Do Patients Gain Weight After Thyroidectomy for Thyroid Cancer?

Abstract

Background:

Patients who undergo thyroidectomy often complain of weight gain, which they frequently attribute to inadequate thyroid hormone replacement. To assess the weight changes associated with thyroid hormone replacement or suppressive therapy after thyroidectomy, we measured the weights of patients before and after thyroidectomy and compared them to the weights of euthyroid patients with thyroid nodules who were being followed for many years.

Methods:

The weights and heights of 67 women and 35 men who underwent total thyroidectomy for thyroid cancer were recorded before and for a mean of 8.3 years after thyroidectomy. All patients received either suppressive or replacement doses of levothyroxine. As a comparison group, 70 women and 22 men with goiter or thyroid nodules and were euthyroid had serial measurements of height and weight. They were followed for a mean of 7.6 years. The body mass index (BMI) and age-adjusted BMI percentiles were calculated. The weight, BMI, and BMI percentile changes were compared both unadjusted and adjusted for age, gender, thyrotropin (TSH) level, and duration between measurements.

Results:

At baseline, patients with thyroid nodules were older (mean 50.4 years) than those with thyroid cancer (mean 45.8 years). There were no significant differences in baseline weight, BMI, or BMI percentile. The baseline TSH levels were lower for patients with thyroid cancer (mean 0.8 mIU/L) than for those with nodules (mean 1.8 mIU/L) (p=0.002). There were no significant differences between the changes in weight, BMI, or BMI percentile from the start to the completion of the study whether unadjusted or after adjustment for age, gender, TSH, and duration of follow-up.

Conclusions:

Despite the perception of many patients that their thyroidectomy and thyroid hormone replacement or suppressive therapy is responsible for their subsequent weight gain, there were no significant differences in weight gain over time in comparison to a control group of euthyroid patients with thyroid nodules or goiter.

Introduction

T here exists a common perception among patients who undergo thyroidectomy that they experience a weight gain that is out of proportion to age-related weight changes. Several studies have shown that individuals with hyperthyroidism who are rendered euthyroid with radioactive iodine, thioamides, or surgery experience a 4–12-kg increase in body weight over 2 years in comparison to the weight at the time of therapy (1 –6). However, the vast majority of hyperthyroid patients will have lost a substantial amount of weight by the time that they present to their physicians and a diagnosis of hyperthyroidism is made. Indeed, Hoogwerf and Nuttall studied 84 men and 3 women with hyperthyroidism, and noted that at an average of 48 months after treatment their patients were only 101.1% of their pre-hyperthyroid weight, and at 96 months after therapy, they were 102.5% of the baseline weight (7). Similarly, Kromas and colleagues weighed patients before and 1 year after thyroidectomy for benign goiter and found an average of 0.5 kg weight increase 1 year after the surgery (8). These results were quite similar to a group of patients with thyroid cancer examined by Tigas et al., who noted a 0.6-kg weight gain 1 year after thyroidectomy, radioactive iodine, and thyroid hormone suppressive therapy (3).

To re-examine this issue, we measured the weight changes and body mass index (BMI) over a period of several years in patient undergoing thyroidectomy and thyroid hormone replacement or suppression for thyroid cancer. Patients with thyroid nodules or goiter either not receiving thyroid hormone or receiving replacement, but not suppressive, doses of thyroid hormone served as the control population.

Materials and Methods

Patients

The charts from 102 patients (67 women and 35 men) with thyroid cancer and 92 patients (70 women and 22 men) with goiter or thyroid nodules were reviewed and information regarding age, gender, ethnicity, weight, height, and BMI at various time intervals was recorded. The distribution of stage in the thyroid cancer patients was 73% stage I, 9% stage II, 11% stage III, 5% stage IVA, 1% stage IVB, and 1% stage IVC according to American Joint Committee on Cancer 7th Edition system (9). Seventy-eight percent of these patients had received radioactive iodine for remnant ablation/treatment using an endogenous thyroid hormone withdrawal protocol, and 23% received more than one dose. All of the patients with thyroid cancer were taking levothyroxine and 11 were also taking triiodothyronine. The average dose of levothyroxine for those taking only this form of thyroid hormone was 2.15±0.52 (standard deviation, SD) μg/kg body weight. Forty-two percent of the patients with benign disease were receiving levothyroxine (average dose 1.52±0.5 [SD] μg/kg body weight), and an additional 2% were taking desiccated thyroid.

The National Health and Nutrition Examination Survey national anthropometric reference data for all ages of the U.S. population in 2003–2006 were used to convert weight and BMI into BMI percentiles (10). Comparison between the thyroid cancer and thyroid control groups at baseline was carried out by Chi-square test or Fisher's exact test for categorical variables and two-sample t-test or Wilcoxon rank sum test for continuous variable when appropriate.

