Original Contributions

Postoperative Complications of Thyroidectomy

for Differentiated Thyroid Carcinoma

João Gonçalves Filho, MD, and Luiz Paulo Kowalski, MD, PhD

Objective: This study evaluates the incidence and risk factors of complications in patients
submitted to thyroidectomy for differentiated thyroid carcinoma in a cancer hospital with
residency training.
Study Design: A retrospective chart and complications review of 316 consecutive patients
who underwent thyroidectomy for differentiated thyroid carcinoma.
Results: Of the 316 patients, the main postoperative complications were transient hypocal-
cemia in 87 (27.5%), permanent hypocalcemia in 16 (5.1%), transient vocal cord palsy in 4
(1.2%), and permanent vocal cord palsy in 2 (0.6%). Neck dissection and paratracheal lymph
node dissection when associated with total thyroidectomy were significantly related to
transitory and permanent hypocalcemia.
Conclusion: Thyroid surgery can be performed safely in a hospital with medical residency
training program under direct supervision of an experienced surgeon with acceptable
morbidity. Hypocalcemia is the most significant complication. Neck and paratracheal lymph
node dissections were the most significant predictors of hypocalcemia in patients submitted
to total thyroidectomy.
(Am J Otolaryngol 2004;25:225-230. © 2004 Elsevier Inc. All rights reserved.)

According to data from the American Can-
cer Society, the number of new cases of dif-
ferentiated thyroid carcinoma has been in-
creasing in the last decades.1 The same has
also been observed in São Paulo where the
incidence rates have increased from 6.3/
100.000 persons in 1983 to 20.4/100.000 in
1998.2 Thyroidectomy is the main form of
initial treatment for thyroid gland cancer.
However, the extension of the resection
performed in the treatment of differentiated
thyroid carcinoma remains controversial,3,4
particularly because the incidence of postop-
erative complications is directly related to the
From the Department of Head and Neck Surgery and
Otorhinolaryngology, Centro de Tratamento e Pesquisa
Hospital do Câncer A C Camargo, São Paulo, Brazil.
Address correspondence to: Luiz Paulo Kowalski,
MD, PhD, Department of Head and Neck Surgery and
Otorhinolaryngology, Centro de Tratamento e Pesquisa
Hospital do Câncer A C Camargo, Rua Professor Antonio
Prudente, 211, 01509-900, São Paulo, Brazil. E-mail:
lp_kowalski@uol.com.br.
© 2004 Elsevier Inc. All rights reserved.
0196-0709/$ - see front matter
doi:10.1016/j.amjoto.2004.02.001
type of the thyroidectomy (lobectomy, near
total, or total).
At the beginning of the thyroidectomy his-
tory, the major complications were hematoma
and postoperative infection and most of these
pioneer authors reported some postoperative
mortality.5-7 Currently, the main postop-
erative complications are vocal cord palsy
because of recurrent laryngeal nerve dysfunc-
tion and hypocalcemia.5-10 Postoperative
death is now rare or even unrecorded.11-16
The occurrence of such complications has
been attributed to same particular aspects of
the surgical technique, extension of the oper-
ation, reoperation, neck dissection, and expe-
rience of the surgical team.5,17-20 The main
purpose of this study is to evaluate the inci-
dence and risk factors of complications in
patients undergoing thyroidectomy for differ-
entiated thyroid carcinoma, with a view to
propose preventive measures that can result
in improvements in the functional outcomes.
PATIENTS AND METHODS
From January 1990 to December 2000, 1020 pa-
tients were submitted to thyroidectomy in the Head

American Journal of Otolaryngology, Vol 25, No 4 (July-August), 2004: pp 225-230 225

