Evaluation of endothelial function in subclinical thyroid dysfunction.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to F Monzani.
Explore the source record for details and available documents.
BACKGROUND: Conflicting data have been reported on the association between interferon (IFN)-beta therapy of multiple sclerosis (MS) patients and thyroid disease development. AIMS: The goals of this study are as follows: to assess the actual occurrence of thyroid dysfunction and autoimmunity during long-term IFN-beta therapy; to establish the possible presence of predictive factors for thyroid dysfunction development and duration; and to suggest an effective follow-up protocol for patients receiving long-term IFN-beta therapy. STUDY PROTOCOL: A total of 106 MS patients (76 women) underwent IFN-beta 1a or 1b therapy for up to 84 months (median, 42 months). Thyroid function and autoimmunity were assessed at baseline and every 3-6 months throughout the treatment course. RESULTS: Baseline thyroid autoimmunity was detected in 8.5% of patients and hypothyroidism in 2.8%. Thyroid dysfunction (80% hypothyroidism, 92% subclinical, 56% transient) developed in 24% (68% with autoimmunity) of patients and autoimmunity in 22.7% (45.5% with dysfunction), without significant differences between the two cytokines; 68% of dysfunctions occurred within the first year. Autoimmunity emerged as the only predictive factor for dysfunction development (relative risk, 8.9), whereas sustained disease was significantly associated with male gender (P < 0.003). CONCLUSIONS: Both incident thyroid autoimmunity and dysfunction frequently occur in MS patients during IFN-beta therapy, particularly within the first year of treatment. Thyroid dysfunction is generally subclinical and transient in over than half of cases; preexisting or incident autoimmunity emerged as the only significant predictive factor for thyroid dysfunction development. Thyroid function and autoimmunity assessment is mandatory within the first year of IFN-beta therapy; thereafter, serum TSH measurement only in patients with thyroid disease could be sufficient.
We evaluated in primary human thyrocyte cultures the effect of interferon (IFN)-alpha and -beta on the expression of thyroid peroxidase (TPO), sodium/iodide symporter (NIS), and thyroglobulin (Tg) as well as T(4) release. Human thyrocyte cultures were carried out with fresh normal thyroid tissue. Gene and protein expression of Tg, TPO, and NIS were assessed by RT-PCR and Western blot analysis after 24, 48, and 72 h of treatment with TSH alone (10 mIU/ml) and in combination with IFN alpha or -beta (10(4) U/ml). IFN inhibited the TSH-stimulated gene expression of Tg, TPO, and NIS in a time-dependent manner without significant differences between IFN alpha and -beta. Moreover, the addition of both type I IFNs clearly reduced the TSH-stimulated protein expression of Tg, TPO, and NIS after 72 h of exposure. Finally, this down-regulation was associated with a reduction of T(4) release by almost 50%. In conclusion, our study shows that both IFN alpha and -beta down-regulate the TSH-stimulated expression of Tg, TPO, and NIS as well as T(4) release. Indeed, the development of hypothyroidism during type I IFN therapy may be related, at least in part, to an abnormal expression and function of key proteins involved in iodine uptake and organification.
Type I interferons are currently used for the treatment of chronic viral hepatitis, multiple sclerosis and several hematological and solid tumors. Side effects are not uncommon, and include multiple alterations in thyroid function, some of which are unrelated to autoimmunity. Review of the literature revealed an overall mean prevalence of incident thyroid dysfunction of 6.2%, hypothyroidism occurring more frequently (3.9%) than hyperthyroidism (2.3%). Destructive thyroiditis characterized by early transient thyrotoxicosis followed by hypothyroidism has also been described. Thyroid dysfunction was mainly subclinical, and spontaneous resolution occurred in almost 60% of patients with or without withdrawal of interferon. Risk factors for developing thyroid abnormalities were female sex and the presence of pre-existing autoimmune thyroiditis. Whether prolonged interferon therapy will increase the likelihood of experiencing thyroid dysfunction, as well as the relationship between incident thyroid autoimmunity and the efficacy of interferon therapy, are still open questions. Although the most-likely explanation for thyroid disease occurring with type I interferon therapy remains an autoimmune reaction or immune system dysregulation, a direct inhibitory effect on thyrocytes may be presumed in patients who developed hypothyroidism without autoimmunity. However, the mechanisms of thyroid damage induced by type I interferons have not yet been clarified in detail. We recommend routine evaluation of serum thyroid-stimulating hormone during interferon therapy. A systematic thyroid assessment is useful only for those patients with pre-existing thyroiditis or incident dysfunction. Although discontinuation of interferon therapy is seldom required, it may be necessary in patients who develop Graves' disease and overt hyperthyroidism.
