Summary of doctoral dissertation: Research on the value of sentinel lymph node biopsy in thyroid cance

MINISTRY OF EDUCATION MINISTRY OF HEALTH HANOI MEDICAL UNIVERSITY 

NGUYEN XUAN HAU RESEARCH THE VALUE OF SENTINEL LYMPH

NODE BIOPSY IN THYROID CARCINOMA

Specialization :Oncology : 62720149 Code SUMMARY OF DOCTORAL DISSERTATION

HANOI - 2019

THIS STUDY WAS COMPLETED IN: HANOI MEDICAL UNIVERSITY

Spervisor: Assoc. Prof. MD. Le Van Quang

Reviewer 1:Assoc. Prof. MD. Le Ngoc Ha

Reviewer 2:Assoc. Prof. MD. Ngo Thanh Tung

Reviewer 3:Assoc. Prof. MD. Le Trung Tho The thesis will be defended before the Examining Board at university level in Hanoi Medical University At …….. o’clock……… Date ……. Month ……. Year… This thesis could be found at - National Library - Centre Medical Information Library

LIST OF PUBLICATION RELATED TO THE DISSERTATION

1. Nguyen Xuan Hau, Tran Thi Hau, Le Thi Hang (2017). Clinical, subclinical characteristics and histopathology of early thyroid cancer at Hanoi Medical University Hospital. Vietnam Medical Journal, Number 1, volume 456.

2. Nguyen Xuan Hau, Le Van Quang, Duong Chi Thanh (2017). The result of visualization of sentinel lymph node biopsy with Methylene Blue in thyroid carcinoma. Vietnam Medical Journal, Number 2, volume 456.

3. Nguyen Xuan Hau and Le Van Quang (2018). The role of sentinel lymph node biopsy in thyroid carcinoma and the factors affecting the value of the method. Vietnam Medical Journal, Number 2, volume 472.

4. Le Van Quang, Nguyen Van Hieu, Nguyen Xuan Hau, Nguyen Van Hung (2018). Role of sentinel lymph node biopsy in papillary thyroid carcinoma in Vietnam. International Journal of Hematology and Oncology, number 4, Volume 28.

INTRODUCTION

the

increase

1. Reason to choose the thesis. Thyroid cancer is the most common disease in endocrine cancer, accounting for 3.6% of all cancers in general. Surgery is the most important treatment in thyroid cancer. rate of lymphadenectomy Thyroidectomy plus complications after surgery, but the disease recurrent mainly in cervical lymph node and lymphadenectomy is the first priority choice. Surgery to remove central lymph node in recurrent thyroid cancer increase the risk of complications such as paralysis of the recurrent laryngeal nerve and hypoparathyroidism. To limit systemic cervical lymph node dissection, finding a method to detect occult lymph node metastasis is necessary to determine the treatment strategy. Sentinel lymph node is defined as the first lymph node in the lymphatic drainage area to be drained from the primary tumor, which reflects the condition of the remaining lymph nodes with or without lymph node. This result contributes to the detection of metastatic lymph nodes that are smaller than 2-3 mm in diameter - difficult to detect by other techniques such as high resolution ultrasound. In the world, there have been many studies on the sentinel lymph node of thyroid cancer, however, there has not been any research in this field in Vietnam. Therefore, we carried out the thesis: "Research on the value of sentinel lymph node biopsy in thyroid cancer" with the following two objectives:

1. Comment on the results of the method of visualization and sentinel lymph node biopsy in thyroid carcinoma by Methylene blue in Hanoi Medical University Hospital from 11/2014 to 9/2018 and analyze the some factors affecting the results of method.

2. Evaluate the value of sentinel lymph node biopsy in the diagnosis of

occult cervical lymph node metastasis. 2. New finding of the thesis:

- This is the first study in Vietnam, using Methylene Blue to detect sentinel lymph node in thyroid cancer, to help determine the exact occult lymph node metastasis, thereby giving a strategy the cervical

the

time

same

reasonable

lymph node dissection with thyroidectomy. - The rate of sentinel lymph node detection with Methylene Blue is very high: 98.2%; Most lymph nodes are in the pre-tracheal group (group 6): 90.4%. - Percentage of non-metastatic lymph nodes on pathology: 55.7%; metastasis on pathology: 44.3%, the rate of occult cervical lymph node metastasis is very high: 51.5%. - Sensitivity, specificity, positive predictive value, negative predictive value, overall accuracy and false negative rate of the method are: 86%, 100%, 100%, 87,1%, 92.8% and 14% respectively.

