Italy is an area of moderate iodine deficiency, where goiter is rather common. Goiter due to iodine deficiency progresses from a diffuse enlargement of the thyroid in young adults to a multinodular structure with advancing age 1. Patients with large multinodular goiter (MNG) frequently complain of signs and symptoms that range from a slight sense of constriction to more severe compressive disorders and impaired respiration 1. While the functional activity of MNG is variable, in elderly subjects TSH sometimes declines as a consequence of autonomous hormonal activity of the nodules, with clinical features ranging from euthyroidism to sub-clinical or overt hyperthyroidism. In these patients, surgery is the standard procedure but it may be contraindicated because of cardiac, pulmonary, or other chronic disorders. When surgery is not practicable or is refused by the patient, radioiodine (RAI) therapy may be used to reduce thyroid volume (TV) in enlarged MNG 2345; however, the rather low RAI uptake (RAIU) and the need for intense RAI activity can make this therapeutic procedure less effective.

Recombinant human TSH (rhTSH) is now used in the diagnosis and therapy of differentiated thyroid cancer, and may be of value in some other conditions 67. One instance is the use of rhTSH to increase RAIU in patients with benign MNG, as recently reported in a review by Fast et al. 8. RhTSH pre-treatment has been shown to increase thyroid RAIU and to modify the regional distribution of RAI by stimulating RAIU in hypofunctioning regions within a goiter 9. Huysmans et al. 10 showed that a single 0.01–0.03 mg rhTSH dose administered 24-h before a diagnostic RAI dose was able to double RAIU in patients with MNG. Subsequently, it was shown that two 0.1 mg doses of rhTSH administered on two successive days increased the mean RAIU 4.5-fold, a result that closely parallels the reported effectiveness of a single 0.3 mg dose 11. Several observational and prospective controlled studies in the last 4 years have demonstrated that rhTSH combined with RAI treatment can reduce TV by up to 57% 111213141516171819.

Current literature data seem to indicate that using rhTSH as an adjuvant to RAI therapy is more effective than using RAI alone, when the same radioiodine activities are compared. When used for the purpose of calculating the activity of the RAI dose administered, rhTSH allows lower amounts of radiation to be used. However, the optimal rhTSH dose and the minimum effective RAI activity needed to treat non-toxic MNG remain to be established. In addition, the absence of a correlation between TV reduction and post-rhTSH RAIU, area under the curve of TSH after rhTSH, baseline TV or RAI activity actually exerted suggests that goiter reduction may depend on other factors related to rhTSH pre-stimulation, and not only on the RAI activity applied 7. Finally, it should be borne in mind that treatment with rhTSH and RAI has many undesirable consequences, such as transient hyperthyroidism, thyroid pain and acute goiter swelling 6.

In Italy, the maximum RAI activity allowed for therapy in out-patients is 600 MBq. In a previous study 16 we assessed 20 elderly patients with large goiters and compared treatment with RAI activities (370–555 MBq, fixed activity) following two consecutive 0.2 mg doses of rhTSH with treatment with ¹³¹I alone. In comparison with patients who were treated with ¹³¹I alone, those who received rhTSH had higher transient (lasting 2 weeks) elevation of thyroid hormones, a greater reduction in TV, and a similar incidence of hypothyroidism about 2 years after therapy 16.

In the present study, we report the long-term outcome of a larger group of elderly out-patients with large MNG treated with weighted ¹³¹I activities up to 600 MBq after pre-treatment with two consecutive 0.1 mg rhTSH doses. To assess safety, TV reduction and the duration of thyroid function, the results were compared with those observed in a group of out-patients matched for age and TV in whom MNG was conventionally treated with radioiodine up to 600 MBq.

While surgery remains the gold standard in the treatment of MNG, several recent studies have demonstrated that RAI therapy with adjuvant rhTSH pre-treatment can be an attractive alternative 111213141516171819. The prevalence of MNG rises in the elderly, who also often suffer from comorbities 23. Although many patients enrolled in the several observational, prospective and controlled studies were more than 60 years old, few studies involving this new therapeutic protocol have so far focused on elderly subjects 1516.

