Thyroid dysfunction after radiation therapy in head and neck cancer patients

Thyroid Dysfunction After Radiation Therapy in Head and Neck Cancer Patients THOMAS A. TAMI, MD, LCDR PATRICK GOMEZ, MC, USNR, CAPT GREGG S. PARKER, MC, USN, MANU B. GUPTA, BS, AND LCDR DEBORAH A. FRASSICA, MC, USN Introduction: The reported incidence of hypothyroidism following surgery and/or radiation therapy for head and neck cancer varies widely. Most patients undergo thyroid lobectomy during laryngectomy. Standard radiation treatment portals often include the thyroid gland. The insidious development of hypothyroidism may be misdiagnosed. This study examines the incidence of thyroid dysfunction in the setting of head and neck cancer therapy. Materials and Methods: Thyroid function tests were performed on 100 consecutive patients treated in the head and neck tumor clinic. Statistical inferences on proportions were made using chi-square analysis. Results: Therapy included surgery only (10 patients), radiation therapy only (28 patients), and combined therapy (62 patients). These patients experienced thyroid dysfunction in O%, 29%, and 45% of individuals respectively. These differences were statistically signicant (P < .05). The highest rate of dysfunction (69%) was associated with patients undergoing laryngectomy and radiation therapy. When laryngectomy was not performed, thyroid dysfunction occurred in 28%. Conclusion: The likelihood of thyroid dysfunction after radiation therapy is high particularly when combined with surgery in which thyroid lobectomy is performed and the contralateral lobe is potentially devascularized. These results suggest that radiation therapy is a primary factor in alteration of thyroid function. We recommend that routine thyroid function testing be part of follow-up of all head and neck cancer patients. Copyright 0 1992 by W.B. Saunders Company

Although the reported incidence of hypothyroidism following surgery and/or radiation therapy (XRT) for head and neck cancer varies widely, thyroid dysfunction is often not considered a major risk. The standard radiation treatment portals often include the thyroid gland. Hemithyroidectomy is often performed, especially during laryngectomy, fur-

From the Department of Otolaryngology-Head and Neck Surgery, University of California, San Francisco, CA; Department of Otolaryngology-Head and Neck Surgery, Naval Hospital, Camp Lejeune, NC; and the Departments of Otolaryngology-Head and Neck Surgery and Radiation Oncology, Naval Hospital, Portsmouth, VA. The views and assertions contained herein are those of the authors and are not to be construed as official or as reflecting the views of the Department of the Navy or the Department of Defense. Address reprint requests to Thomas A. Tami, MD, Department of Otolaryngology, 400 Parnassus Ave A-71 7, San Francisco, CA 94143-0342. This is a US government work. There are no restrictions on its use. 0196-0709192/l SOS-0006$0.00/0 American

Journal

of Otolaryngology,

ther magnifying this risk.* Because it is generally assumed that the remaining thyroid tissue will adequately compensate for this surgical loss, pre- and post-treatment thyroid function tests are not routinely obtained. Although problems with wound healing, especially pharyngocutaneous fistulas, have been associated with hypothyroidism,‘v3 frank myxedema can also occur. Symptoms of hypothyroidism commonly include a generalized feeling of malaise and depression, accompanied by decreased mental and physical performance disproportionate to the physical disability.4 In 1981, Palmer reported that “psychological depression often found during the first months after laryngectomy is partly caused by a transient hypothyroidism.5” Thyroxine supplements often resulted in complete reversal of all treatment-associated symptoms Because the features of hypothyroidism often have an insidious onset and can be mistaken for physical or psychological sequelae of the neoplasm, hypothyroidism fre-

Vol 13, No 6 (November-December),

1992: pp 357-362

357

358

TAMI ET AL

quently goes undiagnosed and symptoms persist. Appropriate evaluation and monitoring of post-treatment thyroid function could prevent the potential morbidity associated with this clinical syndrome. This study examines the incidence of thyroid dysfunction in the setting of head and neck cancer therapy and explores the association of tumor and treatment variables with the severity and onset of hypothyroidism.