3. The structure of the thesis:

The thesis conclude 122 pages, with 4 main chapters: Introduction 2 pages, Chapter 1 (Overview) 40 pages, Chapter 2 (Subjects and Research Methods) 18 pages, Chapter 3 (Research results) 29 pages , Chapter 4 (Discussion) 30 pages, Conclusion and Recommendations 3 pages. The thesis has 33 tables, 7 pictures and 5 charts, 155 references (26 Vietnamese references, 129 English references).

Chapter 1: OVERVIEW

1.1. Introduction Sentinel lymph node is defined as the first lymph node in the lymph nodes receiving lymph nodes from primary tumors. Since the past 19 years, approved sentinel lymph node biopsy is a technique to determine the metastatic presence of melanoma cancer and early breast cancer. Recently, sentinel lymph node biopsy has been proposed for use in other cancers, including thyroid cancer. The role of this method is being clarified by the authors. The results of lymph node biopsy help the surgeon make a decision right away in terms of regional lymphadenopathy and appropriate management or systemic behavior. 1.2. Sentinel lymph node biopsy in thyroid carcinoma. 1.2.1. Technology

Blue dye is an injection used most often with methylene blue, isosulphan blue. This method involves injecting a blue indicator into sites with the same lymphatic drainage system with the tumor then a certain period of time will be performed to detect lymph nodes by direct observation of the pathways. Blue lymphatic drainage leads to lymph node. All blue dyes disappear during the process of making the specimen and do not affect histological analysis. About radioisotopes, different forms of 99m-Technetium. The lymph nodes in these cases are detected by skin markers for lymph nodes or it can be detected by radioactive signals using a gamma probe during surgery. After exposition of the thyroid gland, four quadrants around the tumor were injected with a total of 1ml of methylene blue with tuberculosis syringe. At this time, stop the operation for one minute to allow diffused dye. The lymphatic pathways catch a blue stain (the stain of the blue dye, then continue into the central cavity and the blue stain nodes are taken and sent to frozen section. 1.2.2. Factors affecting the results of the method. Although the studies have shown that the rate of visualization of the sentinel lymph node with Methylene blue is very high, there is still a small proportion of not showing the lymph nodes when perform the method.

To explain the absence of this sentinel lymph node, many studies suggest that the lymph node system from the tumor to the lymph node is disrupt during the process of thyroid disclosure, the lymphatic pathway is blocked by invasive tumors or The main lymphatic vessels that were not detected after injecting the color indicator may be that it is behind the esophagus or after the thyroid site.

When studying other factors affecting the results of sentinel lymph node such as age, sex, tumor position in lobe, tumor size, tumor stage, multifocality, all studies showed that there was no the relationship between the above factors with the result of the sentinel lymph node.

Chapter 2: OBJECTIVES AND METHODOLOGY

2.1. Research subjects The patient was diagnosed with thyroid cancer and surgery at the Oncology-Palliative Care Department, Hanoi Medical University Hospital from November 2014 to September 2018. 2.1.1. Sample size

n = Z2

1-α/2

Sample size is determined by the formula: p(1 – p) (pε)2

Where n: Minimum sample size in the study Z1-α/2: coefficient of confidence with a probability of 95% ( ε=

0.05) → Z = 1.96.

Select ε: deviation of p, limit of 10% (ε = 0.1) p: percentage of lymph nodes without metastasis (p = 0.7)

according to Cabrera's study (2014)

From the above formula, we selected 170 patients

2.1.2. Criteria for selecting research subjects - Patients diagnosed with thyroid cancer based on clinical, ultrasound and cytology. Cases of unknown cancer in cytology and / or ultrasound will be frozen section in surgery. - Diagnosis of clinical stage T1,2,3,4; N0; M0 classified by AJCC 2010. - Injection of methylene blue around the tumor. - Surgical removal of selective lymph nodes group VI and group II, III, VI on both sides. - Results of postoperative pathology are differentiated thyroid carcinoma 2.1.3. Exclusion criteria - Metastatic lymphoma of the lymph nodes clearly clinical, distant metastasis.

