Recurrent laryngeal nerve palsy andsubsternal goiter. An Italian multicenterstudy
M. Testini a,∗, A. Gurradoa, R. Bellantoneb,P. Brazzarolac, R. Cortesed, G. De Tomae,I. Fabiola Francoa, G. Lissidini a, C. Pio Lombardib,F. Minervaf, G. Di Meoa, A. Pasculli a, G. Piccinnia,L. Rosatog
a Department of Biomedical Sciences and Human Oncology, Unit of Endocrine, Digestive andEmergency Surgery, University Medical School ‘‘A. Moro’’ of Bari, Bari, Italyb Department of Surgery, Unit of Endocrine Surgery, University Medical School ‘‘Cattolica delSacro Cuore’’, Rome, Italy
Please cite this article in press as: Testini M, et al. Recurrenmulticenter study. Journal of Visceral Surgery (2014), http://d
c Department of Surgery, Unit of General Surgery, University Medical School of Verona,Verona, Italyd Department of Neurological anUniversity Medical School ‘‘A. Me Department of Surgery, Unit oSapienza’’, Rome, Italyf Department of Biomedical ScieUniversity Medical School ‘‘A. Mg Department of Surgery, Endoc
KEYWORDSRecurrent laryngealnerve palsy;Total thyroidectomy;Substernal goiter;Cervico-mediastinalgoiter;Thyroid surgery
Summary The aim of this retrnal goiter and the type of surgicaduring total thyroidectomy. Bettomy. Patients were divided intgoiters treated through collar icervical approach; group C (n =
d Psychiatric Sciences, Unit of Physical and Rehabilitation,
t laryngeal nerve palsy and substernal goiter. An Italianx.doi.org/10.1016/j.jviscsurg.2014.04.006
oro’’ of Bari, Bari, Italyf General Surgery ‘‘P. Valdoni’’, University Medical School ‘‘La
nces and Human Oncology, Unit of Medicine ‘‘A. Murri’’oro’’ of Bari, Bari, Italy
rine Surgical Unit, Ivrea Hospital, Ivrea, Italy
ospective multicenter study was to verify whether the subster-l access could be risk factors for recurrent laryngeal nerve palsyween 1999—2008, 14,993 patients underwent total thyroidec-o three groups: group A (control; n = 14.200, 94.7%), cervicalncision; group B (n = 743, 5.0%) substernal goiters treated by50, 0.3%) in which a manubriotomy was performed. Transientoccurred significantly more frequently in B + C vs. A (P ≤ .001)nt bilateral palsy was significantly more frequent in B + C vs. A6). Permanent bilateral palsy was significantly more frequent
vs. A (P ≤ .037). Extension of the goiter into the mediastinum of recurrent nerve palsy during total thyroidectomy.rights reserved.
Figure 1. A multidetector computed tomography scan with mul-tiplanar reformatting and volume-rendering reconstruction of theneck and chest shows an extension of the thyroid mass into theupper mediastinum.
patients underwent TT performed by highly experiencedsurgeons (> 100 thyroidectomies per year during the last5 years), using a standardized capsular dissection technique.Hemostasis was achieved with individual vascular controland division of the thyroid vessels using conventional lig-ature or with the Harmonic scalpel (Harmonic Wave and
ARTICLE+ModelVS-393; No. of Pages 7
ntroduction
lthough the standardized capsular dissection technique isharacterized by low morbidity and virtually no mortality inatients with cervical goiters [1—4], especially when per-ormed in high-volume centers, there is still a relativelyigher reported incidence of postoperative hypoparathy-oidism and recurrent laryngeal nerve (RLN) palsy in casesf substernal goiter [5—7]. RLN palsy following total thy-oidectomy (TT) is a serious complication, which not onlyas a large clinical impact, but also has a medicolegal ones it can lead to accusations of malpractice [4,8].
The definition of substernal goiter is not standardized5,9,10]. Some studies considered only patients with anxtension of > 50% of the thyroid gland beyond the thoracicnlet [11], while others included cases where any part ofhe gland extended below the thoracic inlet [5,9]. There-ore, due to the differences in defining criteria, the recordedncidence of substernal goiter varies from 3% to 20% ofatients undergoing thyroidectomy [5—7,9—12]. The naturalvolution of substernal goiter is progressive growth, causingmpingement on surrounding anatomical structures, onsetf compressive symptoms, and higher risk of malignancy.herefore, TT, performed through a cervical or extracervi-al approach, is mandatory for the treatment of substernaloiter.