General linear models were used to compare the change of weight, height, BMI, and BMI percentile both unadjusted and adjusted for age, gender, thyrotropin (TSH) level, and/or duration between measurements. The linear mixed effects models for longitudinal measures, which allow baseline heterogeneity and take into account the correlation between the two outcome measurements per subject, were carried out to evaluate the group difference between thyroid cancers and controls in weight, height, BMI, and BMI percentile. The mixed model included a random intercept, group variable, time variable, and an interaction between time and group, and with subject as a random effect factor. Age, gender, TSH level, and/or duration between measurements were included in the model as covariates. A compound symmetry structure was assumed. Goodness-of-fit of the mixed model was assessed by investigating the distribution of the residuals.

Results

Table 1 provides the baseline data between groups, indicating that overall there were no significant differences in duration of follow-up, BMI, BMI percentile, or weight. Seventy-six percent of the patients with thyroid nodules were women, as were 66% of the patients with thyroid cancer (p=0.112). The data for the thyroid cancer patients were obtained over an average of 8.3 years (median 5.9 years), whereas that for the nonthyroid cancer patients were obtained over a mean of 7.6 years (median 3.8; no significant difference). As a group, patients with thyroid cancer were significantly younger on average than the patients with benign disease (45.8 vs. 50.4 years; p=0.02). Although both males and females with thyroid carcinoma were younger than those without cancer, the difference did not reach statistical significance (p=0.125 and p=0.069, respectively). Eighty-nine percent of the thyroid cancer patients and 97% of the thyroid nodule patients were Caucasian (p=0.049).

Table 1.

Comparison Between Groups at Baseline

	Nodules/goiter (n=92)		Thyroid cancer (n=102)
	Mean±SE	Median	Mean±SE	Median	p-Value
Age (years)	50.4±1.5	51.8	45.8±1.3	45.1	0.020
Male	52.9±3.5	53.9	46.5±11.4	47.0	0.125
Female	49.6±1.6	51.7	45.4±1.6	43.7	0.069
Duration of follow-up (years)	7.6±0.8	3.8	8.3±0.6	5.9	0.509
Male	5.6±1.6	3.1	7.1±0.6	5.7	0.392
Female	8.2±1.0	4.3	8.9±0.9	6.3	0.598
Weight (kg)	68.1±1.8	71.3	70.8±1.8	70.7	0.297
Male	87.0±4.4	91.1	87.4±3.2	92.0	0.938
Female	62.2±1.3	66.3	62.1±1.3	67.5	0.954
BMI (kg/m²)	24.3±0.5	23.8	24.8±0.5	24.0	0.457
Male	26.9±1.0	25.9	27.6±0.9	27.4	0.546
Female	23.5±0.5	22.8	23.3±0.5	22.1	0.782
BMI percentile	37.5±27.8	37.5	39.1±27.9	25	0.684
Male	51.8±5.6	50	52.9±4.8	50	0.875
Female	33.1±3.0	25	31.9±3.1	25	0.795
TSH^a (mIU/mL)	1.7±0.1	1.5	1.1±0.1	0.9	0.001
Male	1.8±0.2	1.6	1.3±0.3	1.2	0.137
Female	1.7±0.1	1.5	1.0±0.2	0.7	0.002

Excluding values outside of the normal range.

BMI, body mass index; TSH, thyrotropin; SE, standard error.

As shown in Table 2, there were no significant differences between the increase in weight, BMI, and BMI percentile in patients with and without thyroid cancer, with the mean increases being 1.51 kg in weight and 0.84 kg/m² increase in BMI in the group with thyroid cancer and 1.69 kg and 0.88 kg/m² increase in BMI in the control group. There were no differences between the groups when subdivided by gender. When the data were reanalyzed after removing patients with TSH levels outside of the normal range of 0.39–4.5 mU/L, the results were unchanged (1.49 kg for patients with thyroid nodules vs. 0.97 kg for patients with thyroid cancer, p=0.657). Similarly, there were no differences when the data were adjusted for gender, age, duration of follow-up, or TSH level (Table 3). There were no significant differences in weight, BMI, or BMI percentile change between patients in the nodules/goiter group who were taking thyroid hormones (n=39) in comparison to those who were not taking thyroid preparations (n=53) (p=0.187, 0.160, and 0.235, respectively) Also, there were no significant differences in changes in weight, BMI, or BMI percentile in patients with thyroid cancer whose TSH was suppressed (n=71) versus those whose TSH levels were within the normal range (n=31) (p=0.899, 0.702, and 0.540, respectively).