- There is a history of allergy to Methylene blue. - Thyrois cancer is not the epithelial cell type. - Patient was undergoing thyroid surgery at the other hospital. - Patients disagree to participate in the study. 2.2. Research Methods 2.2.1. Research design Descriptive research study 2.2.2. Research Methods 2.2.2.1. Exploiting clinical and subclinical information. 2.2.2.2.Technical process of visualization and sentinel lymph node biopsy. In this study, we used the method of sentinel lymph node biopsy with Methylene Blue. - The patient is in the supine position, his hands pressed close to his body, neck up to the maximum, pillows placed under the shoulders to increase the ability of the neck. - Endotracheal anesthesia, antiseptic neck area down through depressions, above the chin. - Incision skin across the lower neckline, about two fingers away from the noch of breastbone, cutting the platisma muscle, dissection the two skin flaps up and down. Open along the neck fascia before the trachea, clearly showing the thyroid lobe containing the tumor. - Find and protection the parathyroid glands and recurrent laryngeal nerves before injecting Methylene Blue, avoiding difficulties in finding later. - Inject 1 ml of Methylene Blue into 4 positions around the tumor: 3 hours, 6 hours, 9 hours and 12 hours. Be careful not to let Methylen around the operation space to avoid confusion when making a judgment. If gauze and gloves are slotted with a color indicator, they must be replaced to avoid penetrating the surrounding organization. - Gently massage the thyroid lobe that is injected within 1 minute with the fingertip. Dissection around the thyroid lobe to detect blue lymphatic drainage. This process should be carried out gently, meticulously, hemostatic to facilitate maximum observation. Follow all the detected

lymphatic pathways to the stain lymph nodes (lymph nodes). Take all the stain lymph nodes and record the sentinel lymph node in separate groups. - Duration is 10 - 15 minutes from the injection of color indicator. Past that time, if the lymphatic or lymphatic channels are not detected, the procedure is classified as not recognizing the lymph node. - Total thyroidectomy and neck lymphadenectomy selected at the center neck group and lateral neck group or lobe thyroidectomy plus isthmusectomy and combined with selective neck lymph nodes remove are conducted according to modern surgical standards for thyroid cancer. 2.2.2.3. Evaluation parameters

Final pahtology Total Sentinel lymph node biopsy result Metastasis No metastasis

Metastasis a b a+b Frozen section No metastasis c d c+d

b+d Total

established when P <0.05 or Fisher’s exact test.

a+c a+b+ c+d - Detection rate = Number of detected cases with sentinel lymph node / total number of cases performed - Sensitivity = Number of sentinel lymph node (+) / Number of patients with lymph node metastasis on final pathology= a / (a + c). - Specificity = Number of sentinel lymph node (-) / Number of patients with no lymph node metastasis on final pathology = d / (b + d). - Total accuracy = (frozen section positive positive lymph node + true negative frozen secton) / Number of lymph nodes detected = (a + d) / (a + b + c + d). - The rate of false negative = Number of sentinel lymph node (-) / Number of patients with lymph node metastasis on final pathology = c / (a + c). 2.3. Data analysis Data were collected and statistically processed with SPSS 16.0 software. according to research criteria. The statistical method used includes: - Descriptive statistics: average, standard deviation. - Comparison of the ratio: test χ2, statistical significance level was

- Univariate evaluation method, Logistic multivariate regression

model.

Chapter 3: RESEARCH RESULT 3.1. Results of the method of visualization the sentinel lymph node biopsy with methylene blue in thyroid cancer and factors affecting the results of the method 3.1.1. Patient characteristics Age: Average age: 40.3 ± 10.64 (20 - 68) years old. Gender: The disease is mainly seen in women, the rate of women

/ men: 7.5 / 1

Clinical tumor characteristics: The percentage of

tumor palpation on clinical examination is highest, accounting for over 60%. Most patients have tumors in a thyroid lobe, accounting for nearly 90%. The majority of tumors have a firm density, accounting for 81.9%, and mobility is easily accounted for over 95%.

Results of thyroid ultrasound: - The position of tumors in one of the most common lobes accounts for nearly 90%, evenly distributed in both right and left lobes. About 4% of the tumors are located in the isthmus and 7% are in both lobes. In the lobe, tumors in the most common middle third account for over 55%. On ultrasound, hypoechoic is the most common accounts for over 80%, microcalcifiation accounts for 54.1%, angiogenesis accounts for 34.7%, unknown boundaries account for only over 25%.

- Most patients are classified TIRADS 4-5 on ultrasound, accounting for more than 98%. Among them, the most common TIRADS 4b accounts for more than 40%. However, 1.2% of thyroid cancer on ultrasound is TIRADS 3 image.

- Most patients on ultrasound have only 1 tumor accounting for over 77%, 22.4% of patients have more than 1 tumor on ultrasound, of which 7% have 3 tumors.

- Most patients have tumor size <2 cm, accounting for 88.8%,

tumor size> 4 cm accounts for only 1.2%.

Cytology: 100% of patients were performed fine needle aspiration before surgery. The results of cytology diagnosed with

cancer or suspected of cancer account for 93%. only about 7% of thyroid cacner but benign cytology results.

Frozen section: More than 85% of patients have the frozen

section tumor intraopetatively, 100% of results are cancer.

