Risk Factors of Stomal Recurrence After Laryngectomy: A Systematic Review and Meta-analysis

Abstract

Objective:

We wished to investigate the risk factors for stoma recurrence following laryngectomy.

Methods:

PubMed, Cochrane Library, and Embase were searched to identify primary research studies published between January 1, 1967, and October 31, 2016. We only included observational epidemiological studies and used risk ratios (RRs) and 95% confidence intervals (CI) to summarize the primary risk factors associated with recurrence following laryngectomy.

Results:

A total of 44 articles, including 11 928 patients, were included in the current meta-analysis. The pooled estimates of the stomal recurrence rate and the local recurrence rate following laryngectomy were 6.60% (95% CI, 5.40-7.90) and 19.40% (95% CI, 14.00%-24.80%), respectively. For stomal recurrence, we confirmed a series of earlier identified factors, including tumor site, prior tracheotomy, tracheotomy timing, T-classification, lymph node metastases, postoperative pharyngoperistomal fistula, and a positive surgical margin. For local recurrence, postoperative radiotherapy, overexpression of p53, and overexpression of p21 were significant, while preoperative tracheostomy and postoperative pharyngoperistomal fistula had an unexpectedly low impact on risk.

Conclusion:

This study has confirmed a series of earlier identified factors for stomal recurrence and local recurrence following laryngectomy. Our results will provide important insights for clinical practice.

Keywords

laryngeal cancer local recurrence stomal recurrence risk factor systematic review

Introduction

Laryngeal cancer is the most common cancer of the head and neck, with almost 13 400 new cases diagnosed in the United States annually and accounts for 3.5% to 5.1% of all new malignancies worldwide each year.^1,2 The treatment options for laryngeal cancer include surgery, radiation therapy, and chemoradiation therapy, alone or in combination. Total laryngectomy (TL) is usually reserved for advanced laryngeal cancer and cases that recur after radiotherapy or chemoradiotherapy due to its good local control.³ However, tumor recurrence at the tracheal stomal following total laryngectomy for a laryngeal cancer is still a well-recognized phenomenon, with an overall incidence ranging from 2% to 15%.⁴ Stoma recurrence is one of the most severe complications of laryngeal cancer and commonly considered to be incurable. Even with aggressive surgical therapy and radiotherapy, these patients usually succumb to respiratory obstruction or massive haemorrhage.

To the best of our knowledge, only 1 systematic review of this topic has been published, in 2013 in a Chinese journal, which reported that transglottic location, tumor T-staging, and preoperative tracheotomy were important risk factors of recurrence after total laryngectomy.⁵ Discouragingly, authors from other countries cannot access this review due to the lack of an English version of this review. We wished to update the systematic review of this topic in English and provide the information for more authors in the world. We also wished to include papers focused on distinct recurrence and more risk factors and add some important subgroup analyses to provide a full-scale and novel information on this topic.

Materials and methods

Study Selection

We searched PubMed, Cochrane Library, and Embase to identify primary research studies published between January 1, 1967, and October 31, 2016, that reported risk factors of cancer recurrence after laryngectomy for laryngeal carcinoma. The following keywords were used in this study: stomal recurrence, tracheostomal recurrence, peristomal recurrence, local recurrence, local control, tumour recurrence, and laryngectomy. To identify other studies, we also hand searched references from the retrieved articles. We used only data from published articles and did not include data from meetings or conference abstracts. No language restrictions were applied to the search.

For inclusion, studies had to meet the following criteria:

observational epidemiological studies of the risk factors of tumor recurrence after laryngectomy for laryngeal carcinoma;

reported stomal recurrence or local recurrence after laryngectomy;

reported at least 1 risk factor potentially related to tumor recurrence;

reported the prevalence or the relative risk (RR) estimates, along with either corresponding 95% confidence intervals (CIs) or raw data enabling the calculation of 95% CIs.

We excluded studies that contained overlapping data. When there were multiple publications that used the same data set, only the largest study was included. We excluded studies utilizing animal experiments, review articles, letters, and conference abstracts. We also excluded papers reporting only cancers of the head and neck that did not clearly define stomal recurrence or local recurrence, did not include laryngectomy in laryngeal cancer patients, and had a sample size of less than 30 patients.

Two authors searched the electronic databases independently using the same methodology. The same authors determined whether abstracts and full-text articles met the inclusion criteria independently. If discrepancies occurred, a third author performed an additional assessment.

Data Extraction

Two authors extracted the data independently using a standard protocol, and a third author reviewed the results. Disagreements were resolved by discussion with our research team. For each study, the following information was extracted: first author of the paper, year of publication, type of study, country, risk factors, total sample size, mean age of the patients, definition of the patients, treatment modality, tumor recurrence types, number of tumor recurrences, and total patients in different risk groups.

We used the Agency of Healthcare Research and Quality (AHRQ) for cross-sectional study quality to rate the methodological quality of the eligible studies. The ARHQ methodology checklist is applicable for observational studies, including 11 items with a yes/no/unclear response option: Yes would be scored 1, no or unclear was scored 0, and an individual study can have a maximum of 11 quality points. Studies were deemed to be low quality (0-3), moderate quality (4-7), and high quality (8-11).^6,7

Primary Outcome Measures

Laryngeal cancer recurrences were categorized as local, regional, loco-regional, or metastases at distant sites. Local recurrence was defined as any case of recurrent cancer in the central neck, around the stoma, or inside the neopharynx, esophagus, or base of tongue. Local recurrence was divided into stomal recurrences and pharyngeal/base of tongue recurrences according to recurrence site. Stomal recurrence was defined as a diffuse infiltrate of neoplastic tissue at the junction of the trachea and skin after a total laryngectomy, involving the stoma or the peristomal soft tissues in the central neck.

