Surg Today (2008) 38:487–494DOI 10.1007/s00595-006-3673-z

Reprint requests to: A. Ríos, Avenida de la Libertad no. 208, Casillas, 30007 Murcia, SpainReceived: November 14, 2005 / Accepted: November 6, 2006

Original Articles

Results of Surgical Treatment in Multinodular Goiter with an Intrathoracic Component

ANTONIO RÍOS1, JOSÉ M. RODRÍGUEZ

1, PEDRO J. GALINDO1, JUAN TORRES

2, MANUEL CANTERAS3,

MARÍA D. BALSALOBRE1, and PASCUAL PARRILLA

1

1 Departamento de Cirugía, 2 Servicio de Cirugía Torácica, and 3 Departamento de Bioestadística, Universidad de Murcia, Hospital Universitario Virgen de la Arrixaca, Murcia, Spain

AbstractPurpose. (1) To determine the clinical profi le of intra-thoracic multinodular goiter (IMG); (2) to evaluate the results of surgery, and (3) to analyze the incidence of malignancy and its evolution.Methods. Two hundred and forty-seven operated cases of IMG were reviewed. These cases of IMG had all been diagnosed according to Eschapase’s defi nition (>3 cm below the sternal manubrium). The morbidity and postoperative evolution were analyzed. A compar-ative study was carried out on a group of 425 cases of nonintrathoracic goiter. We applied the χ2 test, Student’s t-test, and a logistical regression analysis.Results. Intrathoracic MG occurs in patients over 60 years of age, with goiter which has a long evolution time (>12 years), and more than 60% are symptomatic. Oral tracheal intubation was diffi cult in 10% (n = 24) of the cases, and 7 required the use of a fi brobronchoscope. In 8 cases (3%) a thoracic approach was necessary. Mor-bidity occurred in 24% (n = 59), most notably 29 recur-ring lesions (12%), of which 2 were defi nitive (0.8%), and 31 hypoparathyroidisms (13%), of which 1 was defi nitive (0.4%). No signifi cant difference was found in the postsurgical morbidity between the intrathoracic MG and the nonintrathoracic cases. Regarding the remission of the symptoms, the results were excellent. In 14 cases (5.7%) thyroid carcinoma was related with, most of these being papillary microcarcinoma. In 10 of the 49 cases of partial surgery (20%) a relapse of the goiter was observed.Conclusions. Intrathoracic MG is usually asymptomatic and it occurs in goiter with a long time of evolution. Surgery is a good therapeutic option given that the

goiter can be removed via the neck, with low morbidity, a remission of the symptoms, malignancy is ruled out, and recurrence can be avoided if a total thyroidectomy is performed.

Key words Multinodular goiter · Surgery · Intrathoracic goiter · Postsurgical complications · Compressive symptoms · Relapse · Thyroid carcinoma

Introduction

Since its fi rst anatomical description in 1749 by Haller, intrathoracic goiter has been defi ned using many dif-ferent terms, among them retrosternal, subclavicular, mediastinal, etc. Its most frequent etiology is multi-nodular goiter (MG), given that in its natural evolution it tends to develop multinodularity, functional auton-omy and it is usually located in the retrosternal thyroid.1–3 A principal problem is that it involves an entity which is not very well defi ned, without there being any una-nimity over the quantity of the thyroid gland which should be located in the thorax or at what level it is located. This causes an uneven incidence in the differ-ent series depending on the criteria used, with a rate which ranges from 0.2% to 45% of the total number of goiters.2,4–8 In general, the incidence of intrathoracic goiter is correlated to the goiter, and in this way it is more frequent in goiter-endemic areas.

The treatment of choice is surgery, although this has been described as a risk factor for causing defi nitive complications, such as hypothyroidism as well as recur-rent lesions.2,9,10 However, from 3% to 15% can develop malignant degeneration. In addition, its intrathoracic growth can potentially lead to morbidity (sharp respira-tory insuffi ciency due to compression of the airway, etc.).2,4,5,11 The objectives of this study are to: (1) deter-mine the clinical profi le of intrathoracic MG, (2) evalu-

488 A. Ríos et al.: Surgery and Intrathoracic Goiter

ate the results of surgery in these patients, and (3) analyze the incidence of malignancy and its evolution in these cases of intrathoracic MG.