226
Fig 1. Type of thyroidectomy for period. TT, total thyroidectomy; ST, subtotal thyroidectomy; CT, completion of
thyroidectomy; LⴙI, lobectomy with isthmectomy.
and Neck Surgery and Otorhinolaryngology De-
partment of the Centro de Tratamento e Pesquisa
Hospital do Câncer A C Camargo, in São Paulo,
Brazil. The vast majority of the operations were
performed by third- to fifth-year medical residents
under direct the supervision of 10 different expe-
rienced head and neck surgeons. The attending
surgeons were stratified as follows according to the
total number of thyroidectomies they performed or
supervised during the study period: group A (1 to
20 thyroidectomies), group B (21-40), and group C
(over 40).
Three hundred sixteen patients were submitted
to thyroidectomy for differentiated thyroid carci-
noma. There were 263 female patients (83.2%) and
53 male (16.8%), with a mean age of 42 years
(ranging from 8 to 88 years). Preoperative workup
included medical history and physical examina-
tion, thyroid function tests, thyroid ultrasound,
fine-needle aspiration biopsy, and thorax radiogra-
phy. All patients had preoperative evaluation of
vocal cord mobility by means of indirect laryngos-
copy or fibrolaryngoscopy.
The number of thyroidectomy for cancer increased
progressively from 1990 to 2000, with prevalence of
total thyroidectomy. However, in the last years, the
lobectomy with isthmectomy was performed in se-
lected patients with low-risk tumors21 (Fig 1). The
surgical procedure included total thyroidectomy in
180 patients (57%), subtotal thyroidectomy in 11
FILHO AND KOWALSKI
(3.6%), lobectomy with isthmectomy in 92 (29%),
and completion of thyroidectomy in 33 (10.4%).
Neck dissections were performed in 62 patients who
had lymph nodes clinically metastatic: ipsilateral
dissection in 48 patients (5 jugular chain dissections,
38 type III modified radical dissections, and 5 classi-
cal radical dissections) and bilateral dissections were
performed in 12 patients (3 bilateral jugular chain
dissections, 8 bilateral type III modified radical dis-
sections, and 1 type III ipsilateral radical neck dissec-
tion with contralateral jugular chain dissection).
Paratracheal lymph node dissection was performed
in 123 patients (60 ipsilateral and 63 bilateral). The
pathologic examination showed 276 papillary carci-
noma (88%) and 40 follicular carcinoma (12%).
We routinely identified the parathyroids and re-
current laryngeal nerves before performing the li-
gation of the inferior thyroid pedicles. An auto-
transplant of the parathyroid was performed every
time if vascular supply injury was identified or the
glands were accidentally removed. The indication
and type of surgical drains used varied throughout
the study period. Penrose drains were used until
1992 when it was replaced in by Hemovac-type
closed drainage to facilitate postoperative care and
permit earlier discharge from the hospital. After
1995, at the surgeon’s discretion, the operative
wound of 29 selected patients was not drained.
All patients were evaluated for at last 30 days
with regard to the occurrence of postoperative com-
THYROIDECTOMY COMPLICATIONS 227

plications. We routinely performed indirect laryn- TABLE 2. Characteristic of the Patients With
goscopy or nasofibroscopy before the 30th postop- Permanent Hypocalcemia
erative day. Recurrent laryngeal nerve dysfunction,
hypoparathyroidism, and other complications Patient (N)/Gender/
were analyzed according to the number of opera- Age (years) Histology Surgery
tions, type of thyroidectomy, and whether or not
1/F/26 Ca papillary TT ⫹ BPD
drains were used. Postoperative vocal cord palsy or
2/F/47 Ca papillary CT
paresis were defined as the presence of immobility
3/F/19 Ca papillary TT ⫹ BPD
or decreased movement of the vocal cords. A per- 4/F/37 Ca papillary TT ⫹ BPD
sisting vocal cord dysfunction after 6 months was 5/F/32 Ca papillary TT ⫹ IPD
considered permanent. Palsies that resulted from 6/M/53 Ca papillary TT ⫹ BND ⫹ BPD
inferior laryngeal nerve resection because of neo- 7/F/21 Ca papillary TT ⫹ IND ⫹ BPD
plastic involvement were classified as sequelae and 8/F/18 Ca papillary TT ⫹ IND ⫹ BPD
not computed as complications. Hypocalcemia was 9/F/62 Ca papillary TT ⫹ IND
considered present whenever there was a need for 10/F/51 Ca papillary TT ⫹ IPD
exogenous calcium replacement to maintain nor- 11/F/43 Ca papillary TT ⫹ IND ⫹ IPD
mal serum levels (8-10.4 mg/dL) or to eliminate the 12/F/15 Ca papillary TT ⫹ BND ⫹ IPD
clinical signs and symptoms of hypocalcemia. Hy- 13/F/70 Ca papillary TT ⫹ BPD
pocalcemia was considered permanent when cal- 14/F/49 Ca papillary TT ⫹ BND
cium replacement was necessary for over than 6 15/F/51 Ca papillary TT
months. The association between the studied vari- 16/F/42 Ca papillary TT ⫹ BPD
ables with the occurrence of complications was
evaluated by the chi-square test or Fisher exact test Abbreviations: TT, total thyroidectomy; CT, reoperation;
when applicable. Values of P ⬍ 0.05 were consid- IPD, ipsilateral paratracheal dissection; BPD, bilateral para-
ered significant. tracheal dissection; IND, ipsilateral neck dissection; BND,
bilateral neck dissection.

