- Email: [email protected]
Comparison of surgery and radiotherapy in T1 and T2 glottic carcinomas
Comparison of Surgery and Radiotherapy Tl and T2 Glottic Carcinomas Jean Ton-Van, MD, Jean-Louis Lefebvre, MD, Jordan C. Stern, Bernard Cache-De-queant, MD, Bernard Vankemmel,
We retrospectively studied 356 patients who received treatment for Tl and T2 glottie carcinomas. Two hundred and thirty patients were treated with surgery (200 by cordectomy, 15 by vertical partial laryngeetomy, and 15 by subtotal laryngectomy) . Radiotherapy was used to treat 126 patients. There were 206 Tl and 24 T2 lesions in the surgically treated group and 107 Tl and 19 T2 lesions in the radiotherapy group. Sixty-four patients received radiotherapy because it was the treatment of choice (scheduled radiotherapy) and 62 patients received radiotherapy because they had medical contraindications for surgery (default radiotherapy). Actuarial survival rates at 5 years were 84% for patients who underwent surgery and 78% for patients who underwent scheduled radiotherapy. In the surgically treated group, there were 10 local recurrences in 170 patients with tumors of the true vocal cord, eight recurrences in 36 patients with anterior commissure lesions, and 6 recurrences in 24 patients with tumors extending to the arytenoid. In the scheduled radiotherapy group, there were 7 local recurrences in 38 patients with true vocal cord tumors, 6 recurrences in 20 patients with anterior commissure tumors, and 5 recurrences in 6 patients with tumors extending to the arytenoid. We conclude that survival is similar in these patients whether they receive operative treatment or scheduled radiotherapy. However, in the radiotherapy group, local recurrences were more frequent in patients with tumors extending to the arytenoid. We advocate extended functional surgery for patients with Tl and T2 glottic lesions except for those with small tumors arising from the middle third of the vocal cord.
From the Departments of Head and Neck Surgery (JT-V, J-LL, JCS, EB. BV) and Radiotherapy (BC-D), Centrc Oscar Lambret (Northern F&x &ncer Center). L&z, FIX&X. Requests for reprints should be addressed to Jean Ton Van, MD, Department of Surgery, Centre Oscar Lambret, 1, rue F. Combsmale, 59020, Lie, Franoc. Pnscnted at the 37th kulual Meeting of the Society of Head and Neck Surgeons, Maui, Hawaii, May l-4.1991.
THE AMERICAN
in
MD, Etienne Buisset, MD, MD, Lille, France
I
t is widely accepted that early stage squamous cell carcinoma of the vocal cord may be successfully treated by either surgery or radiotherapy. Initial treatment takes many factors into consideration: tumor staging, performance status, associated disease, compliance with close follow-up, and the patient’s acceptance of the pro posed treatment. Thus, some patients with a theoretic indication for surgery have medical contraindications or refuse to undergo the operation. They are then irradiated (default radiotherapy). The purpose of this retrospective study was to compare the outcome of patients undergoing surgery, scheduled radiotherapy, or default radiotherapy, depending on the location of the primary tumor. PATIENTS
AND METHODS
We analyzed data collected from 356 patients with invasive, previously untreated Tl and T2 squamous cell carcinoma of the glottic larynx. These patients were treated either by surgery or radiotherapy between January 1970 and December 1983. The mean patient age was 59 years (range: 26 to 86 years). There were 342 men (96%) and 14 women (4%). All patients were heavy users of alcohol and tobacco. Mean tobacco consumption was 2 1.5 g/day (4 1 pack-year). Mean pure alcohol consump tion before diagnosis was 303 kg. Patients’ tumors were staged according to the 1988 Staging System of the American Joint Committee on Cancer [I]. Patients presenting with a synchronous seu ond primary lesion of the head and neck were excluded. Panendoscopy with fiberoptics was performed on all the patients to evaluate the extension of the tumor. Vocal cord mobility was carefully evaluated to differentiate bs tween a bulky lesion causing restricted motion and a lesion in which there was fmtion secondary to paraglottic space invasion. A total of 313 of 356 patients (88%) presented with Tl tumors. Forty-three of 356 patients (12%) had T2 primary sites. Seventy-five of the 356 tumors (21%) extended to or originated on the anterior commissure. These tumors were classified as Tlb. Twenty-one of 356 tumors (5.8%) extended to the arytenoid cartilage without fmtion of the cricoarytenoid joint. These tumors were classified as T2. The incidence of palpable neck nodes at diagnosis was 1.6% (1.2% for Tl lesions and 2.3% for T2 lesions). Our theoretic treatment protocol was as follows: For T 1 lesions involving only the middle third of the true vocal cord, we used radiotherapy or cordectomy. Patients were advised to have surgery if they were under 50 years, not vocal professionals, would not comply with follow-up, or were difiicult to examine. For Tl lesions extending to the anterior commissure or arytenoid, we performed vertical
JOURNAL
OF SURGERY
VOLUME
162
OCTOBER
1991
337
TON-VANET AL
TABLE I
TABLE II Clinical Staging/Treatment Tla
Surgery p Value* Radiotherapy ‘p
170 = 74%
Tlb 36 = 16%
Local Recurrence Treatment T2
Number of Patients Partial
24 = 10% NS 19 = 15%
Radical
Palliative
Valuebetween surgery and radiotherapy groups.