Patients and Methods

Selection Criteria

A retrospective study was carried out on patients diag-nosed and operated on for MG between January 1970 and January 2000. The selection criteria were: (1) Uni- or bilateral MG, defi ned as that which presents more than one nodule in the thyroid lobe on cervical explora-tion; (2) patients with an intrathoracic component have been selected according to Eschapase’s defi nition,12 which considers the goiter to be intrathoracic when the goiter is located either totally or partially in the medi-astinum, while in the operating position it has its lower edge at least 3 cm below the sternal manubrium; (3) no associated parathyroid pathology; and (4) a diagnosis of MG in the histological study.

Description of Patients

The average age of the selected 247 cases of MG was of 60 ± 12 years and most were women (87%; n = 214), 24% (n = 59) residing in goiter-endemic areas and 6.5% (n = 16) with family medical history of thyroid disease. The average time of evolution was of 148 ± 141 months, with 63% (n = 155) being symptomatic, mainly with compressive symptoms (48%; n = 119) and hyperthy-roidism (18%; n = 45). A simple X-ray was carried out on both patients, showing 81% of the cases to be of tracheal or mediastinal invasion. A computed tomogra-phy (CT) scan was carried out on 49 of these (20%), in order to delimit its anatomical relations. In the cases of dysphonia, laryngoscopy was carried out to identify the presence of recurrent paralysis. Scintigraphy was carried out in the 45 (18%) patients with hyperthyroidism and showed the presence of one or more hot nodules.

Regarding the location of goiter at the thoracic level, 99% (n = 244) had anterior retrosternal spread and the remaining 1% (n = 3) posterior. Two of these cases of posterior location were not related to the cervical com-ponent of the goiter; in other words, they were ectopic mediastinal goiters.

Study Variables

The variables analyzed were as follows.

Socio-Personal VariablesAge; sex; family history of thyroid pathology; residence in goitergenic endemic areas in our community.

Clinical VariablesTime of evolution of MG before surgery (defi ned as the time between the patient reporting the fi rst symptoms of goiter and the time of surgery); being asymptomatic; hyperthyroidism; compressive syndromes; consistency of the goiter on physical examination; the grade of cer-vical goiter [grade 0 (cervical goiter not objectivized), grade I (it is not seen but can be palpated), grade II (it can be seen and palpated), and grade III (affects neigh-boring structures)]; alteration in the simple X-ray.

SurgeryThe indications for surgery; diffi culty for anesthetic intubation; endocrine surgery experience of the surgeon who carried out the fi rst operation, thus meaning that the surgeon had previously carried out more than 100 thyroid surgery operations;13 surgical technique [Total thyroidectomy; Dunhill technique (hemithyroid-ectomy with contralateral subtotal resection leaving a remnant of about 2 g); bilateral subtotal thyroidectomy (bilateral subtotal thyroid resection leaving 2 g thyroid remnants on both sides); hemithyroidectomy; and uni-lateral subtotal hemithyroidectomy (subtotal thyroid resection leaving a 2 g thyroid remnant)]; duration of the surgery; the weight of the goiter; the need for a sternotomy.

Postoperative Morbidity and MortalityThe following postoperative complications14 were noted: (1) Hypoparathyroidism: considered as occurring during the postoperative period with calcium fi gures below 7.5 mg/dl or when the patient presented symptoms of hypocalcaemia with fi gures below 8.5 mg/dl. If the cal-cemia remained below 8.5 mg/dl one year after surgery it was labeled as permanent. (2) Recurrent laryngeal nerve injury: considered to be an alteration in the tone, timbre or intensity of the voice following surgery and with vocal chord paralysis confi rmed by laryngoscopy. It was labeled as permanent when persisting for more than 12 months. (3) Superior laryngeal nerve lesion: normal voice which after speaking reveals a loss of tone and timbre, with normal laryngoscopy. (4) Surgical wound complications. (5) Systemic complications. Postoperative mortality was also evaluated.

Related Thyroid CarcinomaNumber and type; and evolution.

EvolutionRemission of symptoms; relapse of goiter (in patients with partial surgery the relapse is defi ned as the pres-ence of new thyroid nodules in the routine ultrasonog-raphy test); and secondary surgery for relapse.

A. Ríos et al.: Surgery and Intrathoracic Goiter 489

Control Group

To determine the distinguishing features of intratho-racic MG, a comparative study was carried out between the patients with intrathoracic goiter and the group of 425 cases of MG without an intrathoracic component (these patients were operated on during this period of time in our department).

