| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Interventional US Unit, Department of Medicine, S. Giovanni di Dio Hospital,
80059 Torre del Greco (NA), Italy.
2 Present address: via montedoro 43 is. B (parco montedoro), 80059 Torre del
Greco (NA), Italy.
3 VS. Maria della Pietà Hospital, ASLNA3, Casoria (NA), Italy.
4 Division of Surgery, Department of Surgery, Anesthesiology, and Emergency,
Faculty of Medicine and Surgery, Second University of Naples, Naples,
Italy.
Received September 24, 2004;
accepted after revision September 9, 2007.
Address correspondence to L. Tarantino
(luciano_tarantino{at}fastwebnet.it).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One hundred twenty-five patients (88 women, 37 men; age range, 17–76 years; mean age, 53 years) with 127 hyperfunctioning thyroid nodules (volume, 1.2–90 mL; mean, 10.3 mL) were treated with PEI. There were 1–11 PEI sessions per patient (average, 3.9) performed, with injection of 1–14 mL of ethanol per session (total injected ethanol per patient, 3–108 mL; mean, 14.0 mL). Efficacy of the treatment was assessed with color Doppler sonography; scintigraphy; and free triiodothyronine (FT3), free thyroxine (FT4), and thyroid-stimulating hormone (TSH) assays. Follow-up (9–144 months; median, 60 months) was performed with TSH and color Doppler sonography every 2 months for 6 months and every 6 months thereafter.
RESULTS. Three (2.4%) of 125 patients refused completion of PEI
therapy because of pain. Results are reported in 122 patients with 124
nodules. All 122 patients showed posttreatment normal levels of FT3, FT4, and
TSH. A complete cure (absent uptake in the nodule and recovery of normal
uptake in the thyroid parenchyma) was obtained in 113 (93%) of 122
patients—115 (92.7%) of 124 treated nodules. Residual hyperfunctioning
nodular tissue along with decreased thyroid parenchyma uptake (partial cure)
was present in nine patients accounting for nine (7.3%) of 124 nodules. Rates
of complete cure after PEI were: overall nodules, 115 (92.7%) of 124; nodules
10 mL, 63 (94.0%) of 67; nodules > 10 to 30 mL, 32 (91.4%) of 35;
nodules > 30 to 60 mL, 17 (89.5%) of 19; nodules > 60 mL, three
(100%) of three. The overall rate of major complications (transient laryngeal
nerve damage, two patients; abscess and hematoma, one patient each) was four
(3.2%) of 125 patients. Follow-up examinations showed marked shrinkage of 112
treated nodules ranging from 50% to 90% of the pretreatment volume (mean, 66%)
and new growth of hyperfunctioning tissue in four patients at color Doppler
sonography and scintigraphy at 12, 18, 18, and 48 months' follow-up,
respectively. However, all patients remained euthyroid (low or normal TSH and
normal FT3 and FT4) during follow-up.
CONCLUSION. PEI of hyperfunctioning thyroid nodules seems to be an effective and safe alternative to traditional treatment. It also appears to be effective in patients with hyperfunctioning thyroid nodules larger than 30 mL.
Keywords: hyperthyroidism • interventional procedures • percutaneous ethanol injection • sonographic guidance • thyroid • thyroid nodules
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Conventional therapies for hyperfunctioning thyroid nodule include radioiodine and surgery [1]. Both of these treatments are associated with a small but definite risk of hypothyroidism, whereas hypoparathyroidism and vocal cord lesions may occur after surgery [2, 3].
Over the past 16 years, percutaneous ethanol injection (PEI) under sonographic guidance has been proposed as an alternative therapy to surgery and radioiodine [4–8]. Ablation of hyperfunctioning thyroid nodules by PEI is considered successful when TSH, FT3, and FT4 serum levels return to normal levels [4–8]. However, if the patient becomes euthyroid, persistence of an abnormal area of 99mTc pertechnetate uptake and suppressed thyroid parenchyma may occur. This condition has been reported as a "partial cure" [7]. A complete cure by PEI ablation of hyperfunctioning thyroid nodules is considered posttreatment absence of hyperfunctioning tissue and normal uptake of thyroid parenchyma at scintigraphy [4–8].
