Salvage Laryngectomy in Clinically N0 Patients: Is Elective Neck Dissection Indicated?

ABSTRACT

Objectives

This systematic review and meta‐analysis investigated the role of elective neck dissection (END) in the salvage management of clinically N0 necks.

Data Sources

PubMed/Medline, Google Scholar, and the Cochrane Library were systematically searched for relevant studies.

Methods

Both electronic and manual search strategies were conducted within the abovementioned databases and included articles and reviews to find the relevant studies. Rates of occult nodal metastasis, survival outcomes, and postoperative complications were analyzed in N0 patients undergoing salvage laryngectomy with END. Fixed and random effects models were used to calculate pooled estimates of overall survival metastasis rates with 95% confidence intervals, and heterogeneity was assessed using tau², I ², and Cochran's Q test. Statistical analyses were performed in R software with p < 0.05 as significant.

Results

The included eight studies reported occult metastasis rates ranging from 4% to 10% in clinically N0 necks post‐radiotherapy. Overall survival was approximately 52%, with mixed evidence on the survival advantage of END. Disease‐free survival and recurrence‐free survival rates ranged from 71.7% to 95.5% at 5 years. Complication rates associated with END were high, with up to 47.4% of patients experiencing postoperative morbidity. The metastasis rate was 10%–13% highlighting the ability of END to detect occult metastases.

Conclusion

The survival benefits of END are not consistently supported by current evidence, while the significant morbidity associated with the procedure raises concerns about its routine use. A conservative approach may be more appropriate for clinically node‐negative patients, particularly when weighed against the elevated risks of complications. END demonstrates moderate overall survival rates and an ability to detect occult metastases, but its role in improving long‐term outcomes remains unclear. Larger prospective studies and randomized trials are needed to better define the indications and outcomes of END in the salvage management of clinically N0 necks.

A systematic review and meta‐analysis of eight studies on elective neck dissection (END) in clinically N0 necks post‐radiotherapy found a low incidence of occult nodal metastasis and varied survival outcomes, with some showing no significant benefit over control groups. Complication rates for END were substantial, reaching up to 47.4%, raising safety concerns. The findings suggest that a conservative management approach may be more suitable for clinically node‐negative patients, with further prospective research needed to clarify END's role.

1. Introduction

Squamous cell carcinomas are the most prevalent type of head and neck malignancy and the seventh most common cancer worldwide. It develops from the mucosal epithelium in the larynx, pharynx, and oral cavity, the major regions of head and neck cancer [1]. The incidence of head and neck squamous cell malignancy continuous to increase and is expected to rise by 30% in 2030 [2].

Several potential risk factors are linked to the development of head and neck squamous cell carcinoma, including cigarette smoking, alcohol consumption, betel quid chewing, poor nutrition, and inadequate oral hygiene. Additionally, infections from viruses like human papillomavirus (HPV), especially in oropharyngeal cancers, and Epstein–Barr virus (EBV), particularly in nasopharyngeal cancers common in Asia, as well as infections from Candida albicans , are also associated with an increased risk of head and neck squamous cell carcinoma [3, 4]. Furthermore, the condition of the lymph nodes is a key factor in determining the prognosis of patients with head and neck squamous cell carcinoma, as even a single positive lymph node can reduce survival by as much as 50% [5].

Patients who had a neck recurrence after first‐line treatment had lower rates of regional control and died more frequently from uncontrolled neck disease [6, 7, 8, 9, 10]. Hence, management for the neck becomes an important aspect of the treatment approach of patients with head and neck malignancy. The treatment of patients in the N0 stage is debatable. After failed chemo and radiotherapy, salvage surgery treating head and neck cancer is a challenging and increasingly important situation with high stakes for the patients [11]. Nevertheless, there is no consensus on the relative risks and advantages of elective neck dissection (END) against observation as a treatment approach in situations when the cervical lymph nodes are not clinically or radiologically involved. If the tumor has been surgically removed, options between elective neck treatment and observation would be conflicting for the adequacy and safety of the patient. Many studies have tried to address this dilemma over the past two decades, with varied findings. Some authors advocate neck dissection in N0 patients, while others argue that neck dissection should only be pursued in select cases [12, 13]. Others, however, oppose END because of the high morbidity of neck dissection‐related problems and poor survival outcomes, particularly in previously irradiated patients [14].