RESULTS
One hundred twenty-three patients (38.9%)
had postoperative complications, with hy-
pocalcemia as the most frequent one. Other
less frequent complications were vocal fold
palsy, hematoma, seroma, and wound infec-
tion (Table 1). The mean duration of hospital
stay was 2 days (ranging from 1 to 18 days).
Thyroidectomies were performed or super-
vised by surgeons classified as follows accord-
ing to the number of thyroidectomies: group A
(22%), group B (18%), and group C (60%).
The overall rate of vocal cord palsy was
1.8% (6/316) (Table 1). Transitory palsy oc-
curred in 4 patients with full recovery occur-
ring within 6 months. Two patients had per-
manent vocal cord palsy. There were no
bilateral palsies, nor was there a need for
TABLE 1. Postthyroidectomy Complications
Procedures
Complications (N) (%)
Transitory hypocalcemia 87 (27.5)
Permanent hypocalcemia 16 (5.1)
Transitory vocal fold palsy 4 (1.2)
Permanent vocal fold palsy 2 (0.6)
Hematoma 4 (1.2)
Seroma 15 (4.7)
Surgical site infection 7 (2.2)
Chyle leaks 2 (0.2)
tracheotomy. The type of thyroidectomy, the
surgeon’s experience, patient’s gender, diam-
eter of the nodule, and the association or not
with neck dissection did not have any signif-
icant association with the incidence of vocal
cord palsy.
Postoperative hypocalcemia occurred in
103 patients (32.6%) (Table 2). Transitory hy-
pocalcemia was found in 87 patients (27.5%),
all with full recovery occurring within 6
months. Sixteen patients (5.1%) had perma-
nent hypocalcemia. Table 2 presents the char-
acteristics of the patients with hypocalcemia.
The incidence of postoperative hypocalcemia
did not have any significant correlation with
patient’s gender, age, diameter of the nodule,
and surgeon’s experience. However, the ex-
tension of thyroidectomy and the association
with neck or paratracheal lymph node dissec-
tion were significantly associated with the
risk of transitory and permanent hypocalce-
mia (Table 3). The incidence of postoperative
hypocalcemia was also higher when thyroid-
ectomy was performed with en bloc neck dis-
section or paratracheal lymph node dissection
(Fig 2). In the group of patients who under-
went thyroidectomy only, the incidence of
hypocalcemia was significantly lower when
compared with the cases that a neck dissec-
tion was associated (19% v 33.3%) (P ⫽ .001)
or paratracheal lymph node dissection (19% v
228 FILHO AND KOWALSKI

