Thyroid cancers. I. Papillary, follicular, and Hürthle cell.

The well differentiated thyroid carcinomas, papillary, follicular, and Hürthle cell, comprise the majority of malignant thyroid diseases. This article presents guidelines for the evaluation and treatment of these malignancies. Prognostic factors, surgical treatment of the thyroid and regional lymph nodes, postoperative detection of recurrence, and treatment of distant metastases are discussed.

Management of the recurrent laryngeal nerve in suspected and proven thyroid cancer.

Vocal cord paralysis occurs with and without infiltration of the recurrent laryngeal nerve. Patients with no paralysis may have recurrent laryngeal nerve infiltration. We studied 262 patients with invasive thyroid carcinoma and an additional 480 patients that we operated on, and we offer an approach to the management of the recurrent laryngeal nerve during surgery for suspected or proven thyroid cancer. Because complete excision of papillary carcinoma with resection of the recurrent laryngeal nerve did not improve survival over incomplete excision, we recommend incomplete excision of papillary carcinoma infiltrating a functioning recurrent laryngeal nerve with postoperative I-131 and thyroid-stimulating hormone suppression. We observed (1) lymphoma and Reidel's thyroiditis causing paralysis with infiltration of the recurrent laryngeal nerve with recovery of function, (2) benign and malignant nodules causing paralysis without infiltration of the recurrent laryngeal nerve with recovery of function, and (3) infiltration of the recurrent laryngeal nerve without paralysis in lymphoma, Graves' disease, and thyroiditis. In surgery for a suspicious nodule with paralysis, identify the recurrent laryngeal nerve, resect it if it is infiltrated by carcinoma, and preserve it if not infiltrated or if infiltrated by any pathology besides carcinoma.

Sequential agonist activation and site-specific mediation of acute cyclosporine constriction in rat renal arterioles.

Evidence suggests that acute and chronic cyclosporine (CsA) nephropathy may be related to its renal vasoconstrictor effects. While the mechanism of CsA-induced renal vasoconstriction is uncertain, several studies indicate that endogenous constrictor agonists including endothelins (ET), platelet activating factor (PAF), and thromboxane A2 (TXA2) play a mediating role. In this study, two possible mechanisms explaining the participation of multiple constrictor agonists in CsA vasoconstriction were investigated: sequential activation of agonists initiated by CsA and site-specific mediation of CsA constriction by different agonists. The acute constrictor effects of CsA were examined in isolated rat renal afferent (AA) and efferent arterioles (EA) without and with specific receptor antagonists of ETA (BQ123, 10(-7) M), PAF (L-659,989, 10(-7) M), and TXA2/PGH2 (SQ29,548, 10(-7) M) in the bathing media. Both BQ123 and L-659,989 completely inhibited CsA, constriction in AA, but had no significant inhibiting effect in EA. Constriction to ET-1 was also blocked by the PAF antagonist L-659,989 in AA, but not EA. There was no effect of SQ29,548 on CsA constriction in AA--however, there was partial attenuation of CsA constriction in EA. Based on these results in isolated rat renal arterioles, it is suggested that CsA-induced constriction in AA is likely mediated by sequential activation of ET and PAF. However, CsA constriction of EA involves a different mechanism or mediator that, in part, may involve TXA2/PGH2 stimulation.

Angiotensin II-induced changes in smooth muscle calcium in rat renal arterioles.

The isolated rat renal arteriole technique was adapted for use in a fluorescence ratio imaging system in which angiotensin II (AII)-induced changes in lumen diameter and smooth muscle cell (SMC) cytosolic calcium [(Ca2+]i were serially determined for 4 min with Fura-2. Selective fluorescence acquisition from SMC was guided by direct image visualization of the vessel walls. A maximal constricting concentration of AII (10(-8) M) caused similar abrupt and sustained increases in SMC [Ca2+]i in afferent arterioles (AA) at 80 mm Hg and efferent arterioles (EA) at 30 mm Hg. When lumen pressure was reduced to 0 mm Hg, 10(-8) M AII caused abrupt peak increases in SMC [Ca2+]i in 15 s in both AA and EA, which declined rapidly thereafter--patterns distinctly different from pressurized vessels (P < 0.02). With diltiazem (10(-5) M) in the bathing media, 10(-8) M AII caused an abrupt rise and decline in SMC [Ca2+]i in AA, but a sustained elevation in EA (P < 0.02). In low-Ca(2+)-EGTA media, there was an abrupt peak and rapid decline in SMC [Ca2+]i to 10(-8) M AII in AA and EA; the abrupt peaks were attenuated by the prior addition of dantrolene (5 x 10(-5) M) to the low-Ca(2+)-EGTA media. When half-maximal constricting (EC50) AII for AA (4 x 10(-11) M) was added, there was a slow, progressive increase in SMC [Ca2+]i that was distinctly different from the abrupt peak and decline with EC50 AII (5 x 10(-12) M) in EA. Collectively, these findings indicate that maximal AII stimulates both Ca2+ entry and storage mobilization in AA and EA; EC50 AII stimulates primarily Ca2+ entry in AA, but storage mobilization in EA. Lumen pressure modifies the AII SMC [Ca2+]i response profiles.