MATERIALS AND METHODS Thyroid function tests were performed on 100 consecutive patients observed in the Head and Neck Tumor Clinic at the Naval Hospital in Portsmouth, VA. Charts were reviewed to determine age, sex, site and stage of cancer, type and date of surgery, and dates and dosages of XRT. Patients were excluded if they had preexistent thyroid disease or previous thyroid surgery. Overt hypothyroidism was defined as a thyroid-stimulating hormone (TSH) level greater than 8 mUL (normal value TSH <8-O) and a T4 level ~70 nmol/L (normal value T4 70 to 160 nmol/L). Decreased thyroid reserve was defined as an elevated TSH with normal T4 levels. Secondary hypothyroidism was represented by low T4 levels with an associated low TSH. Statistical inferences on proportions were made using chisquare analyses.

RESULTS Of the 100 patients, 10 underwent surgery alone, 28 had XRT only, and the remaining 62 had combined therapy. The rates of thyroid dysfunction among these treatment groups (Table 1) are significantly different (P < .05). Patients having surgery only (no XRT) showed no incidence of thyroid dysfunction. When

TABLE1.

Treatment

XRT was used alone [no surgery), a 29% incidence was encountered (21% had decreased thyroid reserve and 7% displayed overt hypothyroidism). Patients who underwent combined therapy had an overall dysfunction rate of 45% (18% had decreased thyroid reserve and 24% were overtly hypothyroid). Although the difference in thyroid dysfunction rates between the radiation only and the combined therapy groups was striking, it approached, but did not achieve, statistical significance (P = .056). A further breakdown of the 62 patients who underwent combined therapy is shown in Table 2. The highest rate of dysfunction (69%) was observed in patients undergoing laryngectomy. When laryngectomy was not performed, the overall rate of dysfunction was only 28%. This difference was highly statistically significant (P = .OOl). The rate of overt hypothyroidism was also significantly higher when laryngectomy was performed (P = .005). No statistically significant relationship to the performance of hemithyroidectomy was observed. Stratifying the 90 patients who underwent XRT according to age and sex did not suggest a significant correlation with thyroid dysfunction. Similarly, no statistical relationship between total radiation dose or time since radiation and the occurrence of thyroid dysfunction could be discerned. Table 3 stratified patients who underwent XRT according to the stage of their disease. Although a trend toward more thyroid dysfunction in higher-staged disease was observed, this difference was not statistically significant.

Modality No. of Patients Surgery Only

Total Thyroid function Normal Thyroid dysfunction Decreased reserve Overt hypothyroidism Secondary hypothyroidism * P = .036 t P = ,056

XRT

XFIT and

Only

Surgery

10

28

62

10 (100%)

20 (71%) 8’ (29%) 6 (21%) 2t (7%) 0

34 28’ 11 15t 2

0* 0 0 0

(55%) (45%) (18%) (24%) (3%)

THYROID

DYSFUNCTION

TABLE 2.

Surgery

AFTER RADIATION

359

THERAPY

Type No. of Patients Laryngectomy

Total Thyroid function Normal Thyroid dysfunction Decreased reserve Overt hypothyroidism Secondary hypothyroidism

Without

With

Hemithyroid

Hemithyrold

14

12

5 9 3 5 1

(36%) (64%) (21%) (36%) (7%)

3 9 2 6 1

and XRT Surgery Other Than Laryngectomy and XRT

Total

26 (25%) (75%) (17%) (50%) (8%)

36

8 18’ 5 11 t 2

(31%) (69%) (19%) (42%) (8%)

26 10’ 6 4t 0

(72%) (28%) (17%) (11%)