PURPOSE: To evaluate genetic damage and oxidative stress following a single therapeutic dose of 131I in Graves' disease patients monitored up to 180 days after treatment. MATERIALS AND METHODS: Genetic damage induction was estimated as the increase in micronuclei in peripheral lymphocytes of patients. As indicators of radiogenic oxidative stress, vitamin E and lipoperoxide levels were assessed in the plasma of patients, as well as the release of plasmic clastogenic factors measured by the induction of micronuclei in vitro in peripheral lymphocytes of a healthy donor. RESULTS: Vitamin E depletion lasted at least 3 days and the basal level was restored within 7 days. No statistically significant variations were observed in lipoperoxide plasma levels. A sharp increase of micronuclei in the peripheral lymphocytes of patients was correlated (p < 0.001) with the release of clastogenic factor in the plasma. The highest micronucleus value was negatively correlated (p < 0.03) with the lowest vitamin E level observed in each patient. CONCLUSIONS: Micronuclei induction was the direct consequence not only of the energy deposition of 131I on the genetic material, but also of oxidative stress, likely via the release of clastogenic factor.
Subclinical hypothyroidism (sHT) is associated with dyslipidemia and enhanced cardiovascular risk. We assessed carotid artery intima-media thickness (IMT, high-resolution ultrasonography) and lipoprotein profile in 45 sHT patients (aged 37 +/- 11 yr) at baseline and after 6 months of randomized, placebo-controlled L-T(4) replacement. In comparison with 32 age- and sex-matched controls, sHT patients had elevated total and low-density lipoprotein (LDL) cholesterol and ApoB levels (P = 0.002, P = 0.0007, and P = 0.01, respectively) and higher mean-IMT values (P < 0.0001). In stepwise regression analysis, mean-IMT was positively related (r(2) = 0.71, P < 0.0001) to age, TSH, and LDL cholesterol. L-T(4) replacement significantly reduced both total and LDL cholesterol (P < 0.0001 for both) and mean-IMT (by 11%, P < 0.0001). The decrement in IMT was directly related to the decrements of both total cholesterol and TSH (P = 0.02 and P = 0.0001, respectively). We conclude that early carotid artery wall alterations are present in sHT patients. Whether such IMT increase is related to an early atherosclerotic involvement of the arterial wall cannot be clearly decided on the basis of the present results. However, the fact that L-T(4) replacement therapy was able to improve both the atherogenic lipoprotein profile and intima-media thickening suggests that lipid infiltration of arterial wall may represent a major mechanism underlying IMT increase in subclinical hypothyroidism.
Laparoscopy and laparoscopic ultrasonography (LUS) have been proposed for the diagnosis and treatment of pancreatic insulinoma. We present for cases of pancreatic insulinoma approached by laparoscopy guided by LUS. In three cases, insulinomas were in the pancreatic body and in one case in the pancreatic head. All lesions were detected preoperatively by abdominal US and confirmed by computed tomography. Laparoscopy was performed under general anesthesia. LUS was performed using a 10-mm flexible probe. In two cases the adenoma was enucleated using scissors and electrocoagulation, major vessels were controlled using clips, and enucleation was completed using a 30-mm endo-GIA. In one case a laparoscopic distal pancreatectomy with spleen preservation was performed. In one case the adenoma was deep in the pancreatic head; minilaparotomy was performed and the adenoma enucleated. Patients were discharged in good health 5-7 days after surgery. The postoperative course was complicated in one case of enucleation by peripancreatic fluid collection that was treated percutaneously. Our experience confirms that accurate localization followed by excision of tumors via the laparoscopic approach constitute a significant advance in the management of insulinoma.
