Prediction of thyroid hormone supplementation after thyroid lobectomy

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 3 ( 2 0 1 5 ) 2 7 3 e2 7 8

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.JournalofSurgicalResearch.com

Prediction of thyroid hormone supplementation after thyroid lobectomy Doh Young Lee, MD, Jungirl Seok, MD, Woo-Jin Jeong, MD, PhD, and Soon-Hyun Ahn, MD, PhD* Department of Otolaryngology Head & Neck Surgery, Seoul National University Bundang Hospital, Seongnam-Si, Korea

article info

abstract

Article history:

Background: Thyroid function, as assessed by thyroid-stimulating hormone (TSH) levels, was

Received 17 April 2014

evaluated in patients after thyroid lobectomy. These assessments were analyzed against

Received in revised form

perioperative measurements to determine if any of these preoperative values were predictive

3 June 2014

of postoperative hypothyroidism and the need for postoperative levothyroxine treatment.

Accepted 1 July 2014

Methods: In a retrospective study, data from 276 thyroid lobectomy patients were examined.

Available online 5 July 2014

These surgeries occurred over the period from January 2003eDecember 2012. Age, sex, volume of resected thyroid, thyroiditis, preoperative free T4, TSH, and microsomal antibody

Keywords:

levels were analyzed for correlation with postoperative levothyroxine supplementation.

Hemithyroidectomy

Results: The overall percentage of the patients taking postoperative levothyroxine was

Thyroid lobectomy

23.6%. The preoperative TSH level showed strong correlation with TSH levels measured

Papillary thyroid carcinoma

1-mo postoperatively (P < 0.001). Preoperative TSH levels >2.5 mIU/L and positive micro-

Thyroid-stimulating hormone

somal antibody showed significant correlation with postoperative levothyroxine supple-

Hypothyroidism

mentation (P < 0.001; relative risk, 8.933, and 3.438, respectively). By stratifying the patients based on preoperative TSH levels and presence of microsomal antibodies, in the low-risk group with TSH <2.5 mIU/L and negative microsomal antibody, 7% of patients received postoperative levothyroxine replacement but in the high-risk group with TSH >2.5 mIU/L and positive microsomal antibody, 77.8% required levothyroxine replacement (P < 0.001). Conclusions: The most significant preoperative predictors for levothyroxine supplementation are preoperative TSH level and presence of microsomal antibodies. Patients with preoperative TSH <2.5 mIU/L showed a low risk of requiring postoperative levothyroxine supplementation. ª 2015 Elsevier Inc. All rights reserved.

1.

Introduction

Thyroid lobectomy is a valid treatment option for managing patients with benign thyroid disease and may even be considered curative for patients with early differentiated thyroid carcinoma [1]. The needs of thyroid hormone supplementation for life-long period is an important consideration when deciding

the extent of surgery in a patient with thyroid disease, especially when the patient’s disease is on the borderline between the indication of total thyroidectomy and thyroid lobectomy. In literature, thyroid hormone supplementation is required in approximately 10%e50% of patients after thyroid lobectomy [2e6]. Numerous recent studies have focused on the prediction of thyroid hormone supplementation after thyroid lobectomy.

* Corresponding author. Department of Otorhinolaryngology, Seoul National University College of Medicine, Seoul National University Bundang Hospital, 300 Goomi-Dong, Bundang-Gu, Seongnam-Si, Gyeonggi-Do 463 707, South Korea. Tel.: þ82 31 787 7403; fax: þ82 31 787 4057. E-mail address: [email protected] (S.-H. Ahn). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.07.003

274

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 3 ( 2 0 1 5 ) 2 7 3 e2 7 8

However, a universally accepted definition of hypothyroidism has not been established and therefore the specific indications for thyroid hormone supplementation have not been agreed on across all institutions. Therefore, it is important for surgeons to set up their own criteria for levothyroxine supplementation after thyroid lobectomy and thoroughly counsel the patients preoperatively regarding the probability of the need for postoperative levothyroxine supplementation. In this study, we evaluated the correlation between preoperative data and postoperative levothyroxine supplementation to give predictive information that a practitioner could use to provide preoperative counseling to the patients regarding their chances of needing postoperative levothyroxine supplementation after thyroid lobectomy.