Meta-analysis Methods

We used odds ratios (ORs) and 95% CIs to summarize the primary risk factors associated with stoma recurrence or local recurrence following laryngectomy. We also used weighted mean difference (WMD) and 95% CI to compare the difference in the average time interval between the performance of emergency tracheostomy and laryngectomy in the patients who did and did not develop stomal recurrences (days). The degree of heterogeneity was assessed by χ² test (P < .05) combined with the I² method (I² > 50% was regarded as high level of heterogeneity).⁸ If the included studies have very divergent ORs of risk factors for stomal recurrence or local recurrence, we used a random effect model for meta-analysis. Otherwise, we used a fixed effect model.⁹ We performed subgroup analyses by transglottic tumor (glottic-supraglottic, glottic-subglottic), intraoperative tracheostomy (intraoperative tracheostomy at start of operation, intraoperative tracheostomy at a late stage of the operation), and surgery combined with radiotherapy (preoperative radiotherapy, postoperative radiotherapy) to identify the most important risk factors of tumor recurrence. We also used meta-regression analyses to determine whether the year of publication influenced the association between risk factors and tumor recurrences. Publication bias was assessed by visual inspection of the funnel plot. If there is publication bias, for example because smaller studies without statistically significant effects remain unpublished, this will lead to an asymmetrical appearance of the funnel plot with a gap in a bottom corner of the graph.¹⁰ Then, the funnel plot asymmetry was tested using the Egger regression test; if there is asymmetrical funnel plot, the regression line does not run through the origin, and the intercept provides a measure of asymmetry. The larger the deviation of the intercept from zero, the more pronounced the asymmetry.¹¹ The analyses were performed using the software Stata version 10 (StataCorp LP, College Station, Texas, USA). All P values were 2-sided. A P value of less than .05 was considered to be statistically significant.

Results

Characteristics of Eligible Studies

The study selection process is shown in Figure 1.^12
-55 The search terms yielded 2375 potentially relevant papers, 2331 of which were excluded after the abstracts, and full papers were reviewed. Lastly, a total of 44 articles, including 11 928 patients, were included in the current meta-analysis. The majority of the studies (88.64%) were retrospective observations and varied in size from 43 to 1616 patients. Of these, 19 were carried out in European countries, 14 were carried out in Asian countries, 9 were carried out in American countries, and 1 was carried out in an African country. A total of 29 studies provided data for the stomal recurrences, whereas the remaining 15 analyzed the risks of local recurrences following laryngectomy.

Figure 1.

Prisma 2009 flow diagram literature search and study selection. PRISMA diagram showing the different steps of systematic review, starting from literature search to study selection and exclusion. At each step, the reasons for exclusion are indicated.

According to ARHQ, 23 studies were rated as high quality, 21 studies were rated as moderate quality, while no study was rated as low quality (Table 1).

Table 1.

Characteristics of the Studies Included in the Meta-analysis.

