Surgery for thyroid goiter in western India. A prospective analysis of 334 cases.RD Bapat, P Pai, S Shah, SD Bhandarkar
Dept of General Surgery, Seth GS Medical College and KEM Hospital, Parel, Bombay, Maharashtra.
334 consecutive cases of thyroid swellings operated by a single surgical unit over 9 years have been analysed prospectively. There was a female preponderence (4.39:1). The swellings were clinically differentiated into uninodular (39.52%), multinodular (47.31%) and diffuse (13.17%). Hyperthyroidism was manifested in 49 cases (14.67%). Pressure symptoms were present in only 1.5% cases. FNAC detected malignancy in 14 of 162 cases (8.64%). The initial 100 cases were operated upon by standard Lahey's technique and the latter 234 by modified technique described by Bapat et al for benign thyroid disease. Operations performed included nodulectomies (5.39%), hemithyroidectomies (41.92%), partial thyroidectomies (25.75%), subtotal (25.45%) and near total thyroidectomies (1.5%). Post-operative complications were higher in the first group and included unilateral cord palsies-5 (5%). hypocalcemia-4 (4%) hypoparathyroidism-1 (1%) haemorrhage-1 (1%) and mortality-1 (1%) vis a vis cord palsies-2 (0.85%), hypocalcemia-3 (1.28%), hypoparathyroidism-1 (0.43%) and there was no mortality. Histopathology revealed 83 (24.85%) colloid goiters, 193 (57.78%) nodular goiters, 21 (6.29%) follicular adenomas, 7 (2.10%) cases of thyroiditis and 30 (8.98%) malignancies. This study reveals the lower incidence of RLN palsy after modified thyroidectomies, and a low incidence of malignancy.
Keywords: Adolescent, Adult, Aged, Biopsy, Needle, Child, Comparative Study, Female, Goiter, classification,epidemiology,pathology,surgery,Human, Incidence, India, epidemiology,Male, Middle Age, Population Surveillance, Postoperative Complications, epidemiology,etiology,Prospective Studies, Sex Factors, Thyroidectomy, methods,statistics &numerical data,Treatment Outcome,
Thyroidectomies account for approximately 3% of total nonemergency surgeries done by general surgery units in one year. We present the results of a prospective study of thyroidectomies performed by one general surgery unit from January 1984 to January 1993.
This study was aimed at establishing a correlation between clinical presentation, FNAC and histopathology; and assess prospectively the advantage of the modified technique of thyroidectomy described by Bapat et al  over the standard Lahey technique for the treatment of benign thyroid disease.
334 consecutive cases of thyroidectomies performed in the 9 year period from January 1984 to January 1993 were analysed prospectively. The patients were assessed clinically and investigated by the Endocrine service with hormonal assays, thyroid scans and FNAC. A note of the patients' home town and family history was also made.
The goiters were clinically differentiated into uninodular, multinodular and diffuse. Presence of hyperthyroidism and pressure symptoms were also noted. Hormonal assays were performed in cases clinically suspected to be non-euthyroid. Thyroid scans were performed in uninodular swellings, suspected malignancies and hyperthyroid cases. FNAC was performed routinely in all cases after 1985 when it was made available. In some cases, FNAC was performed under ultrasonographic guidance.
The initial 100 cases were operated upon by the standard Lahey technique while the later 234 were operated by the modified technique described by Bapat et al which relies on minimal lateral dissection [so as to stay far away from the recurrent laryngeal nerve (RLN)], clevascularisation of the inferior thyroid artery on the main glandular tissue itself and to keep back a wedge section of the thyroid so that enough thyroid tissue is left back over the capsule to allow for the rare chance of the nerve passing through the capsule and for better approximation and good homeostasis. The operations were also performed by junior residents under supervision. Operations performed include nodulectomies, hemi-, partial, subtotal and near total thyroidectomies. A detailed post-operative record was maintained and all complications noted. Patients were followed up for at least 6 months after surgery.
There was a female preponderence (4.39: 1) in the 334 patients. The ages ranged from 12 years to 75 years with a median of 35 years. The patients were found to come mainly from the foothills of the Saputhara range in Western India.
Clinically, the thyroid swellings presented as 158 (47.31%) cases of multinodular goiter, 132 (39.52 %) cases of uninodular goiter and 44 (13.17 %) cases of diffuse goiter. Hyperthyroidism manifested in 49 cases (14.67%) and pressure symptoms in 5 (1.5 %) cases. Hyperthyroidism was found in 26 cases of multinodular goiter, 20 diffuse and only 3 uninodular goiters.
Thyroid scans were performed in 196 cases. Ninety-four cases showed a cold area, 65 a patchy uptake and 14 a warm area. An increased uptake was seen in 14 cases but decreased uptake in 1 case. 16 cases showed a normal scan.