The aim of this study was to evaluate whether extensionf a goiter into the mediastinum was a risk factor for RLNnjury during TT performed using cervical or extracervicalpproaches.
atients and methods
his study was a multicenter retrospective analysis per-ormed on a population of patients that underwent TT.etween January 1999 and December 2008, patients whonderwent TT in five Italian Departments of Surgery withigh-volume of thyroidectomy (> 100 per year) were ana-yzed. Exclusion criteria were redo-surgery, concomitantrimary hyperparathyroidism, anaplastic carcinoma, need
Please cite this article in press as: Testini M, et al. Recurrenmulticenter study. Journal of Visceral Surgery (2014), http://d
or lymph node dissection, extensive surgery (i.e., laryn-ectomy plus thyroidectomy), subtotal and near totalhyroidectomy, lobo-isthmectomy, minimally invasive video-ssisted thyroidectomy (MIVAT), use of nerve monitoringystems, and patients lost during the follow-up.
After application of these exclusion criteria, 14,993atients were included in the study. Only patients inhom > 50% of the gland was located below the clavicleere considered to have a substernal goiter. The diag-osis was always confirmed by a multidetector computedomographic (CT) scan with multiplanar reformatting andolume-rendering reconstructions of the neck and chestFig. 1). All patients underwent a preoperative workuphat included measurement of thyroid function and auto-ntibodies, serum measurements of calcium, inorganichosphorus, and magnesium, plain chest and neck radiogra-hy. An evaluation of preoperative vocal cord function waserformed in all patients.
The study population was divided into three groups:roup A (control; n = 14,200, 94.7%), including patients withervical goiter treated by TT through a cervical collarncision (Fig. 2); group B (n = 743, 5.0%) with substernaloiter treated through a cervical incision (Fig. 3); group
(n = 50, 0.3%) in which TT was performed through aanubriotomy (Fig. 4). After written informed consent,
Fs
t laryngeal nerve palsy and substernal goiter. An Italianx.doi.org/10.1016/j.jviscsurg.2014.04.006
igure 2. Total thyroidectomy performed for a cervical goiter:chematic drawing.
Substernal goiter and recurrent laryngeal nerve palsy
specifically, the mean age ± standard deviation was 55 ± 16
Figure 3. Total thyroidectomy performed for a substernal goiterthrough a cervical approach: schematic drawing.
Please cite this article in press as: Testini M, et al. Recurrenmulticenter study. Journal of Visceral Surgery (2014), http://d
Harmonic Focus, Ethicon EndoSurgery, Cincinnati, OH), andwas improved using bipolar scissors, and forceps, and bio-surgical agents, as also reported in our previous experience[13]. Both monopolar and bipolar electrocautery was alwaysavoided in close proximity to the nerves or parathyroidglands. When an intraoperative RLN transection was rec-ognized, it was repaired with a direct neurorrhaphy, andthe anastomosis was made with three or four stitches of 8-0or 9-0 nylon suture using microsurgical instruments and anoperating microscope.
In each patient, follow-up at 1 year was accomplishedby clinical visits, telephone or e-mail interviews. An inde-pendent and blinded ENT specialist performed a clinicalvoice assessment and fiberoptic laryngoscopy on all patientswith hypo-functioning vocal cords at the time of extubationor with an immediate or late postoperative voice change,dyspnea, or dysphagia. Vocal cord function was defined asnormal, unilateral or bilateral paresis (hypomotility withoutparamedian paralysis) or paralysis (absence of motility andparamedian position). In the analysis, paretic or paralyticvocal cords were grouped together as RLN palsy. When RLNpalsy was identified, the patient was seen in weekly follow-up in association with speech therapy and steroid medication
Figure 4. Total thyroidectomy performed for a substernal goiter throof the technique.
PRESS3
for the first 3 months, and thereafter every 4 weeks untilrecovery was obtained. The RLN palsy was considered per-manent, if it persisted beyond 6 months.