First Author	Year	Location	Study Design	Patients	Surgery Treatment	Period of Treatment	Recurrent Cancer Types	Recurrence Rate	Risk Factors	Study Quality
Burnam	1967	USA	ROS	153	Laryngectomy		SR	12.00	Bilateral extent of the primary lesion, metastatic nodes, site of the primary lesion, prior tracheostomy	9
Condon	1969	England	ROS	110	Laryngectomy under endotracheal anesthesia		SR	5.50	Tracheostomy timing	6
Stell	1971	Netherlands	ROS	130	Laryngectomy	1965-1971	SR	4.08	Site of tumor, type of operation, prior tracheostomy, endotracheal anesthetic	7
Bonneau	1975	USA	ROS	115	Laryngectomy	1961-1972	SR	11.96	Site of tumor, tracheostomy timing	6
Schneider	1975	USA	ROS	246	Total laryngectomy	1948-1970	SR	12.60	Postoperative radiation, radiotherapeutic technique	6
Weisman	1979	USA	ROS	251	Laryngectomy	1995-1977	SR	5.60	Preoperative tracheostomy, lymph node metastases, thyroidectomy, type of operation, site of the tumor	7
Bignardi	1983	Italy	ROS	681	Total larygnectomies	1965-1978	SR	2.49	Tracheostomy timing; site of the tumor	9
Komiyama	1984	Japan	ROS	112	Total laryngectomy	1970-1979	SR	9.82	Prior tracheotomy, endotracheal anesthesia after tracheotomy, preoperative chemoradiotherapy	9
Amatsu	1985	Japan	ROS	340	Total laryngectomy with or without radical neck dissection	1969-1982	SR	5.88	Prior tracheotomy, site of tumor	10
Rubin	1990	USA	ROS	444	Total or partial laryngectomy	1976-1988	SR	3.38	Site of tumor, primary stage, previous therapy, neck pathologic status, treatment of the neck, postoperative adjunctive therapy, tracheotomy timing	9
Laccourreye	1991	France	ROS	416	Partial laryngectomy	1972-1984	LR	15.70	Clinical T classification, site of tumor, thyrotomy and cordectomy, vertical partical laryngectomy	7
Rockley	1991	UK	ROS	91	Laryngectomy	1951-1987	SR	12.09	Prior tracheotomy	7
Esteban	1993	Spain	ROS	209	Total laryngectomy, either alone or associated to neck dissection		SR	8.13	Prior tracheotomy, tumor site, N,classification, T classification	7
Hosal	1993	Turkey	ROS	488	Total laryngectomy and neck dissection	1968-1990	SR	2.66	Retrosternal dissection; prior tracheotomy	6
Fagan	1996	South Africa	ROS	43	Total laryngectomy	1984-1993	SR	16.28	Preoperative emergency tracheostomy	7
Wang	1996	China	ROS	95	A combination of laryngectomy and neck dissection	1978-1990	LR	11.58	Surgical wound infection	7
Yotakis	1996	Greece	ROS	352	Total laryngectomy	1983-1993	SR	5.97	Primary location, preoperative emergency tracheostomy, stage of primary lesion, lymph node metastases	7
Yuen	1996	Hong Kong	ROS	334	Total laryngectomy under endotracheal intubation	1976-1990	SR	5.28	T classification, nodal stage, site of tumor, airway obstruction	6
Zbaren	1996	Switzerland	ROS	130	Total laryngectomy	1978-1990	SR	10.00	Treatment modality, T classification, site of primary lesion, N classification, timing of tracheotomy, postoperative pharyngoperistomal fistula	7
Adamopoulos	1997	Greece	ROS	92	Horizontal supraglottic laryngectomy with or without neck dissection	1987-1993	LR	7.14	T classification	6
Halfpenny	2001	UK	ROS	265	Total laryngectomy		SR	1.13	Timing of tracheostomy	9
Reddy	2001	USA	ROS	114	Total laryngectomies or hemilaryngectomies	1976-1990	SR	25.00	Sites involved at first recurrence, p53 status, subglottic extension, prior tracheostomy, surgery methods, lymph node involvement, thyroid cartilage involvement	7
Imauchi	2002	Japan	ROS	69	Total laryngectomy	1985-1995	SR	8.70	Primary site, timing of tracheotomy, T classification, N classification, paratracheal lymph node metastasis, preoperative radiation	7
Santoro	2003	Italy	ROS	464	Total laryngectomy	1980-1997	SR	5.39	Site of the laryngeal lesion, timing of tracheostomy	6
Markou	2004	Greece	ROS	804	Total laryngectomy with or without neck dissection	1992-2001	LR	31.30	Postoperative pharyngocutaneous fistula	9
Petrovic	2004	Serbia	ROS	402	Total laryngectomy	1982-1996	SR	9.20	Localization, T classification, regional extent, emergency tracheostomy, postoperative radiotherapy, metastatic nodes in the neck, positive paratracheal nodes	7
Hassanabadi	2007	Iran	CS	83	Total laryngectomy	2002-2004	SR	10.84	Subglottic invasion, T classification, primary glottic tumor, thyroid lobectomy, preoperation radiation	9
Sartini	2007	Brazil	ROS	47	Total laryngectomy	1995-2004	SR	10.64	Tumor site, previous tracheotomy, T classification, N classification, postoperative radiation	7
Zhao	2009	China	ROS	589	Total laryngectomy with neck dissection	1995-2004	SR	8.76	Primary location, T-classification, lymph node metastasis, tracheotomy timing, histological differentiation, surgical margin	9
Pezier	2013	UK	ROS	60	Total laryngectomy	2003-2010	SR	8.06	Preoperative tracheotomy	7
Brandstorp-Boesen	2016	Norway	CS	1616	Total laryngectomy	1983-2010	SR	3.52	Site of recurrence	11
Mantravadi	1981	USA	ROS	507	Total laryngectomy along or with radical neck dissections	1960-1981	SR	5.13	Primary location, T-classification, lymph node metastasis, tracheotomy time, prior emergency tracheotomy, histological differentiation, surgical margin	9
Martinez	1997		ROS	369	Laryngectomy		SR	2.10	Age, sex, primary location, T classification, N classification, histological grading, time from emergency tracheotomy	7
Nikolaou	2005	USA	ROS	308	Total laryngectomy	1992-1999	LR	31.17	Treatment modality, lymph node metastasis, primary location	9
Saarilahti	2003	Finland	ROS	90	Laryngectomy following with postoperative radiation therapy	1981-1998	LR	35.16	Radiotherapy course, Ki-67 expression, Cyclin A expression	10
Cho	2011	Korea	ROS	92	Supracricoid partial laryngectomy	1994-2006	LR	8.54	Preoperative T understaging	9
Liu	2013	China	ROS	183	Partial laryngectomy	2005-2009	LR	20.22	Age, sex, primary location, T classification, N classification, smoking, drinking, histological grading, treatment modality, postoperative radiotherapy, surgical margin	10
Zhang	2013	China	ROS	84	SCPL-CHEP, VPL	2002-2007	LR	10.71	Treatment modality	9
Markou	2011	Greece	ROS	255	Total laryngectomy	1992-2007	LR	28.63	Tumor size, T stage	9
Wei	2000	China	CCS	255	Total laryngectomy	1982-1992	LR		Overexpression of p53 and p16 in surgical margin	9
Yang	2014	China	CCS	85	Laryngectomy	2004-2010	LR		Overexpression of p53, p21, and Cdc2 in surgical margin	9
Zeng	2016	China	CS	98	Laryngectomy	2005-2010	LR		Overexpression of p53, p21, PCNA, and COX2 in surgical margin	9
Zeng	2001	China	ROS	97	Partial laryngectomy	1982-1993	LR	11.49	Surgical margin	10
Zhou	2004	China	ROS	160	Partial laryngectomy	1991-1999	LR	22.50	Surgical margin size, T classification, site of the tumor, postoperative radiotherapy	10

Abbreviations: CCS, case-control study; CS, cohort study; LR, local recurrence; ROS, retrospective observational study; SCPL-CHEP, supracricoid partial laryngectomy cricohyoidoepiglottopexy; SR, stomal recurrences; VPL, vertical partial laryngectomy.

The Pooled Estimate of Recurrence Rate Following Laryngectomy

The pooled estimate of the stomal recurrence rate following laryngectomy was 6.60% (95% CI, 5.40-7.90); however, high levels of heterogeneity were observed between studies (Q = 169.42, P < .01; I² = 83.5%). Figure 2 showed that there may be significant asymmetrical appearance of the funnel plot with a gap in a bottom corner of the graph, indicating that smaller studies with lower stomal recurrence rate may have remained unpublished (t = 7.68, P < .01). Approximately 19.40% (95% CI, 14.00%-24.80%) of patients reported local recurrence following laryngectomy; however, high levels of heterogeneity were observed between studies (Q = 123.28, P < .01; I² = 91.1%). We found no significant publication bias (t = 0.19, P = .86).

Figure 2.

Funnel plots of stomal recurrence rate and local recurrence rate after laryngectomy among patients with laryngeal cancer. (a) Outcome was stomal recurrence rate. (b) Outcome local recurrence rate.

The stomal recurrence rate was highest in African countries (16.30%; 95% CI, 5.20%-27.30%) followed by American (9.70%; 95% CI, 6.30%-13.10%), Asian (7.10%; 95% CI, 5.30%-8.90%), and European countries (5.30%; 95% CI, 3.90%-6.80%). The stomal recurrence rate varied widely from 27.03% (95% CI; 13.37%-47.05%) in glottic-subglottic to 4.28% (95% CI; 3.42%-5.34%) in supraglottic. The stomal recurrence rate was 4 times higher in patients with preoperative tracheostomy at least 24 hours prior to surgery (15.90%; 95% CI, 11.40%-20.40%) than in patients with intraoperative tracheostomy at the time of laryngectomy (3.80%; 95% CI, 2.10%-5.60%). The stomal recurrence rate was positively associated with T staging, with the highest rate (11.80%; 95% CI, 8.30%-15.20%) occurring in patients with T4 laryngeal cancer. The patients with metastatic nodes in the neck, positive surgical margin, postoperative pharyngoperistomal fistula, and p53 overexpression all had a stomal recurrence rate of >20%, but these results should be interpreted with caution due to the small number of studies (Figure 3a).