Table 2.

Differences Between Measurements at Start and End of Study

	Nodules/goiter (n=92)		Thyroid Cancer (n=102)
All patients	Mean±SE	Median	Mean±SE	Median	p-Value
Weight (kg)	1.69±0.64	0.7	1.51±0.78	1.3	0.857
Male	1.59±0.95	0.7	0.97±1.51	2.7	0.728
Female	1.73±0.79	0.7	1.80±0.89	0.7	0.954
BMI (kg/m²)	0.88±0.23	0.5	0.84±0.30	0.7	0.908
Male	0.53±0.26	0.5	0.33±0.56	0.8	0.747
Female	0.99±0.29	0.5	1.10±0.35	0.7	0.807
BMI percentile	3.49±1.72	0.0	2.30±1.93	0.0	0.647
Male	3.00±2.24	0.0	−2.00±3.47	0.0	0.232
Female	3.64±2.16	0.0	4.55±2.28	0.0	0.773

	Nodules/goiter (n=85)		Thyroid cancer (n=27)
For patients with TSH within the normal range	Mean±SE	Median	Mean±SE	Median	p-Value
Weight (kg)	1.49±0.66	0.6	0.97±0.96	2.2	0.657
Male	1.59±0.95	0.7	−0.90±1.94	−0.2	0.272
Female	1.45±0.83	0.6	1.90±1.03	3.1	0.736
BMI (kg/m²)	0.81±0.24	0.5	0.58±0.31	0.9	0.555
Male	0.53±0.26	0.5	−0.15±0.5	−0.1	0.252
Female	0.91±0.31	0.5	0.94±0.36	1.3	0.941
BMI percentile	3.13±1.79	0.0	2.81±2.55	0.0	0.920
Male	3.00±2.24	0.0	−0.44±4.19	0.0	0.482
Female	3.17±2.29	0.0	4.44±3.21	0.0	0.749

Table 3.

Difference Between Measurements at the Start and End of Study Adjusted for Age, Gender, Thyrotropin, and Duration of Follow-Up

All patients	Nodules/goiter (n=92)	Thyroid cancer (n=102)	p-Value ^a	p-Value ^b
Weight (kg)	2.00±0.73	1.24±0.69	0.758	0.462
BMI (kg/m²)	0.97±0.27	0.75±0.26	0.782	0.564
BMI percentile	3.49±1.91	2.30±1.81	0.590	0.659

For patients with TSH within the normal range	Nodular goiter (n=85)	Thyroid cancer (n=27)	p-Value ^a	p-Value ^a
Weight (kg)	1.45±0.59	1.07±1.07	0.562	0.759
BMI (kg/m²)	0.79±0.21	0.65±0.38	0.448	0.754
BMI percentile	2.81±1.64	3.82±2.99	0.810	0.773

p-Value is from linear mixed effects model.

p-Value is from general linear model.

Discussion

Contrary to the perception of many of our patients who undergo thyroidectomy for thyroid cancer, with or without radioactive iodine ablation of residual thyroid tissue, followed by replacement or suppressive doses of thyroxine, there is on average no excessive weight gain over the expected age-related increase in weight. These results are similar to those found by Tigas et al. in their study on 25 patients with thyroid cancer treated in the same manner or the 8 postmenopausal women described by Kromas and co-workers who underwent a thyroidectomy and thyroxine replacement therapy for benign goiters (3,8).

The genesis of the concern about weight gain may actually be twofold. First, some may question if increased weight may be a result of relative deficiency of triiodothyronine (T3) in patients with total thyroidectomy and lone thyroxine replacement. However, it appears that this parameter, as has been shown by others, appears unchanged with levothyroxine treatment alone, at least in short-term studies (11 –14). In addition, Jonklaas and co-workers showed that normal T3 levels are achieved with levothyroxine treatment alone that results in a normal or suppressed TSH level after near-total or total thyroidectomy (15). Second, the marked weight increase and occasional overshoot that is found in patients treated for hyperthyroidism cause such patients to implicate inadequate thyroxine replacement as a cause for their weight issues. In fact, hypothyroid patients often do experience a weight loss in the first 6 months of treatment, possibly due to mobilization of myxedematous fluid, followed by a return to baseline by 1 year (7). In contrast, there is uniformly a weight increase in patients who lost weight due to hyperthyroidism during and after the treatment of the thyrotoxicosis (1 –6). Most of the studies that examined the weight and body composition changes in these patients compared patients' weights just before or at the time of treatment of the hyperthyroidism to weights at various time intervals after treatment. However, they did not compare the final weights to the patients' weights before their developing hyperthyroidism. Hoogwerf and Nuttall did compare the prehyperthyroid weights to the post-treatment weights in a group composed primarily of men and found that 8 years after treatment, the weight was only 102.5% of the prehyperthyroid weight, an increase that falls well within the expected range for the population (7).