In the present study we prospectively evaluated 40 outpatients with a mean age of 71 years. Within the limits of our protocol (the patients were consecutively recruited without randomization, and rhTSH administration was not blinded/placebo-controlled) the results obtained documented a greater reduction in TV when rhTSH treatment preceded RAI therapy than when similar RAI activities were applied alone. A significant, effective and permanent TV reduction is the main outcome of RAI therapy for MNG in the elderly. In our previous study 16, fixed doses of RAI (370 to 555 MBq) alleviated symptoms in all but one patient following rhTSH administration (0.2 mg on two consecutive days), and yielded a mean 50% reduction in TV after about 2 years. In the present study, TV reduction was greater in the rhTSH-treated group 6–12 months after RAI therapy; thereafter, a slow decline in TV was maintained. On long-term follow-up (median 36 months) evaluation, a 62% reduction was found, which was 15% greater than after RAI alone. So far, rhTSH-pre-treated patients have not required any further treatment to reduce TV, while some patients treated with RAI alone have required further RAI therapy or surgery.

It should be noted that the sample of patients we enrolled comprised both patients with euthyroid MNG and patients with MNG and subclinical hyperthyroidism, as in other series 1114161819). More patients in group 2 (no rhTSH pretreatment) had subclinical hyperthyroidism; consequently, more group 2 patients were treated with low-dose MMI before RAI. This constitutes a potential bias of the study. However, in our opinion, the results of RAI are fairly comparable between the 2 groups, bearing in mind that baseline RAIU and TV were very similar.

It is not easy to compare the data from the various studies, owing to differences in sample populations and therapeutic protocols (different amounts of rhTSH from 0.03 to 0.45 mg; different RAI doses from 370 to 5550). However, Cohen et al. 14 reported a 30% reduction in TV after a 6-month follow-up, which is roughly comparable to that found in our patients. We observed a 50% reduction in TV after 12 months, which is similar to that seen in other studies 11141819.

It is not clear how long rhTSH and RAI exert their reducing effect on TV. In a follow-up study lasting up to 12 months, Paz-Filho et al. 24 reported an average 50% reduction in TV after RAI (1110 MBq) in patients with an initial TV of 100 ml pre-treated with rhTSH (0.1 or 0.2 mg); most (about 70%) of this reduction occurred in the 1^st year after treatment. On the other hand, Cardia et al. 25 reported data from a 3-year follow-up of patients involved in a previously published study 14. At the last evaluation of available patients, ongoing TV reduction was observed from the 1^st to the 4^th year after RAI therapy; moreover, the initial difference in TV reduction in favor of rhTSH-pre-treated patients persisted (on average, 70% with rhTSH plus RAI vs 60% with RAI alone 25. The trend in our data suggests that a similar percentage reduction in TV can be achieved by using lower RAI activities (up to 600 MBq instead of up to 5550 MBq). On the whole, we think that the expected degree of final TV reduction after rhTSH plus RAI is in any case greater than that observed in long-term follow-up studies after RAI alone 345.

Forecasting TV reduction is an unresolved issue. We did not find any correlation between TV reduction and age, BMI, initial TV, thyroid function, baseline RAIU or the RAI dose administered. Similar results have been reported by Paz-Filho et al. 19. According to Bonnema et al. 18, TV reduction may be related to initial TV if the goiter is very large, which is the most frequent situation in elderly patients eligible for rhTSH plus RAI therapy. However, they found that the reduction in the volume of the goiter after rhTSH and RAI was unrelated to the thyroid dose retained, an observation in line with other studies conducted on patients with smaller goiters 17. Our data, which were obtained at low RAI activity, suggest that the effect of rhTSH depends on factors independent of thyroid irradiation. It is plausible that rhTSH might induce the reactivation of dormant thyroid tissue, or render thyrocytes more vulnerable to ionizing radiation 18. Indeed, the existence of such preconditioning by TSH has been suggested in studies of hyperthyroid patients treated with RAI 26.

Hypothyroidism and the need for L-T4 treatment might be seen in our protocol as an unwelcome effect of rhTSH pre-treatment. During follow-up, we observed hypothyroidism in 63% of rhTSH-pre-treated patients, while L-T4 treatment was necessary in only 14% of patients treated with RAI alone. This difference could be explained by a difference in functional status between the two groups on enrolment (more thyroid hyperfunction in the group treated with RAI alone), while a different autoimmune response is not proven. Moreover, the efficacy of RAI treatment in MNG is usually hampered by the low and irregular RAIU 1, and it has been shown that rhTSH increases RAIU (9, 13), thereby enhancing the adsorbed activity 27.