Figure 3.

Estimated stomal recurrence rate and local recurrence rate after laryngectomy among patients with laryngeal cancer. (a) Outcome was stomal recurrence rate. (b) Outcome was local recurrence rate.

The distribution tendency of local recurrence rates following laryngectomy was similar to the stomal recurrence rate. The local recurrence rate was highest in American countries, followed by European and Asian countries. The patients with laryngeal cancer involving the subglottic also showed a higher local recurrence rate than those who did not (Figure 3b).

We also performed meta-regressions analyses to explore the temporal trend of stomal recurrence rates and local recurrence rate following laryngectomy, and the results indicated that neither the stomal recurrence rate (t = −0.11, P = .91) nor the local recurrence rate (t = 1.18, P = .26) was affected over time (Figure 4).

Figure 4.

Meta-regression of recurrence rate with publication years as a moderator. (a) Outcome was stomal recurrence rate (t = −0.11, P = .91). (b) Outcome local recurrence rate (t = 1.18, P = .26). Circles present the recurrence rate reported in each study. Circle size is directly proportional to the relative weight of each study.

Meta-analyses of the Risk Factors Associated With Stomal Recurrence After Laryngectomy

The primary outcomes of the meta-analyses are shown in Table 2. Compared to the patients with supraglottic cancer, those who with transglottic cancer (OR = 3.85; 95% CI, 2.54-5.81), subglottic cancer (OR = 4.80; 95% CI, 2.73-8.46), and glottis-subglottic cancer (OR = 5.36; 95% CI, 2.65-10.87) were more likely to suffer from stomal recurrence after laryngectomy. Compared to the patients with T1-2 laryngeal cancer, those with T4 laryngeal cancer (OR = 2.13; 95% CI, 1.06-4.27) were more likely to suffer from stomal recurrence after laryngectomy.

Table 2.

Summary Estimates for the Risk Factors of Stomal Recurrence and Local Recurrence After Total Laryngectomy.