Subclinical hypothyroidism (sHT) affects 5-15% of the general population; however, the need of lifelong L-T(4) therapy is still controversial. As myocardium is a main target of thyroid hormone action, we investigated whether sHT induces cardiovascular alterations. Twenty sHT patients were randomly assigned to receive placebo or L-T(4) therapy and were followed for 1 yr. Twenty sex- and age-matched normal subjects served as controls. Doppler echocardiography and videodensitometric analysis were performed in all subjects. Myocardium textural parameters were obtained as mean gray levels, which were then used to calculate the cyclic variation index (CVI; percent systolic/diastolic change in mean gray levels). Patients had a significantly higher isovolumic relaxation time (3.1 +/- 0.5 vs. 2.6 +/- 0.6; P < 0.03), peak A (0.77 +/- 0.16 vs. 0.56 +/- 0.13 m/s; P < 0.01), and preejection/ejection time (PEP/ET) ratio (0.72 +/- 0.05 vs. 0.57 +/- 0.06; P < 0.03) and a lower CVI (P < 0.0001) than controls. CVI was inversely related to TSH level (P < 0.0001) and PEP/ET ratio (P < 0.01). L-T(4)-treated patients showed a significant reduction of the PEP/ET ratio (P < 0.05), peak A (P < 0.05), and isovolumic relaxation time (P < 0.05) along with a normalization of CVI. Conversely, no changes were observed in the placebo-treated group. In conclusion, sHT affects both myocardial structure and contractility. These alterations may be reversed by L-T(4) therapy.
BACKGROUND: The authors previously reported on the development of thyroid dysfunction and autoimmunity during 1-year treatment of patients with MS with interferon-beta 1b (IFN beta-1b). OBJECTIVE: To evaluate the evolution of incident thyroid disease and the possible development of more thyroid disease during longer term therapy. PATIENTS: The authors studied 31 patients (aged 34 +/- 7 years; 21 women) with relapsing-remitting MS during 3 years of IFN beta-1b treatment. Systematic thyroid assessment was performed every 3 or 6 months, depending on the development of thyroid disease. RESULTS: After the first year of IFN beta-1b treatment, no further cases of thyroid disease were observed. Among the six patients with early incident subclinical hypothyroidism, thyroid dysfunction persisted only in those with baseline autoimmune thyroiditis (n = 2). The three patients who developed transient hyperthyroidism remained euthyroid throughout the treatment course. A positive autoantibody titer was continually detected in only two out of five patients without baseline autoimmunity. CONCLUSIONS: The risk of thyroid disease seems related to IFN beta-1b treatment during the first year only, particularly in patients with preexisting thyroiditis. Furthermore, incident thyroid dysfunction is generally transient and mild in degree. Indeed, we recommend a routine systematic thyroid assessment only in patients with baseline thyroiditis. During the first year of therapy, serum thyroid-stimulating hormone measurement should suffice as first line test; a systematic thyroid assessment is only useful for those patients with incidental and persistent dysfunction. Further studies with many patients will be necessary to confirm our suggestions as broad clinical guidelines.
BACKGROUND: In subclinical hypothyroidism (sHT), a condition in which impaired hormone synthesis is compensated by thyroid-stimulating hormone (TSH) hypersecretion, previous studies have suggested the presence of disturbances in left ventricular (LV) function. OBJECTIVES: Our goal was to investigate LV structure and function through the combined use of conventional Doppler echocardiography and ultrasonic videodensitometry. METHODS: We studied 16 patients with sHT (aged 32+/-12 [mean +/- SD] years) who had raised TSH levels (> 3.6 mIU/L) but normal levels of free thyroid hormones (free thyroxine [FT(4)] and free triiodothyro-nine [FT(3)]), and 16 carefully age- and sex-matched euthyroid subjects. Transmitral flow Doppler analysis and quantitative analysis of the echocardiographic digitized images were performed in all study subjects. Textural parameters of the septum and posterior wall were obtained as mean gray levels, which were then used to calculate the cyclic variation index (CVI), that is, the percent change in mean gray levels between diastole and systole. RESULTS: Patients with sHT had a significantly higher LV mass index (92 +/- 16 versus 76 +/- 16 g.m(2), P<.01) and isovolumic relaxation time corrected for heart rate (IVRTc) (2.9 +/- 0.6 versus 2.5 +/- 0.6, P<.04) than did controls. On videodensitometry, patients had lower CVIs both for the septum (-5% +/- 22% versus 33% +/- 9%, P<.0001) and the posterior wall (10% +/- 26% versus 49% +/- 18%, P<.0001). IVRTc discriminated only 25% of the patients from the controls, whereas CVI analysis correctly identified 85% of the patients with sHT (P<.002). Furthermore, CVI values were found to be significantly related to serum FT(4) and FT(3) concentrations in a direct fashion, and to serum TSH levels in an inverse fashion. CONCLUSIONS: Subclinical hypothyroidism is associated with changes in videodensitometric myocardial structure. These changes, which are not accurately detected by conventional or Doppler echocardiography, are quantitatively related to loss of thyroid function and could represent an early sign of myocardial damage in hypothyroidism.
