Abstract
Objective: To examine existing evidence, trends and possible factors that may have affected the incidence of papillary thyroid cancer (PTC) among patients undergoing thyroidectomies in an iodine-sufficient population of Greece. Study design: All histology records from the patients who had undergone thyroid surgery at the Department of Surgery Laiko Hospital, Athens, Greece from January 1991 to December 2006 were retrospectively analyzed. Records were placed in a database which included patients' demographics, history, and medical condition, clinical and surgical parameters. Patients and Methods: One thousand four hundred and twenty-six patients (265 males and 1161 females) had undergone thyroidectomy during the above period of time. All surgeons favoured total thyroidectomy with resection of pro- and paratracheal lymph nodes. Thyroid tumors were classified according to the WHO classification system and were staged according to the TNM staging system. Results: In 278 patients, PTC was histologically diagnosed. From 1999 onwards, thyroid surgery shifted towards total thyroidectomy, while statistically significantly increased incidence of PTC and papillary microcarcinoma detection and decreased incidence of PTC greater than 10 mm detection in the whole population were noticed. Moreover, from 1999 onwards, smaller size of primary tumors, higher incidence of T1 tumors, lower incidence of T4 tumors, lower incidence of metastatically infiltrated peritracheal lymph nodes, higher incidence of stage I tumors and lower incidence of stage IV tumors were documented. Finally, a higher incidence of PTC in males, females and the whole population aged 51-70 years compared to the other age groups since 2003 was noticed. Conclusion: The increased incidence of PTC clearly correlated to the increased incidence of papillary microcarcinoma detection, reflecting the proportion for total thyroidectomy as well as changes in the diagnostic approach boosted by more careful pathological examination, rather than the effect of environmental factors such as the Chernobyl accident. Whether the Chernobyl accident has any predisposing effect on the increased incidence of PTC remains to be proven.
It can be postulated that the concentration of radioactive materials both in the atmosphere and soil of the South-Eastern Europe countries (including Greece) might have increased over the last twenty years due to two important events: i) the Chernobyl nuclear power station accident (28-04-1986) and ii) the use of depleted uranium weapons during the Balkan war in former Yugoslavia (1990-1991).
Following these events, western or south-western winds drifted the radioactive cloud west, gradually affecting other European countries, while the rainfall which exacerbated radionuclide fallout gave rise to contamination of groundwater and soil. Both these events may have important health consequences, either immediate, such as radiation sickness (1), or future ones, such as carcinogenesis, leukemia or birth defects in the general population of affected countries (1, 2). Although there is no clear evidence that depleted uranium is related to future carcinogenesis (3), several epidemiological studies have found an increased incidence of differentiated thyroid cancer during childhood both in the former Soviet Union Republics (2, 4-6) and in other European countries (7, 8) following the Chernobyl accident (1, 2, 9). It has therefore been proposed that this form of radiation constitutes a predisposing factor for thyroid cancer development (1, 2, 9).
Greece is one of the countries which are thought to have been affected by the Chernobyl accident. The effect of ionizing radiation in Greece following the accident, based on the 137Cs deposition on surface soil samples, is presented in Figure 1. The present study constitutes a retrospective analysis of the iodine sufficient population of Greece (light grey area in Figure 1) who underwent thyroidectomy, fulfilling the criteria for thyroid surgery. The aim of the study was to examine existing evidence, trends and possible factors that may have affected the incidence of thyroid cancer.
Patients and Methods
We retrospectively analyzed all histology records from patients (n=1426) who had undergone thyroid surgery in the Department of Surgery of an academic tertiary referral medical center (University of Athens, Medical School, Laiko Hospital, Athens, Greece) from January 1991 to December 2006. There were 1324 patients who had undergone total thyroidectomy, 85 patients who had undergone subtotal thyroidectomy and 17 patients who had undergone lobectomy as first operation, proceeding to lobectomy of the remaining lobe between 1991 and 2006 after the histological confirmation of a malignancy
Records were placed in a database, which included patients' demographics, history, and medical condition, clinical and surgical parameters. There were 265 male patients with a median age of 43 years (interquartile range (IR): 36.5-58) and 1161 female patients with a median age of 47 years (IR: 36-60). All surgeons favoured total thyroidectomy with resection of pro- and paratracheal lymph nodes.
