Impact of preoperatative tracheostomy on tracheostome recurrence and overall survival in patients undergoing laryngectomy

Introduction Tracheostome recurrence (TR) following primary total laryngectomy (TL) for squamous cell carcinoma ( SCC) occurs in 2–15% [1,2] and has been defi ned as ‘a diff use infi ltration of neoplastic tissue at the junction of the trachea and skin’ [1]. It is associated with very poor prognosis despite aggressive surgery or radiotherapy. Factors that are thought to predispose patients to TR include advanced T stage and N stage, subglottic involvement, and preoperative tracheostomy [3]. Other predisposing factors include insuffi cient tracheal resection margins and thyroid gland invasion [4].


Introduction
Tracheostome recurrence (TR) following primary total laryngectomy (TL) for squamous cell carcinoma ( SCC) occurs in 2-15% [1,2] and has been defi ned as 'a diff use infi ltration of neoplastic tissue at the junction of the trachea and skin' [1]. It is associated with very poor prognosis despite aggressive surgery or radiotherapy. Factors that are thought to predispose patients to TR include advanced T stage and N stage, subglottic involvement, and preoperative tracheostomy [3]. Other predisposing factors include insuffi cient tracheal resection margins and thyroid gland invasion [4].
Most cases are diagnosed within the fi rst year following TL, but some cases present as late as 3.5 years after surgery [4]. It is associated with 90% mortality, with more than 80% of the patients dying within the fi rst 24 months [5].
Preoperative tracheostomy ( POT) is one of several factors associated with an increased risk for TR. Several articles have reported a poor outcome with POT [6][7][8], whereas others could not fi nd a signifi cant relation [9,10]. It seems plausible that POT might disrupt the primary tumor or seed the tracheal tract and lead to reduced local control.
Th e diff erent treatment modalities for these recurrences include surgery, radiotherapy, and radiochemotherapy. All these modalities have not been satisfactory in controlling the recurrence and therefore, special attention should be given to the prevention of such pathologies [2,11,12].

Impact of preoperatative tracheostomy on tracheostome recurrence and overall survival in patients undergoing laryngectomy
Mohamed S. Taha, Reda Sabra, Tarek A. Hamdy, Magdy A. Riad

Background
The development of tracheostomerecurrence after total laryngectomy is a complication with a dismal prognosis .The average survival rate of tracheostome recurrence (TR) is 7.5%, ranging from 1.7% to 40%The TR consists of diffuse in ltration of tumor into the soft tissues of the neck and mediastinum; hence the control of this disease is dif cult.

Objective
To analyze the impact of preoperative tracheostomy ontracheostome recurrence (TR), regional recurrence (RR), and overall survival (OS) in patients undergoing primary laryngectomy.

Material and methods
Thirty three consecutive patients, who underwent primary total laryngectomy for locally advancedlaryngeal squamous cell carcinoma, were enrolled in this study. Patients whounderwent primary chemo-radiotherapy, partial laryngectomyor those treated palliatively wereexcluded from the study.Patient factors analyzed included age,gender, primarytumor site, TNM classi cation, type of procedure, thyroid gland management, extent of neck dissection and preoperative tracheostomy (POT). The timeinterval between tracheostomy and de nitive surgery was calculated.

Results
A total of 33 patients underwent total laryngectomy(TL) for squamous cell carcinoma of the larynx.Thirty patients (90.9%) were males, while 3 (9.1%) patients were females. Their mean age was 57.7 (±11.6) years, 26 (78.8%) of them were smokers, their follow up mean time was 23 (±6.6) months. Tracheostomy recurrence occurred in 3 (9.1%) patients and regional recurrence in another 3(9.1%) patients. The 2 years survivalfor the whole patients was 84.8% and overall survival was 81.1%. Nineteen patients had POTbetween 10 to 21days (median 15)prior to TL surgery. Fourteen patients had their tracheostomy at the time of surgery .There was no statistical signi cant difference between both groups as regard overall survival, stomal recurrence and regional recurrence.
Many methods have been proposed to prevent recurrence in the tracheostomy area, such as a broader surgical margin by resecting the trachea at a lower level, excision of lymph-nodes located in the paratracheal area, emergency laryngectomies rather than preoperative tracheostomy if the patients present with signifi cant airway obstruction, and postoperative radiotherapy including the tracheostomy and the upper mediastinum [1,6,13,14]. Th eoretically, any delay between the POT and the defi nitive TL would allow the seeded tumor cells to more eff ectively establish themselves in the fresh tracheostomy.
In this study, we present our experience with a series of laryngectomy patients, with the aim of analyzing the impact of POT on TR, regional recurrence, and overall survival (OS) in patients undergoing primary laryngectomy.