2.

Materials and methods

A retrospective chart review was authorized and drawn from patients undergoing unilateral thyroid lobectomy between January 1, 2003 and December 31, 2012. The study protocol was approved by the institutional review board at our institution (IRB No. B-1402-238-103). All surgeries were performed by one author (S.-H.A.). Cases of complete thyroidectomy, incomplete preoperative thyroid function tests, completion thyroidectomy after lobectomy, preoperative levothyroxine supplementation, and follow-up <4 mo were excluded from this study. Parameters investigated included patient sex, age, preoperative thyroid function levels (free T4, thyroid-stimulating hormone [TSH], and microsomal antibodies), postoperative diagnosis, volume of resected thyroid, thyroiditis on pathology report, postoperative free T4, and TSH levels (1 wk, 1 mo, 4 mo, 1 y, and 2 y, postoperatively). Positive test for microsomal antibodies was defined as the presence of microsomal antibodies over 60 U/mL. Patients with TSH levels over 7.0 mIU/L were given levothyroxine supplement during the follow-up period. In addition, patients with TSH levels over 4 mIU/L and complaining of hypothyroidism symptoms (tiredness, lack of energy, discrete cognitive disorders, and mood disturbances) were also treated with levothyroxine replacement. Serial measurements of TSH levels and distribution of TSH levels within the patient group (0e2.5 mIU/L, 2.5e4.0 mIU/L, 4.0e7.0 mIU/L, 7.0e10.0 mIU/L, and >10.0 mIU/L) were evaluated during the follow-up period. The percentage of patients requiring levothyroxine supplementation stratified according to the preoperative TSH level and duration between the operation and levothyroxine supplementation was also evaluated. All the statistical analyses were performed using the SPSS for Windows version 20.0 software package (SPSS Inc, Chicago, IL). Pearson bivariate correlation was examined between postoperative 1-mo TSH level and age, resected thyroid volume, preoperative free T4, and TSH level. The mean of postoperative 1-mo TSH level was compared with independent sample t-test according to the sex, thyroiditis, and microsomal antibody. After grouping, univariate analysis about the risk of thyroid hormone replacement was performed using KaplaneMeier analysis and with selected variables, Cox regression analysis with enter method was performed for the multivariate analysis.

3.

Results

A total of 276 patients qualified for our study. Table 1 shows demographic and perioperative data for these patients. In general, 65 patients out of 276 (23.6%) required levothyroxine replacement therapy over the postoperative follow-up period. Average starting point of levothyroxine treatment was 3.2 mo after surgery (1 wke21.8 mo). Fifty-seven percent of patients started their levothyroxine regimen within 2 mo after surgery. At 6 mo after surgery, 90% of the patients requiring levothyroxine during the follow-up period had started the medication (Fig. 1). Among these, 65 patients were treated with levothyroxine, 17 patients (26.2%) could discontinue levothyroxine therapy after an average of 16.4 mo (range, 3.1e43.5 mo) due to normalization of TSH levels. Table 2 shows the correlation between TSH levels measured 1-mo postoperatively and preoperative findings and pathologic findings. The resected thyroid volume showed significant negative correlation with postoperative TSH levels at 1 mo. The preoperative TSH level showed strong positive correlation with postoperative TSH levels at 1 mo. Sex, thyroiditis, or presence of microsomal antibodies did not show a significant difference in TSH levels at 1 mo. Postoperative TSH levels were stratified into the following groups: 0e2.5 mIU/L, 2.5e4.0 mIU/L, 4.0e7.0 mIU/L, 7.0e10.0 mIU/L, and >10.0 mIU/L. The number of patients in each group at each postoperative time point is shown in Figure 2. Postoperatively, the percentage of patients with normal TSH level (<4 mIU/L) ranged from 52.1%e66.1%. At postoperative 1 wk, 17 out of 221 cases (7.7%) showed >7.0 mIU/L of TSH level. With follow-up, 18 out of 268 cases (6.7%) after 1 mo, 12 out of 269 cases (4.4%) after 4 mo, and 6

