0021-972X/91/7305-1051$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1991 by The Endocrine Society
Vol. 73, No. 5 Printed in U.SA.
Inverse Relation between Iodine Intake and Thyroid Blood Flow: Color Doppler Flow Imaging in Euthyroid Humans A. B. ARNTZENIUS, L. J. SMIT, J. SCHIPPER, D. VAN DER HEIDE, AND A. E. MEINDERS Departments of General Internal Medicine (A.B.A., L.J.S., A.E.M.) and Radiology (J.S.), University Hospital, Leiden; and the Department of Human and Animal Physiology, Agricultural University (D.v.d.H.), Wageningen, The Netherlands
ABSTRACT. High intake of iodine inhibits iodide trapping, iodide organification, and hormone release from the human thyroid. We investigated whether iodine intake also affects thyroid blood flow, as was suggested by a recent study in euthyroid rats. With a Color Doppler device we made 14 consecutive DuplexDoppler registrations of both superior thyroid arteries in 10 euthyroid volunteers during baseline iodine intake (1 week), iodine restriction (2 weeks), return to baseline (1 week), and iodine excess (1 week; 80 ^mol sodium iodide/day). Vessel diameters and mean flow velocity were measured on videotape recordings by a "blinded" observer. Baseline iodide excretion was 0.88 ± 0.38 (±SD) /imol/day.
T
Mean flow velocity was 13.9 ±4.1 cm/s, and vessel diameter was 1.07 ± 0.22 mm. Blood flow was 7.7 ± 3.8 mL/min • superior thyroid artery. During the low iodine diet, excretion dropped to 0.49 ± 0.16 /imol/day, and blood flow increased to 11.0 ± 5.0 mL/min (P < 0.001), remaining elevated (10.3 ± 4.4 mL/min) during the second baseline diet. During high iodide intake, blood flow averaged 5.8 ± 3.4 mL/min (P < 0.001), and the expected decrease in thyroid hormone levels and increase in TSH were seen. We conclude that thyroid blood flow responds inversely, and independently from TSH, to changes in iodine intake in euthyroid humans. (J Clin Endocrinol Metab 73: 1051-1055, 1991)
OXIC goiters are highly vascular (1). Since Plummer introduced iodide (Lugol's solution) treatment for Graves' hyperthyroidism (2), surgeons have debated whether preoperative iodide treatment reduces blood loss from the thyroid (3). Firm evidence that iodide really inhibits thyroid blood flow in Graves' hyperthyroidism has come from studies using isotope techniques (4, 5) and Duplex-Doppler (6). In euthyroidism, the effect of iodide on thyroid blood flow has, to our knowledge, only been studied in rats. Michalkiewicz et al. (7), who measured thyroid blood flow in healthy rats with a labeled microsphere injection technique, found a decrease in flow to 60% of baseline after 2 weeks on a high iodine diet, but (more interestingly) an increase to 250% after 2 weeks on a low iodine diet. Duplex-Doppler is an important new noninvasive method for measuring flow dynamics (8). A recent development is Color Doppler Flow Imaging, which permits better identification of small vessels, such as the superior thyroid artery (STA). We
used this improved Duplex-Doppler technique to investigate the effect of changes in iodine intake on thyroid blood flow in euthyroid humans.
Received January 28,1991. Address all correspondence and requests for reprints to: A. B. Arntzenius, Department of General Internal Medicine, Building 1, ClR41, University Hospital, P.O. Box 9600, 2300 RC Leiden, The Netherlands.
Diet phases
Subjects and Methods Subjects and study design Ten volunteers were selected on the basis of a negative previous medical history, clinical and biochemical euthyroidism, and negative microsomal and thyroglobulin antibodies. Seven women and 3 men, between 19-30 yr of age, entered the study. A dietary history, including daily bread consumption, was obtained. The protocol entailed 14 measurement sessions over a 36-day period, divided into 4 diet phases. At each session, a blood sample, a 24-h urine collection, and a Duplex-Doppler registration were obtained at an individually fixed time of the day. Urinary iodide was measured with the cerium-arsenite method (9). The effect of diet on the individual results (compared to the first baseline diet, Bl) was evaluated by 2-way analysis of variance.
