DOI: https://doi.org/10.21203/rs.3.rs-73562/v1
Background: Thyroid cancer is the most common malignant disease of endocrine system and its incidence is rapidly increasing globally (about three to four times higher among females and accounts the sixth most common malignancy diagnosed in women). Among the four major types of thyroid cancer, papillary thyroid cancer (PTC) accounts 85-90% from all thyroid cases, followed by follicular thyroid cancer (FTC) which accounts for 5-10% of cases.
Materials and Methods: A retrospective data of 260 patients who underwent thyroid surgery from 2015 to 2019 were included in the study and the status of the disease (clinical manifestations, laboratory/pathology findings (thyroid function tests (TFTs) and biopsy) were reviewed from patient cards. The pattern of thyroid cancer was discriminated based on histo-pathological biopsy result. Descriptive and analytical statistics were applied to express the finding and reported by tables, figures and narration. Cross tabulation and logistic regression was applied to determine the association between thyroid CA and predictors. A p-value of <0.05 was declared as statistically significant.
Results: From the total sample of 260 patients underwent thyroid surgery, majority of them were females 219(84.2%), belong to age interval of 31-40 years (34.6%), dwellers of Oromia region (91.2%) and Jimma zone (83.8%). Colloid goiter was the most prevalent (74.6%) pattern of thyroid disease followed by follicular CA (4.2%), papillary CA (1.9%) and medullary CA (0.77%) while biopsy results of 48(18.46%) patients were not known. In general, about the 194(91.5%) of the thyroid lesion was identified as benign type and malignancy accounts for 18(8.5%) from the total conducted biopsy results of 212 patients. About six variables (duration of the disease, marital status, family history of the disease, nodularity, surface and border of the thyroid mass) were the candidate variables in binary logistic regression (p-value <0.25) and finally, three variables (duration of the disease (>10 years), surface (rough) and border (irregular) of the thyroid mass) were identified as the predictors of thyroid malignancy with AOR 0.05(0.004-0.60, P-v=0.016; 1.9(1.17-5.8), P-v=0.012 and 2.5(1.13-16.16), P-v<0.001 respectively.
Conclusion and recommendation: The burden of malignancy was higher and alarming among thyroid diseases and warrants early screening and management.
Thyroid cancer (TCA) is a relatively rare disease accounting for approximately 1–2% of all cancers worldwide. However, it is the most common (95%) endocrine malignancy associated with more deaths annually than all other endocrine cancers combined [1, 2].
The incidence of thyroid cancer has continuously and sharply increased globally due to advanced technology to detect/screen in early preclinical stage, increased gender differences and a possible consequence of increased population exposure to radiation and to other still unrecognized carcinogens [3–5].
TCA commonly affects women (about 3–4 times) than men, can occur at any age but most tumors are diagnosed during the third to the sixth decades of life. It is the second and the fifth most common cancer in women in Italy and USA respectively. In USA only about 62, 000 new cases occurred in men and women in 2015 [6].
Despite, early diagnosis and better treatment of thyroid cancer, its mortality rate is increasing and in USA among patients diagnosed with thyroid cancer from 1974–2013, the overall incidence of thyroid cancer increased by 3.6% annually [7, 8].
A cohort study conducted in Australia among 2,260 patients underwent thyroidectomy performed over 40 years, thyroid cancer was diagnosed among 3.6 to 7.5% (P < 0.05) and confirmed an increase in the incidence of thyroid cancer over 4 decades [9]. The study conducted in Turkey among patients underwent thyroidectomy reported 14% of thyroid cancer rate [10].
Most thyroid malignancies are well differentiated thyroid cancers (DTC) and have an excellent prognosis with survival rates of more than 95% at 20 years although the recurrence or persistent rate is still high [11–15]. However, a notable proportion of cases are undifferentiated carcinomas and exhibit aggressive characteristics associated with a significantly higher recurrence and mortality rates [16–19].
Most primary thyroid cancers are epithelial tumors that originate from thyroid follicular (acinar) cells [20–24]. The three major pathologic types of thyroid carcinomas that arise from follicular cells are: papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC) and anaplastic thyroid carcinoma (ATC). Medullary thyroid carcinoma (MTC) arises from thyroid para-follicular (C) cells [11, 14, 25, 26].
PTC is the most common type and it accounts for 85–90% of all thyroid cancer cases [27–30] followed by FTC which is the second common carcinoma constituting about 5–10% of all thyroid cancers [31–33]. Medullary thyroid tumors accounts about 5–10% of all thyroid carcinoma and are familial in approximately 25% of cases [34]. ATC which is an undifferentiated thyroid carcinoma accounts for less than 2% of thyroid cancers and typically diagnosed in the elderly patients and is associated with poor prognosis [35–37].