Patients and methods
Th is study was conducted after approval by the Institutional Review Board of Ain Shams University Hospitals and obtaining informed consent from all participants.
Th is is a prospective study, carried out between January 2011 and October 2013. Th irty-three consecutive patients who underwent primary TL for locally advanced laryngeal SCC were enrolled in this study. Patients who underwent primary chemoradiotherapy or partial laryngectomy, and those treated palliatively were excluded from the study.
Patient factors analyzed included age, sex, primary tumor site, tumor, node, metastasis classifi cation, type of procedure, thyroid gland management, extent of neck dissection, and POT.
Th e time interval between tracheostomy and defi nitive surgery was calculated. All patients who had undergone POT had the tracheal window excised and sent for histological analysis. All patients underwent primary TL, with or without unilateral or bilateral neck dissection. Histological features, such as the degree of diff erentiation, and the adequacy of the pathological margin of excision were documented. Th e decision to administer postoperative radiotherapy or radiochemotherapy to the patients was recorded.
Statistical analysis was carried out on a personal computer using MedCalc© version 11.4 (MedCalc© Software, Mariakerke, Belgium).
Th e Kolmogorov-Smirnov goodness-of-fi t test was used to test the normality of numerical data distribution.
Normally distributed numerical data are presented as mean and SD, and diff erences between the two groups were compared using the unpaired Student t-test. Skewed numerical data are presented as median and interquartile range, and the Mann-Whitney U-test was used to compare intergroup diff erences. Categorical data are presented as number and percentage, and diff erences between the two groups were compared using the Pearson 2 -test and the 2 -test for trends for nominal and ordinal data, respectively. Fisher's exact test was used in place of t he 2 -test if greater than 20% of cells in any contingency table had an expected count of less than 5.
Th e Kaplan-Meier method was used to create curves for survival and for the time to stomal or regional recurrence in the two groups. Th e Kaplan-Meier curves of both groups were compared using the logrank test. All P-values are two-sided. P less than 0.05 is considered statistically signifi cant.

Results
A total of 33 patients underwent TL for SCC of the larynx. Th irty patients (90.9%) were male, whereas three (9.1%) patients were female. Th eir mean age was 57.7 (±11.6) years; 26 (78.8%) of them were smokers; their mean follow-up period was 23 (±6.6) months. TR occurred in three (9.1%) patients, two of them were Sisson's stage and one was Sisson's stage IV, and regional recurrence occurred in another three (9.1%) patients. Th e 2-year survival for the entire patient group was 84.8%, and the OS was 81.1%.
Nineteen patients had undergone POT betwe en 10 and 21 days (median 15) before TL surgery. Fourteen patients underwent tracheostomy at the time of surgery [intraoperative tracheostom y (IOT)]; demographic data are shown in Table 1.
Th ere were no statistically signifi cant diff erences between the two groups in age, sex, smoking habits, T and M stage, tumor diff erentiation, or tumor site. Signifi cant diff erences were observed in the tumor stage, with the POT group having a higher clinical stage and cervical nodal metastases (Table 2).
Further subgroup analysis was carried out using the T-classifi cation. Again POT was not predictive of  Data are presented as number (%). On the basis of Kaplan-Meier survival curves for the IOT and POT groups, there was no statistically signifi cant diff erence between both groups with regard to OS [hazard ratio = 0.717, 95% confi dence interval (CI) = 0.142 to 3.631, P = 0.680; Figure 1 and Table 4]. In other words, POT was not statistically signifi cantly associated with poorer outcome in terms of OS.
Kaplan-Meier survival curves for stomal recurrence and regional recurrence in the IOT and POT groups showed that there was no statistically signifi cant diff erence between the two groups (hazard ratio = 0.342, 95% CI = 0.034-3.422, P=0.357 and hazard ratio = 0.373, 95% CI = 0.038-3.672, P = 0.401) (Figs. 2 and 3) and (Table 3). Th is result indicates that there was no statistically signifi cant association between POT and regional or stomal recurrence.