There is a scarcity of high‐quality information on this subject, and the majority of the prior studies have been limited to retrospective studies with a modest patient population. Given the possible therapeutic and/or prognostic consequences of performing END in the setting of salvage laryngectomy, this topic warrants exploratory research. The aim of this study is to evaluate the role of END in the salvage management of clinically N0 necks following radiotherapy, focusing on its impact on survival outcomes, the detection of occult nodal metastases, and postoperative complications. By conducting a systematic review and meta‐analysis, the study aims to provide clarity on the clinical relevance and safety of END, while addressing the variability in reported outcomes and highlighting areas requiring further research.

2. Methodology

2.1. Search Strategy

Relevant literature was searched in multiple databases, including PubMed/Medline, Google Scholar, and the Cochrane Library. The following specific keywords were used alone or in combination for the search: neck dissection, nodal dissection, and laryngectomy. Boolean operators (AND, OR, NOT) were also used to increase the sensitivity of the search. Our search strategy was limited to the title and abstract of the search results to utilize all the relevant studies only. All of these results were exported to an Endnote library to identify and execute all duplicates between the different searched databases. Furthermore, we manually searched all similar article sections in PubMed and included studies and relevant reviews for possible detection of any missed studies by the main electronic search strategy. References within each study were explored to identify other studies. In this systematic review, the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) criteria were used [15].

2.2. Study Inclusion Criteria

Utilizing the PICOS (population, intervention, comparator, outcome, study design) framework, the population of interest is adults aged > 18 years who underwent salvage laryngectomy and neck dissection in the squamous cell carcinoma of N0 neck patients with or without a comparator group (observation). Unilateral and bilateral END were eligible for inclusion. The study includes patients who underwent primary radiotherapy or chemoradiotherapy for laryngeal cancer and subsequently underwent salvage laryngectomy. Study objectives included the incidence of occult nodal metastasis, complications, and survival results of patients undergoing neck dissection, and the study design was all study types and case series with more than 10 patients.

2.3. Screening and Extraction

Articles with irrelevant titles were excluded from consideration. In the subsequent phase, both the abstracts and full texts of the papers were meticulously reviewed to ensure compliance with the inclusion criteria. To streamline the process, titles and abstracts were organized, assessed, and checked for duplicates using reference management software (EndNote X8). To ensure the highest quality of selection, a dual screening approach was employed: one stage involved evaluating titles and abstracts, and the other consisted of a comprehensive examination of the full texts. Once all relevant articles were identified, a structured extraction sheet was created to capture key information aligned with our specific objectives. Two independent researchers conducted the data extraction process separately.

2.4. Quality Assessment

We utilized the methodological index for the non‐randomized studies (MINORS) for the quality assessment of the studies [16]. The MINORS contained 12 items listed in Table 1. The items were scored 0 if not reported; 1 when reported but inadequate; and 2 when reported and adequate. The global ideal score was 16 for non‐comparative studies and 24 for comparative.

TABLE 1.

Summary of the results of quality assessment of the included studies using methodological index for the non‐randomized studies (MINORS) tool.

Author	A clearly stated aim	Inclusion of consecutive patients	Prospective collection of data	Endpoints appropriate to the aim of the study	Unbiased assessment of the endpoint	Follow‐up period appropriate to the aim of the study	Loss to follow‐up less than 5%	Prospective calculation of the study size	Total score
Dagan et al. 2010 [17]	2	2	0	2	2	2	2	0	12
Basheeth et al. 2013 [18]	2	2	0	2	2	2	2	0	12
Pezier et al. 2014 [19]	1	2	0	2	2	2	2	0	11
Freiser et al. 2014 [20]	2	2	0	2	2	2	1	0	11
Birkeland et al. 2016 [21]	2	2	0	2	2	2	0	0	10
Mazerolle et al. 2018 [22]	2	2	0	2	2	2	2	0	12
Bernard et al. 2018 [23]	2	2	0	2	2	2	2	0	12
Gouzos et al. 2019 [24]	2	2	0	2	2	2	2	0	12

2.5. Data Synthesis

Data extraction was performed by two independent investigators, covering study characteristics, participant details, intervention specifics, outcome measures, and funding sources. The meta‐analysis included studies that reported overall survival (OS) and metastasis rates following END in clinically N0 patients who underwent salvage laryngectomy. Both fixed effects and random effects models were applied to calculate pooled estimates, allowing for a balanced analysis that accounted for study sizes and variability. Heterogeneity among studies was assessed using the I ² statistic and p values, revealing moderate variability in the results. Finally, pooled survival and metastasis rates were estimated with 95% confidence intervals, ensuring the reliability and statistical significance of the findings. All statistical analyses were performed using R software, and a p value of less than 0.05 was considered statistically significant.