				Test of Heterogeneity		Egger’s Test
Comparisons	No. Studies	No. Patients	Odds Ratio (95% CI)	P	I²	t	P	Y-Axis Intercept: a (95% CI)
Stomal recurrences
Site of the tumor
Supraglottic	—	—	Reference	—	—	—	—	—
Glottic	15	3765	1.61 (1.21 to 2.13)	.08	36.7%	−0.93	.37	−0.66 (–2.20 to 0.87)
Subglottic	13	1754	4.80 (2.73 to 8.46)	<.01	57.9%	2.04	.07	2.27 (–0.18 to 4.72)
Transglottic	13	1492	3.85 (2.54 to 5.81)	.41	3.9	−0.95	.37	−0.71 (–2.39 to 0.97)
Glottic-supraglottic	2	112	0.68 (0.16 to 2.94)	.40	0	—	—	—
Glottic-subglottic	5	488	5.36 (2.65 to 10.87)	.49	0	1.50	.23	1.35 (–1.53 to 4.24)
Postcricoid area	1	291	3.26 (1.18 to 8.99)	—	—	—	—	—
Pyriform sinus	3	458	1.22 (0.42 to 3.53)	.98	0	0.72	.61	0.27 (–4.52 to 5.06)
Prior tracheotomy
Without prior tracheotomy	—	—	Reference	—	—	—	—	—
Prior tracheotomy	16	3401	3.09 (2.12 to 4.49)	.09	33.9%	−0.15	.88	−0.14 (–2.14 to 1.86)
Tracheotomy timing
Preoperative tracheotomy at least 24 h prior to surgery	—	—	Reference	—	—	—	—	—
Intraoperative tracheotomy	13	3456	0.34 (0.23 to 0.49)	.01	53.3%	−0.37	.72	−0.58 (–4.05 to 2.89)
Intraoperative tracheotomy at start of operation	3	1272	0.16 (0.05 to 0.46)	.03	72.4%	−0.39	.76	−6.27 (–210.80 to 198.27)
Intraoperative tracheotomy at a late stage of the operation	6	1148	0.37 (0.27 to 0.52)	.11	44.6%	−0.13	.91	−0.28 (–6.39 to 5.83)
T-classification
T1-2	—	—	Reference	—	—	—	—	—
T3	10	2137	1.46 (0.94 to 2.27)	.41	3.3%	0.39	.71	0.44 (–2.17 to 3.04)
T4	10	1401	2.13 (1.06 to 4.27)	.04	48.8%	−0.06	.95	−0.10 (–3.64 to 3.44)
Surgery treatment modality
Only total laryngectomy	—	—	Reference	—	—	—	—	—
Total laryngectomy with radical neck	3	1133	0.65 (0.35 to 1.22)	.21	36%	0.91	.53	6.06 (–79.00 to 91.12)
Total laryngectomy with thyroidectomy	2	274	2.92 (0.06 to 150.54)	<.01	85.8%	—	—	—
Total laryngectomy with repair by stomach, skin, or colon	1	488	0.38 (0.12 to 1.22)	—	—	—	—	—
Radiotherapy
Without radiotherapy	—	—	Reference	—	—	—	—	—
Surgery combined with radiotherapy	12	2602	1.17 (0.91 to 1.51)	<.01	82.1%	1.52	.15	2.83 (–1.22 to 6.89)
Postoperative radiotherapy	3	704	0.75 (0.47 to 1.20)	.15	46.9%	0.78	.58	1.79 (–27.58 to 31.17)
Preoperative radiotherapy	7	682	0.85 (0.18 to 4.04)	<.01	88.4%	0.13	.90	1.36 (–27.50 to 30.23)
High-dose radiotherapy followed by laryngectomy
No	—	—	Reference	—	—	—	—	—
Yes	1	130	3.72 (1.21 to 11.44)	—	—	—	—	—
Lymph node metastases
Without neck lymph node metastasis	—	—	Reference	—	—	—	—	—
Lymph node metastases before operation	15	3642	1.83 (1.25 to 2.67)	.02	48.2%	−0.72	.48	−0.70 (–2.79 to 1.39)
Metastatic nodes in the neck	2	449	4.54 (2.52 to 8.18)	.73	0%	—	—	—
Paratracheal lymph node metastasis	2	469	2.10 (0.99 to 4.43)	.29	11%
N-classification
N0	—	—	Reference	—	—	—	—	—
N1	4	871	0.82 (0.42 to 1.59)	.16	414%	2.16	.15	2.16 (–1.87 to 6.19)
N2	4	843	0.86 (0.39 to 1.88)	.95	0	0.61	.25	0.61 (–1.04 to 2.28)
N3	3	7325	4.19 (0.26 to 68.22)	.56	0	2.74	.50	2.74 (–32.14 to 37.63)
Postoperative pharyngoperistomal fistula
No	—	—	Reference	—	—	—	—	—
Yes	1	130	3.17 (1.10 to 9.10)	—	—	—	—	—
p53 overexpression
Negative	—	—	Reference	—	—	—	—	—
Positive	1	20	1.33 (0.19 to 9.31)
Surgical margin
Negative	—	—	Reference	—	—	—	—	—
Positive	1	548	3.43 (2.01 to 5.83)	—	—	—	—	—
Local recurrence
Site of the tumor
Supraglottic	—	—	Reference	—	—	—	—	—
Glottic	2	328	0.77 (0.49 to 1.20)	.02	81.8%	—	—	—
Subglottic	1	55	1.59 (0.26 to 9.77)	—	—	—	—	—
Transglottic	2	469	2.54 (1.66 to 3.89)	.77	0%	—	—	—
Overexpression of p53, p16, p21 in surgical margin
No	—	—	Reference	—	—	—	—	—
Overexpression of p53	3	283	2.61 (1.50 to 4.56)	<.01	81.2%	2.54	.24	8.60 (–34.47 to 51.66)
Overexpression of p21	2	183	7.27 (2.96 to 17.89)	.56	0%	0	>.99	—
Overexpression of p16	1	100	1.00 (0.4 to 2.55)	—	—	—	—	—
Overexpression of p53 and p16	1	100	6.17 (1.29 to 29.49)	—	—	—	—	—
Overexpression of p53 and p21	1	70	6.86 (2.78 to 16.96)	—	—	—	—	—
T-classification
T1	—	—	Reference	—	—	—	—	—
T2	3	611	2.11 (1.40 to 3.17)	.28	20.5%	1.10	.47	1.29 (–13.59 to 16.17)
T3	3	127	4.29 (1.58 to 11.68)	.85	0%	—	—	—
T4	3	104	5.78 (1.93 to 17.27)	.86	0%	—	—	—
Estimate on T-staging
Accurate preoperative T-staging	—	—	Reference	—	—	—	—	—
T-stages were underestimated preoperatively	1	82	11.17 (2.39 to 52.16)	—	—	—	—	—
Cervical lymph node metastasis
No	—	—	Reference	—	—	—	—	—
Yes	1	238	1.95 (1.28 to 2.98)
Comparisons of surgery treatment
Vertical thyrotomy and cordectomy vs vertical partical larynectory	1	416	0.81 (0.51 to 1.29)	—	—	—	—	—
Treated for recurrence following previous treatment vs treated with primary total laryngectomy	1	308	1.62 (1.16 to 2.27)	—	—	—	—	—
Vertical partial laryngectomy vs partial laryngectomy cricohyoidoepiglottopexy	1	84	6.93 (0.91 to 53.01)	—	—	—	—	—
Comparisons of radiotherapy
Postoperative radiotherapy vs without radiotherapy	2	251	0.47 (0.27 to 0.82)	.96	0%	—	—	—
Continuous radiotherapy course vs split-course radiotherapy	1	90	0.58 (0.25 to 1.33)	—	—	—	—	—
Surgical wound infection
No	—	—	Reference	—	—	—	—	—
Yes	1	95	6.56 (2.13 to 20.22)	—	—	—	—	—
Surgical margin
Negative	—	—	Reference	—	—	—	—	—
Positive	2	270	4.14 (2.33 to 7.35)	.37	0%	0	>.99	1.91
Surgical margin size
<3 mm	—	—	Reference	—	—	—	—	—
4-5 mm	1	85	0.36 (0.09 to 1.37)	—	—	—	—	—
>5 mm	1	144	0.38 (0.20 to 0.74)	—	—	—	—	—
Postoperative pharyngocutaneous fistula
No	—	—	Reference	—	—	—	—	—
Yes	1	377	1.13 (0.74 to 1.71)	—	—	—	—	—

The pooled relative risk of stomal recurrence was significantly higher among patients with prior tracheotomy than among patients without prior tracheotomy (OR = 3.09; 95% CI, 2.12-4.49). Considering the tracheotomy timing, the pooled relative risk of stomal recurrence was significantly lower among patients with intraoperative tracheostomy than patients with preoperative tracheostomy at least 24 hours prior to surgery (OR = 0.34; 95% CI, 0.23-0.49). We also performed the analyses to determine the effect of treatment modality on the risk of stomal recurrence after laryngectomy. We found no significant difference between total laryngectomy only and total laryngectomy combined with radiotherapy, radical neck, thyroidectomy, and repair by stomach, skin, or colon. Interestingly, the pooled relative risk of stomal recurrence was significantly higher among patients with high-dose radiotherapy followed by laryngectomy than among patients without (OR = 3.72; 95% CI, 1.21-11.44).

The pooled relative risk of stomal recurrence was significantly higher among patients with lymph node metastases before operation, including regional cervical lymph nodes and distant metastases to the lung or other locations, than among those without (OR = 1.83; 95% CI, 1.25-2.67). Furthermore, when only including the patients with metastasis to regional cervical lymph nodes, the pooled relative risk of stomal recurrence was significantly higher among patients with metastatic nodes in the neck than among those without (OR = 4.54; 95% CI, 2.52-8.18).

Other risks, for example, postoperative pharyngoperistomal fistula (n = 1), wild type p53 overexpression (n = 1), and positive surgical margin (n = 1), were also analyzed in this study. However, these results should be interpreted with caution due to the small number of studies.

Meta-analyses of the Risk Factors Associated With Local Recurrence After Laryngectomy

The primary outcomes of the meta-analyses are shown in Table 2. We found that the overexpression of p53 (OR = 2.61; 95% CI, 1.50-4.56) and the overexpression of p21 (OR = 7.27; 95% CI, 2.96-17.89) were significantly associated with an increased risk of local recurrence after laryngectomy but had an unexpectedly low impact on risk on stomal recurrence.