Maintenance of endothelin-induced renal arteriolar constriction in rats is cyclooxygenase dependent.

Influence of arachidonate cyclooxygenase (COX) products on endothelin (ET)-evoked renal vasoconstriction was assessed. In microperfused rat afferent (AA) and efferent arterioles (EA), indomethacin had no effects on the maximal contraction of both AA and EA by ET, but reduced the duration of ET-induced constriction in both arterioles. ET infusion to rats in vivo resulted in a selective increase in efferent but not afferent arteriolar resistance, leading to a dramatic increase in transcapillary hydraulic pressure difference. Glomerular filtration rate (GFR), which fell progressively during infusion of ET alone, was markedly preserved by COX inhibition, but not during selective thromboxane A2 antagonism. In isolated glomeruli, release of prostaglandin (PG) F2 alpha in response to 10(-6) mol/l ET exceeded that the PGE2 by a ratio of 3.2. Collectively, these data provide strong evidence that locally released COX products, possibly PGF2 alpha, play a key role in sustaining ET-induced renal arteriolar constriction. COX inhibition leads to acute vasorelaxation of AA despite continued ET administration, without affecting EA constriction in vivo, thereby resulting in a dramatic reversal of the effects of ET on GFR.

KCl and angiotensin responses in isolated rat renal arterioles: effects of diltiazem and low-calcium medium.

Studies in intact renovascular models have shown that calcium entry blockers inhibit angiotensin (ANG II)-induced vasoconstriction in afferent (AA) but not efferent arterioles (EA), suggesting that increases in smooth muscle cell cytosolic calcium, the initiating intracellular message, result from entry through potential-operated channels in AA, but from organelle storage mobilization or entry through nonpotential-operated channels in EA. The present study examined the effects of diltiazem (10(-5) M) on the constrictor responses to KCl (50 mM) and half-maximal constricting concentrations (EC50) of ANG II and the effects of low-calcium bathing medium on EC50 ANG II responses in isolated rat AA and EA. KCl caused slightly greater decreases in lumen diameter in AA than in EA (P < 0.05) that were completely inhibited by diltiazem in both. Vasoconstriction to ANG II was significantly inhibited by diltiazem (29 +/- 12 vs. 67 +/- 31%; P < 0.02) in AA. However, constrictor response to ANG II in EA was unchanged by diltiazem (42 +/- 32 vs. 41 +/- 31%). Constriction to ANG II of AA in low-calcium medium was significantly attenuated (8 +/- 13 vs. 54 +/- 12%; P < 0.01); however, EA constrictor response was not affected (43 +/- 22 vs. 51 +/- 19%). These data indicate that EC50 ANG II-induced AA constriction requires calcium entry primarily through potential-operated channels. While potential-operated calcium entry channels can be functionally expressed in EA, intracellular calcium mobilization is the primary mechanism for ANG II-induced constriction.

Effects of atriopeptin III on isolated rat afferent and efferent arterioles.