The histological classification of the thyroid tumors was based on the WHO classification system (10). All tumors were staged according to the AJCC/UICC, 6th edition, and 2002 TNM staging system (11).
In the attempt to evaluate the thyroid malignancies further, the study period was divided into several chronological periods and the results were compared. Six hundred and eighty-two thyroid operations were performed between 1991 and 1998, and 744 were performed between 1999 and 2006. Subtotal thyroidectomy was performed in 77 patients (11%) in the former period, but only in 8 patients (1%) in the latter.
Statistical analysis. Univariate analyses between groups were determined by the Fisher exact test. P<0.05 was considered statistically significant. All statistical calculations were performed using SPSS (Statistical Package for the Social Sciences, version 12.0, SPSS, Chicago, IL, USA) for Windows XP (Microsoft).
Results
In 315 out of the 1426 patients, thyroid cancer was histologically detected. Papillary carcinoma (PTC) represented the commonest histological type of thyroid cancer, accounting for 88% (278 out of 315) of all malignancies, while other malignant histological types were follicular carcinoma (n=21), Hürthle cell carcinoma (n=6), medullary carcinoma (n=7), anaplastic carcinoma (n=1), primary lymphoma (n=1) and metastatic carcinoma (n=1).
No statistically significant differences in the incidence of follicular and Hürthle cell carcinomas between sex, different age groups or different periods of the study were detected.
Overall incidence of PTC. The total number of PTC cases (both microcarcinomas as well as PTC>10 mm) which were diagnosed annually is presented in Figure 2. PTC was detected in 58 male patients (22%) with a median age of 49.5 years (IR: 38-62) and in 220 female patients (19%) with a median age of 46 years (IR: 36-57). Comparing the distribution of PTC by sex between 1991-1998 and 1999-2006, a statistically significantly increased incidence of PTC detection in surgically resected thyroid glands among both the female patients (p=0.0015) and the whole population (p<0.001), but not among the male patients (p=0.099) since 1999 was documented.
Incidence of papillary microcarcinomas. Overall, papillary microcarcinomas (≤10 mm) were detected in 59 out of 682 (9%) thyroid operations performed between 1991 and 1998 and in 118 out of 744 (16%) thyroid operations performed between 1999 and 2006 (p<0.001).
In the second half of the study period (from 1999 onwards), papillary microcarcinomas were responsible for PTC detection in 118 out of 172 patients compared to the 59 out of 106 patients in the first period of the study (p=0.029). This finding was unrelated both to the size of the primary tumor and to the age of the patient at the time of the operation, but was correlated to the surgical technique (118 out of 736 versus 59 out of 605 total thyroidectomies), (p<0.001).
Incidence of PTC>10 mm. Excluding the papillary microcarcinomas from the analysis, a statistically significantly decreased incidence of PTC>10 mm (54 out of 172 versus 47 out of 106) (p=0.029) was documented from 1999 onwards. This finding was unrelated to the size of the primary tumor, the age of the patient at the time of operation, as well as the surgical technique.
Histological findings of PTC. Regarding the microscopic findings of PTC, comparing the period 1991-1998 to the period 1999-2006, the present study disclosed that the latter was characterized by statistically significantly: (i) smaller size of primary tumors (median 6 mm, IR 3-14 mm versus median 10 mm, IR 4-20 mm) (p=0.028); (ii) higher incidence of T1 tumors (142 out of 172 patients versus 72 out of 106 patients) (p=0.0047); (iii) lower incidence of T2 tumors (21 out of 172 patients versus 24 out of 106 patients) (p=0.021); (iv) lower incidence of T4 tumors (2 out of 172 patients versus 6 out of 106 patients) (p=0.028); (v) lower incidence of metastatically infiltrated peritracheal lymph nodes (12 out of 172 patients versus 15 out of 106 patients) (p=0.031); and (vi) higher incidence of stage I tumors (142 out of 172 patients versus 75 out of 106 patients) (p=0.020). Distribution of PTC by sex. Among the female patients (Table I), the period 1991-1998 was characterized by the statistically significantly increased incidence of T4 tumors (p=0.024) and metastatically infiltrated peritracheal lymph nodes (p=0.043), as well as a trend for increased incidence of stage IVa tumors (p=0.058) compared to the period 1999-2006.