partial laryngectomy (VPL) or extended cordectomy. These procedures were subsequently replaced by subtotal laryngectomy (STL) or frontal-anterior laryngectomy [2,3]. For T2 lesions with infiltration of the paraglottic space, we performed STL. Patients with medical contraindications to surgery or who refused surgery were treated by “default radiotherapy.” Clinically palpable nodes were always treated by neck dissection and surgery on the primary lesion, whatever the T stage. Cordectomy was performed by laryngofissure and excision of the vocal cord and muscle and the inner perichondrium of the thyroid ala. When necessary, we performed an extended cordectomy (i.e., excision of either the anterior commissure or the arytenoid). In VPL, we removed the anterior angle of the thyroid cartilage and excised the diseased vocal cord, the anterior commissure, and a fourth of the contralateral vocal cord. STL consisted of a supracricoid laryngectomy with crico-hyoidopexy, as described by Piquet et al [ 21. Table I shows the clinical staging as a function of initial treatment. Tl and T2 lesions were equally divided between the surgery and radiotherapy groups, although there were significantly more Tlb lesions in the radiotherapy group (p
15), the mean length of hospitalization was 17 days. All the patients underwent tracheotomy for an average of 13 days, and decannulation could not be achieved in one patient. Mean length of time before oral feeding could be resumed was 14 days. Radiotherapy was used to treat 126 patients (Table I). Fifty-four percent of the tumors were staged as Tla, 3 1% as Tl b, and 15% as T2. Sixty-four patients received radiotherapy because they were candidates for definitive radiotherapy. Sixty-two patients were irradiated because of medical contraindications or because they refused surgery. In the radiotherapy group, we did not include any patients who had lesions that were unsuitable for a voicesparing operation. All patients were treated using megavoltage equipment (60Co, Electrons), as a continuous course, in once-a-day fractionation, five fractions per week. Patients received a median total dose of 60 Gy (range: 50 to 65 Gy). No major complications occurred.
338
162
THE AMERICAN
JOURNAL
OF SURGERY
VOLUME
RESULTS The follow-up period ranged from 5 to 18 years. No patient was lost to follow-up. Local recurrences are shown in Table II. Local recurrences developed in 24 of 230 patients who underwent surgery and in 38 of 126 patients who received radiotherapy (18 of 64 in the scheduled radiotherapy group and 20 of 62 in the default radiotherapy group). Among the 24 patients in the surgical group in whom treatment failed, 21 underwent salvage surgery (15 total laryngectomies, 6 STLs). In the scheduled and default radiotherapy groups, 15 patients in each group were treated by salvage surgery. Table RI shows the effect of tumor extension on local failure rate. Neck recurrences developed in 5.6% of patients (13 of 230) in the surgery group and in 6.3% (8 of 126) in the radiotherapy group. The incidence of neck recurrence in the literature ranges from 0.75% to 7.8% [4-71. These failures were controlled by neck dissection combined with radiotherapy in 66% of cases (14 of 21). Table IV shows the actuarial survival rate calculated by the Kaplan-Meier method [8]. Differences in survival were tested using log-rank analysis. There were no signiflcant differences between the surgery and the scheduled radiotherapy groups. Fifty-two deaths occurred in the surgically treated group. Six patients died of nodal recurrence, two of local failure, and one of nodal and local failure. Six distant metastases were observed. Fourteen patients developed and died of another carcinoma of the upper aerodigestive tract. Three patients died of intercurOCTOBER
1991
SURGERY AND RADIOTHERAPY IN GLOTTIC CARCINOMA
rent die and 18 of unknown causes. There were also two postoperative deaths. A total of 74 patients in the radiotherapy group died. Four deaths were due to nodal recurrence, nine to local failure, and one to nodal and local failure. Five distant metastases were observed. Eleven patients developed a second unrelated carcinoma of the head and neck. Sixteen patients died of intercurrent diise and 28 of unknown causes. COMMENTS In considering factors that affect local failure, most investigators emphasize the adverse influence of tumor volume [9,1O], impairment of vocal cord mobility [4,9,1 I], and extension of the tumor to the anterior commissure [7]. On the other hand, other investigators believe that, when the treatment used is radiotherapy, the local control rate is not related to vocal cord mobility [5,10,12] or extension to the anterior commissure [5,6,22]. In our series, the overall recurrence rate was 10% in the surgically treated group and 28% in the scheduled radiotherapy group (p
TABLE
III
Local Recurrences/Extension Number of Patients Without Extension RT
Extension to Arytenoid
Extension to Ant. corn.