Procedure of

thyroid surgery: The patient had

total thyroidectomy, accounting for 89.4%, lobe thyroidectomy plus isthmusectomy accounted for 10.6%.

Pathology result of thyroid tumors: Papillary adenocarcinoma accounts for the majority of cases over 99%. Follicular carcinoma accounted for only 0.7%.

Thyroid inflammation enclosed: Among patients with thyroid

cancer, more than 11% of patients had thyroiditis with cancer. T stage: Clinically, the majority of patients diagnosed with T1 stage account for about 90%. However, on pathlogy, the proportion of T1 patients accounted for only over 52%, the rate of T3 stage tumors increased to 37.6%.

3.1.2. The results of the method of visualization the sentinel lymph nodes biopsy by methylene blue and the factors affecting the results of the method The rate of detection of sentinel lymph nodes with methylene blue:

The rate of detection of sentinel lymph nodes with methylene blue is

very high, accounting for 98.2%.

Some factors affecting the result of sentinel lymph node.

Table 3.1. Some factors affecting the result of the method

Factor % OR 95% CI p SLN (+) (n=167) SLN(-) (n=3)

Age 109 2 98,2 1,06 0,094 – 12 1 <45 ≥45 58 1 98,3

Tumor positon

on the lobe (n=163)*

1/3 uper 23 95,8 1 -- -- 0,145 1/3 midle 91 100 0

1/3 lower 46 95,8 2

Tumor size

<1cm 84 98,8 1 0,5 0,04 – 5,6 1 ≥1cm 83 97,6 2

Gender Male 19 95,0 1 3,9 0,33-45 0,32 Female 148 98,2 2

T stage T1,2 98 97 3 -- -- 0,27 T3,4 69 100 0

number unifocal 131 99,2 1 7,3 0,64 – 82,6 0,13 Multifocal 36 94,7 2

* 7 tumors on the isthmus cases; SLN: sentinel lymph node

Comment: The percentage of sentinel lymph node result of the position of the middle third tumor has a 100% compared to the group of patients with tumor position in uper third of 95.8% and lower third 95.8 %, this difference is not statistically significant p = 0.145. The patients at T1 and T2 stage had a lower incidence of sentinel lymph node than patients in the T3 and T4 stage (97% versus 100%), this difference was not statistically significant. p= 0.27. The number of sentinel lymph nodes detected: The average number of sentinel lymph nodes 3.4 ± 1.99; The maximum number of nodes is 12, at least 1. Most patients detected more than 1 lymph node accounted for more than 85%. The position of the lymph node: Sentinel lymph node detected in the pretracheal group (group 6) is the most common, accounting for

more than 90% (150/167). There were 6.0% (11/167) patients with sentinel lymph nodes found in the both pretraheal and lateral group, while the percentage of sentinel lymph nodes detected in the ipsilateral group: 3.6% (6/167).

The position of the sentinel lymph node in the pretracheal group: In the pretracheal group, the highest percentage of sentinel lymph node was detected in the recurrent chain group above 85%. The rate of detection of sentinel lymph node in the group pretracheal is only about 5%, prelaryngeal (Denphien lymph nodes) only accounts for about 1%. There were 3.7% of sentinel lymph node detected in both groups of the recurrent chain group and the prelaryneal.

The number of cervical lymph nodes: The average number of

cervical lymph nodes were 21,89 ± 9.09 nodes, the highest is 64 nodes, at least 4 nodes.

3.2. The value of sentinel lymph node biopsy in diagnosis of cervical lymph node metastasis

3.2.1. Status of sentinel lymph node metastasis on frozen section

Table 3.2. sentinel lymph node metastasis status on frozen section

Sentinel lymph node Patients (n) Rate %

metastasis 66 39,5

No metastasis 101 60,5

Total 167 100

Comments: In 167 patients with senetinel lymph node detected by

methylene blue, all these lymph nodes were sent to prozen section.

The rate of SLN metastasis accounts for 39.5%, 60.5% of the SLN

negative on frozen section.

3.2.2. Status of sentinel lymph node metastasis on pathology

Table 3.3. sentinel lymph node metastasis status on pathology

Sentinel lymph node Patients (n) Rate %

metastasis 44,3 74

No metastasis 55,7 93

Total 100 167

Comment: On pathology, the number of patients with SLN

metastasis accounting for more than 44%, the percentage of negative SLN on routine pathology was 55.7%.

3.2.3. The number of metastatic SLN on final pathology

The average number of SLN metastasis: 1.97 ± 1.05 nodes. In 74 cases of SLN on pathology: the average number of SLN metastatic were 1.97 ± 1.05, at most 5 metastases, at least 1 lymph node.

