Standard treatment regimes for supraglottic larynx carcinoma worldwide involve the options of partial or total laryngectomy and/or (chemo)radiotherapy. However, transoral laser microsurgical procedures, especially for advanced stage supraglottic carcinoma, are still controversial. In the last two decades, there have been several publications reporting successful laser microsurgical treatments in Europe [
7,
14‐
16] and in the United States [
5,
17,
18]. Successful treatment can be measured with the parameters of primary control, percentage of organ preservation, neck control, and survival. Nonetheless, any discussion of the oncologic outcomes is challenging, since evaluation of the evidence supporting the effectiveness of one treatment over the other is complicated by different stages, outcome measures, use of statistics, and investigations of different laryngeal sites (supraglottic vs. glottic vs. all sites).
Local and loco-regional control
In our series patients with cancer of the supraglottic larynx and all T categories were managed by transoral laser surgery with or without selective neck dissection and with or without (chemo)radiotherapy. Five-year overall local control rate for stage I and stage II was 87.5 % and for stage III and IV was 81.1 %.
Local control rates of conventional supraglottic laryngectomy range from 90–100 % for T1, 80–97 % for T2, and 71–94 % for T4 carcinomas. In a retrospective review of 202 patients with supraglottic squamous cell carcinoma of all stages, Lutz et al. [
19] achieved high local control rates with primary failures in 2 % of patients (3 of 130) following total laryngectomy and in 1 patient of 72 (1 %) treated by conventional supraglottic partial resection. Bocca et al. [
20] reported a 2-year local recurrence rate of 16.5 % (stage I and II) and 21.5 % (stage III and IV) for open surgery. The vast majority in his series showed pT2 cancer (70 %), and additional 13 % pT1 and 17 % pT3. So the results were achieved with 83 % of the patients presenting with pT1 or pT2 tumor. In comparison, our patients showed pT1 or pT2 tumor in 42 % and pT3 or pT4 tumor in almost 48 % and we achieved a 2-year local control rate of 93 % for pT1/2, 93 % for pT3, and 84 % for pT4 tumors.
Primary radiotherapy can achieve local control rates of 77–100 % in supraglottic T1 cancers and 62–83 % in T2 cancers [
21‐
25]. For supraglottic T3 carcinomas local control rates of 50–76 % are published [
26‐
28]. Mendenhall et al. [
23] reported 5-year local control rates after radiotherapy of 83 % for T2 and 68 % for T3 tumors. Our corresponding results were 85 and 82 %. While results are comparable in smaller T2 tumors, transoral laser surgery has a clear advantage in pT3 cases. In a series of 274 patients treated for supraglottic carcinoma by radiotherapy alone or radiotherapy with neck dissection, Hinerman et al. [
26] reported an actuarial probability of local control of 100 % for T1, 86 % for T2, 62 % for T3, and 61 % for T4.
Using TLM, Ambrosch et al. [
14] achieved a local control of 89 % in early T1-2 supraglottic carcinoma. Iro et al. [
29] found stage-dependent 5-year local control rates of 86.1 % for stage I, 74.6 % for stage II, 75.4 % for stage III, and 78.4 % for stage IV. Grant et al. [
30] investigated results of 38 patients undergoing TLM for carcinoma of the supraglottic larynx of all T categories. Twenty-six patients (68 %) underwent neck dissections. Thirteen patients (34 %) received adjuvant radiotherapy. The 2-year Kaplan–Meier estimates for local control were 97 %. In our study, we achieved 91 % for early disease and 85 % for advanced disease. Beside small groups the slightly better results presented by Grant et al. [
30] may be explained by the fact, that 42 % of the patients presented with pT3/4 tumors, whereas in our series 64 % of the patients showed advanced disease.
Davis et al. [
31] reported a local control of 97 % for 48 T2 tumors using TLM in combination with adjuvant (chemo)radiotherapy in 38 patients, which is in line with our results and data presented by Eckel [
15] and Zeitels et al. [
17].
Organ preservation in our study could be achieved in 26 of 26 pT1 cases (100 %), in 88 of 92 pT2 cases (96 %), in 97 of 104 pT3 cases (93 %), and in 46 of 55 pT4 cases (84 %). Overall, this results in larynx preservation of 93 %. In contrast to these results, Johansen et al. [
32] report about salvage laryngectomy in 33 of 117 patients in T1 supraglottic carcinomas after primary radiotherapy. This translates into organ preservation of less than 72 % in T1 cancer. Hinerman et al. [
26] treated patients with radiotherapy for supraglottic T3 carcinoma and were able to preserve the larynx in 68 %, Nakfoor et al. in 72 % of the cases [
27].
