z-score in the 12 months after TA, but not after, reporting a catch-up growth in Uw subjects, while undesirable weight gain was experienced only by eight participants (17% of the TA group) [39].Another RCT showed that the BMI z-score remained similar at baseline and follow-up for both groups, and waist circumference did not increase in the TA group [40].
In two studies, Jensen et al. and Kirkham et al. presented a reanalysis of a multicenter randomized controlled trial (CHAT) comparing watchful waiting to early TA in 464 prepubertal OSA children between 2008 and 2011, with a 7-month follow-up [35,36].This trial did not demonstrate differences in unwanted weight gain seven months after TA versus watchful waiting.Children in both arms experienced undesirable weight gain over seven months but did so similarly in the two arms (45% adenotonsillectomy vs. 41% watchful waiting; p = 0.400) [35,36].A first analysis performed by Katz et al. in 2014 showed more significant weight increases seven months after TA in all weight categories, with children who failed to thrive that TA normalized weight but with increased risk for obesity in overweight children [34].Years later, Jensen et al. and Kirkham et al. repeated the statistical analysis, and after excluding the thinnest children based on BMI < 20th percentile, regression models with a sensitivity analysis that excluded children who were underweight at baseline no longer reported a significant effect of TA on BMI z-score [35,36].
In our systematic review, there was heterogeneity in the outcomes of the ten prospective or retrospective studies that we rated as having moderate quality of evidence, primarily due to variations in the population covered and the study design.Firstly, there were differences in the study population enrolled in the studies regarding BMI percentile before surgery.The percentage of patients categorized by BMI percentile as Uw was not high (1.3-14.9%) in the studies that reported these data [22,24,27,32]; however, after TA, they showed a catch-up growth that influenced the mean BMI z-score of the whole cohort (as demonstrated by correlation analysis).We emphasize the fact that if 4.5% of Uw in the CHAT RCT were not excluded in the regression model, they were able to show or not a significant effect of TA on BMI z-score [35].Secondly, considering the study design, the sample size was fundamental in these studies to determine whether the BMI z-score comparison was statistically significant.In the study of Ha EK et al. [23], the BMI z-score after TA of 0.41 ± 0.02 compared to 0.18 ± 0.01 in the control group was significant in this large Korean population study on 3172 children with TA and 31,663 controls, but this could have had no clinical impact at the individual level [23].Vice versa, other studies included in this review presented sample sizes in the low hundreds in total and were then divided among the various weight categories; therefore, the study could be underpowered for some conclusions, and results should be viewed with caution [25].The length of follow-up and the national growth charts used for BMI percentiles in various nations are two additional differences that may impact direct comparisons of study results.
The second outcome of this systematic review was identifying subgroups of children and adolescents at risk of weight gain.Not all the studies with moderately high levels of evidence were concordant, but different studies revealted that those who were underweight at baseline [22,24,27,32,35,36] or younger than 4-6 years [16,24,32] gained more weight after TA.Weight gain in children with failure-to-thrive after TA has been previously reported [43].Age at risk is important to consider in clinical practice, as greater preoperative counseling, closer follow-up, and extra measures might be required for these young patients receiving TA to manage their obesity.Otolaryngologists and primary care providers should counsel families that increased weight gain could be possible in the first 6-12 months after TA, particularly in those under 6 years old [32].In different cohort studies not included in this systematic review, a higher BMI pre-tonsillectomy was reported and associated with a history of repeated use of antibiotics, which probably affected weight by altering gut microbiota [44][45][46].
As a third outcome, we aimed to elucidate mechanisms beyond increased caloric intake that underlie weight gain.This was not possible due to the absence of caloric intake and physical activity reports before and after TA in most of the studies included in this review.Only one study that we had, classified as low-quality level of evidence, explored this outcome [13].Dietary and physical records could help determine whether weight gain is caused by surgery or by parents who support specific recovery practices (overfeeding and restricting physical activity), and they could determine whether there is more to the story (i.e., hormonal, and metabolic changes), which could be intriguing for future research directions.
Previous studies, not included in this review because of their publication date being before 2013 and/or their failure to report weight variation within 24 months, showed conflicting results pre-and post-TA.In these studies, authors used diet records.Even though they are not a precise measurement of food intake, they are a practical and frequently used method of assessing caloric intake, type of food eaten, and the source of calories, and they have been shown to provide reliable assessments of food intake in children [12].