The effects of atriopeptin III (AP III) on in vitro prepared afferent (AA) and efferent arterioles (EA) from rat kidneys were tested in a system in which lumen diameter could be measured. AP III (10(-13)-10(-7) M) had no effect on lumen diameter of AA that were not preconstricted. When AA were preconstricted with either angiotensin II (ANG II) or norepinephrine (NE), however, AP III increased lumen diameter in a concentration-dependent manner to the preconstriction baseline value. Maximal vasodilation occurred at 10(-10) M AP III. Unlike AA, EA constricted by 50% to 10(-10) M AP III further constricted EA that were pretreated with ANG II or NE. Dilation in ANG II-preconstricted AA to AP III was not inhibited by indomethacin. Constriction of EA to AP III was not altered by [Sar1-Ala8] ANG II, enalapril, OKY 046, or phentolamine. Results indicate that in isolated renal arterioles AP III dilates preconstricted AA but constricts EA that have either not been pretreated or have been preconstricted with other agonists. The effect of AP III on preconstricted AA does not require vasodilator prostaglandin mediation. The constrictor effect of AP III on EA is not dependent on angiotensin, thromboxane, or alpha-adrenergic mediation.

Atrial natriuretic peptide and dopamine in established acute renal failure in the rat.

Recent studies have shown that atrial natriuretic peptide (ANP) alone or combined with dopamine (D) markedly improved renal function when given immediately after an ischemic insult. However, the potential beneficial effect of ANP or ANP-D in the established phase of acute renal failure (ARF) and the duration of this effect have not been examined. In the present study atriopeptin III (APIII) and D, sufficient to maintain mean arterial pressure between 100 and 110 mm Hg, or normal saline were given i.v. for four hours to awake Munich-Wistar rats (N = 6 each group) two days after ARF induction with intrarenal norepinephrine (NE). Renal function and morphology were then examined two days after treatment (Day 4). Serum creatinine (SCr) was similarly increased in both groups at the time of APIII-D or saline infusion (Day 2). By Day 4 SCr had decreased to the pre-ARF induction level in APIII-D-treated rats (P less than 0.005), but continued to rise in saline-treated animals (P less than 0.025). Day 4 renal blood flow and glomerular filtration rate (GFR) were higher in APIII-D-compared to the saline-treated group (9.13 +/- 1.14 vs. 4.41 +/- 2.25 ml/min and 0.787 +/- 0.066 vs. 0.370 +/- 0.185 ml/min, respectively, both P less than 0.005).(ABSTRACT TRUNCATED AT 250 WORDS)

Serial glomerular and tubular dynamics in thyroidectomized rats with remnant kidneys.

Serial measurements were performed in Munich-Wistar rats with five-sixths nephrectomy that had undergone prior selective thyroidectomy (Tx group) or thyroidectomy with thyroxine replacement (TxT4 group) to determine the effects of Tx on glomerular and tubular dynamics in relation to Tx attenuation of renal failure progression. At 1 week, inulin clearance rates (Cin) in TxT4 and Tx rats were 0.367 +/- 0.171 and 0.120 +/- 0.036 mL/min, respectively, different at P less than 0.01. Corresponding single-nephron filtration rate (SNGFR), glomerular plasma flow (QA), glomerular transcapillary hydraulic pressure (delta P), and proximal tubular reabsorption (Jv) were all reduced in Tx compared with TxT4 rats (P less than 0.01). Protein excretion (UPROT) was 151 +/- 40 in TxT4 rats, and 9 +/- 5 mg/d in Tx animals. Glomerular mesangial matrix expansion and focal tubulointerstitial changes were more frequent in TxT4 than Tx rats. By 4 weeks, Cin, SNGFR, QA, glomerular ultrafiltration coefficient (Kf) and Jv were similar in Tx and TxT4. Only glomerular capillary pressure (PGC) remained lower in Tx rats (35 +/- 3 v 50 +/- 3 mm Hg in TxT4, P less than 0.001). UPROT was 161 +/- 24 in TxT4 and 17 +/- 12 mg/d in Tx rats. While 7% +/- 4% of glomeruli showed focal sclerosis in TxT4 rats, there was none in the Tx group. Maximal glomerular planar area increased between 1 and 4 weeks in the TxT4 group, but not in the Tx group. However, this measurement was not significantly different between TxT4 and Tx glomeruli at 1 or 4 weeks. Minimal focal tubulointerstitial changes were found in TxT4, but there were not progressive from those observed at 1 week. The reduced PGC at 1 week was the result of a disproportionately greater increase in afferent (RA) than efferent arteriolar resistance (RE) in Tx rats (P less than 0.025); however, at 4 weeks, both RA and RE had decreased to values identical to those in TxT4 animals and the lower PGC in Tx rats was the result of a reduced mean arterial pressure. In conclusion, a reduced PGC was the sole functional correlate of decreased proteinuria and glomerulosclerosis afforded by Tx in this partial nephrectomy model. Suppression of either nephrectomy-related hypertrophy or tubulointerstitial injury by Tx could not be excluded as at least partially protective factors.