Thyroid cancer, particularly undifferentiated thyroid carcinoma, is a cause of significant mortality and morbidity. Death from differentiated thyroid carcinoma, although rare, can also occur; early diagnosis and treatment is the key for good prognosis.
The study was conducted at Jimma medical center (JMC) which is one of the known teaching referral hospitals of the country located at distance of 335 km from the capital of Ethiopia, Addis Ababa to southwest direction. It was established in 1930 and owned initially by Jimma health science (Jimma University currently). It provides health services for approximately 9000 in patient and 80,000 outpatient attendances per year with a very wide catchment population (about 15 million people) in southwest Ethiopia. The study was conducted from November December, 2019 by reviewing patient cards that underwent thyroid surgery from 2015 to 2019 by employing a cross-sectional study design among sampled 260 patients. But, the histo-pathologic results of 48 patients were not available (probably not done/missed) and finally only 212 patients were analysed for pattern of CA. A structured questionnaire was used to record patient information including previous laboratory results.
Data was checked for missed values and outliers, cleaned, entered into Epidata version 4.3.1 and finally exported to SPSS version 22 for analysis. Both descriptive and analytical statistics were applied and the finding of the study was reported by tables, graphs and narration. Cross tabulation and logistic regression were applied to assess the association of outcome variable (thyroid cancer) and associated factors. P-value < 0.05 was declared as statistically significant.
Ethical clearance was obtained from Jimma University ethical committee after the proposal was approved by Jimma University ethical review board and support letter was collected from the University and hospital. All the information obtained from the records was anonymous.
Among total sample of 260 patients underwent thyroid surgery, majority of them belongs to female 219(84.2%), at age interval of 31–40 years (34.6%), Oromia region (91.2%) and Jimma zone (83.8%) as detailed in Table 1.
| Variable | Category | Frequency | Percent (%) |
|---|---|---|---|
| Sex | Male | 41 | 15.8 |
| Female | 219 | 84.2 | |
| Age | < 20 | 19 | 7.3 |
| 21–30 | 73 | 28.1 | |
| 31–40 | 90 | 34.6 | |
| 41–50 | 41 | 15.8 | |
| > 51 | 37 | 14.2 | |
| Marital status | Single | 57 | 21.9 |
| Married | 195 | 75.0 | |
| Divorced | 2 | 0.8 | |
| Widowed | 6 | 2.3 | |
| Region | Oromia | 237 | 91.2 |
| SNNP | 23 | 8.8 | |
| Zone | Jimma | 218 | 83.8 |
| Bedelle | 9 | 3.5 | |
| Ilu Abbabor | 10 | 3.8 | |
| Bonga | 14 | 5.4 | |
| Mizan Tepi | 9 | 3.5 | |
| Duration of illness in years | < 1 | 24 | 9.2 |
| 1–5 | 105 | 40.4 | |
| 6–10 | 91 | 35.0 | |
| > 10 | 40 | 15.4 |
The clinical profile of thyroid disease was recorded/ retrieved from patient cards including patient history, physical examination, laboratory results (T3, T4 and TSH) and pathologic results (FNAC and biopsy) as seen in Table 2.