As previously reported, the surgical removal of tonsils might also improve the gag reflex and, therefore, eating patterns [47,48].Rarely, adenoids will regrow after adenoidectomy, above all in young children, which could cause a plateau in weight gain [49,50], and the prevalence has been reported to be 8% [49].
A few studies compared children whose tonsils were removed and those who retained their tonsils.TA was not associated with higher levels of caloric intake and weight gain according to diet records reported in one of these studies [6].In other studies on prepubertal children, significant weight gain associated with decreased mean caloric expenditure was reported [51].In the latter study, the composition of the diet was similar in both groups, with no significant differences in the percentage of calories derived from protein, carbohydrates, or fat [51].
A few studies compared caloric intake pre-and post-TA in prepubertal children and found an increased caloric intake after TA.Gkouskou et al. found that, pre-operatively, the total calories received by children with tonsillar hypertrophy did not differ significantly from the control group.Six months after TA, they increased the calories they consumed daily, taking in even more significant amounts of food than the control group, resulting in significant weight gain without leading to overweight (catch-up growth) [52].Selimoglu et al. found that, pre-operatively, healthy children had higher mean energy intake, weight, and height z-scores than children with adenotonsillar hypertrophy, and mean energy intake, weight, and height significantly increased after TA, again without leading to overweight [53].Nachalon et al., in the study that we included in this review, reported increased caloric intake, weight, and height z-scores (without leading to overweight) after TA, as assessed by the Short Food Frequency Questionnaire (SFFQ), before and after TA, without considering a control group [13].
Regarding the diet composition reported in these studies, Gkouskou et al. found that the percentages of calories received from consuming sugar products, soft drinks, and edible fats were greater in children who had TA than in the control group.In contrast, the total calories received from other foods (meat, vegetables, or legumes) did not differ between children with TA and the control group [52].The authors interpreted these differences as being due to parents' recovery practices: they detected a possible inadequate caloric intake, so they provided the rest of the calories in soft drinks and candies as a more accessible and more pleasant form of food to consume [52].Selimoglu et al. analyzed protein intake, and it was not statistically different in healthy children compared to children with adenotonsillar hypertrophy before and after TA [53].Nachalon et al. reported a rise of 2% in protein and a decline of 4.49% in fat consumption.Although these changes were found to be significant, their actual effect on growth patterns was not apparent [13].
All these data suggest an increased caloric intake in children after TA as a possible mechanism underlying catch-up weight gain.Still, data on diet composition in children after TA are conflicting, and dietary interviews in the first few months after TA could elucidate this aspect.Furthermore, adenoids and tonsils represent vital constituents of the immune system, carrying profound implications also for immunology, physiology, and the microbiome [48].The intricate interplay between adenoids, tonsils, and various aspects of human health has been reported [48], and their removal or reduction via surgery can profoundly affect growth through different pathways [48].
The main strengths of this review are the inclusion of all subjects who underwent tonsillectomy (not only for OSA), making our data transferable to the general population; a significant period of observation (10 years); the application of the PICOS model for studies' selection and the GRADE system for assessment of evidence; and the presence of RCTs on the topic.Limitations are the heterogeneity of the analyzed studies reported above regarding the population analyzed and study design.

Conclusions
Our findings suggest that TA may not be an independent risk factor for unfavorable weight gain in children undergoing TA.RCT studies support this evidence by distinguishing the effect of TA from the natural history in an obesogenic context.However, children who were Uw pre-operatively or younger than six years are reported as gaining more weight after TA than expected.Therefore, otolaryngologists and primary care providers should counsel families that increased weight gain could occur in the first 6-12 months after TA, particularly in those under six.
This systematic review highlights the need for further studies, including caloric intake reports, to better understand the mechanisms underlying weight gain in these categories of patients and to ascertain a possible role of microbiome change with tonsillectomy.
(up to 24 months after) growth (weight, height, BMI (body mass index), and relative percentiles for age, body composition, metabolic and hormonal changes and/or comparison with a control group Study design

Table 1 .
Literature analysis after PICOS selection: summary of the studies and grading of evidence by the GRADE system.

Table 2 .
Summary of studies with moderate-high level quality of evidence.

Table 3 .
Summary of studies with low quality of evidence.