TABLE 3. Postoperative Hypocalcemia According DISCUSSION

to Clinical Variables

Hypocalcemia (%) Thyroidectomy is a very common surgical

procedure worldwide and is performed by
Variable No. Transient P Permanent P
surgeons with varied training and back-
Gender grounds such as general surgery, thoracic sur-
Male 53 13 (24.5) .362 3 (5.7) .738 gery, endocrine surgery, otorhinolaryngology,
Female 263 74 (28.1) 13 (7.1)
Age (years)
oncologic surgery, and head and neck surgery.
⬍20 24 6 (25) .545 3 (12.5) .175 Our department is part of an oncological sur-
20-50 119 37 (31) 4 (3.4) gery training program, where most of such
⬎50 173 44 (25.4) 9 (5.2) operations are performed by 3rd to 5th year
Nodule (cm)
ⱕ4 207 60 (29) .254 9 (3.3) .292 medical residents in surgical oncology or
⬎4 109 27 (24.8) 7 (6.4) head and neck surgery fellows under the di-
Surgeon group rect supervision of one experienced head and
A 69 20 (29) .675 5 (7.2) .603 neck surgeon. Although some reports in liter-
B 57 13 (22.8) 2 (3.5)
C 190 54 (28.4) 9 (4.7) ature had shown that the complication rates
Thyroidectomy were associated with the surgeon’s experi-
Partial 103 2 (1.9) .001 0 (0) .008 ence,5,20 there are also several others reports
Reoperation 33 8 (24.2) 1 (3) that point out the safety of this operation
Total 180 77 (42.8) 15 (8.3)
Neck dissection performed at residency training centers, with
No 256 63 (24.6) .007 9 (3.5) .002 acceptable complication rates.5,14,22-25 In our
Unilateral 48 22 (45.8) 4 (8.3) study, 60% of the thyroidectomies were per-
Bilateral 12 2 (16.7) 3 (25)
formed or supervised by 2 surgeons, 22% by 6
Paratracheal
dissection surgeons, and 18% by 2 surgeons, and no
No 193 35 (18.1) .001 4 (2.1) .003 statistically significant differences on the rates
Unilateral 60 24 (40) 4 (6.7) of complications were observed between the
Bilateral 63 28 (44.4) 8 (12.7)
groups. Similarly, Bergamaschi et al,18 in a
series of 1,192 thyroidectomies performed by
11 surgeons, also did not observe significant
differences between patient’s volume per sur-
48%) (P ⫽ .0001). In the same way, the asso- geon and the complication rates.
ciation of neck dissection with paratracheal There has been a significant reduction in
lymph node dissection was significantly cor- the incidence of complications and mortality
related with the highest rate of postoperative in thyroid surgery since the beginning of
hypocalcemia (57.7%). the 20th century, currently making thyroid-
Seroma was a postoperative complication ectomy a surgical procedure with low ac-
found in 15 patients (4.7%). Seroma devel- ceptable morbidity and mortality rates. Post-
oped in 2 of 39 patients drained with a Pen- operative mortality with thyroidectomy has
rose drain (5.1%) and 13 of the 248 drained become an extremely rare complication in
with Hemovac (5.2%). None of the 29 patients various reports in the literature.7,10-16,20 Post-
in whom no drain was placed presented with operative death was not observed in our se-
this type of complication. Seven patients ries. Pederson et al23 had 0.4% deaths and
(2.2%) developed postoperative surgical-site Herranz-Gonzalez et al,26 in a series of 335
infections. Four patients (1.3%) were reoper- thyroidectomies, observed a mortality rate of
ated on as a result of a hematoma that oc- 0.8%. The rates of postoperative complica-
curred in 1 of the 29 patients in whom no tions reported in literature are variable (7.4%
drain was placed and in 3 patients drained to 53%).7,8,11,14,23 In our series, we observed
with Hemovac. Chyle leaks occurred in 2 pa- postoperative complications in 123 patients
tients (0.6%) who underwent neck dissection. (38.9%) and the extension the thyroidectomy
No significant differences were observed be- was correlated with an incidence of compli-
tween these complications and the use of cations. In the same way, we observed a
drain, type of thyroidectomy, and number of higher incidence of postoperative complica-
thyroidectomies performed or supervised by tions in a series of patients treated with locally
the surgeon (P ⬎ .05). invasive differentiated thyroid carcinoma, the
THYROIDECTOMY COMPLICATIONS
Fig 2. Incidence of postoperative hypocalcemia when neck or paratracheal dissection was associated with