3. Results

3.1. Search Results

A total of 52 abstracts were screened from the 120 publications obtained using the search strategy with English as the only language (Figure 1). If abstracts were found relevant, the full‐text article was reviewed. Of the 22 articles assessed, eight studies [17, 18, 19, 20, 21, 22, 23, 24] met all of the criteria for inclusion and were included in the review, with a total of 840 patients presented in the analysis.

FIGURE 1.

PRISMA flow diagram.

3.2. Results of Quality Assessment

The MINORS tool highlights several key findings. Most studies demonstrated strong methodological aspects, particularly in having a clearly stated aim, inclusion of consecutive patients, endpoints appropriate to the study's aim, unbiased assessment of endpoints, and follow‐up periods suitable for the study objectives, with these criteria consistently receiving high scores (2 points). However, none of the studies collected data prospectively, as they all scored 0 on this criterion. Additionally, only a few studies managed to achieve less than 5% loss to follow‐up, while none performed a prospective calculation of the study size, reflecting gaps in sample size planning. The total scores across the studies ranged from 10 to 12, with five studies (Dagan, Basheeth, Mazerolle, Bernard, and Gouzos) achieving the highest score of 12, indicating relatively strong methodological rigor despite some common limitations. These results suggest that while the studies generally followed robust methodologies, improvements in prospective data collection and sample size calculation could further enhance their quality (Table 1).

3.3. Characteristics of the Included Studies

The baseline characteristics of the eight included studies reveal several key trends. All studies were retrospective, with data collection periods ranging from 1965 to 2017, and they were conducted in countries such as the United States, Ireland, the United Kingdom, Germany, the Netherlands, and Australia. The total sample size across these studies was 840 patients, with most participants being middle‐aged to older adults, as the median or mean ages were predominantly in the 60s, highlighting the focus on older populations. There was a notable gender imbalance, with male participants making up 86.3% of the cases (725 males to 115 females). This was particularly evident in Gouzos's study, which included only male participants. All the baseline characteristics of these studies are shown in Table 2.

TABLE 2.

Baseline characteristics of the included studies.

Author	Country	Study type	Study period	Total sample size	Age	Male/female
Dagan et al. 2010 [17]	USA	Retrospective	1965–2006	57	62.2 a	49/8
Basheeth et al. 2013 [18]	Ireland	Retrospective	1996–2011	45	64.3 b	36/9
Pezier et al. 2014 [19]	UK	Retrospective	2003–2010	32	61 a	29/3
Freiser et al. 2014 [20]	USA	Retrospective	2000–2012	125	70 a	118/7
Birkeland et al. 2016 [21]	USA	Retrospective	1997–2014	233	61.6 b	198/35
Mazerolle et al. 2018 [22]	Germany	Retrospective	2005–2015	239	61 a	200/39
Bernard et al. 2018 [23]	Netherland	Retrospective	2002–2014	86	65 b	72/14
Gouzos et al. 2019 [24]	Australia	Retrospective	2009–2017	23	Case; 64.1 a , control; 68.5 a	23/0

^a Median.

^b Mean.

3.4. Node Metastasis

The incidence of nodal metastasis in patients undergoing END after primary radiotherapy or chemoradiotherapy varies among studies. Dagan et al. [17] reported a 10% rate of pathological nodal metastasis in their END group. Similarly, Basheeth et al. [18] found an 8% incidence, indicating that occult metastatic disease in N0 necks is relatively rare in this population. In contrast, Birkeland et al. [21] highlighted that histologically positive occult nodes were more prevalent in T4 tumors and supraglottic subsites, underscoring the importance of careful tumor evaluation when considering the risk of nodal involvement. Freiser et al. [20] reported a 10.2% rate of pathological evidence of nodal metastasis, while Mazerolle et al. [22] noted a lower incidence of 9%. Overall, these studies suggest that while node metastasis exists, it may not warrant routine END in all cases.