Statistical Analysis

In the statistical analysis, the χ2 test was used, with Yates’ correction or Fisher’s exact test if necessary, and Student’s t-test. In the cases in which it was necessary to apply a nonparametric test, Wilcoxon’s test was applied. For determining and evaluating multiple risks, a logistical regression analysis was carried out using the

variables which were statistically signifi cant in the bivar-iate analysis. The analyses were performed using the SPSS statistical package, 10.0 version. Differences were considered to be signifi cant for levels of P < 0.05.

Results

Clinical Profi le of Intrathoracic MG

Intrathoracic MG has a series of characteristics which distinguishes it from other types of MG (Tables 1 and 2). Therefore, in a bivariate analysis (Table 1) we observed that it occurs more often in older patients (60 versus 47 years; P < 0.00001), it is more frequent in males (13% versus 5%; P = 0.0002), the goiter has a longer time of evolution (148 versus 80 months;

Table 1. Variables related to intrathoracic multinodular goiter (MG): bivariate analysis

Variable Intrathoracic MG (n = 247) Nonintrathoracic MG (n = 425) P

Age (years) 60 ± 12 47 ± 15 <0.00001Sex 0.0002 Female (n = 617) 214 (87%) 403 (95%) Male (n = 55) 33 (13%) 22 (5%)Family medical history of goiter 0.5361 No (n = 623) 231 (94%) 392 (92%) Yes (n = 49) 16 (6%) 33 (8%)Residence in goiter-endemic area 0.3362 No (n = 525) 188 (76%) 337 (79%) Yes (n = 147) 59 (24%) 88 (21%)Evolution time (months) 148 ± 141 80 ± 100 <0.00001Asymptomatic <0.00001 No (n = 274) 155 (62.8%) 119 (28%) Yes (n = 398) 92 (37.2%) 306 (72%)Hyperthyroidism 0.4105 No (n = 560) 202 (82%) 358 (84%) Yes (n = 112) 45 (18%) 67 (16%)Compressive syndromes <0.00001 No (n = 515) 129 (52%) 387 (91%) Yes (n = 157) 119 (48%) 38 (9%)Consistency of the goiter 0.0769 Hard (n = 124) 37 (15%) 87 (20%) Elastic (n = 548) 210 (85%) 338 (80%)Alteration in the X-ray <0.00001 No (n = 396) 47 (19%) 349 (82%) Yes (n = 276) 200 (81%) 76 (18%)

Table 2. Variables related to intrathoracic MG: multivariate analysis

Variable Regression coeffi cient (β) Standard error Odds ratio P

Age 0.03202 0.0073 <0.00001Sex 0.0032 Female (n = 617) 1 Male (n = 55) 0.49280 0.1689 1.64 (2.3–1.2)Compressive syndromes <0.00001 No (n = 515) 1 Yes (n = 157) 0.98615 0.1190 2.68 (3.4–2.1)

490 A. Ríos et al.: Surgery and Intrathoracic Goiter

P < 0.00001), most are symptomatic (63% versus 28%; P < 0.00001) with a high incidence of compressive syn-dromes (48% versus 9%; P < 0.00001), and they often have alterations in the simple X-ray fi ndings (81% versus 18%; P < 0.00001). In a multivariate analysis (Table 2), the variables with an independent relation-ship are age, male sex (relative risk = 1.6), and the pres-ence of compressive symptoms (relative risk = 2.7).

Surgery and Related Morbidity and Mortality

All the patients were operated on. During anesthesia there was diffi culty in intubation in 24 of the 247 cases (10%), with 7 of these requiring the use of a fi beroptic bronchoscopy for intubation. In all of these cases showing diffi culty with intubation, a simple X-ray showed a deviation or compression of the tracheal air column. In the cases of goiter without an intrathoracic component there was intubation diffi culty in only 4 patients (1%), and none of them required the use of a fi beroptic bronchoscopy (Table 3). In 8 cases (3%) it was necessary to carry out a thoracic approach to reach the thyroid to be able to extract it. In six cases it was done by a half sternotomy and in the two remaining cases, which correspond to ectopic goiters with poste-rior mediastinal location, a right posterolateral thora-cotomy was carried out. The technique used in 79% of the patients (n = 195) was a total thyroidectomy, while in the three cases of intrathoracic goiter which had already had previous thyroid surgery, a thyroidectomy

was carried out, and in the remaining patients partial techniques were carried out, as can be seen in Table 3. In 7 cases (2.8%) it was concluded that one of the para-thyroid glands had become devascularized or had been extirpated, and therefore a parathyroid auto transplant was carried out in the sternocleidomastoid muscle. The average length of surgery was 124 ± 43 min.