During the baseline (Bl; days 1-8) diet phase, all subjects continued their usual diet, while avoiding seafood and maintaining constant bread consumption. During the low iodine 1051
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 March 2016. at 03:22 For personal use only. No other uses without permission. . All rights reserved.
ARNTZENIUS ET AL.
1052
intake (Lo; days 9-23) phase, they received iodine-free bread and water substitutes, and were requested to avoid a list of (relatively) iodine-rich foods. The second baseline (B2; days 24-29) diet was the same as the first (Bl). The high iodine intake (Hi; days 30-36) study phase was characterized by additional oral intake of sodium iodide (6 mg twice daily; 80 /xmol/day) as a 2 mg/ml solution, beginning on the 29th day at 1800 h. Duplex-Doppler registrations and blood flow measurements All Doppler registrations were performed by one of us (L.S.), using a Philips Angiodynograph 2 Color Doppler Flow Imaging system. This system processes the signal from a 7.5-MHZ linear phased array transducer to a real-time Grey scale image with colored areas, representing moving blood cells. In addition, the real-time doppler shift (i.e. frequency shift; Af) at any point on the screen can be "sampled" with the sample-volume cursor. After the angle (a) between the ultrasound beam and the vessel has been indicated on the screen with the angle cursor (required because velocity is calculated from Af and cosa), the "flow velocity envelope" curve is displayed on the monitor. Mean (i.e. time-averaged) flow velocity (MFV) can be calculated for any selected portion of this curve. After the subject had rested for 10 min, the left STA was identified, and the sample-volume cursor was positioned within the axis of a straight stretch of the vessel. The spectral wave form was stored digitally on videotape for at least 30 consecutive cardiac cycles, and this was repeated for the right STA and carotid artery. All registrations were unlabeled except for a code number, randomly assigned beforehand. The actual measurements were performed on these tapes at a later date by an unbiased "blinded" observer. Vessel diameter (d) was determined by using diameter callipers on the recorded screen image. A sequence of 5 homogeneous cardiac cycles of optimal quality and with optimal alignment of the angle cursor was selected for the measurement of MFV. Volume blood flow was calculated by multiplying MFV by the cross-sectional area of the vessel (l/47rd2).
Results Urinary iodide excretion The differences between individuals in mean iodide excretion (four 24-h collections each) during the baseline diet correlated well with differences in their previously stated bread consumption (0.14 jumol iodide/slice of bread; P < 0.005; Fig. 1). The 24-h creatinine excretions were constant throughout the study (Fig. 2). During the iodine-restricted diet, mean iodide excretion dropped in all individuals except one, who inadvertently used an iodide-containing antiseptic on the 14th day (*, Fig. 2). We excluded this subject from the analysis of the low iodine phase (see Table 1). Mean iodide excretion returned to initial values during the second baseline diet and rose steadily during the final (high iodine intake) phase (Table 1), reaching the intended intake on the final day.
JCE & M • 1991 Vol73«No5
Serum hormone levels Serum T4, free T4 index, T3, and TSH levels remained within normal limits throughout the study. Changes took place only during the period of high iodine intake, when the free T4 index decreased from 106 ± 13 to 88 ± 14 (mean ± SD; P < 0.005) and T 3 decreased from 1.94 ± 0.28 to 1.77 ± 0.25 nmol/L (P < 0.05). This was mirrored by a gradual increase in TSH from 1.6 to 3.3 mU/L on the final day (Fig. 4); the mean TSH for this diet phase was 2.5 mU/L (Table 1). STA blood flow
The mean values and SDs of flow velocity (MFV), diameter, and calculated blood flow per STA are shown for each diet phase in Table 1. The baseline interindividual differences in flow were significant (P < 0.001) and did not correlate with sex, baseline iodine intake, body weight, or surface. During iodine restriction, the MFV, diameter, and calculated blood flow of the STA all increased significantly (Table 1 and Figs. 3 and 4). After the return to baseline iodine intake, they decreased, but remained significantly higher than baseline levels. During the final week of high iodine intake, they decreased rapidly, reaching their minimum value after 3 days (Fig. 3). The average MFV and blood flow of the STA were lower than those during the first baseline diet; the diameter was not. The MFV of the carotid artery did not change between diet phases.