10/19
12139 NS 8136 NS 6/20
16168
Surgery Scheduled RT Ant. corn. = anterior
lChi-square
p <0.001*
p = 0.05*
10/170 p <0.02* 7138
6124 p
commissure; RT = radiotherapy.
test.
tWith Yates’ mrrection.
TABLE
IV
Actuarial Sufvlval Rate (Kaplan-Meier) 1
Year
3 Years
5 Years
10 Years
Surgery
97%
87%
84%
Scheduled RT
95%
84%
78%
Default RT RT
84% 90%
86% 75%
56% 67%
61% NS 54% p <0.01* 26% 38%
RT = radiotherapy. *Log rank test.
mors with impaired mobility, to 100% for Tla lesions [5-71. Conservation surgery remains possible as a salvage procedure only when the recurrence is detected early. In these cases, morbidity is acceptable [6,16]. In our series, although total laryngectomy was the main treatment for salvage surgery after irradiation, we performed partial surgery in highly selected cases. Only 29% of patients (11 of 38) were deemed candidates for conservation surgery after radiation failed. No deaths occurred postoperatively, and morbidity was low. In particular, completion laryngectomy was not necessary in patients with a difficult postoperative course. SUMMARY (1) Treatment of early vocal cord cancer requires close cooperation between radiation therapists and surgeons to ensure a maximal cure rate without loss of laryngeal function. (2) Since local failures after primary radiotherapy can rarely be treated with salvage conservation surgery and only in highly selected cases, we do not advocate primary radiotherapy with surgical salvage in reserve as the treatment of choice for patients with early glottic carcinomas [16-181. (3) Based on the results from this study and reports in the literature, we recommend primary conservation surgery for patients with Tl and T2 glottic lesions except for
JOURNAL
OF SURGERY
VOLUME
162
OCTOBER
1991
339
those with small tumors involving the middle third of the true vocal cord.
REFJZRJZNCES 1. American Joint Committee on Cancer Staging. Manual for staging cancer. 3rd ed. Philadelphia: JB Lippincott, 1988: 39-44. 2. Piquet JJ, Desaulty A, Decroix G. Crico-hyoido-pexie. Technique operatoire et resultats fonctionnels. Ann Qtolaryngol Chir Cervicofac 1974; 91: 681-9. 3. Tucker HM, Wood BG, Levine H, Katz R. Glottic reconstruction after near total laryngectomy. Laryngoscope 1979; 89: 609-18. 4. Laccourreye 0, Brasnu D, Trotoux J, Laccourreye H. L’echec local et ganglionnaire des epitheliomas a point de d6part glottique trait&s par chirurgie partielle laryngk a propos de 432 patients. J Qtolaryngol 1990; 19: 130-5. 5. Mendenhall WM, Parsons JT, Stringer SP, Cassisi NJ, Million RR. Tl-T2 vocal cord carcinoma: a basis for comparing the results of radiotherapy and surgery. Head Neck Surg 1988; 10: 373-7. 6. Woodhouse RJ, Quivey JM, Fu KK, Sien PS, Dedo HH, Phillips TH. Treatment of carcinoma of the vocal cord. A review of 20 years’ experience. Laryngoscope 1981; 91: 1155-62. 7. Amommarn R, Prempree T, Viravathana T, Donavanik V, Wizenberg MJ. A therapeutic approach to early vocal cord carcinoma. Acta Radio1 Oncol 1985; 24: 321-5. 8. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457-80.