We also found that the patients whose T-classification was underestimated preoperatively (OR = 11.17; 95% CI, 2.39-52.16) with cervical lymph node metastasis (OR = 1.95; 95% CI, 1.28-2.98), surgical wound infection (OR = 6.56; 95% CI, 2.13-20.22), and a positive surgical margin (OR = 4.14; 95% CI, 2.33-7.35) were more likely to suffer from local recurrence. However, these results should be interpreted with caution due to the small number of included studies.

Evaluation of Publication Bias

We assessed funnel plot asymmetry using Egger’s linear regression test, and the results are shown in Table 2. We found no significant publication bias for all primary outcomes.

Discussion

To the best of our knowledge, this is the first study to systematically analyze the risk factors of laryngeal cancer recurrence following laryngectomy in English. We subdivided cancer recurrence into stomal recurrence and local recurrence and found distinct risk factors for each. For stomal recurrence, we confirmed a series of earlier identified factors including tumor site, prior tracheotomy, tracheotomy timing, T-staging, lymph node metastases, postoperative pharyngoperistomal fistula, and positive surgical margins. However, postoperative radiotherapy, the overexpression of p53, and the overexpression of p21 were significantly associated with the risk of local recurrence but had an unexpectedly low impact on risk on stomal recurrence. Stomal recurrence is a common pattern of local recurrence and might have different causative factors with other patterns of local recurrence. This meta-analysis presented complete and reliable results regarding the risks of stomal recurrence following total laryngectomy. Over the past several years, there has been a substantial development in the strategies to treat early and advanced laryngeal cancer. The objective of laryngeal cancer management is cure with preserved larynx function. To achieve this, it is vital to assess the risk of recurrence in each patient with laryngeal cancer. Early detection of recurrent laryngeal cancer is a very important contributor to a successful disease outcome. Thus, confirmation of earlier identified factors for recurrence in our study would be highly relevant to the clinician.

We found that the stomal recurrence rate after total laryngectomy was stable over time, while the slight increase of local recurrence rate was observed from the raw figure in spite of no significant difference. Over the past 20 years, there has been a change in the strategies to treat laryngeal cancer. The operation of laryngectomy is more commonly reserved for either very advanced primary tumors or cases that fail primary nonsurgical treatment. Thus, the changes in treatment paradigms for laryngeal cancer likely contributed to stable or increased recurrence rates. We also found a variable geographical distribution of laryngeal cancer recurrence; for example, the stomal recurrence rate in African countries was about 3 times higher than that in European countries. The geographical distribution of laryngeal cancer recurrence may be explained by socioeconomic status, medical and health services, therapeutic level, and so on.

In this study, we found the presence of a cancer in the subglottis (transglottic or primary subglottic tumors) to be the most important risk factor in stomal recurrence. Moreover, the fact that glottic-subglottic cancer was more likely to cause stomal recurrence than subglottis cancer was presented in our study. Several possibilities could explain the high risk of a subglottic tumor in the development of stomal recurrence. First, subglottic cancers occur in a less common primary tumor location, and cases often present with advanced disease because the symptoms are less obvious.^56,57 Second, subglottic cancers are more frequently associated with cervical node metastasis as well as critical airway obstruction, necessitating tracheostomy.⁵⁸ Lastly, subglottic cancers have a tendency toward circumferential growth and cartilage invasion. Spread tends to occur outside the larynx through the cricothyroid membrane to the trachea or hypopharynx.

Prior tracheostomy is the most frequently cited risk factor for stomal recurrence after total laryngectomy, and the impact of prior tracheostomy on local control after total laryngectomy is controversial. A total of 16 articles, including 3401 patients, were included in the current meta-analysis, and the increased incidence of stomal recurrence associated with prior tracheotomy was identified. Additionally, we also found that intraoperative tracheotomy was significantly associated with decreased risk of stomal recurrence compared to preoperative tracheotomy at least 24 hours prior to surgery. Our results indicated that tracheotomy might not be associated with an increased risk of stomal recurrence when performed during the surgical operation. In clinical practice, preoperative tracheotomy at least 24 hours prior to surgery is generally required for more advanced cancer and tumors with subglottic extension, while intraoperative tracheotomy is usually performed for the administration of anasthesia before laryngectomy, which might provide a partial interpretation of the different risk of stomal recurrence between preoperative tracheotomy and intraoperative tracheotomy.

The most common and most frequently reported complication of salvage total laryngectomy is pharyngocutaneous fistula, which occurs in 5.6% to 73% of all patients who undergo salvage total laryngectomy, according to a recent systematic review.⁵⁷ Our study uncovered an increased incidence of stomal recurrence in patients with a postoperative pharyngoperistomal fistula. This could be the result of viable dislodged tumor cells having had the opportunity to implant in the perifistular tissue near the tracheostoma. However, this result should be interpreted with caution due to the small number of included studies. Furthermore, our review also found some tumor genetic factors, for example, the overexpression of p53 and p21, were linked to an increased risk of local recurrence after laryngectomy. Discouragingly, the associations between these genetic factors with stomal recurrence were not identified due to the lack of relevant papers.

Some of the factors used in our multivariate analysis, such as postoperative infection, lack unified definition or identifying standard across studies. “Not similar enough” means the included studies are difficult to combine in a meta-analysis. Therefore, we have performed rigorous quality control by specifying eligibility criteria, developing data collection form, and assessing the risk of bias in individual primary studies based on the Non-Randomised Studies Methods Group (NRSMG) recommendations.¹⁰ Especially, we have studied very carefully the extent and origin of heterogeneity between included studies when deciding whether to combine the effect values. If included studies were not sufficiently homogeneous to combine in a meta-analysis, we gave up pooled estimate. Considering the observed heterogeneity in the included studies, we used a random-effects model to stay on the safe side.⁵⁹

There are multiple limitations to this review. First, stomal recurrence is surely too simple a concept, while local recurrence may take the form of stomal recurrence. Therefore, we cannot ensure whether stomal recurrence and local recurrence have been counted as one in some of the publications because many authors have not provided clear definition for local recurrence in their papers, which may produce obvious information bias in our primary analyses. Second, the results should be treated with caution due to the relatively small number of included studies, highlighting the need for future studies. Third, the majority of studies in this review were retrospective and observational in nature, and patient data were collected over a long time period, producing inevitable inclusion bias. In the end, risk factors associated with laryngeal cancer, including gender, smoking, drinking, overweight/obesity, and physical inactivity, were not assessed due to a lack of data.