3.2.4. Relation between the SLN on frozen section and the

pathology

All lymph nodes after frozen section are sent for final pathology.

Of the 74 patients with SLN metastasis on pathology, 8 patients with frozen section results were negative. The rate of false negative were 10.8% (8/74).

3.2.5. Results of cervical lymph node pathology: The rate of cervical

lymph node metastasis in the study group was 51.5%

3.2.6. Relationship between the SLN results on pathology and cervical lymph node result.

Table 3.4. Relationship between the SLN results on pathology and

cervical lymph node result.

SLN on Cervical lymh node Total pathology metastasis No metastasis

metastasis 74 0 74

No metastasis 12 81 93

Total 86 81 167

Sensitivity = 74 / (74 + 12) = 86%

Specificity = 81 / (81 + 0) = 100%

Positive predictive value = 74 / (74 + 0) = 100%

Negative predictive value = 81 / (81 + 12) = 87.1%

Total accuracy = (74 + 81) / (74 + 0 + 12 + 81) = 92.8%

False negative rate = 12 / (12 + 74) = 14%

3.2.7. The relationship between SLN metastasis on frozen section and the some factors

Table 3.5. The relationship between SLN metastasis on frozen

section and the some factors

SLN metastasis OR 95% CI p Factor Yes No

<45 52 57 Age 2,87 1,4-5,8 0,03 ≥45 14 44

Male 11 8 Gender 2,3 0,88-6,1 0,08 Female 55 93

T1-2 31 67 T stage 0,45 0,24-0,85 0,01 T3-4 35 34

<1cm 27 57 Tumor size 0,5 0,3-1,0 0,05 ≥1cm 39 44

Multifocal 9 27 0,4 0,2-0,99 0,04 Tumor number Unifocal 57 74

Anormal 2 10 TSH 0,13 0,06-1,34 0,13 concentration Normal 64 91

Papillary 64 93 Pathology Follicular variant 1 8 -- -- 0,1 subtype. Follicular 1 0

Yes 5 15 Thyroiditis 0,15 0,16-1,4 0,15 No 61 86

Comment: There was the relationship between SLN metastasis on

frozen section with factors: age, stage T, and tumor number (p: 0.03, 0.01 and 0.04 respectively).

When multivariate analysis of SLN metastasis on frozen section

and some factors: patients <45 years of age had a risk of SLN metastasis on frozen section 2.9 times higher than patients ≥ 45 years old (95% CI : 1.33 - 6.44, p = 0.008). Patients in T1.2 stage reduced the risk of SLN metastasis on frozen section 1 / 0.39 = 2.56 times (95% CI: 0.2 - 0.8, p = 0.012). Patients with multifocal cancer reduced the risk of SLN metastasis on frozen section 1 / 0.35 = 2.86 times (95% CI: 0.14 to 0.88, p = 0.03).

3.2.8. The relationship between SLN metastasis on final pathology

and the some factors

Table 3.6. The relationship between SLN metastasis on final pathology and the some factors

SLN metastasis Factor OR 95% CI p Yes No

<45 2,35 1,2-4,59 0,012 53 56 Age ≥45 40 18

Male 1,86 0,71-4,88 0,205 8 11 Gender Female 85 63

T1-2 0,43 0,23-0,8 0,008 63 35 T stage T3-4 30 39

<1cm 0,6 0,32-1,11 0,104 52 32 Tumor size ≥1cm 41 42

Multifocal 0,4 0,18-0,9 0,024 26 10 Tumor number

Unifocal 67 64

Anormal 0,23 0,05-1,09 0,045 10 2

TSH concentration Normal 83 72

Pathology Papillary -- -- 0,213 86 71

subtype.

FVPC 2 7

Follicular 1 0

Yes 6 14 0,5 0,18-1,37 0,17 Thyroiditis No 68 79

Comment: When analyze SLN metastasis on final pathology and the

some factors showed: age, stage T, number of tumors and TSH concentration are statistically significant with p: 0.012, 0.008, 0.024 and 0.045 respectively.

When analyzing multivariate SLN metastasis on final pathology

and the some factors showed: patients <45 years of age have a risk of SLN metastasis on final pathology 2.3 times higher than patients ≥ 45 years old (95% CI : 1.1 - 4.8; p = 0.026). Patients in T1.2 stage reduced the risk of SLN metastasis on final pathology 1 / 0.37 = 2.7 times (95% CI: 0.18 - 0.75, p = 0.026). Patients with multifocal cancer reduced the risk of SLN metastasis on final pathology 1 / 0.32 = 3.1 times (95% CI: 0.13 - 0.77; p = 0.01). Patients with abnormal TSH levels reduced the risk SLN metastasis on final pathology 1 / 0.16 nodes = 6.25 times (95% CI: 0.03 - 0.84; p = 0.03).