Morbidity occurred in 24% (n = 59), including 29 recurrent lesions (12%), 2 of them defi nitive (0.8%), and there were 31 hypoparathyroidisms (13%), 1 of them defi nitive (0.4%). Three (1.2%) had cervical hemorrhages, with one of these requiring an urgent tracheotomy. No signifi cant difference were found in post-surgical morbidity between the patients operated on for intrathoracic goiter and those operated on for nonintrathoracic goiter (P = 0.336). In those patients with a parathyroid autotransplant transitory hypopara-thyroidism was observed. Table 4 shows the distribu-tion of the hypoparathyroidisms and recurrent lesions according to the surgical technique carried out. A higher incidence of transitory lesions is seen for a total thyroidectomy in comparison to other techniques (P < 0.0001).

In 3 patients (1.2%) the characteristics of the trachea during surgery (weakness of the tracheal wall and partial collapse during surgery) led to a suspicion of tracheo-malacia. Therefore, extubation was done under strict supervision with a team ready for a possible emergency reintubation. In two patients extubation was carried out in the fi rst 12 h without incident, using a normal control

Table 3. Intraoperative and histological fi ndings of intrathoracic MG

Variable Intrathoracic MG (n = 247) Nonintrathoracic MG (n = 425) P

Anesthetic intubation diffi culty <0.00001 No (n = 644) 223 (90%) 421 (99%) Yes (n = 28) 24 (10%) 4 (1%)Surgeon with experience 0.0044 No (n = 140) 37 (15%) 103 (24%) Yes (n = 532) 210 (85%) 322 (76%)Thoracic approach 0.0002 No (n = 664) 239 (97%) 425 (100%) Yes (n = 8) 8 (3%) 0 (0%)Surgical technique <0.00001 TT (n = 413) 195 (79%) 218 (51%) TS (n = 43) 19 (8%) 24 (6%) Dunhilla (n = 54) 10 (4%) 44 (10%) Complete thyroidectomy (n = 28) 3 (1%) 25 (6%) Hemithyroidectomy (n = 109) 20 (8%) 89 (21%) Sub-hemithyroidectomy (n = 25) 0 (0%) 25 (6%)Surgical time (min) 124 ± 43 100 ± 34 <0.00001Weight of extracted thyroid (grams) 171 ± 169 75 ± 75 <0.00001Related cancer <0.00001 No (n = 613) 233 (94%) 380 (89%) Yes (n = 59) 14 (6%) 45 (11%)

a Hemithyroidectomy with contralateral subtotal resection leaving a remnant of about 2 g

A. Ríos et al.: Surgery and Intrathoracic Goiter 491

laryngoscope. In the third patient extubation caused dyspnea with stridor, with the laryngoscope confi rming the diagnosis of tracheomalacia. This case eventually resolved after a prolonged orotracheal intubation (36 h).

Evolution

The results of surgery showed an excellent remission of the symptoms. In this way, the symptoms derived from hyperthyroidism remitted immediately in all cases (n = 45) after intervention, although in cases with ophthal-mopathy they disappeared in a progressive way during the two to three months after surgery. With regard to compressive symptoms, tracheal compressive syndrome (n = 74) and esophageal compressive syndrome (n = 53) remitted immediately after surgery. In the 23 patients with preoperative dysphonia due to elongation–compression of the recurrent laryngeal nerve, the dys-phonia improved progressively until it completely disappeared over a period of 2–30 days, except in one of the cases in which it persisted as a defi nitive recurrent lesion. Patients with superior vena cava syndrome (n =

7) and Claude–Bernard–Horner syndrome (n = 1) had various simultaneous compressive syndromes, namely between three and four, with all of them becoming asymptomatic before the 7th postoperative day.