The best results of PEI ablation of hyperfunctioning thyroid nodules have been observed for small to midsized solitary nodules (nodular volume smaller than 30 mL), whereas, in patients with a nodular volume larger than 40 mL, PEI seems to be rather ineffective [4–12]. However, some authors have reported very good results with PEI treatment in patients with hyperfunctioning thyroid nodules larger than 40 mL [13, 14].
In the present study, we report the results on the long-term efficacy of PEI treatment in terms of nodule volume reduction, normalization of both serum thyroid hormone levels, and scintigraphy in a series of patients with hyperfunctioning thyroid nodules treated with PEI and followed up for 13 years.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients had been advised of PEI and referred to our unit from several endocrinologists working in other institutions. Enrollment criteria were single or multiple (up to three) hyperfunctioning nodules at scintigraphy (Fig. 1A), low serum TSH levels, and contraindication to or refusal of surgery and radioiodine therapy. Large size of the nodules (e.g., > 30 mL) was not considered an exclusion criterion.
|
|
|
|
All patients gave their informed consent before starting PEI treatment. The ethics committee at our institution fully approved the treatment protocol.
Methods
PEI sessions and cycles—A PEI session is a single outpatient
procedure entailing one or more punctures and ethanol injection
(Fig. 2). A cycle consisted of
multiple PEI sessions performed every other week. A cycle of PEI sessions was
considered completed when all portions of the nodule seemed to be perfused
from injected ethanol and no color Doppler signal was detectable (Figs.
1B and
1C). These findings are
associated with disappearance of hyperfunctioning thyroid nodules at
scintigraphy (Fig. 1D),
recovery of FT3 and FT4 levels in almost all cases, and recovery of normal TSH
levels in most cases. However, if subsequent thyroid scintigraphy showed still
hyperfunctioning tissue in the lesion, a second cycle of PEI sessions was
performed soon afterward, with the objective of ablation of the still active
portion of the nodule. Because our treatment schedule entailed the complete
necrosis of the lesion, additional cycles could have been performed to
accomplish our objective.
|
Before PEI treatment, all nodules were evaluated with sonography for the following parameters: volume, which was automatically calculated with the software available in the sonography machine by measuring the three orthogonal diameters of the nodule on the vertical and horizontal sectional images; echogenicity (solid, mixed, or fluid and iso-, hypo-, or hyperechoic) in contrast with the adjacent parenchyma; and pattern of vascularization at color power Doppler sonography. The threshold for color-flow signal was set individually for each patient by raising the threshold control to a level slightly above the point at which random color noise disappeared. Images were recorded on both hard copy and digital videotape.
The same physician performed B-mode sonography, color Doppler sonography, and FNB in all patients. FNB was performed with a 22-gauge, 90-mm spinal needle (Yale Spinal, Becton Dickinson) and a freehand technique under sonographic guidance. A single experienced pathologist evaluated the cytologic samples.
Procedure—A single physician performed all procedures using a freehand technique. PEI sessions were performed with the patient in the supine position with the neck slightly hyperextended. All patients underwent multiple sessions of PEI treatment. In each session, the patients underwent multiple punctures to inject the ethanol into the different portions of the nodules. The punctures were performed with a freehand technique under sonographic guidance by using a commercially available syringe filled with 99.8% ethanol (FE RO, SA) with a 22-gauge x 30-mm needle. A longer 22-gauge x 90-mm needle (Yale Spinal) was used to inject the alcohol into the deepest portions of very large nodules.
As a general rule, for both FNB and ethanol injection, after imaging the nodule on sonography, the needle was positioned on an oblique angle under the probe so as to image its long axis.
Just before each PEI session, to evaluate treatment efficacy and to plan further PEI sessions, two sonographic parameters were considered: changes in echogenicity in a part of or in the whole nodule at B-mode sonography and persistence of intranodular vascular signals at color power Doppler sonography.