Few data exist on the operative and pathological findings in patients having undergone previous percutaneous ethanol injection (PEI) therapy of thyroid nodules. We report here our experience with 13 patients operated on by the same surgical team. Two pathologists, both blinded to the previous PEI treatment, carried out histological evaluation. Reasons for surgery included PEI failure, suspicion of malignancy, and tracheal compression. The operations did not pose any special problem from the technical point of view, and the postoperative courses were uneventful. One patient who had a second operation developed hypoparathyroidism, and laryngeal nerve palsy was never observed. The histological diagnosis was hyperplastic or adenomatous nodule in 12 cases and papillary thyroid cancer in 1. No difficulty was found in evaluating the nodule capsule and surrounding vessels. In two lesions, nuclear enlargement and clearing were identified in thyroid follicles immediately adjacent to necrotic or scarred areas. These changes were considered reactive. In conclusion, patients previously treated by PEI were operated on without special technical problems. Histological diagnosis was not hindered, and there was no difficulty in ruling out malignant lesions. PEI, however, should be performed only by skilled operators, and incidental ethanol seepage throughout the nodule capsule must be carefully avoided.
DESIGN, PATIENTS AND MEASUREMENTS: The presence of neuromuscular symptoms was ascertained by questionnaire in 33 consecutive patients with subclinical hypothyroidism (sHT) as compared to 44 age- and sex-matched controls. Blood was sampled for PTH, magnesium, phosphate, and total and ionized calcium determination. Patients reporting three or more symptoms were also studied by surface electromyography (sEMG). The study was repeated following a six-month L-T4 course. RESULTS: Neuromuscular symptoms were significantly more frequent in patients than in controls (P = 0. 0001), and correlated with TSH values (r = 0.52; P = 0.0001). Among patients showing three or more symptoms (n = 11), sEMG documented the presence of repetitive discharges in 8 patients. L-T4 therapy led to a significant improvement of symptoms (P = 0.0001); persistent repetitive discharges were no longer observed. Total and ionized calcium values, always within the normal limits, were significantly lower in patients than controls (P < 0.0001). An inverse relationship was observed between ionized calcium and: TSH values (r = -0.69, P = 0.0001); the number of neuromuscular symptoms (r = -0.53, P = 0.0001). L-T4 replacement induced a significant increase in both total and ionized calcium levels (P < 0.01 and P < 0.0001, respectively). CONCLUSIONS: Neuromuscular symptoms and dysfunction are rather common in subclinical hypothyroidism, and may be associated with abnormalities in serum calcium balance and surface electromyography. The ability of L-T4 treatment to reverse all these changes suggests that subclinical hypothyroidism patients may require early therapy not only to prevent progression to frank hypothyroidism, but also to improve their neuromuscular dysfunction.
Recent research has greatly expanded the domain of insulin action. The classical action of insulin is the control of glucose metabolism through the dual feedback loop linking plasma insulin with plasma glucose concentrations. This canon has been revised to incorporate the impact of insulin resistance or insulin deficiency, both of which alter glucose homeostasis through maladaptive responses (namely, chronic hyperinsulinaemia and glucose toxicity). A large body of knowledge is available on the physiology, cellular biology and molecular genetics of insulin action on glucose production and uptake. More recently, a number of newer actions of insulin have been delineated from in vitro and in vivo studies. In sensitive individuals, insulin inhibits lipolysis and platelet aggregation. In the presence of insulin resistance, dyslipidaemia, hyper-aggregation and anti-fibrinolysis may create a pro-thrombotic milieu. Preliminary evidence indicates that hyperinsulinaemia per se may be pro-oxidant both in vitro and in vivo. Insulin plays a role in mediating diet-induced thermogenesis, and insulin resistance may therefore be implicated in the defective thermogenesis of diabetes. In the kidney, insulin spares sodium and uric acid from excretion; in chronic hyperinsulinaemic states, these effects may contribute to high blood pressure and hyperuricaemia. Insulin hyperpolarises the plasma membranes of both excitable and non-excitable tissues, with consequences ranging from baroreceptor desensitisation to cardiac refractoriness (prolongation of QT interval). Under some circumstances insulin is vasodilatory-the mechanism involving both the sodium-potassium pump and intracellular calcium transients. Finally, by crossing the blood-brain barrier insulin exerts a host a central effects (sympatho-excitation, vagal withdrawal, stimulation of corticotropin releasing factor), collectively resembling a stress reaction. Description and understanding of these new roles, their interactions, the interplay between insulin resistance and hyperinsulinaemia, and their implications for cardiovascular disease have only begun.