The weight of the extracted piece was 171 ± 169 g, and in 14 cases (5.7%) a thyroid carcinoma was related, 11 of these being microcarcinomas. All were well differen-tiated and papillary, except for one which was follicular. Of the papillary cases, there was only one case of cap-sular invasion, without any cases of vascular or ganglia invasion, although four cases were multifocal. On the other hand, in the case of follicular carcinoma, vascular, capsular, and ganglia invasion was observed (Table 5). Currently, all the patients are disease free, as can be seen in Table 5.

In 10 of the 49 patients (20%) in which partial surgery was carried out, there was relapse of the goiter, with an average relapse time of 93 ± 32 months after surgery (Table 6). The rate of relapse after 5 years was 10% for the Dunhill technique, 6% for the bilateral subtotal thyroidectomy, and 7% for hemithyroidectomy, with these rates increasing after 10 years up to 14%, 43% and 21%, respectively, as can be seen in Table 6.

Table 4. Distribution of the recurrent lesions and of hypothyroidism according to the surgical technique carried out

Surgical technique

Postoperative hypoparathyroidism

(n = 31)Hypoparathyroidism or defi nitive (n = 1)

Postoperative recurrent lesion

(n = 29)

Defi nitive recurrent lesion

(n = 2)

Total thyroidectomy (n = 195) 29 1 28 2Bilateral subtotal thyroidectomy (n = 19) 0 0 0 0Dunhill technique (n = 10) 2 0 0 0Complete thyroidectomy (n = 3) 0 0 0 0Hemithyroidectomy (n = 20) 0 0 1 0

Table 5. Evolution of thyroid carcinomas related to intrathoracic MG

Case Carcinoma SizeNo. of

fociVascular invasion

Capsular invasion

Ganglia invasion

Evolution time (months) Current situation

1 Papillary 3 1 — — — 34 Disease free 2 Papillary 3 1 — Yes — 34 Disease free 3 Papillary 0.9a 3 — — — 24 Disease free 4 Papillary 0.6 1 — — — 72 Disease free 5 Papillary 0.6 1 — — — 73 Disease free 6 Papillary 0.6 1 — — — 123 Disease free 7 Papillary 0.5a 3 — — — 36 Disease free 8 Papillary 0.4 1 — — — 133 Disease free 9 Papillary 0.3 1 — — — 36 Disease free10 Papillary 0.3 1 — — — 24 Disease free11 Papillary 0.3 1 — — — 24 Disease free12 Papillary 0.3a 3 — — — 97 Disease free13 Papillary 0.1a 3 — — — 73 Disease free14 Follicular 1.1a 3 Yes Yes Yes 78 Disease free

a Largest tumoral foci size

492 A. Ríos et al.: Surgery and Intrathoracic Goiter

Discussion

As mentioned in the Introduction, there is no consen-sual defi nition for intrathoracic goiter. In our service, in the same way as other authors,7,15 we have opted to accept Eschapase’s defi nition12 so that a goiter which exceeds the eternal manubrium by 3 cm in hypertension may potentially produce the same complications as any other which enters the mediastinum. It has to be remembered that compressive syndrome is generally produced by compression at the level of the entry point to the mediastinum, as the goiter compresses the structures between the manubrium and the vertebral column. Multinodular goiter tends to become intratho-racic if it is allowed to evolve. Lahey and Swinton16 described the anatomical factors which encourage its retrosternal development. They observed that the thyroid gland is limited by rigid structures (vertebrae, cervical fascia, thyroid and cricoid cartilage, etc), except at its lower limit, thus representing the part of less resistance. In addition, it benefi ts from negative intrathoracic pressure, descending traction produced by deglution and the weight of the thyroid gland (gravity mechanism).11

In our MG global series, there is an intrathoracic component in 37%, characterized by patients with an older average age, around 60 years, due to the fact that the goiter takes time to reach this size.2,17 In this way, the average goiter evolution time is greater than 12 years. In our geographical area, being a male with MG is a risk factor for developing an intrathoracic compo-nent, with a relative risk of 1.6 with respect to women, as can be seen in the statistical analysis in Tables 1 and 2. In addition, compressive symptoms are characteristic in more than 50% of the cases, most commonly dyspnea and dysphagia, which is different from cervical MG which lacks compressive symptoms in most cases.