thyroidectomy. TT, total thyroidectomy; ND, neck dissection; PD, paratracheal dissection.
most frequently found complication being hy-
pocalcemia.27
The most common and feared complica-
tions in thyroid gland surgery are vocal cord
palsy and hypocalcemia, and there are
various factors involved in their occur-
rence.7,11,14,18,26,28-30 The incidence of recur-
rent laryngeal nerve injury reported in litera-
ture vary from 0% to 4.8%,8,12,14,23,25,26 being
higher in extensive resections and in reopera-
tion series.10,11,13,16 In our series, only 2 pa-
tients (0.6%) showed permanent vocal cord
palsy. Pezzullo et al6 found permanent vocal
cord palsy in 2.8% of the cases. Chao et al16
observed a 2.6% incidence of transitory palsy
and a 1.7% rate of permanent vocal cord
palsy. In our group, the dissection and iden-
tification of the recurrent laryngeal nerve is
performed as a routine manner before the li-
gation of the inferior pedicle vessels, thus
reducing the risk of nerve injury. This early
identification of the recurrent laryngeal nerve
is also advocated by several other au-
thors.5,8,10 Cernea et al29 reported the impor-
tance of the external branch of the superior
laryngeal nerve and the rate of nerve injury
during thyroidectomy. The identification and
preservation of this branch is also of para-
mount importance.
229
Hypocalcemia was an important complica-
tion in our series (27.5% transitory and 5.1%
permanent). On reviewing recent thyroid-
ectomy literature we found an incidence
of symptomatic postoperative hypocalcemia
ranging from 4% to 42%.7,8,10,13,18,23,26,30 Per-
manent hypocalcemia occurs with a lower
incidence (0% to 8%).5,7,8,14,15,23,26,30 Neck
dissection and paratracheal dissection (ipsi-
lateral or bilateral) were the most important
risk factors for the occurrence of hypocalce-
mia. In this series, the age did not present any
significant association with postoperative hy-
pocalcemia. Although in previous study, we
observed a higher incidence of hypocalcemia
associated with patients under the age of 18
years.31 In the same way, other authors re-
ported a higher rate of post-thyroidectomy
hypocalcemia in childhood and adolescent
patients.32-34 Paratracheal dissection has been
implicated on the increase of the risk of both
vocal cord palsy and postoperative hypocal-
cemia.6,8,26 In our series, postoperative hy-
pocalcemia was significantly more common
when neck or paratracheal dissections were
associated with thyroidectomy (19% v 33.3%)
and (19 v 48%), respectively. In the same way,
Cheah et al35 reported a higher incidence of
postoperative hypocalcemia when a neck dis-
230
section was associated with thyroidectomy
than when it was performed without neck
dissection (60% v 17%).
In view of the results of this study, we
conclude that thyroidectomy, whether or not
associated with neck dissection, is an opera-
tion that has acceptable morbidity with hy-
pocalcemia being the most important one.
Thyroidectomy may be safely performed in
resident training centers, provided that it is
performed under the supervision of an expe-
rienced surgeon. Paratracheal lymph node
dissection was the most significant predictor
of hypocalcemia in cancer patients who un-
derwent total thyroidectomy. The contraindi-
cation of elective paratracheal dissection can
reduce the occurrence of the main postopera-
tive complications of thyroidectomy in se-
lected low-risk patients. However, this chain
of lymph node should be examined carefully
during the surgical exploration.
REFERENCES
1. O’Brien K, Cokkinides V, Jemal A, et al: Cancer statis-
tics for Hispanics. CA Cancer J Clin 53:208-226, 2003
2. Mirra AP, Latorre MR, Veneziano DB: Incidência de
câncer no municı́pio de São Paulo, Brasil 1997-1998:
Mortalidade de câncer no municı́pio de São Paulo: ten-
dência no perı́odo 1969-1998. São Paulo, Ministério da
Saúde, 2001
3. Jossart GH, Clark OH: Well differentiated thyroid
cancer. Curr Prob Surg 31:935-1011, 1994
4. Gagel RF, Goepfert H, Callender DL: Changing con-
cepts in the pathogenesis and management of thyroid
carcinoma. CA Cancer J Clin 46:261-283, 1996
5. Shaha A, Jaffe BM: Complications of thyroid surgery