| Variable | Category | Frequency | Percent (%) |
|---|---|---|---|
| Duration of illness in years | < 1 | 24 | 9.2 |
| 1–5 | 105 | 40.4 | |
| 6–10 | 91 | 35.0 | |
| > 10 | 40 | 15.4 | |
| Pain | Painless | 255 | 98.1 |
| Painful | 5 | 1.9 | |
| Voice change | Yes | 8 | 3.1 |
| No | 252 | 96.9 | |
| Dyspnea | Yes | 113 | 43.5 |
| No | 147 | 56.5 | |
| Dysphagia | Yes | 4 | 1.5 |
| No | 256 | 98.5 | |
| Family history of the thyroid disease | Yes | 93 | 35.8 |
| No | 167 | 62.2 | |
| History of smoking | Yes | 3 | 1.2 |
| No | 257 | 98.8 | |
| Previous radiation exposure | Yes | 0 | 0.0 |
| No | 260 | 100.0 | |
| Size of the swelling (longest diameter in cm) | < 5 | 33 | 12.7 |
| 6–10 | 86 | 33.1 | |
| > 10 | 141 | 54.2 | |
| Border of the mass | Regular | 238 | 91.5 |
| Irregular | 22 | 8.5 | |
| Nodularity | Simple nodular` | 192 | 73.8 |
| Multinodular | 68 | 26.2 | |
| Consistency | Soft | 53 | 20.4 |
| Firm | 200 | 76.9 | |
| Hard | 7 | 2.7 | |
| Surface | Smooth | 237 | 91.2 |
| Rough | 23 | 8.8 | |
| TSH | Low | 13 | 5.0 |
| Normal | 244 | 93.8 | |
| High | 3 | 1.2 | |
| T3 | Low | 2 | 0.8 |
| Normal | 258 | 99.2 | |
| High | 0 | 0.0 | |
| T4 | Low | 1 | 0.4 |
| Normal | 259 | 99.6 | |
| High | 0 | 0.0 | |
| Metastasis | LAP | 4 | 1.5 |
| Bone | 8 | 3.1 | |
| Lung | 0 | 0.0 | |
| FNAC result | Colloid goiter | 235 | 91.2 |
| Papillary CA | 7 | 2.7 | |
| Follicular neoplasm | 13 | 5.0 | |
| Medullary CA | 2 | 0.8 | |
| Others (hyperplastic) | 1 | 0.4 | |
| Biopsy result | Colloid goiter | 194 | 74.6 |
| Papillary CA | 5 | 1.9 | |
| Follicular CA | 11 | 4.2 | |
| Medullary CA | 2 | 0.8 | |
| Not applicable | 48 | 18.5 |
Colloid goiter was the most prevalent (74.6%) pattern of thyroid disease followed by follicular CA (4.2%), papillary CA (1.9%) and medullary CA (0.77%). But, biopsy results of 18.46% patients were not known. In general from total patients of 212 who had biopsy results, the burden of benign and malignancy accounted for 91.5% and 8.5% respectively (Fig. 1).
Logistic regression was applied to evaluate the associated factors of thyroid malignancy based on histopathological biopsy result. Accordingly, about six variables (duration of the disease, marital status, family history of the disease, nodularity, surface and border of the thyroid mass) were the candidate variables in binary logistic regression (p-value < 0.25) and finally, three variables (duration of the disease (> 10 years), surface (rough) and border (irregular) of the thyroid mass) were identified as the predictors of thyroid malignancy with AOR 0.05(0.004-0.60, P-v = 0.016; 1.9(1.17–5.8), P-v = 0.012 and 2.5(1.13–16.16), P-v < 0.001 respectively as detailed in Table 4.
| Variable | Category | Status of thyroid lesion; No (%) | ||||||
| Benign | Malignancy | COR(CI) | P-value | AOR(CI) | P-value | |||
| Age | < 30 | 70(33.0) | 5(2.4) | 1 | - | - | ||
| ≥ 30 | 124(58.5) | 13(6.1) | 1.5(0.5–4.3) | 0.483 | - | - | ||
| Sex | Male | 32(15.1) | 2(0.9) | 1 | - | - | ||
| Female | 162(76.4) | 16(7.5) | 1.6(0.4–7.2) | 0.555 | - | - | ||
| Region | Oromia | 175(82.5) | 17(8.0) | 1.8(0.2–14.6) | 0.562 | - | - | |
| SNNP | 19(9.0) | 1(0.5) | 1 | - | - | |||
| Zone | Jimma | 160(75.5) | 13(6.1) | 0.6(0.2–1.6) | 0.289 | - | - | |
| Others | 34(16.0) | 5(2.4) | 1 | - | - | |||
| Marital status | Married | 146(68.9) | 17(8.0) | 5.6(0.7–43.1) | 0.099* | 5.4(0.5–55) | 0.158 | |
| Others | 48(22.6) | 1(0.5) | 1 | 1 | ||||
| Duration of illness in years | < 1 | 17(8.0) | 4(1.9) | 1 | 1 | 24(9.2) | ||
| 1–5 | 76(35.8) | 7(3.3) | 0.4(0.1–1.5) | 0.169* | 0.4(0.06–2.2) | 0.280 | ||
| 6–10 | 69(32.5) | 4(1.9) | 0.3(0.05–0.9) | 0.064* | 0.08(0.009-0.7) | 0.022* | ||
| > 10 | 32(15.1) | 3(1.4) | 0.4(0.08–1.9) | 0.242* | 0.05(0.004-0.6) | 0.016* | ||
| Voice change | Yes | 7(3.3) | 1(0.5) | 1.6(0.2–13.5) | 0.681 | - | - | |
| No | 187(88.2) | 17(8.0) | 1 | - | ||||
| Dyspnea | Yes | 83(39.2) | 9(4.2) | 1.3(0.5–3.5) | 0.556 | - | - | |
| No | 111(52.4) | 9(4.2) | 1 | - | - | |||
| Family history | Yes | 70(33.0) | 4(1.9) | 0.5(0.2–1.6) | 0.245* | 1.5(0.3–7.4) | 0.617 | |
| No | 124(58.5) | 14(6.6) | 1 | 1 | ||||
| Size of neck swelling in cm | < 5 | 22(10.4) | 1(0.5) | 1 | - | - | ||