A simple X-ray, together with a good physical exami-nation, were the two main factors which allowed for a diagnosis of intrathoracic goiter, and the CT was reserved for those goiters where a large mediastinal tumor could be seen in a X-ray. This diagnostic protocol allowed us to diagnose most of the patients and to plan surgery. We did not consider it necessary to routinely carry out a CT scan in all the patients with an intratho-racic component as recommended by other authors.11 In

our series CT is only carried out to clarify relationships with neighboring structures, above all for evaluating infi ltration when there is suspicion of a possible carci-noma, and for limiting its spread towards the media-stinum and its possible extension. In this situation, we would thus perform a sternotomy instead of a cervicotomy.

With regard to surgical handling, oral tracheal intuba-tion diffi culty during anesthesia in 10% of the cases, with 7 of these (2.8%) needing the use of a fi beroptic bronchoscopy. Therefore, we consider it to be necessary to have the surgeon ready in the case of an emergency intubation, since it is necessary in less than 3% of cases, as most cases are resolved by reducing the size of the anesthetic intubation tube or using a guiding wire.

A thoracic approach for reaching the thyroid was necessary in 3% of patients, with the choice of tech-nique being the half sternotomy, except in cases of ectopic goiter with posterior location. Currently, this surgical approach remains controversial in posterior ectopic goiter, and although it can be extirpated through the neck,18 most authors opt for a tracheotomy or a combined cervicotomy–thoracotomy approach.19 If it is a superior mediastinal goiter originating from a contra-lateral cervical goiter, the best approach described is the combined one, as it allows for optimum control of the mediastinal and cervical blood vessels, protection of the recurrent nerve, and a complete extirpation of the cervical and thoracic component of the goiter.19 In our series there were no cases with these characteristics.

The chosen technique for an extirpation of intratho-racic goiter in our service is a total thyroidectomy, which has been carried out in nearly 80% of the cases, with this being the technique used in more than 97% of goiters in the last 10 years. Furthermore, in the cases with partial techniques there has been relapse in more than 20% of the cases. Currently, the only partial resec-tion technique that we use is a hemithyroidectomy. It is reserved for exceptional cases of multinodular goiter that are predominantly unilateral and in which the hemithyroid gland that is not extirpated is cervical and does not have any evident sign of nodular disease.

With respect to morbidity, there is a high rate of postsurgical complications, although most are transi-tory. The rate of defi nitive complications was low, 0.4% for hypoparathyroidisms and 0.8% for recurrent lesions,

Table 6. Goiter relapse in patients operated on by means of partial surgical techniques

Technique 2 years 5 years 7 years 10 years

Dunhill technique (n = 10) 0% (0/10)a 10% (1/10) 11% (1/9) 14% (1/7)Subtotal bilateral thyroidectomy (n = 19) 0% (0/19) 6% (1/18) 12% (2/17) 43% (6/14)Hemithyroidectomy (n = 20) 0% (0/20) 7% (1/15) 13% (2/15) 21% (3/14)

a Percentage relapse (cases of relapse / cases with post-surgical follow-up according to number of years)

A. Ríos et al.: Surgery and Intrathoracic Goiter 493

which are on par with the best results obtained with partial techniques. No signifi cant differences in compli-cations have been observed between intrathoracic MG and MG without an intrathoracic component. This is possibly because of a higher percentage of surgeons with experience in endocrine surgery treating intratho-racic goiter compared to those treating the goiter group without this component.

Tracheomalacia (Herzog’s Syndrome) is character-ized by a loss of thyroid cartilage integrity due to pro-longed tracheal compression or invasion.20 The incidence of tracheomalacia in MG ranges from 0.001% to 1.5%11,21 and it is more frequent in intrathoracic and large goiters.2 As it is uncommon, and many authors such as Wade,22 Melliere et al.,23 and Shaha et al.24 have not seen it in years of experience, routine tracheopexy or tracheot-omy procedures are not recommended when it is sus-pected.9 The most accepted therapeutic technique when there is suspicion of tracheomalacia is intubation for 24–48 h and extubation under controlled conditions with strict patient supervision,2,24 and occasionally a tracheotomy.2,11,25 Some authors prophylactically rec-ommend tracheotomy,11 and others such as Allo and Thompson26 advocate a tracheopexy. This occurred in one patient in our series, thus accounting for less than 0.5% of the cases, although it was suspected in three. What is most important is to suspect the possibility of tracheomalacia in order to undertake a later extubation under strict control, thus avoiding problems of acute respiratory insuffi ciency after extubation, and particu-larly to have everything ready for an emergency intuba-tion so as not to delay a situation that can jeopardize the patient’s life.