Discussion Estimations of resting thyroid blood flow vary from 12.2% of cardiac output, depending on the technique used and the species studied (10). With preoperative electromagnetic flowmetry during surgery, blood flow rate has been estimated at 29 mL/min for normal human thyroids, divided evenly over the superior and inferior thyroid arteries (11). With Duplex-Doppler, Hodgson et al. (8) found a higher estimate of 74 mL/min. In our study a Color Doppler system is used to obtain a velocity wave form in the STA. The use of Color Doppler Flow Imaging enables better visualization of small vessels such as the STA, so both sample volume cursor positioning and angle correction can be optimized. With this technique, our estimate of basal thyroid blood flow is 31 mL/min (7.7 mL/min-artery). Although this figure seems plausible, one must realize that there are a number of uncertainties involved in calculating absolute thyroid blood flow from such measurements. The procedure used to position the diameter callipers and the assumptions implicit in the calculation (that the velocity profile across the vessel lumen is homogeneous and that flow volumes in superior and inferior thyroid arteries are equal) may all lead to
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 March 2016. at 03:22 For personal use only. No other uses without permission. . All rights reserved.
1053
IODINE AND THYROID BLOOD FLOW 2.00 CO
•o
o E
1.50
female FIG. 1. Relation between each subject's mean iodide excretion (average of four measurements the during baseline diet phase) and his/her stated daily bread consumption.
I
LOO
(2 subjects)
0)
male
o
X
Vol. 73, No. 5 Printed in U.SA.
Inverse Relation between Iodine Intake and Thyroid Blood Flow: Color Doppler Flow Imaging in Euthyroid Humans A. B. ARNTZENIUS, L. J. SMIT, J. SCHIPPER, D. VAN DER HEIDE, AND A. E. MEINDERS Departments of General Internal Medicine (A.B.A., L.J.S., A.E.M.) and Radiology (J.S.), University Hospital, Leiden; and the Department of Human and Animal Physiology, Agricultural University (D.v.d.H.), Wageningen, The Netherlands
ABSTRACT. High intake of iodine inhibits iodide trapping, iodide organification, and hormone release from the human thyroid. We investigated whether iodine intake also affects thyroid blood flow, as was suggested by a recent study in euthyroid rats. With a Color Doppler device we made 14 consecutive DuplexDoppler registrations of both superior thyroid arteries in 10 euthyroid volunteers during baseline iodine intake (1 week), iodine restriction (2 weeks), return to baseline (1 week), and iodine excess (1 week; 80 ^mol sodium iodide/day). Vessel diameters and mean flow velocity were measured on videotape recordings by a "blinded" observer. Baseline iodide excretion was 0.88 ± 0.38 (±SD) /imol/day.