The effect of ionizing radiation in Greece following the Chernobyl nuclear accident based on the 137Cs deposition on surface soil samples. Source: Department of Nuclear Engineering, National Technological University of Athens, April 1995 (with permission).
Annual cases of papillary microcarcinomas and PTC>10 mm.
Among the male patients (Table I), the period 1991-1998 was characterized by larger primary tumors (p=0.035), lower incidence of T1 (p=0.003) and higher incidence of T2 tumors (p=0.003), lower incidence of stage I (p=0.008) and higher incidence of stage II tumors (p=0.018) compared to the period 1999-2006.
Correlation between sex and several parameters of papillary cancer.
Distribution of PTC by age. A higher incidence of papillary carcinoma detection in males (p=0.011), females (p<0.0001) and the whole population (p<0.0001) aged 51-70 years compared to the other age groups since 2003 was noticed. Comparing the age of the female patients to several parameters of the papillary carcinoma (Table II), we disclosed a higher incidence for stage I tumors (p=0.022) with their predominant development in the right lobe of the gland (p=0.004) in patients age between 51 and 70 years from 2003 onwards. Comparing the female group aged 51-70 years between 1991-2002 and 2003-2006, a higher incidence for T1 (p=0.026) and stage I tumors (p=0.011) and a lower incidence for capsular invasion of the tumors (p=0.015) in the period 2003-2006 was found.
Comparing the age of the male patients for several parameters of the papillary carcinoma (Table III), we disclosed a lower incidence for multifocal development (p=0.027) and capsular invasion (p=0.060) of the tumors in patients aged 51-70 years from 2003 onwards. Comparing the male group aged 51-70 years between 1991-2002 and 2003-2006, we did not find any statistically significant difference.
Discussion
The present study disclosed a statistically significantly increased incidence of PTC detection in the surgically resected thyroid glands among Greek females from 1999 onwards and this finding was clearly correlated to the increased incidence of papillary microcarcinoma detection over the same period of time. Moreover, since 1999, thyroid surgery shifted towards total thyroidectomy as the operation of choice, practically abandoning subtotal thyroidectomy and lobectomy as surgical options.
The increased incidence of PTC could be attributed to the well-documented increased prevalence of thyroid cancer in female patients compared to males (12). Benign thyroid disease is more frequent in women, leading to increased awareness of the female population concerning the investigation and diagnosis of thyroid cancer, enabling the detection of small, clinically indolent papillary carcinomas (13). Moreover, a recent report showed a rise in the proportion of total thyroidectomise, favouring the hypothesis of a causal effect linking the increased incidence of thyroid cancer to medical practice and surgery in particular and not to the consequence of possible contamination (14).
Epidemiological studies have examined the possible relationship between several factors and the risk of developing thyroid cancer. Multivariate analyses concluded that external radiation to the head/neck, a history of goiter/nodules and a family history of proliferative thyroid disease constituted independent predisposing factors for thyroid cancer (15). The risk of developing thyroid cancer depends on the age of exposure and the dose of radiation (16-18). Radiation exposure of the thyroid gland is the only well-documented risk factor that increases the incidence of well-differentiated thyroid cancer (17). The hypothesis of an association between iodine intake and the risk for thyroid cancer development was based on the results of experimental animal studies, which demonstrated an increased incidence of thyroid epithelial cell carcinomas after prolonged iodine deficiency, leading to a stimulation of the thyroid gland by thyrotropin and possibly other growth factors (19). Epidemiological studies, however, found contradictory results: Thus, the highest incidence rates of thyroid cancer were observed in Iceland (20) and Hawaii (21) (both iodine-sufficient areas), and exposure to volcanic activity (where the natural radiation is higher and radiation is known to increase the development of thyroid carcinomas) has been suggested as an explanation for the high incidence in these islands. In contrast, a study from Italy demonstrated a two-fold increased incidence of thyroid cancer in iodine-deficient areas compared to iodine-sufficient ones (22). Several studies have demonstrated a shift towards a rise in papillary carcinoma, but no clear relationship between overall thyroid cancer incidence and iodine intake (23-26).