The results are excellent regarding the remission of the symptoms after surgery, with this remission being permanent in patients in which a total thyroidectomy has been carried out. It is noteworthy that there is a recovery from recurrent paralysis in 22 of the 23 cases after it occurs. The paralysis which persisted as defi ni-tive could have occurred because the lesion was already defi nitive, as it was one of the two patients in which dysphonia had a time of evolution of more than 6 months prior to surgery and it is known that this is a negative prognostic factor. It could also have been caused by an intraoperative nerve lesion.

Regarding the partial techniques, the relapse rates vary according to the series, being 20% in our series, although not all were asymptomatic. With these tech-niques it has been seen that the rate of recurrence of hyperthyroidism is directly proportional to the size of the remnant which is left, which causes oscillation in relapse rates of between 5%–30%. On the other hand the size of the remnant is inversely related to the devel-opment of hypothyroidism, in other words, the less relapse there is, the higher the hypothyroidism.27–31

Mittendorf et al.32 considered subtotal thyroidectomy not to be a good technique for treating toxic MG. Therefore, if the remnant is large there will be hyper-thyroidism and another type of treatment will be neces-sary or even surgical reintervention with the consequent risks, and on the other hand, if the remnant is small it will remain hypothyroid and there will be no advantage over total thyroidectomy as there will have to be sub-stitutive treatment with thyroxine. Mittendorf et al.32 consider the Dunhill technique to be a good alternative; however, we have also observed relapse with this tech-nique. The results for compressive symptoms have also been excellent, as far as remission from clinical symp-toms is concerned. Therefore, an immediate remission of all the compressive symptoms was observed, except in one case of dysphonia. These data confi rm those obtained by McHenry and Piotrowski,9 who was able to eliminate 100% of the compressive symptoms in a series of large goiters, with a rate of complications comparable to that of the patients having thyroidectomy for a smaller sized thyroid and without there being mortality, as in our series. We have to point out that in our cases the evaluation of the relapse is through neck explora-tion, using ultrasonography to confi rm relapse. This is assuming that we do not detect nor include any subclini-cal ultrasound relapses in our index.

The rate of malignancy related to these cases of goiter is low, and it mainly corresponds to papillary microcar-cinomas.33 There was in fact one case of follicular car-cinoma with vascular, capsular, and lymphatic invasion. Its evolution has been favorable, without any tumoral relapse.

In conclusion, we can say that patients with intratho-racic goiter are characterized by having a greater evolu-tion of the goiter and having compressive symptoms, with the risk of developing this type of goiter being higher in males. Surgery is a good therapeutic option as it can be carried out in most cases through the neck, with a low level of morbidity, it achieves a remission of the symptoms, rules out malignancy, and if the tech-nique used is a total thyroidectomy, then recurrence can thus be avoided.

References

1. Hurley DL, Gharib H. Evaluation and management of multi-nodular goiter. Otolaryngol Clin North Am 1996;29:527–40.

2. Singh B, Lucente FE, Shaha AR. Substernal goiter: a clinical review. Am J Otolaryngol 1994;15:409–16.

3. Wiersinga WM. Determinants of outcome in sporadic nontoxic goiter. Thyroidology 1992;4:41–3.

4. Rodríguez JM, Hernández Q, Piñero A, Ortiz S, Soria T, Ramirez P, et al. Substernal goiter: clinical experience of 72 cases. Ann Otol Rhinol Laryngol 1999;108:501–4.

5. Madjar S, Weissberg D. Retrosternal goiter. Chest 1995;108:78–82.

494 A. Ríos et al.: Surgery and Intrathoracic Goiter

6. Wong CKM, Wheeler MH. Thyroid nodules: rational manage-ment. World J Surg 2000;24:934–41.

7. Alvarez A, Cerezo F, Baamonde C, Salvatierra A, López FJ. Tratamiento quirúrgico del bocio cervicomediastínico (in Spanish with English abstract). Arch Bronconeumol 1997;33:84–8.

8. Buckley JA, Stark P. Intrathoracic mediastinal thyroid goiter. AJR 1999;173:471–5.

9. McHenry CR, Piotrowski JJ. Thyroidectomy in patients with marked thyroid enlargement: airway management, morbidity, and outcome. Am Surg 1994;60:586–91.