Statistical analysis
Comparisons among groups were made using Student’s t-testfor independent samples. Frequencies were compared usingthe Chi2 test. Odds ratios (ORs) and 95% confidence inter-vals (CIs) were calculated to estimate the association ofRLN palsy in cervico-mediastinal goiters. A P-value < .05 wasconsidered statistically significant. Statistical analyses wereperformed using Stata 12 software (StataCorp LP, CollegeStation, Texas, USA).
Results
Mean age of all patients was 55.8 years (range: 17—86);
t laryngeal nerve palsy and substernal goiter. An Italianx.doi.org/10.1016/j.jviscsurg.2014.04.006
in group A, 64 ± 13 in B + C, 63 ± 13 in B, 68 ± 11 in C,with significant statistical differences between B + C vs. A(P < .001), B vs. A (P < .001), C vs. A (P < .001) and C vs. B(P = .014). In the study population, 4858 (32.4%) patientswere men, distributed as follows: group A, 4565 (32.1%),B + C: 293 (36.9%), B: 269 (36.2%), and C: 24 (48%), with sig-nificant statistical differences between B + C vs. A (P = .005),B vs. A (P = .021) and C vs. A (P = .017). An immediate recon-struction of the injured RLN was performed in 12 patients(0.08%): 7 in group A (0.05%), 5 in group B (0.67%) and nonein group C (0%). A statistically significant difference wasfound when B + C (5 patients, 0.63%) was compared with A(P < .001) and in B vs. A (P < .001), while C vs. A and C vs. Bdid not show any significant difference.
Table 1 summarizes operating time, duration of drainand length of hospitalization. The overall mortality rate was0.007% (n = 1), due to an acute myocardial infarction, on the2nd postoperative day.
Table 2 shows the morbidity related to RLN injury. OverallRLN palsy rate was 4.6% in 14,993 patients. The transientRLN palsy rate was 3.5% and the permanent rate was 1.1%.
Temporary tracheotomy was necessary in three cases oftransient bilateral RLN palsy, and definitive tracheostomy intwo cases of permanent bilateral RLN palsy.
ugh a manubriotomy: schematic drawing and intraoperative image
Table 1 Operating time, duration of drain and length of
Group A(n = 14,200)
Group B + C(n = 793)
Group
(n = 743
Operating time (mina) 87 ± 21 115 ± 38 110 ± 3
Duration of drain (hb) 38 ± 12 73 ± 19 70 ± 1
Length ofhospitalization (hb)
70 ± 10 89 ± 21 86 ± 1
Data are given as mean ± standard deviation.a Minutes.b Hours.c P-value is statistically significant.
iscussion
he incidence of substernal goiter in this study was 5.3%,onfirming the need to identify whether a correlation existsetween this condition and postoperative RLN palsy. Vocalord palsy is a well-recognized morbidity after TT and is
disabling sequel constituting the number one cause ofalpractice litigation following thyroid surgery [8]. The
eported incidence varies widely in the literature, depend-ng on definition, patient selection, surgical volume, typef operation, and admission diagnosis. Transient RLN palsyhows an incidence of 0.5—18%, while permanent RLN palsyccurs less frequently (0—5.8%) [2,4].
Several mechanisms have been proposed to explainnjury to the RLN, including clamping, stretching, liga-ure entrapment, transection, perineural devascularisation,ompression by hematoma and electrothermal effect. How-ver, even when visual confirmation of nerve integritys obtained during surgery, an unexpected rate of post-perative palsy continues to be observed [8,10]. Thisemonstrates the extreme vulnerability of the nerves dur-ng surgical dissection, mostly in the course of prolongedperations such as those for recurrent goiter, advancedalignancy or substernal goiter. Indeed, literature reports
Please cite this article in press as: Testini M, et al. Recurrenmulticenter study. Journal of Visceral Surgery (2014), http://d
how the overall incidence of RLN palsy after TT for sub-ternal goiter to be higher compared to the same operationor cervical goiter [1,4,5,11], reaching up to 20% [12]. In oureries, while the overall RLN palsy rate (4.6%) in group B + Cppears to be similar to that reported in literature [2—6],ith an 3.5% incidence of transient palsy and a 1.1% inci-ence of permanent palsy. Accordingly, transient unilateralLN palsy is the most frequently recorded complication.owever, comparing group B + C with control, this studyhowed that RLN palsy was mainly associated with patientsaving a substernal goiter. Moreover, the significant asso-iation of permanent unilateral, transient unilateral, andermanent bilateral forms of palsy in group B representsery interesting data. We believe these complications coulde the effect of an underestimated disease managed by blindntraoperative finger maneuvers through a cervical incision.