Table 1 e Demographic and perioperative data (N [ 276). Age M:F Preoperative TSH level (mIU/L), n (%) 0e2.5 2.5e4.0 4.0e7.0 7.0e10.0 Preoperative microsomal antibody (n ¼ 252) Positive Negative Diagnosis Papillary thyroid carcinoma Follicular adenoma Adenomatous goiter Follicular thyroid carcinoma Others* Volume of resected thyroid (mean  95% CI) Thyroiditis on pathology report Positive Negative Postoperative levothyroxine supplementation Follow-up duration *

49.0 (20e80) 80:196 198 53 22 3

(71.7) (19.2) (8.0) (1.1)

27 (9.8) 225 (81.5) 176 (63.8) 48 (17.4) 42 (15.2) 6 (2.2) 4 (1.5) 28.75  3.10 cm3 50 225 65 28.8 mo

(18.1) (81.5) (23.6) (3.67e105)

1 case of medullary thyroid carcinoma, 1 case of welldifferentiated tumor with uncertain malignant potential, 1 case of degenerated benign cyst, and 1 case of malignant lymphoma.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 3 ( 2 0 1 5 ) 2 7 3 e2 7 8

275

Fig. 1 e Timing of starting the levothyroxine treatment.

out of 247 cases (2.0%) after 1 y showed TSH >7.0 mIU/L for the first time during follow-up period. There was no new patient showing >7.0 mIU/L at 2-y postoperatively. To determine the cutoff value of preoperative TSH level, the TSH level was divided by 0.5 mIU/L between 1.0 and 4.0 mIU/L and KaplaneMeier analysis was performed (Fig. 3) The rate of thyroid hormone supplementation was significantly different between groups of preoperative TSH level of 2.5 mIU/L and the groups of >2.5 mIU/L. So we decided the cutoff value of TSH level as 2.5 mIU/L. Table 3 shows the analysis of the risk for levothyroxine replacement in our study by KaplaneMeier survival analysis. In this analysis, resected tissue volume failed to show a significant effect on levothyroxine medication rate. However,

Table 2 e Correlation with postoperative 1-mo TSH (N [ 257*). Correlation Age Resected thyroid volume Preoperative free T4 Preoperative TSH T-test Sex Thyroiditis in pathology Microsomal Antibodyy Pathology (n ¼ 253)z

Pearson correlation

P value

0.35 0.125

0.575 0.044

0.015

0.814

Fig. 2 e Distribution of groups according to the TSH level on postoperative follow-up.

positive preoperative tests for microsomal antibodies and preoperative TSH levels showed significant effects on levothyroxine medication rate (Fig. 4). In a multivariate analysis with Cox regression, postoperative TSH levels >2.5 mIU/L and positive tests for microsomal antibodies were correlated with postoperative levothyroxine requirement (P < 0.001 for both factors, relative risk, 8.933, and 3.438, respectively, Table 4). TSH levels and the presence of microsomal antibodies were combined to stratify the patients into risk groups as follows: low-risk ¼ TSH 2.5 mIU/L and () microsomal antibodies; intermediate-risk group 1 ¼ TSH 2.5 mIU/L and (þ) microsomal antibodies; intermediate-risk group 2 ¼ TSH >2.5 mIU/L and () microsomal antibodies; and high-risk ¼ TSH >2.5 mIU/L and (þ) microsomal antibody. There was a significant difference in the percentage of patients

<0.001

0.475

Male (75) Female (182) No (210)

Average TSH (mIU/L) 3.74  0.44 4.08  0.43 3.99  0.37

0.835

Yes (47) Negative (213)

3.91  0.75 3.91  0.34

0.985

Positive (24) Papillary carcinoma (162) Follicular neoplasm (50) Adenomatous goiter (41)

3.90  0.99 3.87  0.40

0.585

0.357

4.23  0.77 4.24  0.95

*

11 cases with hormone supplementation are excluded from analysis. y n ¼ 237. z One-way analysis of variance test.