10. Hall TS, Caslowitz P, Popper C, Smith GW. Substernal goiter versus intrathoracic aberrant thyroid: a critical difference. Ann Thorac Surg 1988;46:684–5.

11. Newman E, Shaha AR. Substernal goiter. J Surg Oncol 1995;60:207–12.

12. Dahan M, Gaillard J, Eschapase H. Surgical treatment of goiters with intrathoracic development. In: Delarue NC, Eschapase H, editors. Thoracic surgery: frontiers and uncommon neoplasms. International trends in general thoracic surgery. St Louis: Mosby; 1989. p. 5.

13. Sosa JA, Bowman HM, Tielsch JM, Powe NR, Gordon TA, Udelsman R. The importance of surgeon experience for clinical and economic outcomes from thyroidectomy. Ann Surg 1998;228:320–30.

14. Reeve T, Thompson NW. Complications of thyroid surgery: how to avoid them, how to manage them, and observations on their possible effect on the whole patient. World J Surg 2000;24:971–5.

15. García MJ, Arnau A, Martínez P, Vazquez A, Martín E, Cantó A. Tratamiento del bocio cervicomediastínico (in Spanish with English abstract). Cir Esp 2000;67:268–72.

16. Lahey FH, Swinton MW. Intrathoracic goiter. Surg Gynecol Obstet 1934;59:627–37.

17. Katlic MR, Wang C, Grillo HC. Substernal goiter. Ann Thorac Surg 1985;39:391–9.

18. DeAndrade MA. A review of 128 cases of posterior mediastinal goiter. World J Surg 1977;1:789–97.

19. Shahian DM, Rossi RL. Posterior mediastinal goiter. Chest 1988;94:599–602.

20. Geelhoed GW. Tracheomalacia from compressing goiter: man-agement after thyroidectomy. Surgery 1988;104:1100–8.

21. Vadasz P, Kotsis L. Surgical aspects of 175 mediastinal goiters. Eur J Cardiothorac Surg 1998;393–7.

22. Wade JSH. Respiratory obstruction in thyroid surgery. Ann R Coll Surg Engl 1980;62:15–24.

23. Melliere D, Saada F, Etienne G, Becquemin JP, Bonnet F. Goiter with severe respiratory compromise: evaluation and treatment. Surgery 1988;103:367–73.

24. Shaha AR, Burnett C, Alfonso A, Jaffe BM. Goiters and airway problems. Am J Surg 1989;158:378–81.

25. Green WE, Shepperd HW, Stevenson HM, Wilson W. Tracheal collapse after thyroidectomy. Br J Surg 1979;66:554–8.

26. Allo MD, Thompson NW. Rationale for the operative manage-ment of substernal goiters. Surgery 1983;94:969–77.

27. Menegaux F, Turpin G, Dahman M, Leenhardt L, Chadarevian R, Aurengo A, et al. Secondary thyroidectomy in patients with prior thyroid surgery for benign disease: a study of 203 cases. Surgery 1999;125:479–83.

28. Mishra A, Agarwal A, Agarwal G, Mishra SK. Total thyroidec-tomy for benign thyroid disorders in an endemic region. World J Surg 2001;25:307–10.

29. Martos JM, Pérez M, del Valle A, Naranjo JR, Perez MI, Sousa JM, et al. Hipoparatiroidismo postquirúrgico (in Spanish with English abstract). Cir Esp 1998;63:25–9.

30. Tweedle D, Colling A, Schardt W, Green EM, Evered DC, Dick-inson PH, et al. Hypothyroidism following partial thyroidectomy for thyrotoxicosis and its relationship to thyroid remnant size. Br J Surg 1977;64:445–8.

31. Rios A, Rodriguez JM, Balsalobre MD, Torregrosa NM, Tebar FJ, Parrilla P. Results of surgery for toxic multinodular goiter. Surg Today 2005;35:901–6.

32. Mittendorf EA, McHenry CR. Thyroidectomy for selected patients with thyrotoxicosis. Arch Otolaryngol Head Neck Surg 2001;127:61–5.

33. Yamashita H, Noguchi S, Watanabe S, Uchino S, Kawamoto H, Toda M, et al. Thyroid cancer associated with adenomatous goiter: an analysis of the incidence and clinical factors. Surg Today 1997;27:495–9.