*P = .OOl t P = ,005

In 1961, Felix et al7 documented the first reported case of hypothyroidism appearing 6 years after treatment for carcinoma of the larynx, and in 1965,Markson and Flatman’ described five patients who developed myxedema after irradiation to the neck for cancers of the nasopharynx, cervical lymph nodes, and breast; they wrote that “as a complication of deep x-ray therapy to the neck in nonthyrotoxic patients . . . hypothyroidism would seem to be very rare.” Much of the earlier literature generally reported a low incidence of hypothyroidism. In 1964, Koulumies et al9 studied 118 patients who had been definitively treated with up to 5,000 rads and found no cases of decreased thyroid function; however, in 1967 Einhorn and Wilkholm” found that 3 of 41 patients developed hypothyroidism 10 or more years after external beam radiation for laryngeal and hypopharyngeal cancer, and many of those patients who were not overtly hypothyroid displayed a reduction in thyroid reserve. Over the last two decades, more attention

The relationship of primary tumor site and the subsequent occurrence of thyroid dysfunction is shown in Table 4. A higher incidence of dysfunction was observed in patients with primary tumors below the level of the hyoid bone; however, this difference was not statistically significant. When combined therapy (radiation and surgery) was used, significantly more patients with primaries below the level of the hyoid developed thyroid dysfunction (64% v 29%, P = .006). Of the 62 patients who underwent combined therapy, the incidence of thyroid dysfunction was higher when radiation therapy preceded surgery (Table 5). This difference was not statistically significant. COMMENT Before 1970, hypothyroidism was rarely reported following the treatment of head and neck cancer. Clinicians at that time were unaware of this association, and the thyroid gland was considered a radioresistant organ.6 TABLE 3.

Stages of Disease No. of Patients Stages

Total Thyroid

I and

II

Stages III and IV

33

52

Non-Staaed

5

function

Normal Thyroid dysfunction Decreased reserve Overt hypothyroidism Secondary hypothyroidism

32 20 10 9 1

(62%) (38%) (19%) (17%) (2%)

17 16 7 8 1

(52%) (48%) (21%) (24%) (3%)

5 (100%) 0 0 0 0

360

TAMI ET AL

TABLE 4.

Primary

Tumor

Site No. of Patients Below

Hyoid

Above

XRT and XRT only

Total

18

Thyroid function Nomal Thyroid dysfunction Decreased reserve Overt hypothyroidism Secondary hypothyroidism

14 4 3 1 0

*P

(78%) (22%) (17%) (6%)

Total

46

10 18* 6 10 2

(36%) (64%) (21%) (36%) (7%)

XRT Only

10

24 22 9 11 2

(52%) (48%) (20%) (24%) (4%)

6 4 3 1 0

Surgery

31 (60%) (40%) (30%) (10%)

22 Y 4 5 0

(71%) (29%) (13%) (16%)

Total

Unknown Primary Total

41

3

28 (68%) 13 (32%) 7 (17%) 6(15%) 0

2 (67%) 1 (33%) 1 (33%) 0 0

= ,006

has been given to the association of thyroid dysfunction and treatment for head and neck cancer. In 1971,Lavelle’ reported that 12 of 21 patients receiving combined laryngectomy and XRT developed increased TSH levels and decreased thyroid reserve, and in 1972, Murken and Duvallll found that of 12 patients treated with 3,800 to 5,000 rads of preoperative radiation followed by surgery, 8 (66%) became clinically hypothyroid. In 1981, Palmer et al5 also reported that 30% of their patients who were treated with XRT and laryngectomy had low thyroxine levels. A similar report by Liening et all2 in 1990 documented a 28% incidence of hypothyroidism in patients who were treated with a surgical procedure not involving the thyroid gland but followed by XRT. Interestingly, the use of XRT in the management of Hodgkin’s disease and supraclavicular nodal metastasis from breast carcinoma has also been associated with the development of thyroid dysfunction,13’14 further imTABLE 5.