Survival
When discussing our results, it has to be mentioned that our patients included a high number of stage III (32 %) and stage IV (32 %) tumors. In our series we measured 5-year overall, recurrence-free and disease-specific survival rates for stage I and II of 77, 81, and 92 %; for stage III and IVa of 59, 65, and 81 %.
In a large review of the National Cancer Data Base, Hoffman et al. [
36] investigated data of 158,426 patients with laryngeal cancer. The authors found that survival has decreased among patients with laryngeal cancer during the past two decades in the United States. During this time there has been an increase in the nonsurgical treatment of laryngeal cancer. Advanced T3 laryngeal cancer 5-year observed and relative survival showed the best outcome for patients whose initial management was surgery, either alone or combined with irradiation. Observed survival rates were 52.9 % for surgery alone, 33.1 % for irradiation, 55.4 for surgery and adjuvant irradiation, and 50.7 % for (chemo)radiotherapy.
Oncologic results of transcervical approaches have been presented previously in several studies [
37‐
39]. For transcervical supraglottic laryngectomy, 5-year survival rates range between 68 and 89 % (all stages); for total laryngectomy between 55 and 80 %. Lutz et al. [
19] investigated the outcome of 202 patients with supraglottic carcinoma of the larynx after conventional surgery. The authors state a 2-year survival rate of 67 % for stage I, 82 % for stage II, 84 % for stage III, and 55 % for stage IV. Our corresponding 2-year overall survival results were 90 % for stage I and II tumors and 83 % for stages II and IVa. The results of Lutz et al. include 130 patients (60 %) who underwent primary total laryngectomy whereas in our series only 20 of 277 patients (7 %) needed secondary total laryngectomy as a salvage procedure.
In a large retrospective study (901 patients), Scola et al. [
33] reported a 5-year uncorrected actuarial survival rate of 84, 81, 76, and 55 % for stages I, II, III, and IV. All 12 stage I cases were treated by conventional or extended supraglottic laryngectomy. Again, there were no significant differences in survival compared with our organ and function preserving results.
Spaulding et al. [
25] reported a series of 162 patients with supraglottic cancer treated with radiotherapy and surgical salvage, preoperative radiotherapy and surgery, total laryngectomy and postoperative radiotherapy, and partial laryngectomy with postoperative radiotherapy. The 3-year determinant survival rate is stated as 65 % for T1 cancer, 81 % for T2, 57 % for T3, and 61 % for T4. Our corresponding 3-year overall survival results were 86 % for early disease (pT1/2) and 73 % for advanced disease (pT3/4) and thus compare favorably.
Primary (chemo)radiotherapy can achieve a corrected 5-year survival rate of 70 % as presented by Sykes et al. [
28] 331 patients with clinically node-negative supraglottic carcinoma of the larynx were treated with radiotherapy. Examined by T-category, corrected survivals were 83, 78, 53, and 61 % for T1, T2, T3, and T4 tumors respectively. Mendenhall et al. [
40] reported 129 patients with supraglottic cancer (all stages) which were treated by radiotherapy alone or followed by neck dissection. The authors observed an absolute 5-year survival rate of 58 % for all stages. Nakfoor et al. [
27] treated 190 patients with carcinoma of the supraglottic larynx with primary (chemo)radiotherapy. For T1, T2, T3, and T4 tumors, relapse-free survival rates were 78, 82, 64, and 40 %.
The first paper of the Göttingen study group was published in 1998 and presented good results for early supraglottic laryngeal carcinoma (pT1/pT2) [
14]. 48 patients with supraglottic carcinoma were treated almost exclusively (96 %) by TLM. The 5-year recurrence-free rate and 5-year overall survival rate were 83 and 76 %, respectively. In our large series of 277 patients, the 5-year recurrence-free and 5-year overall survival rates were 81 and 77 %, respectively. These data are in line with the good results presented by Ambrosch et al. [
14].
Iro et al. [
29] reviewed the medical records of 141 patients with supraglottic carcinomas undergoing transoral laser surgery, facultatively in combination with neck dissection or radiotherapy. Five-year recurrence-free survival rates were 65.7 %; stage I, 85.0 %; stage II, 62.6 %; stage III, 74.2 %; and stage IV, 45.3 %. The local and regional recurrence rates were 16.3 and 9.9 %, respectively. Vilaseca et al. [
18] investigated 147 patients with T3 glottic and supraglottic laryngeal carcinoma and observed a 5-year disease-specific survival rate of 61.8 % for supraglottic carcinoma.
In our series we measured 5-year overall, recurrence-free and disease-specific survival rates for stage I and II of 77, 81, and 92 %; for stage III and IVa of 59, 65, and 81 % which are good results corroborating the relevance of TLM as a valuable option for the treatment of supraglottic laryngeal cancer.