Conclusion

In conclusion, we have confirmed a series of earlier identified factors for stomal recurrence following total laryngectomy. These earlier identified factors included subglottic laryngeal cancer, preoperative tracheotomy, higher classification, positive surgical margin, lymph node metastases before operation, and postoperative pharyngoperistomal fistula. Furthermore, we also confirmed some special additional risks that associated with local recurrence, for example, T-classification that was underestimated preoperatively, overexpression of p53 and p21, surgical wound infection, and small surgical margin. Confirmation of these earlier identified factors for recurrence would provide important insights for clinical practice.

Footnotes

Acknowledgements

The authors would like to thank all researchers and participants for their valuable contributions to this article. This work was supported by the Key Science and Technology Project of Anhui Province (1501041147).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Torre

Bray

Siegel

Ferlay

Lortet-Tieulent

Jemal

Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87-108.

Siegel

Miller

Jemal

Cancer statistics, 2016. CA Cancer J Clin. 2016;66:7-30.

Forastiere

Zhang

Weber

et al . Long-term results of RTOG 91-11: a comparison of three nonsurgical treatment strategies to preserve the larynx in patients with locally advanced larynx cancer. J Clin Oncol. 2013;31:845-852.

Esteller

Aguero

Martel

Lopez

Quer

Leon

Stomal recurrence in head and neck cancer patients with temporary tracheostomy. Auris Nasus Larynx. 2014;41:467-470.

Sheng

Zhang

Song

Chen

Luo

[Meta-analysis on the risk factors for stomal recurrence after total laryngectomy]. Lin Chung Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2013;27:995-999.

Rostom

Dubé

Cranney

. Celiac disease. http://wwwncbinlmnihgov/books/NBK35156/. Accessed April 27, 2017.

Zeng

Zhang

Kwong

et al . The methodological quality assessment tools for preclinical and clinical studies, systematic review and meta-analysis, and clinical practice guideline: a systematic review. J Evid Based Med. 2015;8:2-10.

Higgins

Thompson

Deeks

Altman

DG.

Measuring inconsistency in meta-analyses. BMJ. 2003;327:557-560.

Greenland

Quantitative methods in the review of epidemiologic literature. Epidemiol Rev. 1987;9:1-30.

10.

Higgins

JPT

Green

Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.1. http://training.cochrane.org/handbook. Accessed July 8, 2017.

11.

Egger

Davey Smith

Schneider

Minder

Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629-634.

12.

Adamopoulos

Supraglottic laryngectomy—series report and analysis of results. J Laryngo Otol. 1997;111:730-734.

13.

Amatsu

Makino

Kinishi

Stomal recurrence—etiologic factors and prevention. Auris Nasus Larynx. 1985;12:103-110.

14.

Bignardi

Gavioli

Staffieri

Tracheostomal recurrences after laryngectomy. Arch Otorhinolaryngol. 1983;238:107-113.

15.

Bonneau

Lehman

RH.

Stomal recurrence following laryngectomy. Arch Otolaryngol. 1975;101:408-412.

16.

Brandstorp-Boesen

Sorum Falk

Folkvard Evensen

Boysen

Brondbo

Risk of recurrence in laryngeal cancer. PLoS One. 2016;11:e0164068.

17.

Burnam

Hudson

WR.

Stomal recurrence of malignancy: an evaluation and its significance in the post-laryngectomy patient. South Med J. 1967;60:823-826.

18.

Cho

Sun

Joo

Kim

MS.

Analysis of clinicopathological stage in supracricoid partial laryngectomy patients: need for adjuvant therapy in clinically understaged cases. Auris Nasus Larynx. 2011;38:255-260.

19.

Condon

HA.

Postlaryngectomy stomal recurrence. The influence of endotracheal anaesthesia. Br J Anaesth. 1969;41:531-533.

20.

Esteban

Moreno

Delgado-Rodriguez

Mochon

Risk factors involved in stomal recurrence following laryngectomy. J Laryngol Otol. 1993;107:527-531.

21.

Fagan

Loock

JW.

Tracheostomy and peristomal recurrence. Clin Otolaryngol Allied Sci. 1996;21:328-330.

22.

Halfpenny

McGurk

Stomal recurrence following temporary tracheostomy. J Laryngol Otol. 2001;115:202-204.

23.

Hassanabadi

Shadabi

Sadr-Hosseini

Zarandi

Eghtesadi-Araghi

Assessment of the recurrence rate of laryngeal cancer in tracheostoma in patients undergoing laryngectomy. J Med Sci. 2007;7:630-634.

24.

Hosal

Onerci

Turan

Peristomal recurrence. Am J Otolaryngol Head Neck Med Surg. 1993;14:206-208.

25.

Imauchi

Ito

Takasago

Nibu

Sugasawa

Ichimura

Stomal recurrence after total laryngectomy for squamous cell carcinoma of the larynx. Otolaryngol Head Neck Surg. 2002;126:63-66.

26.

Komiyama

Watanabe

Yanagita

Kuwano

Hiroto

Inhibition of stomal recurrence in laryngectomy with preoperative far therapy. A statistical evaluation. Auris Nasus Larynx. 1984;11:43-49.

27.

Laccourreye

Weinstein

Brasnu

Trotoux

Laccourreye

Vertical partial laryngectomy: a critical analysis of local recurrence. Ann Otol Rhinol Laryngol. 1991;100:68-71.

28.

Liu

ZG.

[Clinical analysis of relevant factors causing postoperative recurrence of laryngeal cancer after partial laryngectomy]. Zhonghua Zhong Liu Za Zhi. 2013;35:377-381.

29.

Mantravadi

Katz

Skolnik

Becker

Freehling

Friedman

Stomal recurrence. A critical analysis of risk factors. Arch Otolaryngol. 1981;107:735-738.

30.

Markou

Goudakos

Triaridis

Konstantinidis

Vital

Nikolaou

The role of tumor size and patient’s age as prognostic factors in laryngeal cancer. Hippokratia. 2011;15:75-80.