The management of hyperthyroidism. A surgeon's perspective.

Surgery for the treatment of hyperthyroidism is rapid and permanent, highly safe, and highly successful and has an important and complementary role with medical therapy and 131I. In Grave's disease cases total thyroidectomy, performed only if parathyroid glands are preserved, prevents recurrent hyperthyroidism. Bilateral subtotal thyroidectomy or total lobectomy with contralateral subtotal lobectomy are done if at least one parathyroid cannot be preserved on each side. In terms of recurrent laryngeal nerve preservation, all three operations are equally safe if the nerve is positively identified and traced throughout its course.

Ischemia-induced loss of epithelial polarity. Role of the tight junction.

In proximal tubular cells ischemia is known to result in the redistribution of apical and basolateral domain-specific lipids and proteins into the alternate surface membrane domain. Since tight junctions are required for the maintenance of surface membrane polarity, the effect of ischemia on tight junction functional integrity was investigated. In vivo microperfusion of early loops of proximal tubules with ruthenium red (0.2%) in glutaraldehyde (2%) was used to gain selective access to and outline the apical surface membrane. Under control situations ruthenium red penetrated less than 10% of the tight junctions. After 5, 15, and 30 min of ischemia, however, there was a successive stepwise increase in tight junction penetration by ruthenium red to 29, 50, and 62%, respectively. This was associated with the rapid duration-dependent redistribution of basolateral membrane domain-specific lipids and NaK-ATPase into the apical membrane domain. Taken together, these data indicate that during ischemia proximal tubule tight junctions open, which in turn leads to the lateral intramembranous diffusion of membrane components into the alternate surface membrane domain.

Atrial natriuretic peptide and dopamine in a rat model of ischemic acute renal failure.

Atrial natriuretic peptide (ANP) has been shown to reverse experimental models of ischemic acute renal failure (ARF). However, infusion of ANP has been associated with systemic hypotension making its use in clinical ARF impractical. Therefore, in this investigation, dopamine (D) was combined with intravenous (i.v.) atriopeptin III (AP III) to determine if this regimen was effective in reversing ARF while preventing systemic hypotension and maintaining renal blood flow (RBF). Four groups of Munich-Wistar rats were studied. Group 1, sham-ARF; Group 2, renal artery (RA) clamp (55 min) followed by i.v. saline; Group 3, RA clamp followed by i.v. AP III-D; and Group 4, RA clamp followed by i.v. D only. All infusions were begun after RA clamp release and continued for four hour. Mean arterial pressure in Group 3 rats given AP III-D were similar to that in Group 2, slightly less than that in Groups 1 and 4 (P less than 0.05), but consistently greater than 100 mm Hg during the four hour infusion. RBF in Group 3 was elevated above the level in Group 1 at P less than 0.05. Glomerular filtration rate (GFR), depressed by 52% in Group 2, was corrected to control (sham-ARF) levels in Group 3. In Group 4 there was a small but significant increase in GFR compared to Group 2 (P less than 0.05), but it remained less than that in sham-ARF or AP III-D treated ARF rats (P less than 0.01). Urine flow rate and urine sodium excretion rate were more than sixfold higher in Group 3 than any other group.(ABSTRACT TRUNCATED AT 250 WORDS)

The protective mechanism of thyroidectomy in a rat model of chronic renal failure.

Selective thyroidectomy (Tx) has been shown to attenuate proteinuria and disease progression in models of chronic renal failure (CRF). In this investigation, four groups of Munich-Wistar rats were studied to determine if glomerular dynamics or another renal metabolic consequence of Tx was responsible for the protective effect as measured by 24-hour protein excretion (UPROT). The groups were TxT4, thyroxine-replaced Tx rats with five-sixths nephrectomy; Tx, Tx rats not receiving replacement thyroxine with five-sixths nephrectomy; TxI, Tx rats not receiving replacement thyroxine with five-sixths nephrectomy that were given continuous intraperitoneal isoproterenol to restore systemic and renal hemodynamics; and TxT4(C), two-kidney Tx rats receiving replacement thyroxine that served as normal controls. Five-sixths nephrectomy was carried out 2 weeks after Tx, and experiments were carried out 1 week later. Serum T4 was profoundly reduced and there was failure to gain weight in Tx and TxI rats, despite similar protein intakes in all groups. Cardiac output was reduced in Tx, but was similar in TxI to levels in TxT4 rats. Whole-kidney glomerular filtration rate was lower in Tx, at 0.145 +/- 0.052 mL/min (P less than 0.05), but similar in TxI (0.305 +/- 0.147 mL/min) to that in TxT4 rats (0.317 +/- 0.135 mL/min). Twenty-four-hour urinary protein, which was 129 +/- 57 mg in TxT4, was reduced in Tx to 9 +/- 3 mg (P less than 0.01) but restored in TxI to 89 +/- 30 mg, a level similar to that in TxT4.(ABSTRACT TRUNCATED AT 250 WORDS)