Discussion
Development of TR after TL is a complication with a dismal prognosis [14]. Th e average survival rate of TR is 7.5%, ranging from 1.7 to 40% [15].
TR includes diff use infi ltration of tumors into the soft tissues of the neck and the mediastinum; hence, control of this disease is diffi cult [15]. Stomal recurrence prevention is therefore of paramount importance and seems to be the only means of reducing incidence [15,16]. It has b een categorized by Sisson, and this classifi cation is known to be correlated with the selection of appropriate management and outcome [16].
Th e study by Keim et al. [1] reported a 14% incidence of stomal recurrence if laryngectomy was performed at the time of emergency tracheostomy, compared with a 41% incidence if laryngectomy was performed more than 2 days after tracheostomy. Other studies stated that there is no relation between the two [9,10].
Th e main aim of this study was to evaluate patients who had undergone TL and study the impact of POT on stomal recurrence, regional recurrence, and OS.
Th irty-three consecutive patients who had undergone TL for SCC of the larynx at our unit were studied. Th irty patients (90.9%) were male, whereas three (9.1%) were female; this indicates the male predominance of the disease, which is in agreement with the fi ndings from other studies [17]. Th e mean age of the patients was 57.7 (±11.6) years; 26 (78.8%) of them were smokers; and their mean follow-up period was 23 (±6.6) months. Stomal recurrence occurred in three (9.1%) patients; this is in agreement with a study conducted by Breneman et al. [18], in which only two (11%) stomal recurrences occurred among 18 laryngectomy patients with POT. In addition, Yotakis et al. [13] in their review of 352 patients reported only 21 (6%) to have developed a stomal recurrence.
Th is study showed that there was no signifi cant diff erence in the development of stomal recurrence between the group that underwent POT and the group that underwent intraoperative tracheostomy during TL. Th is is in agreement with the fi ndings of Pezier Kaplan-Meier survival curves for the IOT and POT groups; there was no statistically signi cant difference between both groups with regard to overall survival (P = 0.680). IOT, intraoperative tracheostomy; POT, preoperative tracheost omy.

Figure 1
Kaplan-Meier survival curves for stomal recurrence in the IOT and POT groups showing that there is no statistically signi cant difference between the two groups (P = 0.357). IOT, intraoperative tracheostomy; POT, preoperative tracheost omy. et al. [19], who concluded that there is no statistically signifi cant diff erence in OS, disease-specifi c survival, and local recurrence-free survival between patients undergoing POT and those undergoing intraoperative tracheostomy during TL. In addition, Yotakis et al. [13] reported that there was no signifi cant diff erence in the rate of stomal recurrence between those undergoing emergency tracheostomy (23.3%) and those undergoing intraoperative tracheostomy (18.2%).
In contrast, Halfpenny and McGurk [20] found in their study that three (1%) of 265 patients developed a stomal recurrence. All recurrences occurred in the group in which tracheostomy had been performed before laryngectomy; all these three patients had N+ disease. Th ey concluded that the timing of tracheostomy placement is important in reducing the risk for stomal recurrence.
Kaplan-Meier survival curves for the IOT and POT groups show that there is no statistically signifi cant diff erence between both groups with regard to overall survival, stomal recurrence, and regional recurrence (P = 0.680, 0.357, and 0.401, respectively; Figs. 1-3). POT was therefore not statistically associated with a poorer outcome in terms of OS, regional recurrence, and stomal recurrence.
Th is article targets an important issue in the management of laryngeal cancer, namely the impact of POT on survival. We need larger series and multicenter studies with longer periods of follow-up for these patients to support the evidence that POT is not a risk factor for poor outcome.

Conclusion
Management of the compromised airway in advanced laryngeal carcinoma remains a challenge; our results are in keeping with more recent studies, which suggest that POT is not necessarily related to higher stomal recurrence or poor oncological outc ome.
Kaplan-Meier survival curves for the IOT and POT groups showing that there is no statistically signi cant difference between the two groups with regard to regional recurrence (P = 0.401). IOT, intraoperative tracheostomy; POT, preoperative tracheost omy.