3.2.9. The relationship between CLN metastasis and the some

factors

Table 3.7. The relationship between CLN metastasis and the factors

CLN

Yếu tố p

metastasis OR 95% CI

Yes No

63 46 <45 2,08 1,09-3,99 0,026 Age 23 35 ≥45

12 7 Male 1,71 0,64-4,6 0,28 Gender 74 74 Female

42 56 0,43 0,23-0,8 0,008 T stage T1-2

T3-4 44 25

<1cm 37 47 Tumor size 0,55 0,3-1,01 0,053 ≥1cm 49 34

Multifocal 15 21 Tumor number 0,6 0,29-1,27 0,183 Unifocal 71 60

Anormal 3 9 TSH 0,29 0,08-1,11 0,057 concentration Normal 83 72

Papillary 83 74

FVPC 2 7 -- -- 0,092 Pathology subtype. Follicular 1 0

Yes 7 13 Thyroiditis 0,46 1,18-1,23 0,116 No 79 68

Comment: There is the relationship between CLN metastasis and

some factors: age, T stage with p: 0.026 and 0.008 respectively.

When multivariate analysis, There is the relationship between the T stage with the CLN metastasis (p <0.01, OR = 0.39 of T1.2 group, 95% CI: 0.2 - 0.8)

3.2.10. The relationship between the false negative rate of the method and the factors.

Table 3.8. The relationship between the false negative rate of the method and the factors.

Factor N=167 CLN+ CLN- FNR % OR 95%CI p

SLN+ 18 0 ≥45 21,7 (n=58) SLN- 5 35 Age 0,63-7,87 0,29 2,22 SLN+ 56 0 <45 11,1 (n=109) SLN- 7 46

SLN+ 11 0 8,33 Male (n=19) SLN- 1 7 Gender 0,52 0,06-4,45 1 SLN+ 63 0 Female 14,86 (n=148) SLN- 11 74

SLN+ 35 0 T1, T2 16,67 (n=98) SLN- 7 56 T stage 1,56 0,45-5,36 0,55 SLN+ 39 0 11,36 T3, T4 (n=69) SLN- 5 25

SLN+ 32 0 <1cm 13,51 (n=84) SLN- 5 47 Tumor 0,94 0,27-3,23 1 size SLN+ 42 0 14,29 ≥1cm (n=83) SLN- 7 34

SLN+ 64 0 1 ổ 9,86 (n=131) SLN- 7 60 0,22 0,06-0,83 0,03 Tumor number SLN+ 10 0 33,33 Đa ổ (n=36) SLN- 5 21

Comment: When studying the relationship between false negative

status of SLN with the factors, there is the relationship between the false negative rate of the method and the multifocal with p = 0.03.

When multivariate analysis of the relationship between the

factors with false negative status, no factors that are independently related to false negative status are statistically significant.

SLN+ 11 0 1/3 trên 8,33 (n=23) SLN- 1 11 Tumor SLN+ 40 0 18,37 -- -- 0,6 position on the 1/3 giữa (n=91) SLN- 9 42 lobe SLN+ 20 0 1/3 dưới 9,09 (n=46) SLN- 2 24

Chapter 4: DISCUSSION 4.1. Results of the method of visualization and SLN biopsy with methylene blue in thyroid cancer and analysis of factors affecting the results of the method. 4.1.1. Patient characteristics

Average age: 40.3 years. Under 45 years old: 68.8%. According to Le Van Quang (2015), the average age: 43.27 years old. Gender. Female / male ratio: 7.5 / 1. This result is similar to other studies of the prevalence in women. Clinical characteristics: The percentage of tumors on clinical examination in our study was 61.8%. Our results are similar to other studies Tumor characteristics on thyroid ultrasound: tumor position is located on two thyroid lobes nearly equal. The most common hypoechogenic tumor accounts for 81.7%. Our results are consistent with the studies of Le Cong Dinh (2013). TIRADS 4b on the most common ultrasound accounted for 42.3%. Our results are similar to those of Tran Van Thong (2014). Cytology: The results of cancer diagnosis accounted for 61.2%, 7.1% of cancer, but the results of benign cytology, these cases all suspected cancer on ultrasound before surgery. Pathology result of thyroid tumors: papillary carcinoma were the most cases 99.3%. This result is consistent with previous studies that papillary carcinoma is the most common. Thyroid inflammation: Proportion of patients with thyroidcancer with thyroiditis is 11.8%. This result is lower than that of Yasuo Fukuo and Takashi Yamakawa has 28.5% of cases of thyroiditis accompanied by cancer. T stage: The proportion T1 stage accounts for 52.4%, and the rate of stage 3 tumor accounts for 37.6%. The percentage of tumors in stage T2 and T4 was low at 7.1% and 2.9% respectively. This result is also consistent with previous studies that suggest that the tumor in the T1 stage accounts for the highest percentage. 4.1.2. Results of the method of visualization and SLN biopsy with methylene blue in thyroid cancer and analysis of factors affecting the results of the method. Percentage of SLN biopsy detected by Methylene blue