Of course, it could be argued that since many of the patients treated for thyroid carcinoma are receiving suppressive doses of levothyroxine, this subclinical hyperthyroidism may account for the lack of excessive weight gain. However, the control group was euthyroid and showed a similar small increase in weight and BMI, a reflection of the normal changes in weight with aging. Also, when we reanalyzed the data after removing patients with a suppressed TSH, the results for weight, BMI, or BMI percentile between the two groups did not change.

In summary, our findings indicate that patients who undergo thyroidectomy for thyroid cancer and who are receiving adequate doses of thyroxine to keep them euthyroid or suppressed do not experience an abnormal weight gain over a median of >5 years. However, since the bulk of our patients were Caucasian, the results may not be able to be extrapolated to other ethnic groups.

Footnotes

Acknowledgment

This study was supported by the Boris Catz, M.D., Summer Student Fellowship Program. We gratefully acknowledge Ronnie Wong, M.P.H., for her helpful assistance in managing our thyroid database.

Disclosure Statement

The authors declare that no competing financial interests exist.

References

Jansson

, Berg

, Lindsedt

, Michanek

, Nystrom

. 1993. Overweight—a common problem among women treated for hyperthyroidism. Postgrad Med J, 69:107–111.

de la Rosa

, Hennessey

. Tucci JR 1997 A longitudinal study of changes in body mass index, total body composition after radioiodine treatment for thyrotoxicosis. Thyroid, 7:401–405.

Tigas

, Idiculla

, Beckett

, Toft

. 2000. Is excessive weight gain after ablative treatment of hyperthyroidism due to inadequate thyroid hormone therapy? Thyroid, 10:1107–1111.

Dale

, Daykin

, Holder

, Sheppard

, Franklyn

. 2001. Weight gain following treatment of hyperthyroidism. Clin Endocrinol, 55:233–239.

Brunova

, Bruna

, Joubert

. Koning M 2003 Weight gain in patients after therapy for hyperthyroidism. S Afr Med J, 93:529–531.

Greenlund

LJS

, Nair

, Brennan

. 2008. Changes in body composition in women following treatment of overt and subclinical hyperthyroidism. Endo Pract, 14:973–978.

Hoogwerf

, Nuttall

. 1984. Long-term weight regulation in treated hyperthyroid and hypothyroid subjects. Am J Med, 76:963–970.

Kromas

, Diamond

, O'Sullivan

, Smerdely

. 1998. Body mass and body composition after total thyroidectomy for benign goiters. Thyroid, 8:773–776.

Edge

, Byrd

, Carducci

, Compton

. AJCC Cancer Staging Manual, Seventh. Springer: New York, 2009.

10.

McDowell

, Fryar

, Ogden

, Flegal

. 2008. Anthropometric reference data for children, adults: United States. 2003–2006. Nat Health Stat Rep, 10:1–48.

11.

Clyde

, Harari

, Getka

, Shakir

KMM

. 2003. Combined levothyroxine plus liothyronine compared with levothyroxine alone in primary hypothyroidism. A randomized controlled trial. JAMA, 290:2952–2958.

12.

Walsh

, Shiels

, Lim

EEM

, Bhagat

, Ward

, Stuckey

BGA

, Dhaliwal

, Chew

, Bhagat

, Cussons

. 2003. Combined thyroxine/liothyronine treatment does not improve well-being, quality of life, or cognitive function compared to thyroxine alone: a randomized controlled trial in patients with primary hypothyroidism. J Clin Endocrinol Metab, 88:4543–4550.

13.

Escobar-Morreale

, Botella-Carretero

, Gomez-Bueno

, Galan

, Barrios

, Sancho

. 2005. Thyroid hormone replacement therapy in primary hypothyroidism: a randomized trail comparing l-thyroxine plus liothyronine with l-thyroxine alone. Ann Intern Med, 142:412–424.

14.

Saravanan

, Simmons

, Greenwood

, Peters

, Dayan

. 2005. Partial substitution of thyroxine (T4) with triiodothyronine in patients on T4 replacement therapy: results of a large community-based randomized controlled trial. J Clin Endocrinol Metab, 90:805–812.

15.

Jonklaas

, Davidson

, Bhagat

, Soldin

. 2008. Triiodothyronine levels in athyreotic individuals during levothyroxine therapy. JAMA, 299:769–777.