FNAC was performed in 162 cases and was found to be benign in 128 cases, inconclusive in 20 cases and malignant in 14 cases.
The operations performed included 18 (5.39%) nodulectomies, 140 (41.92%) hemithyroidectomies, 86 (25.75%) partial thyroidectomies, 85 (25.45%) subtotal and 5 (1.5%) near total thyroidectomies. Two cases were reoperations done for malignancy detected by histopathology. In thyrotoxicosis, the operations performed included hemithyroidectomies (n = 2) in toxic nodules, partial thyroidectomies (n = 5) and subtotal thyroidectomies (n = 42). The initial 100 cases operated by standard Laheys technique showed a higher rate of complications in the form of RLN palsies (3%), hypocalcemia (4%), hypoparathyroidism (1%), haemorrhage (1%) and mortality (1%) whereas the latter 234 operated by the modified technique of Bapat et al had a lower incidence of RLN plasies (0.85%), hypocalcemia (1.28%) and hypoparathyroidism (0.43%). There were no cases of reactionary haemorrhage or mortality. The incidence of hypocalcemia was higher after subtotal thyroidectomies performed for thyrotoxicosis (11. 11 %).
Histopathology revealed 193 (57.78%) cases of nodular goiter, 83 (24.85%) cases of colloid goiter, 30 cases (8.98%) of malignancies (13 follicular, 12 papillary, 3 medullary and 2 anaplastic adenocarcinoma), 21 (6.29%) cases of follicular adenomas and 7 (2.10%) cases of thyroiditis (4 Hashimotos, 3 granulomatous and 1 lymphocytic thyroiditis). Cystic degneration was seen in 23 cases.
A correlation between clinical presentation, FNAC and histopathology is shown in [Table - 1]. It is evident that there were 13 cases of carcinoma in the first group diagnosed clinically as uninodular nontoxic goiters (n = 129). FNAC failed to detect 4 malignant lesions. There were no malignancies in the second group diagnosed clinically as uninodular toxic goiters (n = 3). The third group diagnosed clinically as multinodular nontoxic goiters (n = 132) consisted of 14 malignancies; two were missed on FNAC and 2 were wrongly diagnosed as malignancies on FNAC (1 nodular goiter and 1 follicular adenoma). The fourth group of multinodular toxic goiter ( n = 26) showed malignancy in only 1 case. The fifth group of diffuse nontoxic goiters (n = 24) showed malignancy in only 2 cases of which 1 was missed on FNAC. The sixth group of diffuse toxic goiters did not reveal any malignancies.
Most studies of surgery for thyroid enlargement are retrospective and therefore prone to all the drawbacks of retrospective analysis. Hence, a prospective study and analysis was undertaken.
Many of the studies , showed a higher incidence of uninodular swellings compared to multinodular swellings in a palpably enlarged thyroid gland. The present series showed a higher incidence of multinodular swellings (158: 132 = 1.97: 1).
[Table - 1] shows that FNAC missed malignancies in 6 cases (sensitivity = 66.67%) and was false positive in 2 cases (specificity = 90.61%). The sensitivity increases if the 20 inconclusive cases were to be included as malignancies, but this would be at the cost of specificity.  The reported accuracy of FNAC ranges 4-6 from 50% to 97%,,.
The overall incidence of malignancy in the present series is 8.98% with a 10.60% incidence in multinodular nontoxic goiters, 10.08% in uninodular nontoxic goiters, 8.33% in diffuse nontoxic goiters and 3.85% in multinodular toxic goiters. The incidence of malignancy in uninodular goiters is higher than that reported by us previously (5.17%) and by Bhansali (9.13%). The incidence of carcinoma in multinodular goiters (toxic + nontoxic) was 6.26% correlating with the finding of Koh who found an incidence of 7.5% in Malaysian patients. Thyroiditis (n= 7) was found mainly in the multinodular nontoxic group (n= 6) and diffuse nontoxic group (n =1)
[Table - 2] shows a comparison of complication of RLN palsy, hypocalcemia and hypoparathyroidism occurring post-operatively when the two different techniques were used. It also compares the complication rates reported by other authors,,,,. The complication rates are low even at the hands of junior surgeons as lateral dissection is avoided minimising trauma to the RLN and blood supply to the parathyroids. In addition, devascularisation on the main glandular tissue itself and keeping a wedge section enables enough thyroid tissue over the capsule to allow for the rarest change of the nerve passing through the capsule. The wedge also gives excellent approximation of the two edges providing better hemostdsis. The only limitation of this technique lies in its use in malignant thyroid tissue where surrounding structures acquire secondary importance to complete ablation of the malignant tissue.
We thank Dr (Mrs) PM Pai, Dean, King Edward Memorial Hospital for allowing us to publish this data.
[Table - 1], [Table - 2]