Under constant real-time sonographic monitoring, diffusion of ethanol through all portions of the nodule was evaluated using the intense hyperechogenicity induced by injection as a guide of ethanol perfusion. Sometimes ethanol diffused from the injection point to a large part of the lesion; however, in other cases it accumulated in the injection point. The reasons for such different, unpredictable results of ethanol injection are unknown. To obtain an intralesional diffusion of the ethanol as wide as possible, the needle was moved along its path and inserted in different portions of the nodules without completely withdrawing it.
Pain can be a drawback of ethanol injection in a thyroid nodule and is likely due to ethanol escaping outside the thyroid capsule with subsequent deep neck nerve stimulation or intravascular diffusion or both. This event is unpredictable and cannot be prevented. Sonographic monitoring during injection can detect ethanol escaping from the nodule through thyroid veins or along the needle path. Changing the position of the needle tip in the nodule to avoid IV injection can reduce the ethanol escaping from the target lesion. Ultimately, if the patient complained of local pain during the procedure, the ethanol injection was transiently withdrawn and restarted in a different portion of the nodule after a slight displacement of the needle tip along its path.
Intraprocedure and postprocedure evalu ation of ablation—The number of punctures to perform and the amount of ethanol to inject per session were not planned in advance but were tailored to the patient's compliance. A first cycle of PEI treatment was considered completed when all portions of the nodule had been apparently perfused with ethanol after multiple sessions of PEI and intralesional color power Doppler sonography signals were absent. After a first cycle of PEI sessions, a second cycle of PEI sessions was performed when, at posttreatment scintigraphy, hyperfunctioning tissue in the treated lesion was still present even if intralesional color Doppler signals were no longer detectable. In these cases, together with indications coming from the scintigraphic findings, gray-scale sonography is useful for targeting the still-active tissue. In fact, through our experience, we learned that several days after the PEI session necrotic tissue shrinks and becomes hypoechoic and inhomogeneous in contrast with the residual hyperfunctioning tissue, which appears solid and homogeneously echogenic at sonography. These findings make it easier to identify the parts of the lesion to be treated in the second cycle (Fig. 3A, 3B, 3C, 3D, 3E) and constitute an additional good reason for performing PEI sessions once a week.
|
|
|
|
|
To assess PEI efficacy, scintigraphy with 99mTc pertechnetate and TSH assay were scheduled within 2 months after the end of treatment. Scintigrams were evaluated to detect the absence or persistence of uptake in the treated nodule and the recovery of uptake in the thyroid parenchyma. The treatment end point was the complete ablation of hyperfunctioning thyroid nodule uptake at scinti graphy and reuptake of normal thyroid parenchyma. In case of incomplete necrosis (persistent uptake in the nodule at scintigraphy) after a first cycle of PEI sessions, additional sessions of PEI (second cycle) were performed.
Early postprocedure monitoring—PEI sessions were performed on an outpatient basis in 117 patients, and eight patients (age range, 64–72 years) with heart dysfunction (tachycardia, atrial fibrillation, and other arrhythmias) were hospitalized. All patients were clinically monitored for 2 hours after each session of ethanol injection to assess prolonged pain or hoarseness. In patients with persistent pain or neck swelling only, sonography was performed within 2 hours after the PEI session to detect possible complications.
Follow-up—Follow-up examinations, including clinical evaluation; color Doppler sonography; and FT3, FT4, and TSH hormone assays were scheduled every 3 months during the first year and every 6 months subsequently. During the followup, the same physician performed color Doppler sonography to assess the changes in echogenicity and vascularization. To evaluate the progressive shrinkage of the lesions after PEI treatment, the volume of nodules was measured in all patients at 6 and 12 months. In case of increasing volume of the treated lesion at sonography or decrease of TSH levels at follow-up examinations, scintigraphy was performed.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Procedural data—During or after the first PEI session, three patients (2.4%) with a single nodule out of 125 patients complained of intense pain and refused to continue PEI treatment. These cases were considered failures of PEI. Thus the interventional procedure actually was accomplished in 122 patients with 124 nodules. For this reason, we will report our results in 122 patients (Table 1 and Fig. 2).
|
The overall number of PEI sessions per patient, including the total number of sessions performed in one or more cycles, ranged from 1 to 11 (average, 4.0). We injected 1–14 mL (mean, 3.6 mL) of 99.8% ethanol per session (range of total amount of ethanol per patient, 3–108 mL; mean, 14.3 mL; mean amount of ethanol per nodule, 14.0 mL). In the first cycle of PEI sessions, from one to seven sessions per patient were performed (average, 3.2 sessions per patient; 3.2 sessions per nodule).