T
Mean flow velocity was 13.9 ±4.1 cm/s, and vessel diameter was 1.07 ± 0.22 mm. Blood flow was 7.7 ± 3.8 mL/min • superior thyroid artery. During the low iodine diet, excretion dropped to 0.49 ± 0.16 /imol/day, and blood flow increased to 11.0 ± 5.0 mL/min (P < 0.001), remaining elevated (10.3 ± 4.4 mL/min) during the second baseline diet. During high iodide intake, blood flow averaged 5.8 ± 3.4 mL/min (P < 0.001), and the expected decrease in thyroid hormone levels and increase in TSH were seen. We conclude that thyroid blood flow responds inversely, and independently from TSH, to changes in iodine intake in euthyroid humans. (J Clin Endocrinol Metab 73: 1051-1055, 1991)
OXIC goiters are highly vascular (1). Since Plummer introduced iodide (Lugol's solution) treatment for Graves' hyperthyroidism (2), surgeons have debated whether preoperative iodide treatment reduces blood loss from the thyroid (3). Firm evidence that iodide really inhibits thyroid blood flow in Graves' hyperthyroidism has come from studies using isotope techniques (4, 5) and Duplex-Doppler (6). In euthyroidism, the effect of iodide on thyroid blood flow has, to our knowledge, only been studied in rats. Michalkiewicz et al. (7), who measured thyroid blood flow in healthy rats with a labeled microsphere injection technique, found a decrease in flow to 60% of baseline after 2 weeks on a high iodine diet, but (more interestingly) an increase to 250% after 2 weeks on a low iodine diet. Duplex-Doppler is an important new noninvasive method for measuring flow dynamics (8). A recent development is Color Doppler Flow Imaging, which permits better identification of small vessels, such as the superior thyroid artery (STA). We
used this improved Duplex-Doppler technique to investigate the effect of changes in iodine intake on thyroid blood flow in euthyroid humans.
Received January 28,1991. Address all correspondence and requests for reprints to: A. B. Arntzenius, Department of General Internal Medicine, Building 1, ClR41, University Hospital, P.O. Box 9600, 2300 RC Leiden, The Netherlands.
Diet phases
Subjects and Methods Subjects and study design Ten volunteers were selected on the basis of a negative previous medical history, clinical and biochemical euthyroidism, and negative microsomal and thyroglobulin antibodies. Seven women and 3 men, between 19-30 yr of age, entered the study. A dietary history, including daily bread consumption, was obtained. The protocol entailed 14 measurement sessions over a 36-day period, divided into 4 diet phases. At each session, a blood sample, a 24-h urine collection, and a Duplex-Doppler registration were obtained at an individually fixed time of the day. Urinary iodide was measured with the cerium-arsenite method (9). The effect of diet on the individual results (compared to the first baseline diet, Bl) was evaluated by 2-way analysis of variance.
During the baseline (Bl; days 1-8) diet phase, all subjects continued their usual diet, while avoiding seafood and maintaining constant bread consumption. During the low iodine 1051
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 March 2016. at 03:22 For personal use only. No other uses without permission. . All rights reserved.
ARNTZENIUS ET AL.
1052
intake (Lo; days 9-23) phase, they received iodine-free bread and water substitutes, and were requested to avoid a list of (relatively) iodine-rich foods. The second baseline (B2; days 24-29) diet was the same as the first (Bl). The high iodine intake (Hi; days 30-36) study phase was characterized by additional oral intake of sodium iodide (6 mg twice daily; 80 /xmol/day) as a 2 mg/ml solution, beginning on the 29th day at 1800 h. Duplex-Doppler registrations and blood flow measurements All Doppler registrations were performed by one of us (L.S.), using a Philips Angiodynograph 2 Color Doppler Flow Imaging system. This system processes the signal from a 7.5-MHZ linear phased array transducer to a real-time Grey scale image with colored areas, representing moving blood cells. In addition, the real-time doppler shift (i.e. frequency shift; Af) at any point on the screen can be "sampled" with the sample-volume cursor. After the angle (a) between the ultrasound beam and the vessel has been indicated on the screen with the angle cursor (required because velocity is calculated from Af and cosa), the "flow velocity envelope" curve is displayed on the monitor. Mean (i.e. time-averaged) flow velocity (MFV) can be calculated for any selected portion of this curve. After the subject had rested for 10 min, the left STA was identified, and the sample-volume cursor was positioned within the axis of a straight stretch of the vessel. The spectral wave form was stored digitally on videotape for at least 30 consecutive cardiac cycles, and this was repeated for the right STA and carotid artery. All registrations were unlabeled except for a code number, randomly assigned beforehand. The actual measurements were performed on these tapes at a later date by an unbiased "blinded" observer. Vessel diameter (d) was determined by using diameter callipers on the recorded screen image. A sequence of 5 homogeneous cardiac cycles of optimal quality and with optimal alignment of the angle cursor was selected for the measurement of MFV. Volume blood flow was calculated by multiplying MFV by the cross-sectional area of the vessel (l/47rd2).