Temporal and Radiation

Relationship

Between

Surgery

No. of Patients surgery

Total Thyroid function Normal Thyroid dysfunction Decreased reserved Overt hypothyroidism Secondary hypothyroidism *P

XRT and

Surgery

28

Hyoid

= ,182

Radiation

Before

Before

Radiation

surgery

49 29 20” 7 12 1

13 (59%) (41%) (14%) (24%) (2%)

5 8’ 4 3 1

(38%) (62%) (31%) (23%) (8%)

plicating radiation therapy in this syndrome. The results of our study showed an incidence of post-treatment overt hypothyroidism and decreased thyroid reserve that is comparable to most recent studies and emphasizes the causative role XRT plays in the development of this problem. The likelihood of thyroid dysfunction after XRT appears to be strong, particularly when combined with the potentially devascularizing effects of surgery. Our results indicate that surgery significantly increases the incidence of hypothyroidism compared with radiation alone but that surgery alone does not pose a substantial risk because none of the 10 patients who had surgical therapy alone developed abnormal thyroid function. Although these 10 patients generally had lower staged disease and underwent less aggressive surgical procedures, 2 had undergone laryngectomy (one of these with hemithyroidectomy). These results substantiate those of similar, earlier studies that implicated XRT as the primary factor in the alteration of thyroid function 1,4,5,8,11,15-18 When combined with XRT, laryngectomy appears to be a particularly high-risk procedure for thyroid dysfunction, and this risk is increased by the addition of partial thyroidectomy. Shafer et alI5 reported a 50% incidence of hypothyroidism when irradiation followed laryngectomy and hemithyroidectomy; Liening et all2 reported a 65% incidence and other studies4v5*16also confirm these findings. Although we observed a very high incidence of post-irradiation hypothyroidism when laryngectomy was associated with partial thyroi-

THYROID

DYSFUNCTION

AFTER RADIATION

361

THERAPY

dectomy (75%), the incidence was also high (64%) when hemithyroidectomy was not included. Several previous studies have described an increased susceptibility of the female thyroid gland to the effects of surgery and XRT. Posner et al” described this relationship in 1984, as did Vrabec and Heffron4 in 1981. Because the rate of spontaneous hypothyroidism is also much higher in women, this may indicate a lower thyroid reserve in this population. The total number of women in our study was too small to make any meaningful conclusions. Although the incidence of hypothyroidism might be expected to increase in proportion to the elapsed time since radiation therapy7*10.‘g as well as the total radiation dosage, we found no such relationships. Within the general therapeutic radiation dosage range for head and neck cancer, dosage did not have a significant bearing on the risk of hypothyroidism. We compared patients whose tumors were treated in early stages (I and II] with those with advanced stage tumors (III and IV). Although there was no statistically significant difference between these two groups (38% versus 48%), a trend toward greater dysfunction with larger tumors was observed. This difference was probably due to a greater proportion of patients in the lower stage group receiving XRT only, as well as the more extensive surgical procedures performed in the higher-staged patients. When tumors located below the hyoid bone were compared with those located above the hyoid, a trend toward more dysfunction was observed (48% versus 32%). When compared for patients undergoing combined therapy (surgery and XRT), this difference was statistically significant. This observation might be attributed to two possible factors. First, a higher rate of laryngectomy was performed in the below-the-hyoid group, and, as we and others have previously shown, laryngectomy is a high-risk procedure. Secondly, the thyroid gland would undoubtedly receive a higher radiation dose when the primary tumor is located in this area. These factors suggest that the effects of treatment, and not any inherent tumor characteristic, are probably the major contributing factors.