SLN detection rate(%)

The rate of SLN biopsy detection according to each study is clearly listed in the following table 4.1: Table 4.1. Percentage of SLN biopsy detected by studies STT

Author

n

Kelemen và cs. 1998

100

1

Dixon và cs. 2000

73,3

2

Arch-Ferrer và cs. 2001

90,1

3

Pelizzo và cs. 2001

75,9

4

Fukui và cs. 2001

95,5

5

Tsugawa và cs. 2001

71,1

6

Takami và cs. 2003

92,6

7

Chow và cs. 2004

66,6

8

Peparini và cs. 2006

9

Rubello và cs. 2006

153

69,9

10

Dzodic và cs. 2006

92,5

11

Abdalla và cs. 2006

12

Roh et al. 2008

13

Wang và cs. 2008

14

Bae và cs. 2009

81,9

15

Takeyama và cs. 2009

91,7

16

Anand và cs. 2009

93,3

17

Cunningham và cs. 2010 211

18

Nguyễn Xuân Hậu. 2018

170

98,2

19

In our study, the rate of detection of SLN with methylene blue is very high, accounting for 98.2%. Our research results are consistent with the meta-analysis studies, all studies show that the rate of detection of SLN with Methylene blue is very high. However, when analyzing each study, we found that the results of SLN in our study were much higher than those of the authors: Dixon, Tsugawa, Chow,

there

is no difference between

especially in the study of the author Fake Peparini rate of SLN detection is 0%. Some factors affecting the method In our study, no statistically significant differences were found between tumor position in the lobe; T1,2 stage; age, gender, tumor size, number of tumor. This result is similar to the study by Balasubramanian (2011), the incidence of SLN and related factors. 4.2. The value of SLN in diagnosis of cervical lymph node metastasis 4.2.1. SLN status on frozen section The rate of SLN metastasis accounts for 39.5%, 60.5% of the SLN is negative on frozen section. Ja Seong Bae (2009), the percentage of metastatic SLN on frozen section were 55.5%, non- metastatic is 44.5%. Jong-Lyel Roh (2008), the rate of SLN metastasis was 30.6%, 69.4% negative. 4.2.2. SLN status on final pathology The incidence of SLN on final pathology in our study was 44.3%, the negative rate accounted for 55.7%, the average number of SLN on final pathology was 1.97 ± 1.05 (1-5). This result is lower than Takami's research but higher than the results of the Rubello study. 4.2.4. Results of cervical lymph node pathology: The rate of CLN metastasis in the study was 51.5%, equivalent to the group of non- metastatic CLN accounted for 48.5%. Our results are also consistent with previous studies. 4.2.5. SLN result on routine pathology and Evaluation parameters As a result of our research, the overall accuracy of SLN on routine pathology is 92.8%. The rate of false negative when comparing SLN test results on routine pathology with CLN result is 14%. The false negative rate in our study is higher than the Balasubramanian meta analysis (14% vs. 7.7%).

Our research has a lower sensitivity than the results of a meta- analysis (86% and 91.6%). 4.2.6. The relationship between SLN metastasis on frozen section and factors In our study, when analyzing the relationship between factors with SLN metastasis on frozen section: there is the relationship between the age factors <45 (2.87, 1.4 - 5.8; p = 0.03); stage T (0.45; 0.24 - 0.85; p = 0.01), multifocality (0.4; 0.2 - 0.99; p = 0.04) with SLN metastasis on frozen section. When multivariate analysis evaluated these factors and the SLN on frozen section, we found that: age, T stage and multifocality were the factors affect the SLN metastasis on frozen section. 4.2.7. The relationship between SLN status on pathology with factors The relationship between SLN metastasis status on pathology with: age, T stage, number of tumors and TSH concentration having statistically significant with p respectively: 0.012, 0.008, 0.024 and 0.045; There was the relationship between SLN metastasis on pathology with factors with age: p <0.05; 95% CI: 1.1 - 4.8), T stage (p <0.05; 95% CI: 0.18 - 0.75), multifocality (p <0.05; 95% CI: 0.13 - 0.77) and TSH concentration (p < 0.05; 95% CI: 0.03 - 0.84) when multivariate analysis. Our multivariate analysis results similar the Cunningham study, suggest that age <45 and extrathyroidal invasive increase the risk of SLN metastasis on pathology . 4.2.8. The relationship between CLN metastasis and factors Analysis of the relationship between CLN metastasis and the factors, we found: Age <45 (2.08; 1.09 - 3.99, p = 0.026); T1.2 (0.39; 0.2 - 0.8, p = 0.008) having the relationship with the CLN metastasis. The results of our research are consistent with the research of the authors Jong-Lyel Roh (2008) and Cabrera (2014) all claim that the extrathyroidal extention is a factor related to the CLN metastasis. When multivariate analysis of the relationship between the factors