Complications
Twenty-six patients (9 %) experienced postoperative bleeding, 22 (85 %) in the larynx requiring coagulation and/or clipping in the operating room under general anesthesia. This is slightly higher than the occurrence of postoperative bleeding reported after open supraglottic laryngectomy, which varies between 1.6 and 5 %, according to Herranz-Gonzalez et al. [
38].
In their large series of 901 patients, Scola et al. [
33] state an incidence of 3.8 % for pneumonia and 4.5 % for fistula. We observed five pneumonias (2 %) and no fistulas in our own study. Roh et al. [
41] show the incidence of aspiration pneumonia after TLM of 11.5 % and after transcervical approach up to 40 %. Lower complication rates after TLM may be attributed to the preservation of healthy tissue and functionally important structures.
In contrast to the 5 % incidence of vestibular stenosis after open supraglottic laryngectomy, as stated by Staffieri [
42], or 2.5 % observed by Herranz-Gonzalez et al. [
38], we saw only 1 (0.4 %) stenosis after extensive laser resection of the supraglottis and paraglottic space bilaterally. This low incidence may be attributed to the lack of wound infection and perichondritis and the preservation of the framework of the larynx, contrary to the findings of Laccourreye et al. [
43], who observed significant complications after partial laryngectomy and radiotherapy. Although their patients were highly selected (exclusion of patients with postoperative complications before adjuvant radiotherapy), there were five cases of laryngeal radionecrosis (5.5 %), four laryngeal stenoses (4.4 %), two stenoses of the esophageal inlet (2.2 %), and three pneumonias (3.3 %). More severe complications caused by radiotherapy were reported by Mendenhall et al. [
23] who observed a carotid blowout in one patient, one partial spinal transsection, three laryngectomies and one permanent tracheotomy because of chondronecrosis and another two permanent tracheotomies due to laryngeal edema.
Functional results
The 4 % rate of temporary tracheotomy compares favorably with reports of conventional surgery by Robbins et al. [
39] who reported an incidence of permanent tracheotomies of 6 %, the 6 % failed decannulations observed by Herranz-Gonzalez et al. [
38] and the up to 15 % of delayed decannulation or late recanulation reported by Krespi and Kheterpal [
44] in a literature review. Scola et al. [
33] observed failed decannulation in 9 % and delayed decannulation (more than 1 month postoperatively) in 38 %. Spirano et al. [
45] observed a median time to decannulation of 4.9 months and a persistent aspiration in 9 %. Kollisch et al. [
46] compared functional results following transcervical supraglottic laryngectomy and transoral laser resection. Concerning swallowing function, tracheotomy rate, and aspiration pneumonia, TLM was superior compared with the transcervical approach. Tracheotomy rates following TLM for supraglottic carcinoma range between 0 and 32 %. In our series 4 % of patients received temporary tracheotomy due to aspiration or bleeding. These results are in favor of laser microsurgery since with the open approach temporary tracheotomy is necessary in almost every patient.
Seventy-four percent of our patients had their feeding tubes removed within the first 14 days. An additional 18 % were weaned from tube feeding after a total of 30 days. Prolonged tube feeding (31–90 days) was necessary in 6 % of our patients. This is comparable to delayed swallowing in 5.5 % of open supraglottic laryngectomy as reported by Bocca [
20]. Five of our patients (2 %) needed permanent gastrostomy and a total of three patients had to undergo total laryngectomy because of aspiration. Taking into account that the majority of our patients had stage III and IV disease we think that the result of 3 % (8 patients) with a failed swallowing function restoration after TLM is acceptable.
Bernal-Sprekelsen et al. [
47] made a valuable contribution on aspiration rehabilitation specifically in patients after TLM of 210 patients presenting with malignant tumors of the hypopharynx and larynx. Endoscopic resections included T2 to T4 tumors, and nasogastric feeding tube was used in 23.2 % of small tumors and in 63 % of locally advanced tumors. Twelve patients (5.7 %) had postoperative pneumonia and 59 (28.1 %) had temporary aspiration that correlated with location and locally advanced tumors. The authors concluded that endoscopic resection of laryngeal and hypopharyngeal tumors is associated with good recovery of deglutition.
As a comparison, Schwaab et al. [
48] report a rate of 9 % total laryngectomies (13 patients) in a series of 146 patients treated by supracricoid partial laryngectomy with cricohyoidopexy, because of severe postoperative aspiration. Logemann et al. [
49] identified the impaired larynx sphincter function and the movement of the tongue base, as the two critical factors for recovery of swallowing. The extent of aspiration depends on the achievement of these two functions and on the age of the patient. In addition to that, preservation of the sensory function seems to be important for restituting swallowing function, according to Flores et al. [
50]. It seems therefore feasible to preserve as much hyoid bone, strap muscle, and parts of the superior laryngeal nerve branches as possible.