344l
THE AMERICAN JOURNAL OF SURGERY
9. Mills EED. Early glottic carcinoma: factors affecting radiation failure, results of treatment and sequelae. Int J Radiat Oncol Biol Phys 1979; 5: 811-7. 10. Karim ABMF, Snow GB, Ruys PN, Bosch H. The heterogeneity of the T2 glottic carcinoma and its local control probability after radiation therapy. Int J Radiat Oncol Biol Phys 1980; 6: 1653-7. 11. Chacko DC, Hendrickson FR, Fischer A. Definitive irradiation of Tl-T4 larynx cancer. Cancer 1983; 51: 994-1000. 12. Howel-Burke D, Peters LJ, Goepfert H, Oswald MJ. T2 glottic cancer recurrence, salvage and survival after definitive radiotherapy. Arch Otolaryngol Head Neck Surg 1990; 116: 830-5. 13. Nichols RD, Mickelson SA. Partial laryngectomy after irradiation failure. Ann Qtol Rhino1 Laryngol 1991; 100: 176-80. 14. Rothfield RE, Johnson JT, Myers EN, Wagner RL. Hemilaryngectomy for salvage of radiation therapy failures. Otolaryngol Head Neck Surg 1990; 103: 792-4. 15. Shah JP, Loree TR, Kowalski L. Conservation surgery for radiation failure carcinoma of the glottic larynx. Head Neck Surg 1990; 12: 326-31. 16. Soo KC, Shah JP, Gopinath KS, Jaques DP, Gerold FP, Strong EW. Analysis of prognostic variables and results after vertical partial laryngectomy. Am J Surg 1988; 156: 2648. 17. Robson NLK, Qswal VH, Flood LM. Radiation therapy of laryngeal cancer: a twenty year experience. J Laryngol Qtol 1990; 104: 699-703. 18. Harwood AR, Hawkins NV, Rider WD, Bryce DP. Radiotherapy of early glottic cancer. Int J Radiat Qncol Biol Phys 1979; 5: 473-6.
VOLUME 162 OCTOBER 1991
We retrospectively studied 356 patients who received treatment for Tl and T2 glottie carcinomas. Two hundred and thirty patients were treated with surgery (200 by cordectomy, 15 by vertical partial laryngeetomy, and 15 by subtotal laryngectomy) . Radiotherapy was used to treat 126 patients. There were 206 Tl and 24 T2 lesions in the surgically treated group and 107 Tl and 19 T2 lesions in the radiotherapy group. Sixty-four patients received radiotherapy because it was the treatment of choice (scheduled radiotherapy) and 62 patients received radiotherapy because they had medical contraindications for surgery (default radiotherapy). Actuarial survival rates at 5 years were 84% for patients who underwent surgery and 78% for patients who underwent scheduled radiotherapy. In the surgically treated group, there were 10 local recurrences in 170 patients with tumors of the true vocal cord, eight recurrences in 36 patients with anterior commissure lesions, and 6 recurrences in 24 patients with tumors extending to the arytenoid. In the scheduled radiotherapy group, there were 7 local recurrences in 38 patients with true vocal cord tumors, 6 recurrences in 20 patients with anterior commissure tumors, and 5 recurrences in 6 patients with tumors extending to the arytenoid. We conclude that survival is similar in these patients whether they receive operative treatment or scheduled radiotherapy. However, in the radiotherapy group, local recurrences were more frequent in patients with tumors extending to the arytenoid. We advocate extended functional surgery for patients with Tl and T2 glottic lesions except for those with small tumors arising from the middle third of the vocal cord.
From the Departments of Head and Neck Surgery (JT-V, J-LL, JCS, EB. BV) and Radiotherapy (BC-D), Centrc Oscar Lambret (Northern F&x &ncer Center). L&z, FIX&X. Requests for reprints should be addressed to Jean Ton Van, MD, Department of Surgery, Centre Oscar Lambret, 1, rue F. Combsmale, 59020, Lie, Franoc. Pnscnted at the 37th kulual Meeting of the Society of Head and Neck Surgeons, Maui, Hawaii, May l-4.1991.