In subjects (mean age 63 years) with large MNG (59% toxic MNG) treated with high rhTSH (0.45 mg) and RAI (up to 5550 MBq) dosages, Silva et al. 14 observed hypothyroidism in 35% of subjects one year after therapy, as against only 21% (70% toxic MNG) of subjects treated with RAI alone. In another study, Cohen et al. 15 investigated 17 patients (aged >60 years) with a large percentage of toxic MNG (70% of cases); six months after very low rhTSH doses (0.03 mg) administered 24-h before 1100 MBq of RAI, hyperthyroidism was seen to have disappeared in 82% of cases and hypothyroidism in only 18%. In a study by Paz-Filho et al. 19 of 17 subjects (aged >60 years) with normal (n = 12) or increased (n = 5) thyroid function on enrolment, 53% of subjects displayed hypothyroidism 1–2 years after 0.1 rhTSH pre-treatment and 1100 MBq RAI therapy. More recently, Paz-Filho et al. 24 reported data on two cohorts of patients pre-treated (0.2 mg) or not with rhTSH before RAI (1100 MBq); on 2-year follow-up examination, they found a higher prevalence of hypothyroidism in the pretreated (73%) than in the non-pretreated group (53%).

Rubio et al. 28 have claimed that TPOAb positivity may occur after RAI, regardless of rhTSH pretreatment, and Paz-Filho et al. 19 have suggested that TPOAb positivity and the presence of smaller goiters may increase the risk of developing hypothyroidism. Our data do not support these hypotheses, since only a few new cases of autoantibody positivity appeared after RAI. On the whole, the available data seem to indicate that early hypothyroidism after rhTSH pretreatment is more frequent than after RAI alone, irrespective of rhTSH dosages and RAI activities. However, the possibility cannot be ruled out that the difference observed is due, at least in part, to differences in baseline thyroid function between the two groups of patients.

Elderly patients with large MNG may have underlying cardiovascular disease, and rhTSH and RAI administration may engender a risk of cardiac complications, owing to the increase in thyroid hormones. In a recent report by Barca et al. 29 on patients aged 42 – 80 years who were treated with a fixed 1100 MBq activity of RAI after 0.1 mg of rhTSH, a 3-fold increase in left ventricular diastolic dysfunction was seen in subjects with laboratory evidence of hyperthyroidism. This 29, as well as other previous studies 11, indicates that beta-blockers should be routinely administered to reduce thyrotoxic symptoms when rhTSH and RAI therapy are administered. In our protocol, at least one daily 25 mg dose of atenolol was administered by default; only one patient from the rhTSH and RAI group suffered a severe, albeit transient, cardiovascular adverse event, while in the RAI group an increase in pre-existing dyspnea was sometimes reported. This indicates that, in selected elderly subjects, the adjuvant administration of rhTSH does not worsen RAI-induced cardiovascular adverse events. On the other hand, in our patients the simultaneous administration of anti-inflammatory steroids resulted in less symptomatic swelling of thyroid tissue, a condition which has been reported by 10% to 30% of patients after rhTSH and RAI 18.

On the basis of the present and previous studies, rhTSH pre-stimulation can improve the goiter reduction achieved by RAI therapy. Administering small doses of RAI and rhTSH can limit side-effects without compromising the goiter-reducing effect, while improving MNG-related symptoms in elderly subjects. RhTSH-induced ^131-I uptake in less active thyroid areas of the goiter, and perhaps rhTSH-induced mechanisms other than the increase in thyroid uptake, could explain the better results achieved by using rhTSH as an adjunct to RAI therapy.

The authors declare no conflict of interest. This study was partially supported by a grant from the University of Genoa (Finanziamento Progetti di Ateneo 2007).

MG conceived the study, performed the statistical analysis and drafted the manuscript. MC participated in the design of the study. IC participated in the design of the study. MB carried out the instrumental evaluations. EC carried out the assays. MS participated in the design of the study. MM participated in the coordination of the laboratory data. DF participated in the sequence alignment. FM participated in the conception of the study. MB conceived the study and drafted the manuscript. All authors have read and approved the final manuscript.