Fig. 3 e Prediction of hormone supplement by preoperative TSH level.

276

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 3 ( 2 0 1 5 ) 2 7 3 e2 7 8

Table 3 e Risk factor of thyroid hormone replacement. KaplaneMeier analysis. Factor

Age (y) (n ¼ 276) Sex (n ¼ 276) Resected volume (cm2) (n ¼ 276) Thyroiditis in pathology (n ¼ 275) Microsomal antibody (n ¼ 252) Preoperative TSH (mIU/L) (n ¼ 276)

Pathology (n ¼ 272)

Number

Hormone replacement (%)

<50 (154) 50 (122) Male (80) Female (196) <30 (191) 30 (85)

21.4 27.6 17.6 26.8 25.4 21.5

No (225) Yes (50)

22.5 30.3

0.216

Negative (225) Positive (27)

20.2 52.6

<0.001

TSH 2.5 (198) 2.5< TSH 4.0 (53)

11.2 48.2

<0.001

4.0< TSH 7.0 (22) 7.0< TSH 10.0 (3) PTC (176) FN (54) AG (42)

72.7 100.0 25.1 22.0 22.3

AG ¼ adenomatous goiter; FN PTC ¼ papillary thyroid carcinoma.

¼

follicular

P value 0.241 0.128 0.513

0.622

neoplasm;

requiring levothyroxine treatment between all groups except the two intermediate-risk groups (Table 5). Figure 5 shows the distribution of postoperative 1-y TSH level based on preoperative TSH level and the presence of microsomal antibodies and stratified as described in the previous paragraph. The percentage of patients with euthyroid status (TSH 4.0 mIU/L) showed large differences across the groups, ranging from 11.1% in the high-risk group to 82.1% in the low-risk group.

4.

Discussion

The normal range of TSH suggested by the National Health and Nutrition Examination Survey III (NHANES III) is

0.4e4.0 mIU/L in a disease-free population. [7] As suggested by the NHANES III data, our institution uses 4.0 mIU/L as an upper limit of the normal range. However, the National Academy of Clinical Biochemists suggested 2.5 mIU/L as the normal range. Therefore, in this study, we stratified the TSH levels using 2.5 mIU/L and 4.0 mIU/L. The clinical criteria for treatment of subclinical hypothyroidism are also the subject of debate. Clinical practice guidelines for hypothyroidism released in 2012 also do not provide definite criteria [8]. However, there is general agreement that TSH levels >10 mIU/L should be treated. Perhaps the basis of this strategy may be the data suggesting significantly increased coronary heart disease risk with these higher TSH levels [9]. In the same article, patients with TSH >7.0 mIU/L also showed increased coronary heart disease mortality. Consequently, we used another stratification line of 7.0 mIU/L in our analysis. Based on these studies, we divided the TSH levels as follows: 2.5, 2.5e4.0, 4.0e7.0, 7.0e10.0, and >10 mIU/L. The definition of post-hemithyroidectomy hypothyroidism varies among studies and institutions [10]. Previous studies have identified risk factors for hypothyroidism, including the presence of microsomal antibodies, thyroiditis, multinodular goiter, preoperative thyrotoxicosis, and a thyroid remnant volume measuring <6 mL [11]. Our data revealed that elevated preoperative TSH levels and presence of microsomal antibody (routinely included in preoperative thyroid function tests) were significant factors in predicting postoperative thyroid hormone replacement. In contrary to the finding about microsomal antibody, we found that evidence of thyroiditis in the pathologic findings was not significantly correlated with the need for postoperative levothyroxine supplementation. This seems to be inconsistent with the previously reported relationship between histologic evidence of thyroiditis and risk of postoperative hypothyroidism [2,12] and also with the finding about microsomal antibody. This may be related to the inconsistency in pathologic reporting of thyroiditis or may be due to selection bias recommending that a patient with a history of thyroiditis or thyroiditis in preoperative ultrasonography undergoes total thyroidectomy rather than hemithyroidectomy. Chu and Lang [10] reported that smaller resected tissue weight was significantly correlated with postoperative hypothyroidism. The reason of this may be that the size of the

Fig. 4 e Survival analysis of hormone replacement. (A) microsomal antibody (P [ 0.000), (B) preoperative TSH level (* versus y, * versus z, * versus x; P < 0.001, y versus z; 0.029, y versus x; 0.041, z versus x; 0.359).