After this first cycle, 85 (69.7%) of 122 patients showed complete ablation of their 86 (69.3%) of 124 nodules at scintigraphy, whereas 37 (30.3%) of 122 patients with 38 nodules showed residual hyperfunctioning tissue in treated lesions and still absence of radionuclide uptake in the thyroid parenchyma at posttreatment scintigraphy (among this group, the patient with two hyperfunctioning thyroid nodules showed complete ablation of one nodule and only partial necrosis of the other one). All 37 patients were advised to undergo a second cycle of PEI sessions.
Seven of these 37 patients refused to undergo a second cycle of PEI sessions. However, after the first cycle, all seven patients were euthyroid, with normal FT4 and FT3 levels and with TSH levels only slightly below the normal range.
Thirty patients with 31 nodules of the 37 cases with incomplete necrosis of the nodule after the first cycle underwent a second cycle of PEI treatment in which from one to four PEI sessions (average, 2.9 sessions) were performed. After the second cycle, scintigraphy showed complete absence of 99mTc pertechnetate uptake in the nodule and recovery of thyroid parenchyma in 29 nodules. Two patients with a single nodule each after the second cycle were euthyroid, with normal levels of FT3, FT4, and TSH, although some residual hyperfunctioning tissue was evident at scintigraphy. Both patients refused a third cycle of PEI. None of the patients in our series underwent a third cycle of PEI sessions.
Outcome of PEI treatment—PEI was accomplished in 122
patients with 124 nodules of the 125 enrolled patients
(Table 1 and
Fig. 4) All 122 patients
showed posttreatment normal levels of FT3, FT4, and TSH. At definitive
scintigraphic follow-up examination, a complete cure (absent uptake in the
nodule and recovery of normal uptake in thyroid parenchyma) was obtained in
113 (93%) of 122 patients and 115 (92.7%) of 124 treated nodules. Residual
hyperfunctioning nodular tissue along with decreased thyroid parenchyma uptake
(partial cure) was present in nine patients and nine (7.3%) of 124 nodules.
Rates of complete cure by size of the treated nodule were nodules 10 mL,
63 (94%) of 67; nodules > 10 to 30 mL, 32 (91.4%) of 35; nodules >
30 to 60 mL, 17 (89.5%) of 19; and nodules > 60 mL, three (100%) of
three.
|
After 32 (6.7%) of 480 PEI sessions within 1–4 days from the PEI, 26 (21.3%) of 122 patients (two times in six patients) complained of persistent severe pain (all 32 cases) and dysphagia (18 cases). All of these patients underwent urgent clinical and sonographic follow-up examination. In 30 cases, no complication was evident and the patient regularly underwent the subsequent PEI sessions. Of the other two patients, the first showed a large hematoma of the neck that forced the postponement of the subsequent scheduled PEI session; after 3 weeks the patient resumed the PEI sessions and successfully completed the treatment. The second patient showed a pyogenic abscess in the right lobe of the thyroid. This complication occurred in a patient with a single nodule (volume, 1.8 mL) in the right lobe treated with 2 mL of ethanol in a single session. The patient presented with hyperpyrexia, severe pain, and stiffness of the neck 3 days after PEI treatment. The sonographic finding of a hypoechoic inhomogeneous nodule 2.5 cm in size was consistent with an abscess at the site of the treated lesion. Gram-positive cocci were shown with Gram stain on a specimen obtained by fine-needle aspiration biopsy. The lesion was successfully treated with IV antibiotics and echo-guided percutaneous aspiration with intralesional instillation of gentamicin.