Some authors assert that an extensive dissection of theerves increases the rate of RLN palsy and recommend thaterve exposure be as minimal as possible [14,15]. However,e agree with others [1,4] that careful dissection and expo-
ure of the RLN during TT should be a necessary procedureo preserve the integrity of the nerve, because we believehat you can only safeguard what you see. In this respect,he use of intraoperative neuromonitoring seems advis-ble [4,16]. However, in this reported experience, use of a
ct[uilwbp
smopnebsspio(pr
PRESSM. Testini et al.
italization.
Group C(n = 50)
B + C vs. AP
B vs. AP
C vs. AP
C vs. BP
194 ± 41 < .001c < .001c < .001c < .001c
121 ± 11 < .001c < .001c < .001c < .001c
134 ± 46 < .001c < .001c < .001c < .001c
eurostimulator device represented an exclusion criterionue to the disparity of its use in the different centers, mean-ng that it would lead to bias in the method. Having said this,n our previous experience the use of loupe magnification3×) during thyroid surgery [17] allowed a faster and saferdentification of the nerve, and also allowed an easier dif-erentiation between recurrent nerve and inferior laryngealrtery, thanks to the magnification of vasa nervorum andasa vasorum, respectively. Furthermore, in order to avoidLN injury, excessive traction of the gland during dissectionf the nerve from Berry’s ligament should always be avoided.his is, in fact, the most common site of inadvertent nerveransection or electrothermal damage, as reported in theiterature [2,4], and large goiters, particularly substernalnes, are at major risk [6,10,11].
Postoperative cervical hematoma occurs infrequently buts a potentially life-threatening complication, occurringn 0.3—4.2% of patients after thyroidectomy [15,18—25].lthough many reports have shown no superiority of theystematic use of drains in terms of preventing postopera-ive hematoma, many surgeons continue to drain [26—28].isk factors for postoperative cervical hematoma after thy-oidectomy have been shown to include the use of a drainr hemostatic agent, Grave’s disease, benign pathology,
t laryngeal nerve palsy and substernal goiter. An Italianx.doi.org/10.1016/j.jviscsurg.2014.04.006
oncurrent use of antiplatelet or anticoagulation medica-ions, and increased volume of the pathology specimen21,26,29,30]. In our practice, we prefer to continue tose a drain, especially in patients with retrosternal and/orntrathoracic goiter, or when the patient is on anticoagu-ant treatment [31—33]. We believe that the use of a drainill not prevent a postoperative hematoma but will allowetter monitoring of the postoperative course in case ofostoperative hematoma.
According to previous reports, a cervical approach forubsternal goiter management can be adequate in theajority of cases [1,5,6,34]. In selected cases, the use
f Fogarty catheter to help externalize the intrathoracicortion of the gland could be useful tanned help prevent theeed for a sternotomy, as reported in our previous experi-nce [35]. In this study, sternotomy was never performedecause manubriotomy has allowed surgeons to obtain aafe exposure of surrounding and structures and great ves-els when needed. Manubriotomy was carried out in 6.3% ofatients of this series, and was associated with a significantncrease of operating time, duration of drainage and lengthf hospitalization (Table 1). The incidence of sternotomysternal split) ranged from 1 to 45% in previous reports ofatients with substernal goiter, predominantly for intratho-acic gland, involvement of the posterior mediastinum,
nd malignancy with infiltration into surrounding structures1,5,6,9,11,12]. This study demonstrated that the use ofanubriotomy was not associated with an increased risk ofLN palsy in comparison with excision of mediastinal goi-ers managed through a cervical approach. However, currentata do not support the liberal use of manubriotomy. We rec-mmend a careful preoperative evaluation, using magneticesonance imaging or CT scan with multiplanar reformat-ing and volume-rendering reconstructions of the neck andhest, to evaluate anatomical landmarks and reduce theumber of cases requiring an intraoperative decision toerform a sternal split [1,4,7]. We believe that a multidisci-linary collaboration between endocrine and cardiothoracicurgeons should be promoted based on the preoperative clin-cal and radiologic findings in order to avoid intraoperativemprovisation, hazardous digital dissection of mediastinalomponents, or even improper sternal split.