OBJECTIVE: Interferon-beta (IFN-beta) is a widely used therapy for multiple sclerosis (MS), a demyelinating disease of the central nervous system. This study has evaluated the effect on thyroid function and autoimmunity of a 1-year treatment with IFN-beta1b in patients with MS. PATIENTS: We studied 31 patients (age 34+/-7 years, 21 women) with relapsing-remitting MS during IFN-beta1b treatment of 1 year duration. Systematic thyroid assessment and measurements of serum interleukin-6 (IL-6) levels were performed at baseline and every 3 months during treatment. RESULTS: Sixteen percent of the patients had autoimmune thyroiditis before IFN-beta1b, all positive for anti-peroxidase antibodies. The overall incidence of thyroid dysfunction was 33% over 1 year (10% hyperthyroidism, 23% hypothyroidism). Thyroid autoimmunity developed in 5/26 patients (19%), in one case without dysfunction. In addition to autoantibody positivity at baseline, female gender and the presence of an ultrasound thyroid pattern suggestive of thyroiditis were identified by multiple logistic regression as additional risk predictors for the development of thyroid dysfunction. During IFN-beta1b treatment, serum IL-6 levels rose in a consistent biphasic pattern; there was, however, no difference between patients with or without incident thyroid abnormalities. CONCLUSIONS: We conclude that IFN-beta1b therapy can induce multiple alterations in thyroid function, some of which are unrelated to thyroid autoimmunity. IL-6 measurement is not useful to identify patients prone to develop thyroid abnormalities. Though thyroid dysfunction is generally subclinical and often transient, systematic thyroid assessment should be performed during IFN-beta1b treatment.
The aim of our study was to define the long-term efficacy and safety of percutaneous ethanol injection (PEI) for the treatment of autonomous thyroid nodule (ATN), and to optimise the clinical usefulness of such a therapy. We treated 132 patients with ATN (30 M and 102 F, aged 47.5+/-12.9 years; mean+/-SD), in case other established treatments were refused or contraindicated. Eighty-five patients were affected by toxic adenoma and 47 suffered from pre-toxic nodules. Ethanol was administered weekly under sonographic control, in 7 sessions (range 2-16). During PEI treatment, 26 toxic elderly patients were treated with methimazole and propranolol. Three possible outcomes were identified for statistical analysis: failure (persistent suppression of extra nodular tissue uptake, along with elevated free thyroid hormone and undetectable TSH levels); partial cure (normal free thyroid hormone and low/undetectable TSH levels); complete cure (normal thyroid hormone and TSH levels; restored extra nodular uptake). The patients were followed for up to 8.5 years (median 76 months). PEI therapy was well tolerated by all patients though a mild to moderate local pain occurred in about 30% of sessions. Complete cure was achieved in all pre-toxic patients and in 60 (70.6%) patients with toxic adenoma, while partial cure was observed in 11 cases (12.9%) and failure in 14 (16.5%). A significant shrinkage of nodule volume was observed in all patients (p = 0.0001), while those with toxic nodules larger than 30 mL showed a significantly lower response rate to PEI (p < 0.05). At controls, only one patient developed subclinical hypothyroidism while, among partially cured patients, five relapsed. The administration of methimazole and/or propranolol did not modify PEI outcome. In conclusion, we suggest that PEI therapy may be the treatment of choice in patients with pre-toxic thyroid adenoma where therapy is least necessary- despite the nodule volume. Though ethanol injection therapy of toxic thyroid nodules may be troublesome for the need of multiple sessions, it appears an effective alternative procedure in patients at poor surgical risk, and in younger patients in whom radioiodine is contraindicated. Since a special technical skill in intervention procedures is required, PEI therapy may be suitable only for patients living nearby a trained centre.
Some concern has been expressed about surgical operations in thyroid nodules previously treated by ethanol injection: the reasons are mainly represented both by the possibility of more surgical risks due to the adhesions caused by ethanol and the difficulty of interpreting the histological pattern when a cancer is suspected. During the last 8 years 219 patients underwent ethanol injection: among these 6 were subsequently submitted to surgical operation. Total thyroidectomy was performed in 5 cases and isthmusectomy in one case: No vocal cord palsy was registered in these patients.. They were all normocalcemic after surgery. Histology showed no significant fibrosis in the ethanol treated nodules but only macrofollicles and this did not affect the histological examination in presence of suspected malignancies.