Correlation between the age (years) of the female patients and several parameters of papillary cancer.
Correlation between the age (years) of the male patients and several parameters of papillary cancer.
The increased incidence of papillary microcarcinomas and favorable parameters for the locoregional disease, as well as the decreased incidence of PTC >10 mm, which were documented during the second half of the present study, are in agreement with another report (27) which reported a significantly lower size of all thyroid tumors and significantly increased rate of papillary microcarcinomas since 1995. It is also in agreement with another European report (28) stating increased incidence of papillary microcarcinoma detection, probably reflecting a more strict application of the histopathological WHO classification since 1995. However, a possible argument could be that patients may prefer early surgical intervention, once the diagnosis of multinodular goiters is established, rather than suppressive conservative management for a long period of time.
The present study found an increased detection of thyroid cancer in the age group of 51-70 years in both sexes, as well as in the whole population since 2003. It is well known that thyroid cancer is more common in the elderly (29) and male sex may represent a risk factor for thyroid cancer in this population subgroup (30). Moreover, recent European reports stated that the median age for diagnosis of thyroid cancer in females has increased by 2 to 8 years comparing the period 1974-1978 to the period 1999-2003 (30) and also confirmed a significantly increased age at the time of diagnosis of the disease (27).
The observation of a clear increased incidence of papillary carcinoma in the whole population since 2003 (seventeen years from the Chernobyl accident) is in agreement with previous reports, which stated that thyroid cancer is more likely to develop 10 years after radiation and the risk probably persists for life, but may begin to fall after 20 years (12, 13), while radiation-induced thyroid cancer is predominantly papillary (24), behaves similarly to and should be treated the same as non-radiation-associated thyroid cancer (12).
It could be hypothesized that the documented increased incidence of papillary microcarcinoma detection since 1999 reflects the causative role of the Chernobyl accident, since at least 15 years of low-grade radiation exposure may have initiated new mutations (31) and chromosomal imbalances (32), leading to microcarcinoma formation which became detectable because of the increased awareness of pathologists, thus finally reflecting a true increased incidence in the actual occurrence of the disease. If that was the case, the age at the time of diagnosis should have decreased, as happened in nations obviously affected by the Chernobyl accident (33). However, such a difference was not noticed in the female patients of the present study (median age at the time of diagnosis 46 years for the period 1991-1998 and 45 years for the period 1999-2006, p=0.437). On the other hand, the finding of the increased incidence of the disease in patients older than 51 years (also related to the increased rate of microcarcinomas detection) indirectly reflects the significant increased age at the time of diagnosis and is in agreement with reports from European countries with unknown influence of the Chernobyl accident (27, 30, 34).
In conclusion, the results of the present study addressing the increased incidence of PTC since 1999 clearly correlated to the increased incidence of papillary microcarcinoma detection over the same period of time, rather reflects the proportion for total thyroidectomy and pathological attribution of a more strict application of the histopathological WHO classification than the effect of environmental factors (Chernobyl accident). However, the increased incidence from 2003 onwards of PTC in female patients aged 51 to 70 years suggests that if iodine exposure exerts any carcinogenetic effect, the process lasts about 15 years. Whether the Chernobyl accident has any predisposing effect on the increased incidence of PTC remains to be proven.
- Received July 8, 2009.
- Revision received October 29, 2009.
- Accepted November 5, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