Loss of epithelial polarity: a novel hypothesis for reduced proximal tubule Na+ transport following ischemic injury.

Ischemia results in the marked reduction of renal proximal tubule function which is manifested by decreased Na+ and H2O reabsorption. In the present studies the possibility that altered Na+ and H2O reabsorption were due to ischemia-induced loss of surface membrane polarity was investigated. Following 15 min of renal ischemia and 2 hr of reperfusion, proximal tubule cellular ultrastructure was normal. However, abnormal redistribution of NaK-ATPase to the apical membrane domain was observed and large alterations in apical membrane lipid composition consistent with loss of surface membrane polarity were noted. These changes were associated with large decreases in Na+ (37.4 vs. 23.0%, P less than 0.01) and H2O (48.6 vs. 36.9%, P less than 0.01) reabsorption at a time when cellular morphology, apical Na+ permeability, Na+-coupled cotransport, intracellular pH and single nephron filtration rates were normal. We propose that the abnormal redistribution of NaK-ATPase to the apical membrane domain is in part responsible for reduced Na+ and H2O reabsorption following ischemic injury.

Phosphate restriction reduces proteinuria of the uninephrectomized, diabetic rat.

Proteinuria is the clinical hallmark of diabetic nephropathy and the harbinger of progressive renal disease. Therefore, the present study was designed to examine the effect of phosphate restriction on the proteinuria of streptozotocin-induced diabetes mellitus in the rat. Uninephrectomy was performed in experimental and control groups to worsen the degree of diabetic nephropathy. Proteinuria was prevented in Sprague-Dawley rats treated with the intestinal phosphate binder, dihydroxyaluminum aminoacetate (DHAAA) (24.75 +/- 20.35 mg/d at 3 months v control, 77.45 +/- 44.72 mg/d, P less than 0.001); an effect that was independent of protein and caloric intake, plasma albumin and lipids, severity of diabetes, mean arterial pressures, cardiac output, and renal calcium accumulation. The effect of DHAAA on protein excretion and glomerular hemodynamics was examined in similarly prepared Munich-Wistar rats; these rats did not tolerate long-term studies. Three weeks of DHAAA again caused a consistent fall in proteinuria (5.98 +/- 7.28 v 34.94 +/- 24.28 mg/d) and in transmembrane hydraulic pressure difference (41.1 +/- 1.2 v 46.4 +/- 2.8 mm Hg, P less than 0.005). In conclusion, phosphate restriction significantly decreases the proteinuria of Sprague-Dawley and Munich-Wistar uninephrectomized rats with streptozotocin-induced diabetes mellitus. Micropuncture of Munich-Wistar rats suggests that a reduction of intraglomerular pressure may be at least partially responsible for such an effect.

Temporary postthyroidectomy hypocalcemia.

The causes of temporary hypocalcemia after thyroidectomy are not well understood. In 18 patients undergoing unilateral (UL) and bilateral thyroid lobectomy (BL), an attempt was made to preserve all parathyroid glands with an intact blood supply. Total calcium (bound and free), free calcium (physiologically active form), albumin, parathyroid hormone, and calcitonin levels were serially measured. After UL, free calcium, parathyroid hormone, and calcitonin levels were unchanged, but total calcium level decreased because albumin level decreased. After BL, total calcium level decreased due to a decrease in albumin-bound calcium level. Free calcium level also decreased due to a decrease in parathyroid hormone level. Calcitonin level did not change. Despite careful preservation of the parathyroids and their blood supply, BL is frequently associated with temporary hypoparathyroidism. Techniques for preservation of parathyroid glands with their vascular integrity, correlation of surgical manipulation of parathyroids and calcium level, and calcium binding are discussed.