In this report we describe an unusual patient with hyperfunctioning thyroid adenoma in whom percutaneous ethanol injection (p.e.i.) therapy was followed by typical Graves' disease. His history revealed the presence of a sister with Hashimoto's thyroiditis. 99-mTc thyroid scintiscan showed focal uptake in the nodule, with suppression of extranodular parenchyma. P.e.i. therapy was followed by the development of severe hyperthyroidism. One month after a second p.e.i. cycle, recurrence of hyperthyroidism associated with diffuse 99-mTc uptake by the gland was observed. TSH-receptor and thyroglobulin autoantibodies were undetectable before p.e.i. therapy, appeared during the first cycle, and showed a further increase after the second p.e.i. therapy cycle. Though spontaneous switch to Graves' disease cannot be excluded in patients with toxic nodules, the massive release of thyroid materials from follicular cells, among these TSH-receptor antigenic components partially denatured by ethanol, may indeed trigger an autoimmune response to the TSH-receptor, thus accounting for this observation. Patients with possible autoimmune disposition, as selected by familiar history and/or laboratory markers should be carefully monitored during p.e.i. treatment.
UNLABELLED: Percutaneous ethanol injection (PEI) has been suggested as an alternative to radioiodine and surgery for the treatment of autonomous thyroid nodules (ATN). OBJECTIVE: In this study we have defined the long-term efficacy and safety of PEI for the treatment of ATN, and we have attempted to optimize the clinical usefulness and improve the technical approach to PEI treatment. PATIENTS: One hundred and seventeen patients with ATN, 26 males and 91 females, aged 48 +/- 12.9 years (mean +/- SD), were offered PEI when other established treatments were refused or contraindicated. Seventy-seven patients were affected by toxic adenoma (60 with a single nodule, 17 with a multinodular goitre); 40 patients suffered from a pretoxic single nodule. METHODS: Sterile 95% ethanol was administered weekly under sonographic control by a 20-22 gauge needle without anaesthesia or pharmacological sedation. During PEI treatment, 26 toxic elderly patients were treated with methimazole and propranolol. According to hormone and scintigraphic data, three possible outcomes were identified for statistical analysis: failure (persistent suppression of extra-nodular tissue uptake, along with elevated free T4 (FT4) and free T3 (FT3) and undetectable TSH levels); partial cure (normalization of FT4 and FT3 levels, with low/ undetectable TSH levels; persistent suppression of extra-nodular uptake); complete cure (normal thyroid hormone and TSH levels; restored extra-nodular uptake). RESULTS: The patients were followed for up to 5 years (median 2.5). PEI therapy was well tolerated by all patients. Complete cure was achieved in all pretoxic patients and in 60 (77.9%) patients with toxic adenoma, while partial cure was observed in 7 cases (9.1%) and failure in 10 (13%). PEI treatment proved similarly effective in toxic patients with a single nodule or with multinodular goitre (87 vs 88.2%, respectively). At the end of treatment, a significant shrinkage of nodule volume was observed in all patients (P = 0.0001). Toxic patients with pretreatment volume > 40 ml (n = 8) did not show a significant difference in treatment response rate as compared to those with volume < 40 ml. Recurrence of hyperthyroidism was never observed during follow-up, independently of thyroid status before treatment. Only one patient with significant thyroid autoantibody serum levels before PEI treatment, developed sub-clinical hypothyroidism at 3 years. The administration of methimazole and/or propranolol did not modify PEI outcome. CONCLUSION: Our data confirm the efficacy and safety of percutaneous ethanol injection for the therapy of autonomous thyroid nodules. The very low incidence of hypothyroidism along with the absence of recurrence of hyperthyroidism suggests that percutaneous ethanol injection is the treatment of choice in patients with pretoxic thyroid adenoma. Percutaneous ethanol injection appears an effective alternative procedure in toxic patients with a high surgical risk even if they have large nodules, and in younger ones in whom radioiodine is contraindicated. Patients may be submitted to anti-thyroid drug and/or beta-blocker therapy if it is necessary, but this does not affect percutaneous ethanol injection treatment outcome. Finally, not only single autonomous thyroid nodules but also toxic multinodular goitre may be successfully treated by percutaneous ethanol injection.