3.5. Survival Rate

OS outcomes among patients who underwent END compared to controls reveal mixed results. Dagan et al. [17] reported a 5‐year OS rate of 45% for the END group versus 56% for controls. Freiser et al. [20] observed comparable 5‐year OS rates of 53.2% for the END group versus 53.1% for the control group. Bernard et al. [23] demonstrated an improved 5‐year OS of 62% in the END group compared to 41% in controls, indicating a potential benefit from END in certain populations. In contrast, Mazerolle et al. [22] found a hazard ratio of 0.91 for OS in the END group, indicating a 9% lower risk compared to controls, though this finding was not statistically significant. Overall, these findings indicate that while END may provide some survival benefits, it is not universally effective.

3.6. Other Outcomes

Other important outcomes, such as disease‐free survival (DFS) and recurrence‐free survival (RFS), also show variability across studies. Pezier et al. [19] reported a locoregional RFS of 71.7% and a distant RFS of 95.5% at 5 years, indicating good control over disease recurrence. Dagan et al. (2010) noted an RFS of 75% at 5 years for the entire population. Freiser et al. [20] reported a DFS of 56.4% for the END group versus 59.3% for controls at 2 years. Bernard et al. [23] highlighted a 5‐year RFS of 74% in the END group versus 59% in the control group, further suggesting that END might improve disease control. Overall, while some studies suggest that END may play a role in preventing recurrence, its impact on survival remains debatable.

3.7. Complications

Complication rates associated with END vary significantly, with several studies reporting increased morbidity. Dagan et al. [17] observed a 25% rate of fistula formation in the END group, compared to 18% in controls, along with other severe complications (17% vs. 12% for grade III/IV complications). Basheeth et al. [18] reported overall major complications of 47.4% in the END group versus 29% in the control group. Freiser et al. [20] reported total complications of 42% for the END group compared to 48% for controls, with significant rates of salivary fistulae (26% vs. 33%) and wound infections (12% vs. 7.4%). Mazerolle et al. [22] also highlighted total complications of 42% in the END group compared to 33% in controls (Table 3).

TABLE 3.

Summary of the outcomes of the included studies in this review.

Author	Stage distribution	Primary therapy	No. of patients of END group, n (%)	No. of patients of control group, n (%)	Pathological evidence of nodal metastasis in END group	Complications	OS	Other outcomes	Conclusion
						END vs. control	END vs. control	END vs. control
Dagan et al. 2010 [17]	T1–T4	Radiotherapy	40 (70.18)	17 (28.82)	10%	Fistula: 25% vs. 18% Other grade III/IV: 17% vs. 12% Death: 5% vs. 0%	45% vs. 56% at 5 years	RFS: 75% at 5 years for the entire population	END is unnecessary in salvage surgery for locally recurrent head and neck SCCA
Basheeth et al. 2013 [18]	T1–T4	Radiotherapy or chemoradiotherapy	38 (84.44)	7 (15.56)	8%	Overall major complications: 47.4% vs. 29% Fistula: 39.5% vs. 29% Wound infections: 7.9% vs. 0%	Poorer cause‐specific survival in bilateral ND	NR	Occult metastatic disease in N0 necks after salvage surgery is rare, and END has minimal impact on cancer control
Pezier et al. 2014 [19]	Nil	Radiotherapy or chemoradiotherapy	32	NR	Over all 4%	All patients with wound infections ultimately developed pharyngocutaneous fistulas at an average of 12 days postoperatively	37% at 5 years	Locoregional RFS: 71.7% Distant RFS: 95.5% DSS: 47.4% At 5 years	Opt for conservative management in clinically node‐negative patients post‐radiotherapy due to high accuracy of neck staging and elevated risks of complications with END
Freiser et al. 2014 [20]	T1–T4	Radiotherapy: 83.2% Chemoradiotherapy: 16%	98 (78.40)	27 (21.60)	10.2%	Total complications: 42% vs. 48% Salivary fistula/leak: 26% vs. 33% Wound infection: 12% vs. 7.4% Wound dehiscence: 18% vs. 7.4% Hematoma: 3.0% vs. 7.4% Revision procedure: 15% vs. 19% Flap failure: 3.1% vs. 0% Medical complications: 10.2% vs. 7.4% Carotid blowout: 3.1% vs. 3.7%	53.2% vs. 53.1% at 5 years	DFS: 56.4% vs. 59.3% at 2 years	END in N0 necks during salvage total laryngectomy can offer prognostic information but does not significantly impact complications or survival
Birkeland et al. 2016 [21]	T1–T4	Radiotherapy or chemoradiotherapy	233 (100)	NR	17%	NR	NR	NR	The presence of histologically positive occult nodes in cases, particularly in T4 tumors and supraglottic subsites, highlights the need for careful evaluation of tumor characteristics in managing the risk of occult metastases
Mazerolle et al. 2018 [22]	T1–T4	Radiotherapy	143 (59.83)	96 (40.17)	9%	Total: 42% vs. 33% Salivary fistulae: 34% vs. 31% Death: 4% vs. 3%	The HR of 0.91 indicates a 9% lower risk of OS at 34 months for the END group compared to the control, but this finding is not statistically significant due to overlapping confidence intervals	The HR of 0.91 indicates a 9% lower risk of PFS at 25 months for the END group compared to the control, but this finding is not statistically significant due to overlapping confidence intervals	Routine elective lymph node dissection in irradiated necks should be avoided in patients undergoing surgery for persistent, recurrent, or new primary pharyngo‐laryngeal carcinomas
Bernard et al. 2018 [23]	T1–T4	Radiotherapy or chemoradiotherapy	27 (31.40)	59 (69)	4%	37% vs. 36% Multiple complications: seven patients had more than one complication	62% vs. 41% at 5 years	RFS: 74% vs. 59% at 5 years DSS: 81% vs. 59% at 5 years	Due to the low complication rates of END and the poor prognosis for regional recurrence after prior treatments, END should be considered during salvage laryngectomy to enhance patient outcomes
Gouzos et al. 2019 [24]	T1–T4	Radiotherapy	17 (73.91)	6 (26.09)	6%	Fistula: 18% vs. 17% Surgical repair: 11.8% vs. 0%	71% vs. 67% at 3 years	Over 3 years, DFS matched OS, except for one patient who experiencing a local recurrence at 6 months and dying 14 months post‐surgery	END did not impact survival outcomes or complication rates during salvage laryngectomy, though its true value is uncertain due to the small risk of occult neck metastases