Glomerular dynamics in the hypothyroid rat and the role of the renin-angiotensin system.

Munich-Wistar rats underwent thyroidectomy (TX) with reimplantation of the parathyroid glands. Systemic hemodynamic and micropuncture studies were performed at 1 and 3 wk post-TX, and the results were compared with levothyroxine replaced controls (TXT4). Cardiac output (CO) in TX rats fell progressively and was 40% of that in TXT4 at 3 wk. Renal blood flow declined in parallel with CO. Systemic blood pressure did not fall in TX rats because of a 50% increase in systemic vascular resistance by 3 wk post-TX. Glomerular dynamics were not significantly different between TX and TXT4 rats at 1 wk; however, by 3 wk single-nephron glomerular filtration rate (SNGFR) had fallen to 16.5 +/- 1.1 vs. 34.1 +/- 3.4 nl/min in TXT4 controls (P less than 0.001). In 3-wk TX rats, glomerular plasma flow (QA) was 50.9 +/- 3.7 vs. 108.0 +/- 8.7 nl/min in TXT4 rats (P less than 0.001), net hydraulic ultrafiltration pressure (delta P) was 33 +/- 2 vs. 37 +/- 1 mmHg in TXT4 rats (P less than 0.01), and the ultrafiltration coefficient (Kf) was 0.025 +/- 0.003 vs. 0.084 +/- 0.008 nl X s-1 X mmHg-1 in TXT4 controls (P less than 0.001). Although the changes in systemic and glomerular hemodynamics were progressive over 3 wk, proximal tubular reabsorption fell maximally within 1 wk. Similar patterns of alterations in glomerular dynamics are known physiological consequences of angiotensin II (ANG II).(ABSTRACT TRUNCATED AT 250 WORDS)

Glomerular and tubular dynamics in mercuric chloride-induced acute renal failure.

Mercuric chloride (HgCl2)-induced acute renal failure has received considerable investigative attention but little agreement as to its pathogenesis. A source of some disagreement has been the lack of direct glomerular dynamics measurements in this disorder. We examined glomerular dynamics, tubular integrity, and whole kidney function at 24 hours in Munich-Wistar rats given either HgCl2, 3.5 mg/kg, or a similar volume of 0.9% saline solution intramuscularly. Arterial blood pressure was elevated in HgCl2-injected rats, but renal blood flow and its distribution were similar to those of controls. Inulin clearance, however, was reduced by 89% in HgCl2-injected animals. Glomerular dynamics experiments demonstrated similar glomerular plasma flow (QA) and glomerular capillary and tubular pressures in control and HgCl2-injected animals but a higher net afferent ultrafiltration pressure and lower ultrafiltration coefficient (Kf) in the HgCl2-injected group. Single-nephron glomerular filtration rate (SNGFR), when measured from Bowman's space and HgCl2-injected rats, was similar to that measured from late proximal tubules in controls. However, SNGFR determined from the later proximal tubule in HgCl2-injected rats was only one third of that measured from Bowman's space. QA estimated from glomerular counting was similar to that calculated from Bowman's space SNGFR in HgCl2-injected rats. 3H-inulin microinjection experiments confirmed the presence of tubular fluid backleak suggested by the discrepancy in Bowman's space and late proximal tubular SNGFR measurements. It is concluded that at 24 hours in low-dose HgCl2-induced acute renal failure, tubular fluid backleak is the major pathogenetic factor, with a decline in Kf having a potential secondary role.

Graves' disease associated with histologic Hashimoto's thyroiditis.

The microscopic slides of 16 patients who underwent bilateral subtotal thyroidectomy for hyperthyroid Graves' disease were reviewed and classified into three groups: I, Hashimoto's thyroiditis; II, Graves' disease; and III, both Hashimoto's thyroiditis and Graves' disease. Three patients were classified as group I, 10 as group II, and three as group III. In 38% of the patients with clinical Graves' disease the histologic evidence of Hashimoto's thyroiditis could be found either alone or in combination with histologic evidence of Graves' disease (groups I and III). One patient in group I, four in group II, and three in group III had infiltrative ophthalmopathy (50% of total). Hyperthyroid Graves' disease, Graves' ophthalmopathy, and Hashimoto's thyroiditis can occur all together, in duads, or individually at a specific time in a patient's life.