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 3 ( 2 0 1 5 ) 2 7 3 e2 7 8

277

Table 4 e Multivariate analysis with Cox regression. Factor Preoperative microsomal antibody Preoperative TSH level (mIU/L)

Number

Relative risk

P value

Negative versus positive 2.5 versus >2.5

3.438

<0.001

8.933

<0.001

thyroid gland is usually symmetric and resected volume would be similar to the remaining contralateral thyroid lobe. Another study reported that the patients with the smaller remnant thyroid volume have a higher risk of postoperative hypothyroidism [13]. In this study, the remnant thyroid volume was estimated from the preoperative ultrasonography images. In our study, resected thyroid volume showed correlation with TSH levels at 1-mo postoperatively with an inverse relationship but failed to show significant relation with hormone supplementation in survival analysis. The resected volume may not always represent the volume of the remnant healthy thyroid due to the coexisting thyroiditis and various thyroid pathologies such as adenomatous goiter that were included in our study. The mean elapsed time between the operation and levothyroxine supplementation in our study was 3.2 mo, and the most frequent timing of starting levothyroxine supplementation was postoperative month one or two (57%) followed by postoperative month 4 (35.1%). Unfortunately, the timing of starting levothyroxine supplementation was not correlated with any perioperative factors (data not shown). Based on the results of this study, it would be more advisable to counsel patients with a high risk of postoperative levothyroxine supplementation that the likely start for this medication would be 2 or 3 mo after surgery. Also in our series, no patients with TSH level maintained <4.0 mIU/L until 4 mo after surgery required thyroid hormone supplementation during follow up (data not shown). However, when there is fluctuation of TSH level, at least 1-y follow-up may be required for the complete evaluation of the thyroid hormonal status because new patients with TSH level >7.0 mIU/L emerged until postoperative 1 y. The goal of this study was to predict the need for levothyroxine supplementation after thyroid lobectomy, and we

Table 5 e Risk stratification by microsomal antibody and preoperative TSH level (KaplaneMeier analysis). Risk of hormone replacement

Low risk Intermediate risk

High risk

TSH 2.5 mIU/L and microsomal antibody () (n ¼ 163)* Group 1: TSH 2.5 mIU/L and microsomal antibody (þ) (n ¼ 17)y Group 2: TSH >2.5 mIU/L and microsomal antibody () (n ¼ 62)z TSH >2.5 mIU/L and microsomal antibody (þ) (n ¼ 10)x

Hormone replacement (%) 7.4 35.3 53.8 85.0

P value *versus y, * versus z, * versus x; <0.001, y versus z; 0.136, y versus x; 0.002, z versus x; 0.018.

Fig. 5 e Distribution of postoperative 1-y TSH level according the preoperative TSH level 2.5 mIU/L and microsomal antibody.

found that the preoperative TSH levels and the presence of microsomal antibodies were significant predictive factors. By combining preoperative TSH levels and the presence of microsomal antibodies, we could classify patients into low-, intermediate-, and high-risk groups for requiring postoperative levothyroxine treatment. As the criteria for starting levothyroxine are somewhat subjective, we provide Figure 4 to show the distribution of TSH level by these risk groups. The ratio of patients with euthyroid status was also quite different between groups. Because the criteria for decision of levothyroxine supplementation vary among institutions and practitioners, our data may not apply to those with the different criteria. However, Figure 4 shows the distribution of the patients based on preoperative TSH level and microsomal antibody, and this might be used as a reference for estimating the need for levothyroxine supplementation based on the different definitions of hypothyroidism. Although we classified both intermediate-risk group 1 (TSH 2.5 mIU/L and positive for microsomal antibodies) and intermediate-risk group 2 (TSH >2.5 mIU/L and negative for microsomal antibodies) in the same risk group, the percentage of euthyroid status patients was quite different between the subgroups, ranging from 56.3% in the former subgroup to 23.8% in the latter subgroup. If some practitioners predict the need for thyroid hormone replacement therapy in patients whose TSH level is >4.0 mIU/L, the requirement for levothyroxine medication may be quite different from the results of this study. This result is about the prediction of thyroid supplementation during short-term follow-up period, postoperatively. If considering the risk of developing hypothyroidism during lifelong period, the presence of thyroiditis in pathology or positive microsomal antibody preoperatively may have more meaning that is important.