An interesting observation was the occurrence of thyroid dysfunction when radiation therapy was used postoperatively (41%) compared with preoperative XRT (62%). This difference was not statistically significant but suggested that tissue injury induced by preoperative XRT may amplify surgically induced thyroid damage, perhaps by facilitating and augmenting devascularization of the thyroid gland. Although not statistically significant, an argument for postoperative instead of preoperative XRT might be considered. In summary, we found that thyroid dysfunction is an extremely prevalent, yet easily treatable, source of morbidity in patients undergoing radiation therapy for head and neck cancer. Although nearly half of all patients receiving combined therapy (surgery and XRT) are at risk for hypothyroidism, it also occurs when XRT is used alone (29% in our series). When XRT is combined with surgery, laryngectomy (with or without hemithyroidectomy) appears to place patients at particularly high risk. Primary tumors that occur below the level of the hyoid bone also appear to be associated with a higher rate of dysfunction, and this is undoubtedly related more to the surgical therapy and radiation portals rather than to an intrinsic feature of the tumor. The physician should not dismiss patient complaints of lethargy and easy fatigability following therapy for head and neck cancer, especially since it is easy to attribute these symptoms to physical or psychological factors related to the original cancer process. To maximize the likelihood of early diagnosis and treatment and minimize the potential morbidity associated with this condition, we recommend that routine thyroid function testing be part of the post-treatment care of all head and neck cancer patients. REFERENCES 1. Lavelle RJ: Thyroid function after radiotherapy and total laryngectomy in the treatment of carcinoma of the larynx. Ann Otol Rhino1 Laryngol 80593-598, 1971 2. Alexander MV, Zajchuk JT, Henderson RL: Hypothyroidism and wound healing. Arch Otolaryngol 108: 289-291, 1982 3. Talmi YP, Finkelstein Y, Zohar Y: Pharyngeal fistulas in postoperative hypothyroid patients. Ann Otol Rhino1 Laryngol 98:267-268, 1989 4. Vrabec DP, Heffron TJ: Hypothyroidism following

362

treatment for head and neck cancer. Ann Otol Rhino1 Laryngol 90:449-453, 1981 5. Palmer BV, Gaggar N, Shaw HJ: Thyroid function after radiotherapy and laryngectomy for carcinoma of the larynx. Head Neck 4:13-15,1981 6. Warren S: Effects of radiation on normal tissues. Arch Path01 35:315-318, 1943 7. Felix H, Dupre N, Drape M, et al: Incidence a long terme dune radiotherapie pour cancer du larynx, surl’apparition d’un myxoedema. Lyon Med 93:1043-1050,196l 8. Markson IL. Flatman GE: Mvxoedema after deeu x-ray therapy to the neck. Br Med J 1:1228-1230, 1965 _ 9. Koulumies M, Voutilainen A, Koulumies R: Effect of x-ray irradiation for laryngeal cancer on the function of the thyroid gland. Ann Med Interne Fenniae. 53:89-96, 1964 10. Einhorn J, Wilkbolm G: Hypothyroidism after external irradiation to the thyroid region. Radiology 88:326328, 1967 11. Murken RE, Duvall AJ: Hypothyroidism following combined therapy in carcinoma of the laryngopharynx. Laryngoscope 82:1306-1314, 1972 12. Liening DA, Duncan NO, Blakeslee DB, et al: Hypothyroidism following radiotherapy for head and neck cancer. Otolaryngol Head Neck Surg 103:10-13, 1990

TAMI ET AL

13. Schimpff SC, Diggs CH, Wiswell JG, et al: Radiation-related thyroid dysfunction: Implications for the treatment of Hodgkin’s disease. Ann Intern Med 92:91-98, 1980 14. Bruning P, Bonfrer J, De Jong-Bakker M, et al: Primary hypothyroidism in breast cancer patients with irradiated supraclavicular lymph nodes. Br J Cancer 51:659663,1985 15. Shafer RB, Nuttall FQ, Pollak K, et al: Thyroid function after radiation and surgery for head and neck cancer. Arch Intern Med 135:843-846, 1975 16. De Jong JMA, Van Daal WAJ, Elte JWF, et al: Primary hypothyroidism as a complication after treatment of tumours of the head and neck. Acta Radio1 Oncol 21:299303,1982 17. Adler RA, Corrigan DF, Wartofsky L: Hypothyroidism after X irradiation to the neck: Three case reports and a brief review of the literature. Johns Hopkins Med 1 138: 180-184,1976 18. Bartels EC, Kusakcioglu

0: Hypothyroidism subseauent to x-rav theranv. Lahev Clin Bull 14:64-71.1965 19. Posner-MR, E-&in TJ, Miller D, et al: Incidence of hypothyroidism following multimodality treatment for advanced squamous cell cancer of the head and neck. Laryngoscope 94:451-454, 1984