performed by residents. Surgery 104:1109-1114, 1988
6. Pezzulo L, Dalrio P, Losito NS, et al: Post-operative
complications after completion thyroidectomy for differen-
tiated thyroid cancer. Eur J Surg Oncol 123:215-218, 1997
7. Moulton-Barrett R, Crumley R, Jalilie S, et al: Com-
plications of thyroid surgery. Int Surg 82:63-66, 1997
8. Shindo LM, Sinha UK, Rice DH: Safety of thyroid-
ectomy in residency: A review of 186 consecutive cases.
Laryngoscope 105:1173-1175, 1995
9. Wax KM, Valiulis AP, Hurst MK: Drains in thyroid
and parathyroid surgery. Arch Otolaryngol Head Neck
Surg 121:981-983, 1995
10. Al-Suliman NN, Ryttov NF, Blichert-Toft M, et al:
Experience in a specialist thyroid surgery unit: a demo-
graphic study, surgical complications, and outcome. Eur
J Surg 163:13-20, 1997
11. Thompson NW, Harness JK: Complications of total
thyroidectomy for carcinoma. Surg Gynecol Obstet 131:
861-868, 1970
12. Jacobs JK, Aland JW Jr, Ballinger JF: Total thyroid-
ectomy: A review of 213 patients. Ann Surg 197:542-547,
1983
FILHO AND KOWALSKI
13. Flynn MB, Lyons KJ, Tarter JW, et al: Local com-
plications after surgical resection for thyroid carcinoma.
Am J Surg 168:404-407, 1994
14. Calik A, Kucuktulu U, Ciel A, et al: Complications
of 867 thyroidectomies performed in a region of endemic
goiter in turkey. Int Surg 81:298-301, 1996
15. Sand J, Palkola K, Salmi J: Surgical complications
after total thyroidectomy and resections for differentiated
thyroid carcinoma. Ann Chirurgiae et gynaecologiae 85:
305-308, 1996
16. Chao TC, Jeng LB, Lin JD, et al: Reoperative thyroid
surgery. World J Surg 21:644-647, 1997
17. Shaha AR, Jaffe BM: Completion thyroidectomy: A
critical appraisal. Surgery 112:1148-1152, 1992
18. Bergamaschi R, Becouarn G, Ronceray J, et al: Mor-
bidity of thyroid surgery. Am J Surg 176:71-75, 1998
19. Harness JK, Fung L, Thompson NW, et al: Total
thyroidectomy: Complications and technique. World
J Surg 10:781-786, 1986
20. Sosa JA, Bowman HM, Tielsch JM, et al: The impor-
tance of surgeon experience for clinical and economic out-
comes from thyroidectomy. Ann Surg 228:320-330, 1998
21. Shaha AR, Shah JP, Loree TR: Patterns of failure in
differentiated carcinoma of the thyroid based on risk
groups. Head Neck 20:26-30, 1998
22. Sternes HF, Brooks DC, Pinkus GS: Surgery for
thyroid carcinoma. Cancer 55:1376-1381, 1985
23. Pederson WC, Johson CL, Gaskill HV, et al: Opera-
tive management of thyroid disease. Am J Surg 148:350-
352, 1984
24. Roy van Zuidewijn DBW, Songun I, Kievit J, et al:
Complications of thyroid surgery. Ann Surg Oncol 2:56-
60, 1995
25. Lamade W, Renz K, Klar E, et al: Effect of training
on the incidence of nerve damage in thyroid surgery. Br J
Surg 86:388-391, 1999
26. Herranz-Gonzalez J, Gavilan J, Matinez-Vidal J, et
al: Complications following thyroid surgery. Arch Otolar-
yngol Head Neck Surg 117:516-518, 1991
27. Kowalski LP, Goncalves Filho J: Results of the
treatment of locally invasive thyroid carcinoma. Head
Neck 24:340-344, 2002
28. Lore JM Jr: Surgery of the thyroid gland. Otolaryn-
gol Clin North Am 13:69-83, 1980
29. Cernea CR, Ferraz AR, Furlani J, et al: Identification
of the external branch of the superior laryngeal nerve
during thyroidectomy. Am J Surg 164:634-639, 1992
30. Wingert DJ, Friesen SR, Iliopoulos JI, et al: Post-
thyroidectomy hypocalcemia. Am J Surg 152:606-609, 1986
31. Kowalski LP, Goncalves Filho J, Pinto CA, et al:
Long-term survival rates in young patients with thyroid
carcinoma. Arch Otolaryngol Head Neck Surg 129:746-
749, 2003
32. Zimmerman D, Hay ID, Gough IR, et al: Papillary thy-
roid carcinoma in children and adults: Long-term follow-up
of 1039 patients conservatively treated at one institution
during three decades. Surgery 104:1157-1166, 1988
33. Millman B, Pellitteri PK: Thyroid carcinoma in
children and adolescents. Arch Otolaryngol Head Neck
Surg 121:1261-1264, 1995
34. Dottorini ME, Vignati A, Mazzuchelli L, et al: Dif-
ferentiated thyroid carcinoma in children and adoles-
cence: A 37-year experience in 85 patients. J Nucl Med
38:669-675, 1997
35. Cheah WK, Arici C, Ituarte PH, et al: Complications
of neck dissection for thyroid cancer. World J Surg 26:
1013-1016, 2002