278

5.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 3 ( 2 0 1 5 ) 2 7 3 e2 7 8

Conclusions [2]

The most significant predictors for postoperative levothyroxine supplementation are preoperative TSH levels and the presence of microsomal antibodies. By stratification of posthemithyroidectomy hypothyroidism using TSH levels and the presence of microsomal antibodies, the low-risk group is less likely to need levothyroxine after surgery than the intermediate- or high-risk groups. Although criteria for giving levothyroxine supplementation are different among practitioners and institutions, our data may be valuable when counseling the patient preoperatively about the need for postoperative levothyroxine supplementation.

Acknowledgment Authors’ contributions: D.Y.L., J.S., and S.-H.A. did the data analysis. D.Y.L. and J.S. did the acquisition, manuscript. D.Y.L., W.-J.J., and S.-H.A. did the final editing of the article. S.H.A. did the study design. There were no financial disclosures.

Disclosure

[3]

[4]

[5]

[6]

[7]

[8]

[9]

The authors reported no proprietary or commercial interest in any product mentioned or concept discussed in this article. The authors declared no conflict of interest.

[10]

references

[12]

[11]

[13] [1] American Thyroid Association Guidelines Taskforce on Thyroid NDifferentiated Thyroid CCooper DS, Doherty GM, Haugen BR, et al. Revised American Thyroid Association

management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009;19:1167. Seiberling KA, Dutra JC, Bajaramovic S. Hypothyroidism following hemithyroidectomy for benign nontoxic thyroid disease. Ear Nose Throat J 2007;86:295. Su SY, Grodski S, Serpell JW. Hypothyroidism following hemithyroidectomy: a retrospective review. Ann Surg 2009; 250:991. Stoll SJ, Pitt SC, Liu J, Schaefer S, Sippel RS, Chen H. Thyroid hormone replacement after thyroid lobectomy. Surgery 2009; 146:554. Miller FR, Paulson D, Prihoda TJ, Otto RA. Risk factors for the development of hypothyroidism after hemithyroidectomy. Arch OtolaryngologyeHead Neck Surg 2006;132:36. Verloop H, Louwerens M, Schoones JW, Kievit J, Smit JW, Dekkers OM. Risk of hypothyroidism following hemithyroidectomy: systematic review and meta-analysis of prognostic studies. J Clin Endocrinol Metab 2012;97:2243. Hollowell JG, Staehling NW, Flanders WD, et al. T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 2002;87:489. Garber JR, Cobin RH, Gharib H, et al. Clinical practice guidelines for hypothyroidism in adults: cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Endocr Pract 2012;18:988. Rodondi N, den Elzen WP, Bauer DC, et al. Subclinical hypothyroidism and the risk of coronary heart disease and mortality. JAMA 2010;304:1365. Chu KK, Lang BH. Clinicopathologic predictors for early and late biochemical hypothyroidism after hemithyroidectomy. Am J Surg 2012;203:461. Said M, Chiu V, Haigh PI. Hypothyroidism after hemithyroidectomy. World J Surg 2013;37:2839. Koh YW, Lee SW, Choi EC, et al. Prediction of hypothyroidism after hemithyroidectomy: a biochemical and pathological analysis. Eur Arch oto-rhino-laryngology 2008;265:453. Moon HG, Jung EJ, Park ST, et al. Thyrotropin level and thyroid volume for prediction of hypothyroidism following hemithyroidectomy in an Asian patient cohort. World J Surg 2008;32:2503.