Results Urinary iodide excretion The differences between individuals in mean iodide excretion (four 24-h collections each) during the baseline diet correlated well with differences in their previously stated bread consumption (0.14 jumol iodide/slice of bread; P < 0.005; Fig. 1). The 24-h creatinine excretions were constant throughout the study (Fig. 2). During the iodine-restricted diet, mean iodide excretion dropped in all individuals except one, who inadvertently used an iodide-containing antiseptic on the 14th day (*, Fig. 2). We excluded this subject from the analysis of the low iodine phase (see Table 1). Mean iodide excretion returned to initial values during the second baseline diet and rose steadily during the final (high iodine intake) phase (Table 1), reaching the intended intake on the final day.
JCE & M • 1991 Vol73«No5
Serum hormone levels Serum T4, free T4 index, T3, and TSH levels remained within normal limits throughout the study. Changes took place only during the period of high iodine intake, when the free T4 index decreased from 106 ± 13 to 88 ± 14 (mean ± SD; P < 0.005) and T 3 decreased from 1.94 ± 0.28 to 1.77 ± 0.25 nmol/L (P < 0.05). This was mirrored by a gradual increase in TSH from 1.6 to 3.3 mU/L on the final day (Fig. 4); the mean TSH for this diet phase was 2.5 mU/L (Table 1). STA blood flow
The mean values and SDs of flow velocity (MFV), diameter, and calculated blood flow per STA are shown for each diet phase in Table 1. The baseline interindividual differences in flow were significant (P < 0.001) and did not correlate with sex, baseline iodine intake, body weight, or surface. During iodine restriction, the MFV, diameter, and calculated blood flow of the STA all increased significantly (Table 1 and Figs. 3 and 4). After the return to baseline iodine intake, they decreased, but remained significantly higher than baseline levels. During the final week of high iodine intake, they decreased rapidly, reaching their minimum value after 3 days (Fig. 3). The average MFV and blood flow of the STA were lower than those during the first baseline diet; the diameter was not. The MFV of the carotid artery did not change between diet phases.
Discussion Estimations of resting thyroid blood flow vary from 12.2% of cardiac output, depending on the technique used and the species studied (10). With preoperative electromagnetic flowmetry during surgery, blood flow rate has been estimated at 29 mL/min for normal human thyroids, divided evenly over the superior and inferior thyroid arteries (11). With Duplex-Doppler, Hodgson et al. (8) found a higher estimate of 74 mL/min. In our study a Color Doppler system is used to obtain a velocity wave form in the STA. The use of Color Doppler Flow Imaging enables better visualization of small vessels such as the STA, so both sample volume cursor positioning and angle correction can be optimized. With this technique, our estimate of basal thyroid blood flow is 31 mL/min (7.7 mL/min-artery). Although this figure seems plausible, one must realize that there are a number of uncertainties involved in calculating absolute thyroid blood flow from such measurements. The procedure used to position the diameter callipers and the assumptions implicit in the calculation (that the velocity profile across the vessel lumen is homogeneous and that flow volumes in superior and inferior thyroid arteries are equal) may all lead to
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 March 2016. at 03:22 For personal use only. No other uses without permission. . All rights reserved.
1053
IODINE AND THYROID BLOOD FLOW 2.00 CO
•o
o E
1.50
female FIG. 1. Relation between each subject's mean iodide excretion (average of four measurements the during baseline diet phase) and his/her stated daily bread consumption.
I
LOO
(2 subjects)
0)
male
o
X