THE AMERICAN
in
MD, Etienne Buisset, MD, MD, Lille, France
I
t is widely accepted that early stage squamous cell carcinoma of the vocal cord may be successfully treated by either surgery or radiotherapy. Initial treatment takes many factors into consideration: tumor staging, performance status, associated disease, compliance with close follow-up, and the patient’s acceptance of the pro posed treatment. Thus, some patients with a theoretic indication for surgery have medical contraindications or refuse to undergo the operation. They are then irradiated (default radiotherapy). The purpose of this retrospective study was to compare the outcome of patients undergoing surgery, scheduled radiotherapy, or default radiotherapy, depending on the location of the primary tumor. PATIENTS
AND METHODS
We analyzed data collected from 356 patients with invasive, previously untreated Tl and T2 squamous cell carcinoma of the glottic larynx. These patients were treated either by surgery or radiotherapy between January 1970 and December 1983. The mean patient age was 59 years (range: 26 to 86 years). There were 342 men (96%) and 14 women (4%). All patients were heavy users of alcohol and tobacco. Mean tobacco consumption was 2 1.5 g/day (4 1 pack-year). Mean pure alcohol consump tion before diagnosis was 303 kg. Patients’ tumors were staged according to the 1988 Staging System of the American Joint Committee on Cancer [I]. Patients presenting with a synchronous seu ond primary lesion of the head and neck were excluded. Panendoscopy with fiberoptics was performed on all the patients to evaluate the extension of the tumor. Vocal cord mobility was carefully evaluated to differentiate bs tween a bulky lesion causing restricted motion and a lesion in which there was fmtion secondary to paraglottic space invasion. A total of 313 of 356 patients (88%) presented with Tl tumors. Forty-three of 356 patients (12%) had T2 primary sites. Seventy-five of the 356 tumors (21%) extended to or originated on the anterior commissure. These tumors were classified as Tlb. Twenty-one of 356 tumors (5.8%) extended to the arytenoid cartilage without fmtion of the cricoarytenoid joint. These tumors were classified as T2. The incidence of palpable neck nodes at diagnosis was 1.6% (1.2% for Tl lesions and 2.3% for T2 lesions). Our theoretic treatment protocol was as follows: For T 1 lesions involving only the middle third of the true vocal cord, we used radiotherapy or cordectomy. Patients were advised to have surgery if they were under 50 years, not vocal professionals, would not comply with follow-up, or were difiicult to examine. For Tl lesions extending to the anterior commissure or arytenoid, we performed vertical
JOURNAL
OF SURGERY
VOLUME
162
OCTOBER
1991
337
TON-VANET AL
TABLE I
TABLE II Clinical Staging/Treatment Tla
Surgery p Value* Radiotherapy ‘p
170 = 74%
Tlb 36 = 16%
Local Recurrence Treatment T2
Number of Patients Partial
24 = 10% NS 19 = 15%
Radical
Palliative
Valuebetween surgery and radiotherapy groups.
partial laryngectomy (VPL) or extended cordectomy. These procedures were subsequently replaced by subtotal laryngectomy (STL) or frontal-anterior laryngectomy [2,3]. For T2 lesions with infiltration of the paraglottic space, we performed STL. Patients with medical contraindications to surgery or who refused surgery were treated by “default radiotherapy.” Clinically palpable nodes were always treated by neck dissection and surgery on the primary lesion, whatever the T stage. Cordectomy was performed by laryngofissure and excision of the vocal cord and muscle and the inner perichondrium of the thyroid ala. When necessary, we performed an extended cordectomy (i.e., excision of either the anterior commissure or the arytenoid). In VPL, we removed the anterior angle of the thyroid cartilage and excised the diseased vocal cord, the anterior commissure, and a fourth of the contralateral vocal cord. STL consisted of a supracricoid laryngectomy with crico-hyoidopexy, as described by Piquet et al [ 21. Table I shows the clinical staging as a function of initial treatment. Tl and T2 lesions were equally divided between the surgery and radiotherapy groups, although there were significantly more Tlb lesions in the radiotherapy group (p
15), the mean length of hospitalization was 17 days. All the patients underwent tracheotomy for an average of 13 days, and decannulation could not be achieved in one patient. Mean length of time before oral feeding could be resumed was 14 days. Radiotherapy was used to treat 126 patients (Table I). Fifty-four percent of the tumors were staged as Tla, 3 1% as Tl b, and 15% as T2. Sixty-four patients received radiotherapy because they were candidates for definitive radiotherapy. Sixty-two patients were irradiated because of medical contraindications or because they refused surgery. In the radiotherapy group, we did not include any patients who had lesions that were unsuitable for a voicesparing operation. All patients were treated using megavoltage equipment (60Co, Electrons), as a continuous course, in once-a-day fractionation, five fractions per week. Patients received a median total dose of 60 Gy (range: 50 to 65 Gy). No major complications occurred.