onclusion
n conclusion, this study demonstrated that TT can be per-ormed safely and effectively through a cervical approachn the majority of patients with substernal goiter. Evenhough the mediastinal extension of goiters is associatedith an increased risk of RLN palsy, detailed preoperativeiagnosis and proper surgical management performed inpecialized high-volume centers could improve this specificutcome. The extracervical approach should be reservedor selected cases where the presence of a large thyroidass with bilateral extension into the upper mediastinum,
uspicion of infiltration into surrounding structures, andhen the control of the intraoperative hemostasis becomesifficult in patients with bleeding coming from the medi-stinum.
isclosure of interest
Please cite this article in press as: Testini M, et al. Recurrenmulticenter study. Journal of Visceral Surgery (2014), http://d
he authors declare that they have no conflicts of interestoncerning this article.
eferences
[1] Randolph GW, Shin JJ, Grillo HC, et al. The surgical man-agement of goiter: part II. Surgical treatment and results.Laryngoscope 2011;121:68—76.
[2] Sancho JJ, Pascual-Damieta M, Pereira JA, et al. Risk factorsfor transient vocal cord palsy after thyroidectomy. Br J Surg2008;95:961—7.
[3] Testini M, Rosato L, Avenia N, et al. The impact of singleparathyroid gland autotransplantation during thyroid surgeryon postoperative hypoparathyroidism: a multicenter study.Transplant Proc 2007;39:225—30.
[4] Chiang FY, Lu IC, Tsai CJ, et al. Does extensive dissection ofrecurrent laryngeal nerve during thyroid operation increasethe risk of nerve injury? Evidence from the application ofintraoperative neuromonitoring. Am J Otolaryngol 2011;32:499—503.
[5] Abboud B, Sleilaty G, Mallak N, et al. Morbidity and mor-tality of thyroidectomy for substernal goiter. Head Neck2010;32:744—9.
[6] White ML, Doherty GM, Gauger PG. Evidence-based sur-gical management of substernal goiter. World J Surg2008;32:1285—300.
[
[
[
[
[
[
[
[
[
[
PRESSM. Testini et al.
[7] Pieracci FM, Fahey 3rd TJ. Substernal thyroidectomy is asso-ciated with increased morbidity and mortality as comparedwith conventional cervical thyroidectomy. J Am Coll Surg2007;205:1—7.
[8] Kern KA. Medicolegal analysis of errors in diagnosisand treatment of surgical endocrine disease. Surgery1993;114:1167—73.
[9] Huins CT, Georgalas C, Mehrzad H, et al. A new classificationsystem for retrosternal goitre based on a systematic review ofits complications and management. Int J Surg 2008;6:71—6.
10] Testini M, Nacchiero M, Miniello S, et al. Management ofretrosternal goiters: experience of a surgical unit. Int Surg2005;90:61—5.
11] Testini M, Gurrado A, Avenia N, et al. Does mediastinalextension of the goiter increase morbidity of total thyroidec-tomy? A multicenter study of 19,662 patients. Ann Surg Oncol2011;18:2251—9.
12] Sinclair IS. The risk to the recurrent laryngeal nerves in thyroidand parathyroid surgery. J R Coll Surg Edinb 1994;9:253—7.
13] Testini M, Marzaioli R, Lissidini G, et al. The effectivenessof FloSeal matrix hemostatic agent in thyroid surgery: aprospective, randomized, control study. Langenbecks Arch Surg2009;394:837—42.
14] Bliss RD, Gauger PG, Delbridge LW. Surgeon’s approach to thethyroid gland: surgical anatomy ant the importance of tech-nique. World J Surg 2000;24:891—7.
15] Reeve T, Thompson NW. Complications of thyroid surgery:how to avoid them, how to manage them, and observationson their possible effect on the whole patient. World J Surg2000;24:971—5.
16] Donnellan KA, Pitman KT, Cannon CR, et al. Intraopera-tive laryngeal nerve monitoring during thyroidectomy. ArchOtolaryngol Head Neck Surg 2009;135:1196—8.
17] Testini M, Nacchiero M, Piccinni G, et al. Total thyroidectomyis improved by loupe magnification. Microsurgery 2004;24:39—42.