we found: only the extrathyroidal extention tumor is an independent factor related to the situation of CLN metastasis (0.39; 0.2 - 0.8. Our results are consistent with Seok Koo Bon (2009) u size> 1, OR = 3,048; Kulter (2012) concludes that there are no independent factors related to CLN metastasis. 4.2.9. Impact of factors on the false negative rate of the method Analysis of factors affecting the false negative rate: The relationship between unifocal of tumor with false negative status of the method is statistically significant (0.22; 0.06 - 0.83; p = 0 , 03). When multivariate analysis the relationships between factors: age, sex, T stage, tumor size, number of tumors and tumor position in lobe with false negative status, no related factors independent of false negative status is statistically significant.

CONCLUSION

1. Results of the method of visualization and SLN biopsy with methylene blue in thyroid cancer and analysis of factors affecting the results of the method.

- Average age: 40.3 years; femal/male ratio:7.5/1; TIRADS 4b: 42.3%; unifocal: 77.6%; tumor size <2 cm: 88.8%; carcinoma on cytology : 61.2%. Papillary cancer: 99.3%; T1 stage: 52.4%.

- The rate of detection SLN with Methylene Blue: 98.2%; The average number of SLN were 3.4 ± 1.99; SLN were in the pre- tracheal group (group 6): 90.4%; the rate of SLN detected in the ipsi recurrent chain 85.7%; Average number of cervical lymph node: 21.89.

- There is no relationship between the factors: age, sex, tumor position in the lobe, tumor size, T stage and the number of tumors to the results of SLN with Methylene blue, p> 0.05 .

2. The value of SLN in diagnosis of cervical lymph node

metastasis

- Percentage of SLN metastasis on frozen section: 39.5%; percentage of non-metastatic SLN on pathology: 55.7%; metastasis on pathology: 44.3%; false negative rate: 10.8%; cervical lymph node metastasis rate: 51.5%.

- Sensitivity, specificity, positive predictive value, negative predictive value, overall accuracy and false negative rate were: 86%, 100%, 100%, 87.1%, respectively , 92.8% and 14%.

- The relationship between SLN metastasis on frozen section with factors: age , T stage, the number of tumors with p: 0.03, 0.01 and 0.04 respectively; There was the relationship between SLN metastasis on frozen section with factors: age (p <0.05; 95% CI: 1.33 - 6.44), T stage (p <0, 05; 95% CI: 0.2 - 0.8) and multifocality (p <0.05; 95% CI: 0.14 - 0.88) on multivariate analysis.

- The relationship between SLN metastasis on pathology with age, T stage, number of tumors and TSH concentration having statistically significant with p: 0.012, 0.008, 0.024 and 0.045 respectively; There was the relationship between SLN metastasis on pathology with factors: age: p <0.05; 95% CI: 1.1 - 4.8), T stage (p <0.05; 95% CI: 0.18 - 0.75), multifocality (p <0.05; 95% CI: 0.13 - 0.77) and TSH concentration (p < 0.05; 95% CI: 0.03 - 0.84) on multivariate analysis.

- The relationship between cervical lymph node metastasis with factors: age ,T stage having statistically significant with p: 0.026 and 0.008 respectively; There was the relationship between cervical lymph node metastasis and T stage(p <0.01, OR = 0.39 of T1.2 group, 95% CI: 0.2 - 0.8) on multivariate analysis..

- The relationship between the false negative rate of the method

with the multifocality was statistically significant with p = 0.03.

RECOMMENDATION

dissection,

cervical

lymph

node

Deploying and applying the method of visualization of the sentinel lymph node by Mehtylen gblue at many different hospitals because this method is a simple, feasible and economic. On the other hand, this method gives low false negative rate, high accuracy, accurately reflects the occult cervical lymph node metastasis. Based on the results of sentinel lymph node biopsy, surgeons can decide exactly when the selective avoiding complications of cervical lymphadenectomy in the absence of occult lymph nodes metastasis .