338
162
THE AMERICAN
JOURNAL
OF SURGERY
VOLUME
RESULTS The follow-up period ranged from 5 to 18 years. No patient was lost to follow-up. Local recurrences are shown in Table II. Local recurrences developed in 24 of 230 patients who underwent surgery and in 38 of 126 patients who received radiotherapy (18 of 64 in the scheduled radiotherapy group and 20 of 62 in the default radiotherapy group). Among the 24 patients in the surgical group in whom treatment failed, 21 underwent salvage surgery (15 total laryngectomies, 6 STLs). In the scheduled and default radiotherapy groups, 15 patients in each group were treated by salvage surgery. Table RI shows the effect of tumor extension on local failure rate. Neck recurrences developed in 5.6% of patients (13 of 230) in the surgery group and in 6.3% (8 of 126) in the radiotherapy group. The incidence of neck recurrence in the literature ranges from 0.75% to 7.8% [4-71. These failures were controlled by neck dissection combined with radiotherapy in 66% of cases (14 of 21). Table IV shows the actuarial survival rate calculated by the Kaplan-Meier method [8]. Differences in survival were tested using log-rank analysis. There were no signiflcant differences between the surgery and the scheduled radiotherapy groups. Fifty-two deaths occurred in the surgically treated group. Six patients died of nodal recurrence, two of local failure, and one of nodal and local failure. Six distant metastases were observed. Fourteen patients developed and died of another carcinoma of the upper aerodigestive tract. Three patients died of intercurOCTOBER
1991
SURGERY AND RADIOTHERAPY IN GLOTTIC CARCINOMA
rent die and 18 of unknown causes. There were also two postoperative deaths. A total of 74 patients in the radiotherapy group died. Four deaths were due to nodal recurrence, nine to local failure, and one to nodal and local failure. Five distant metastases were observed. Eleven patients developed a second unrelated carcinoma of the head and neck. Sixteen patients died of intercurrent diise and 28 of unknown causes. COMMENTS In considering factors that affect local failure, most investigators emphasize the adverse influence of tumor volume [9,1O], impairment of vocal cord mobility [4,9,1 I], and extension of the tumor to the anterior commissure [7]. On the other hand, other investigators believe that, when the treatment used is radiotherapy, the local control rate is not related to vocal cord mobility [5,10,12] or extension to the anterior commissure [5,6,22]. In our series, the overall recurrence rate was 10% in the surgically treated group and 28% in the scheduled radiotherapy group (p
TABLE
III
Local Recurrences/Extension Number of Patients Without Extension RT
Extension to Arytenoid
Extension to Ant. corn.
10/19
12139 NS 8136 NS 6/20
16168
Surgery Scheduled RT Ant. corn. = anterior
lChi-square
p <0.001*
p = 0.05*
10/170 p <0.02* 7138
6124 p
commissure; RT = radiotherapy.
test.
tWith Yates’ mrrection.
TABLE
IV
Actuarial Sufvlval Rate (Kaplan-Meier) 1
Year
3 Years
5 Years
10 Years
Surgery
97%
87%
84%
Scheduled RT
95%
84%
78%
Default RT RT
84% 90%
86% 75%
56% 67%
61% NS 54% p <0.01* 26% 38%
RT = radiotherapy. *Log rank test.