Immediately after a PEI session, three (2.4%) of 125 patients complained of hoarseness, probably due to extracapsular escape of ethanol and infiltration into the recurrent laryngeal nerve. In one case, this hoarseness lasted only 24 hours and the patient recovered without any treatment. The other two patients with persistence of this symptom were referred to an otolaryngologist, who diagnosed vocal chord paresis. Corticosteroids and vitamin B12 were administered for 3 weeks to both patients; the complete recovery of the vocal chord was observed after 14 days and 2 months, respectively. Other side effects were not recorded. Hypothyroidism at follow-up was observed in no patients in our series. The overall rate of major complications (hoarseness, abscess) was four (3.2%) of 125 patients.
Follow-Up
The median length of follow-up in the 122 patients who completed PEI
treatment was 60 months (range, 9–144 months). Fifty-two (43%) of the
122 patients dropped out of followup after 12–42 months.
Sonography follow-up of nodule volume—At the 12-month follow-up, we observed a marked shrinkage of 112 treated nodules ranging from 50% to 90% of the pretreatment volume (mean, 66%). The posttreatment sonographic pattern was hypoechoic in 102 (91%) and hyperechoic in 10 (9%) of the 112 shrunken nodules.
Color Doppler sonography and hormonal follow-up examinations—One hundred nine patients showed steady normal TSH, FT3, and FT4 levels and absence of intralesional color Doppler signal at follow-up examinations. The nine patients showing partial cure after completion of PEI treatment were still euthyroid (normal FT3 and FT4) at 16–78 months of follow-up. Four patients with a single nodule completely ablated at posttreatment scintigraphy (Figs. 5A, 5B, 5C, 5D) and with normal TSH levels after PEI relapsed after 12, 18, 18, and 48 months, respectively. In these patients, color Doppler sonography showed solid hypervascular tissue at the periphery of the previously treated nodules (Fig. 5E). Similarly, scintigraphy showed increased uptake of the tracer in peripheral parts of the nodules (Fig. 5F). However, all four patients were actually euthyroid (normal FT3 and FT4 levels) and showed only low TSH levels; none of these four patients underwent additional PEI treatment and, at 36–60 months of follow-up, they remain euthyroid.
|
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results in our series are similar to or even exceed those reported by the previously mentioned authors. We achieved a high rate of complete cure in 113 (90.4%) of 125 patients and a partial cure in only nine (7.2%) of 125 patients. However, all cured patients remained euthyroid during follow-up. PEI treatment failed in three (2.4%) of 125 cases only because of patient intolerance.
Many authors consider the euthyroid state of the patient an acceptable end point of PEI treatment [4–9]. Since we started PEI treatment of hyperfunctioning thyroid nodules in 1993, our aim has been the complete cure of the patient. We think that complete ablation of hyperfunctioning thyroid nodules with PEI and the ensuing reuptake of normal thyroid parenchyma can avoid the additional costs of too frequent follow-up examinations. Furthermore, in our series, complete ablation resulted in a dramatic shrinkage of the nodules, ranging from 50% to 90% of the pretreatment volume within 12 months from PEI treatment. These data compare with other reported results [14, 16]. The dramatic shrinkage of the nodule after PEI treatment, with subsequent normal appearance of the neck, is certainly a desirable result for the patient's feeling of well-being.
For all these reasons, patients showing incomplete ablation at scintigraphy after a first cycle of PEI sessions were always advised to undergo a second cycle of PEI sessions. Probably the euthyroid state could be considered an acceptable end point in an elderly patient; however, complete cure should always be attempted in younger patients, who are more likely to have a recurrence of hyperthyroidism over time [16].
No significant difference between small and large nodules was observed in terms of complete ablation. This result is in keeping with the experience of Monzani et al. [12] who did not find significant differences in response rate to PEI treatment between patients with nodules smaller than 40 mL and those with nodules larger than 40 mL. In a previously published study on 12 patients with nodules larger than 30 mL, there was complete normalization of TSH and free thyroid hormone serum levels; the recovery of tracer uptake in extranodular tissue at scintigraphy; and the reduction of nodule volume of 40–80% after 3 and 6 months of PEI treatment, respectively [11]. Also in the present series, we did not report significantly different results between large and small hyperfunctioning thyroid nodules.
