ABSTRACT
PURPOSE: Cervical chylous fistula (CF) development is a rare complication after neck dissection in patients with thyroid carcinoma. However, CFs are potentially fatal if left untreated. The present study evaluated three CF management protocols in thyroid cancer patients who had undergone neck dissection. METHODS: A total of 22 CF cases developed in the 353 neck dissections performed in 309 thyroid cancer patients over a period of 2 years. The CF cases involved 6 males and 16 females with a median age of 43.3 years (range, 26-63). The patients weredivided into 3 groups for analysis based on treatment modalities: Group A (n=14), conservative treatment only; Group B (n=5), conservative treatment plus Sandostatin® administration (initially, there were 7 patients in this group, but 2 patients were converted to re-surgery) Group C (n=3), re-surgery due to high-output fistula (>500 ml/d), which in some cases did not respond to conservative treatment plus SandostatinⓇ. Each group was analyzed in terms of total drainage volume, duration of hospital stay and response to treatment. RESULTS: Eighteen CFs occurred in left neck dissection patients, and 4 in right neck patients. Chylous drainage was greater in left neck compared to right neck patients (P= 0.033). All right-sided fistulasclosed following conservative treatment only. The chyle drainage period was longer for Group A (7.6 days) than Group B (4.2 days) patients (P= 0.019), and the duration of hospital stay was longer for Group A than Group B patients (P=0.026). In Group C, re-surgery to close the fistula resulted in termination of chyle flow in all cases. The only complication was wound infection in 1 case (4.6 %). There were no recurrences of CFs in any group. CONCLUSION: The majority of CF cases can be successfully controlled using conservative treatment only. The additional use of Sandostatin® can reduce the duration of CF drainage and lead to earlier CF closure. However, in case where fistula output exceeds 500 ml/d, early re-surgery should be considered.