31.

Markou

Vlachtsis

Nikolaou

Petridis

Kouloulas

Daniilidis

IC.

Incidence and predisposing factors of pharyngocutaneous fistula formation after total laryngectomy. Is there a relationship with tumor recurrence?

Eur Arch Otorhinolaryngol. 2004;261:61-67.

32.

Martinez Vidal

Lopez Amado

Herranz Gonzalez-Botas

Vazquez Barro

Garcia Sarandeses

. [Recurrence of peritracheostoma and emergency tracheotomy]. Acta Otorrinolaringol Esp. 1997;48:291-294.

33.

Nikolaou

Markou

Petridis

Vlachtsis

Nalbantian

Daniilidis

Factors influencing tumour relapse after total laryngectomy. B-ENT. 2005;1:1-10.

34.

Petrovic

Djordjevic

Stomal recurrence after primary total laryngectomy. Clin Otolaryngol Allied Sci. 2004;29:270-273.

35.

Pezier

Nixon

Joshi

et al . Pre-operative tracheostomy does not impact on stomal recurrence and overall survival in patients undergoing primary laryngectomy. Eur Arch Otorhinolaryngol. 2013;270:1729-1735.

36.

Reddy

Narayana

Melian

Kathuria

Leman

Emami

Stomal recurrence in patients with T1 glottic cancer after salvage laryngectomy for radiotherapy failures: role of p53 overexpression and subglottic extension. Am J Clin Oncol. 2001;24:124-127.

37.

Rockley

Powell

Robin

Reid

AP.

Post-laryngectomy stomal recurrence: tumour implantation or paratracheal lymphatic metastasis?

Clin Otolaryngol Allied Sci. 1991;16:43-47.

38.

Rubin

Johnson

Myers

EN.

Stomal recurrence after laryngectomy: interrelated risk factor study. Otolaryngol Head Neck Surg. 1990;103:805-812.

39.

Saarilahti

Kajanti

Kouri

Aaltonen

Franssila

Joensuu

Cyclin A and Ki-67 expression as predictors for locoregional recurrence and outcome in laryngeal cancer patients treated with surgery and postoperative radiotherapy. Int Journal of Radiat Oncol Biol Phys. 2003;57:986-995.

40.

Santoro

Franchi

Tempesti

Sardi

Polli

Stomal recurrence following total laryngectomy: clinical and molecular analysis of a series. Ann Otol Rhinol Laryngol. 2003;112:594-599.

41.

Sartini

Fava

de Faria

PH.

Surgical stoma recurrence after total laringectomy. Braz J Otorhinolaryngol. 2007;73:80-86.

42.

Schneider

Lindberg

Jesse

RH.

Prevention of tracheal stoma recurrences after total laryngectomy by postoperative irradiation. J Surg Oncol. 1975;7:187-190.

43.

Stell

van den Broek

Stomal recurrence after laryngectomy: aetiology and management. J Laryngol Otol. 1971;85:131-140.

44.

Wang

Guo

[Wound infection and recurrence after total laryngectomy]. Zhonghua Er Bi Yan Hou Ke Za Zhi. 1996;31:50-52.

45.

Wei

Tang

[The status of protein P16 and P53 in the surgical margins of laryngeal carcinoma and the association with local recurrence]. Zhonghua Er Bi Yan Hou Ke Za Zhi. 2000;35:461-463.

46.

Weisman

Colman

Ward

PH.

Stomal recurrence following laryngectomy: a critical evaluation. Ann Otol Rhinol Laryngol. 1979;88:855-860.

47.

Yang

Liu

Liang

Sun

Over-expression of p53, p21 and Cdc2 in histologically negative surgical margins is correlated with local recurrence of laryngeal squamous cell carcinoma. Int J Clin Exp Pathol. 2014;7:4295-4302.

48.

Yotakis

Davris

Kontozoglou

Adamopoulos

Evaluation of risk factors for stomal recurrence after total laryngectomy. Clin Otolaryngol Allied Sci. 1996;21:135-138.

49.

Yuen

Wei

Hui

Risk factors of tracheostomal recurrence after laryngectomy for laryngeal carcinoma. Am J Surg. 1996;172:263-266.

50.

Zbaren

Greiner

Kengelbacher

Stoma recurrence after laryngectomy: an analysis of risk factors. Otolaryngol Head Neck Surg. 1996;114:569-575.

51.

Zeng

[Expression of p53, p21, PCNA and COX-2 and its relationship with recurrence in the early-stage laryngeal cancer with negative surgical margin]. Lin Chung Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2016;30:349-352, 356.

52.

Zeng

Xia

Chen

Zhang

[Value of surgical margin in directing the treatment of early laryngeal carcinoma after partial laryngectomy]. Zhonghua Er Bi Yan Hou Ke Za Zhi. 2001;36:250-253.

53.

Zhang

Chen

et al . Supracricoid partial laryngectomy cricohyoidoepiglottopexy (SCPL-CHEP) versus vertical partial laryngectomy for the treatment of glottic carcinoma. Eur Arch Otorhinolaryngol. 2013;270:1027-1034.

54.

Zhao

Ren

Zhuo

Zou

Liu

Stomal recurrence after total laryngectomy: a clinicopathological multivariate analysis. Am J Clin Oncol. 2009;32:154-157.

55.

Zhou

Ding

Wang

[The relationship between the surgical margin and local recurrence of laryngeal carcinoma]. Lin Chuang Er Bi Yan Hou Ke Za Zhi. 2004;18:339-340.

56.

Salvador-Coloma

Cohen

Multidisciplinary care of laryngeal cancer. J Oncol Pract. 2016;12:717-724.

57.

Hasan

Dwivedi

Gunaratne

Virk

Palme

Riffat

Systematic review and meta-analysis of the complications of salvage total laryngectomy. Eur J Surg Oncol. 2017;43(1):42-51.

58.

de Bree

Leemans

Silver

et al . Paratracheal lymph node dissection in cancer of the larynx, hypopharynx, and cervical esophagus: the need for guidelines. Head Neck. 2011;33:912-916.

59.

Borenstein

Hedges

Higgins

Rothstein

HR.

A basic introduction to fixed-effect and random-effects models for meta-analysis. Res Synth Methods. 2010;1:97-111.