There is no doubt that surgery is the most effective treatment for hyperfunctioning thyroid nodules [2, 3, 17]. The risk of important complications such as laryngeal nerve damage or hypoparathyroidism is low [17–19]. Nevertheless, the procedure is invasive and contraindicated in patients at high surgical risk [18, 19] and, in case of total thyroidectomy, surgery may lead to hypothyroidism and a life-long requirement of thyroxine intake.
Radioiodine therapy should be regarded as the standard treatment for hyperfunctioning thyroid nodules because it is noninvasive and has low cost and high efficacy [1, 3, 20–22]. However, patients with hyperfunctioning thyroid nodules larger than 3 cm require much higher doses of 131I than those conventionally used to treat smaller hyperfunctioning thyroid nodules and other thyroid diseases [23]. Therefore, posttreatment hypothyroidism is reported at a rate from 4% up to 44% after conventional doses of 131I [24]. Low doses of 131I do not destroy the hyperfunctioning thyroid nodules. Thus in up to 50% of patients successfully treated with radioiodine, scintigraphy still shows autonomous nodules with suppression of the extranodular tissue [25]. This results in long-term, expensive, and frequent follow-up examinations with sonography, blood tests, and scintigraphy [24]. In addition, radioiodine therapy is contraindicated before and during pregnancy and during breast feeding. Moreover, the known effects of radioiodine on spermatogenesis suggest that it is wise to recommend a 120-day waiting period between radioiodine and fertilization [26]. Thus in most institutions, the indications to radioiodine treatment are currently presence of disease contraindicating surgery, patient age > 40 years, and diameter of the nodule < 3 cm [27].
PEI treatment does not entail the drawbacks of surgery and radioiodine. In our and other series [6–12, 14, 16], none of the patients treated with PEI showed hypothyroidism during follow-up. In the case of complete ablation of hyperfunctioning thyroid nodules after PEI, patients do not need administration of levothyroxine or frequent follow-up examinations with blood tests, sonography, and scintigraphy.
PEI was well tolerated in most of the patients. However, three patients refused PEI treatment after the first session, and an additional nine patients did not complete the treatment as scheduled owing to intense pain during treatment. These patients did not share any factors such as site of the nodule, shape of the neck, age, or sex, so the severe pain they had during the injection of the ethanol remains unexplained. Administering analgesic drugs or sedatives could probably help in these cases, but the pain threshold in the single patient cannot be anticipated.
In our series, none of the patients was pretreated with local infiltration anesthesia, cervical plexus block, or pharmacologic sedation before PEI sessions were performed. Reasons for this choice were that since we started PEI of hyperfunctioning thyroid nodules in 1993, we noted that the procedure was very well tolerated; local infiltration of anesthetics in subcutaneous tissue could impair the sonographic visualization of the thyroid during the procedure; and cervical plexus block or sedation of the patient would require the presence of an anesthesiologist and operating room occupancy and would lengthen observation of the patient after treatment for several hours or on a day-hospital basis. In our opinion, PEI treatment of hyperfunctioning thyroid nodules under conscious sedation or general anesthesia without intubation could be proposed only in selected cases.
We observed transient damage of the laryngeal nerve in three patients in our series. Unfortunately, it is impossible to predict the escape of some ethanol from the nodule during injection, so the occurrence of this complication cannot be anticipated. However, in accordance with similar reported cases [4–16], we observed the complete recovery of the nerve with time in all three cases.
In conclusion, PEI treatment is a safe, well-tolerated, effective, and inexpensive therapy for small as well as for large (> 30 mL) hyperfunctioning thyroid nodules. It can be considered a valid alternative to surgery and radioiodine. Our experience confirms its long-term efficacy. Nevertheless, prospective and randomized clinical trials comparing surgery, radioiodine, and PEI treatment alone or in combination are needed to establish the indications for these different therapies.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