| 6–10 | 63(29.7) | 7(3.3) | 2.4(0.3–21) | 0.415 | - | - | ||
| > 10 | 109(51.4) | 10(4.7) | 2.0(0.3–16.6) | 0.513 | - | - | ||
| Nodularity | Simple nod. | 149(70.3) | 11(5.2) | 1 | 1 | |||
| Multinodule | 45(14.2) | 3(1.4) | 2.1(0.8–5.8) | 0.146* | 2.3(0.5–9.7) | 0.266 | ||
| Consistency | Soft | 30(18.8) | 3(1.4) | 1 | 1 | |||
| Others | 164(77.4) | 15(7.1) | 1.1(0.2–3.3) | 0.893 | - | - | ||
| Surface | Smooth | 186(87.7) | 9(4.2) | 1 | 1 | |||
| Rough | 8(3.8) | 9(4.2) | 2.3(1.2–7.4) | < 0.001* | 1.9.(1.17–5.8) | 0.012* | ||
| Border | Regular | 187(88.2) | 7(3.3) | 1 | 1 | |||
| Irregular | 7(3.3) | 11(5.2) | 4.2(1.2–14.1) | < 0.001* | 2.5(1.13–16.16) | 0.001* | ||
| *-statistically significant; Others- explained in Table 1 for all variables | ||||||||
Among total sample of 260 patients underwent thyroid surgery, majority of them were female 219(84.2%) with male to female ratio of 1:5.3, dominant age interval of 31–40 years (34.6%) while the distribution is in harmony with other studies [27–29].
In general, from total patients who had biopsy results of 212, about 194(81.5%) of patients were diagnosed for benign type of thyroid lesion while malignancy accounted for 8.5% where this finding was also supported by other studies that reported similar pattern [30, 31].
Among 18 patients diagnosed with malignancy, follicular CA was the commonest 11(61.1%) followed by papillary CA 5(27.7%) and medullary CA 2(1.1%). The pattern of CA of thyroid lesion in the present study was in agreement with different studies that reported dominance of follicular CA [32–35]. But, the finding was against other studies [36–40] that witnessed the dominance of papillary CA. The difference in pattern of thyroid CA was negotiated by review study of Ogbera and Kuku that justified as a changing trend toward the more frequent occurrence of papillary CA compared to follicular CA may be attributable to widespread iodization programs [41].
Finally, three variables (short duration of the disease, rough surface and irregular border of the thyroid mass) were associated with thyroid malignancy (p-value < 0.05) which was also in agreement with other studies [42,43]. Even though smoking was associated with the thyroid CA, it was not statistically significant in the present study against study conducted in Korea that revealed age, residence, belief in cancer screening, regular health check-ups, smoking, alcohol drinking, and exercise were associated with thyroid cancer screening [44]. The study by Haymart et al [45] reported that likelihood of thyroid cancer increases with higher serum TSH concentration in contrast to the present study probably due to screening modality and population difference.
In general, about the 81.5% of the thyroid lesion was identified as benign type while malignancy accounted for 8.5% from a total of 212 biopsy result analyzed for CA pattern. Among malignancy, follicular CA was the commonest (4.2%) among thyroid lesion patients underwent thyroidectomy. Short duration of the disease, rough surface and irregular border of the thyroid mass were identified as the predictors of thyroid malignancy.
Ethical clearance was obtained from Jimma University ethical committee after the proposal was approved by Jimma University ethical review board and support letter was collected from the University and hospital. All the information obtained from the records was anonymous.
Not applicable.
Needed materials can be provided upon request of corresponding author.
The authors declared that there is not conflict of interests.
Jimma University provided funding as post graduate research in 2019 for data collection.
Conceptualization, BM, YM and SA; Data curation, BM, YM, SA and WRD; Formal analysis, BM and WRD; Funding acquisition, BM; Investigation, BM; Methodology, BM, YM, SA and WRD; Resources, BM.; Software, BM and WRD; Supervision, YM, SA and WRD; Validation, BM, YM, SA and WRD; Visualization, BM, YM, SA and WRD; Writing original draft, BM and WRD; Writing-review and editing:, BM, YM, SA and WRD. All authors have read and agreed to the developed manuscript.
The authors would like to thank Jimma University for providing funding, and all data collectors and study participants.