mors with impaired mobility, to 100% for Tla lesions [5-71. Conservation surgery remains possible as a salvage procedure only when the recurrence is detected early. In these cases, morbidity is acceptable [6,16]. In our series, although total laryngectomy was the main treatment for salvage surgery after irradiation, we performed partial surgery in highly selected cases. Only 29% of patients (11 of 38) were deemed candidates for conservation surgery after radiation failed. No deaths occurred postoperatively, and morbidity was low. In particular, completion laryngectomy was not necessary in patients with a difficult postoperative course. SUMMARY (1) Treatment of early vocal cord cancer requires close cooperation between radiation therapists and surgeons to ensure a maximal cure rate without loss of laryngeal function. (2) Since local failures after primary radiotherapy can rarely be treated with salvage conservation surgery and only in highly selected cases, we do not advocate primary radiotherapy with surgical salvage in reserve as the treatment of choice for patients with early glottic carcinomas [16-181. (3) Based on the results from this study and reports in the literature, we recommend primary conservation surgery for patients with Tl and T2 glottic lesions except for
JOURNAL
OF SURGERY
VOLUME
162
OCTOBER
1991
339
those with small tumors involving the middle third of the true vocal cord.
REFJZRJZNCES 1. American Joint Committee on Cancer Staging. Manual for staging cancer. 3rd ed. Philadelphia: JB Lippincott, 1988: 39-44. 2. Piquet JJ, Desaulty A, Decroix G. Crico-hyoido-pexie. Technique operatoire et resultats fonctionnels. Ann Qtolaryngol Chir Cervicofac 1974; 91: 681-9. 3. Tucker HM, Wood BG, Levine H, Katz R. Glottic reconstruction after near total laryngectomy. Laryngoscope 1979; 89: 609-18. 4. Laccourreye 0, Brasnu D, Trotoux J, Laccourreye H. L’echec local et ganglionnaire des epitheliomas a point de d6part glottique trait&s par chirurgie partielle laryngk a propos de 432 patients. J Qtolaryngol 1990; 19: 130-5. 5. Mendenhall WM, Parsons JT, Stringer SP, Cassisi NJ, Million RR. Tl-T2 vocal cord carcinoma: a basis for comparing the results of radiotherapy and surgery. Head Neck Surg 1988; 10: 373-7. 6. Woodhouse RJ, Quivey JM, Fu KK, Sien PS, Dedo HH, Phillips TH. Treatment of carcinoma of the vocal cord. A review of 20 years’ experience. Laryngoscope 1981; 91: 1155-62. 7. Amommarn R, Prempree T, Viravathana T, Donavanik V, Wizenberg MJ. A therapeutic approach to early vocal cord carcinoma. Acta Radio1 Oncol 1985; 24: 321-5. 8. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457-80.
344l
THE AMERICAN JOURNAL OF SURGERY
9. Mills EED. Early glottic carcinoma: factors affecting radiation failure, results of treatment and sequelae. Int J Radiat Oncol Biol Phys 1979; 5: 811-7. 10. Karim ABMF, Snow GB, Ruys PN, Bosch H. The heterogeneity of the T2 glottic carcinoma and its local control probability after radiation therapy. Int J Radiat Oncol Biol Phys 1980; 6: 1653-7. 11. Chacko DC, Hendrickson FR, Fischer A. Definitive irradiation of Tl-T4 larynx cancer. Cancer 1983; 51: 994-1000. 12. Howel-Burke D, Peters LJ, Goepfert H, Oswald MJ. T2 glottic cancer recurrence, salvage and survival after definitive radiotherapy. Arch Otolaryngol Head Neck Surg 1990; 116: 830-5. 13. Nichols RD, Mickelson SA. Partial laryngectomy after irradiation failure. Ann Qtol Rhino1 Laryngol 1991; 100: 176-80. 14. Rothfield RE, Johnson JT, Myers EN, Wagner RL. Hemilaryngectomy for salvage of radiation therapy failures. Otolaryngol Head Neck Surg 1990; 103: 792-4. 15. Shah JP, Loree TR, Kowalski L. Conservation surgery for radiation failure carcinoma of the glottic larynx. Head Neck Surg 1990; 12: 326-31. 16. Soo KC, Shah JP, Gopinath KS, Jaques DP, Gerold FP, Strong EW. Analysis of prognostic variables and results after vertical partial laryngectomy. Am J Surg 1988; 156: 2648. 17. Robson NLK, Qswal VH, Flood LM. Radiation therapy of laryngeal cancer: a twenty year experience. J Laryngol Qtol 1990; 104: 699-703. 18. Harwood AR, Hawkins NV, Rider WD, Bryce DP. Radiotherapy of early glottic cancer. Int J Radiat Qncol Biol Phys 1979; 5: 473-6.
VOLUME 162 OCTOBER 1991