18] Bergenfelz A, Jansson S, Kristoffersson A, et al. Complicationsto thyroid surgery: results as reported in a database from amulticenter audit comprising 3660 patients. Langenbecks ArchSurg 2008;393:667—73.
19] Burkey SH, van Heerden JA, Thompson GB, et al. Reexplo-ration for symptomatic hematomas after cervical exploration.Surgery 2001;130:914—20.
20] Godballe C, Madsen AR, Pedersen HB, et al. Post-thyroidectomyhemorrhage: a national study of patients treated at the Danishdepartments of ENT Head and Neck Surgery. Eur Arch Otorhi-
t laryngeal nerve palsy and substernal goiter. An Italianx.doi.org/10.1016/j.jviscsurg.2014.04.006
nolaryngol 2009;266:1945—52.21] Lang BH, Yih PC, Lo CY. A review of risk factors and timing
for postoperative hematoma after thyroidectomy: is outpatientthyroidectomy really safe? World J Surg 2012;36:2497—502.
22] Leyre P, Desurmont T, Lacoste L, et al. Does the risk of compres-sive hematoma after thyroidectomy authorize 1-day surgery?Langenbecks Arch Surg 2008;393:733—7.
23] Promberger R, Ott J, Kober F, et al. Risk factors for postopera-tive bleeding after thyroid surgery. Br J Surg 2012;99:373—9.
24] Rosato L, Avenia N, Bernante P, et al. Complications of thyroidsurgery: analysis of a multicentric study on 14,934 patientsoperated on in Italy over 5 years. World J Surg 2004;28:71—6.
25] Rosenbaum MA, Haridas M, McHenry CR. Life-threatening neckhematoma complicating thyroid and parathyroid surgery. Am JSurg 2008;195:339—43.
26] Dunlap WW, Berg RL, Urquhart AC. Thyroid drains and postop-erative drainage. Otolaryngol Head Neck Surg 2010;143:235—8.
27] Colak C, Akca T, Turkmenoglu O, et al. Drainage after totalthyroidectomy or lobectomy for benign thyroidal disorders. JZhejiang Univ Sci B 2008;9:319—23.
28] Herranz J, Latorre J. Drainage in thyroid and parathyroidsurgery. Acta Otorrinolaringol Esp 2007;58:7—9.
29] Zambudio AR, Rodrıguez J, Riquelme J, et al. Prospective studyof postoperative complications after total thyroidectomy formultinodular goiters by surgeons with experience in endocrinesurgery. Ann Surg 2004;240:18—25.
Please cite this article in press as: Testini M, et al. Recurrenmulticenter study. Journal of Visceral Surgery (2014), http://d
ARTICLE IN+ModelJVS-393; No. of Pages 7
Substernal goiter and recurrent laryngeal nerve palsy
30] Campbell MJ, McCoy KL, Shen WT, et al. A multi-institutionalinternational study of risk factors for hematoma after thy-roidectomy. Surgery 2013;154:1283—91.
31] Neary PM, Connor OJ, Shafiq A, et al. The impact of routineopen nonsuction drainage on fluid accumulation after thyroidsurgery: a prospective randomised clinical trial. World J SurgOncol 2012;10:72.
32] Memon ZA, Ahmed G, Khan SR, et al. Postoperative use of drainin thyroid lobectomy — a randomized clinical trial conductedat Civil Hospital, Karachi, Pakistan. Thyroid Res 2012;5(1):9.
[
[
[
t laryngeal nerve palsy and substernal goiter. An Italianx.doi.org/10.1016/j.jviscsurg.2014.04.006
PRESS7
33] Decide U, Altintoprak F, Sertan Kapakli M, et al. Is the useof a drain for thyroid surgery realistic? A prospective ran-domized interventional study. J Thyroid Res 2013;2013:285768,http://dx.doi.org/10.1155/2013/285768 [Epub 2013, May 30].
34] de Perrot M, Fadel E, Mercier O, et al. Surgical managementof mediastinal goiters: when is a sternotomy required? ThoracCardiovasc Surg 2007;55:39—43.
35] Testini M, Piccinni G, Lissidini G, et al. The lifting of sub-sternal goitres using a Fogarty catheter. Ann R Coll Surg Engl2005;87:63—4.