Abbreviations: DFS, disease‐free survival; END, elective neck dissection; ND, neck dissection; NR, not reported; OBSVN, observation; OS, overall survival; PCF, pharyngocutaneous fistula; RFS, recurrence‐free survival.

3.8. Meta‐Analysis for OS Rate

The meta‐analysis evaluated the pooled OS rate of patients undergoing END after salvage laryngectomy in clinically N0 patients. Across five studies, the survival rates ranged from 37.5% to 70.6%, with larger studies contributing more weight to the analysis due to higher precision. The pooled survival rate was approximately 52%, calculated using both the common effect model (51.75%; 95% CI: 44.94%–58.48%) and the random effects model (52.01%; 95% CI: 42.88%–61.00%). Heterogeneity analysis showed moderate variability (I ² = 42.4%), though this was not statistically significant (p = 0.1386), indicating reasonable consistency across studies. While individual study survival rates varied, the overlap in confidence intervals supports the reliability of the pooled estimate. These findings suggest that END is associated with a moderate OS rate in this patient population, highlighting its potential benefit. However, variability in patient characteristics and study methodologies may account for the observed differences (Figure 2).

FIGURE 2.

Meta‐analysis of pooled data on overall survival rate.

3.9. Meta‐Analysis for Metastasis Rate

The meta‐analysis pooled data from seven studies, involving 596 patients, to estimate the metastasis rate after END. Study‐specific rates ranged from 3.7% to 17.2%, with a pooled metastasis rate of 12.8% (95% CI: 10.27%–15.84%) under the common effect model and 10.87% (95% CI: 7.56%–15.40%) under the random effects model. Moderate heterogeneity (I ² = 38.6%) was observed but was not statistically significant (p = 0.1349). These results suggest that 10%–13% of clinically N0 patients undergoing END show pathological nodal metastasis, supporting its role in identifying occult metastases (Figure 3).

FIGURE 3.

Meta‐analysis of pooled data on metastasis rate.

4. Discussion

4.1. Occult Nodal Metastasis

The incidence of occult nodal metastasis in patients undergoing END after primary radiotherapy or chemoradiotherapy varies across studies, though it remains relatively low. Dagan et al. [17] reported a 10% rate of pathological nodal metastasis, and Basheeth et al. [18] found an 8% incidence, indicating that occult disease in N0 necks is generally rare. However, Birkeland et al. [21] observed higher rates of histologically positive occult nodes in T4 tumors and supraglottic subsites, suggesting the importance of careful tumor evaluation in these cases. Freiser et al. [20] reported a 10.2% rate, while Mazerolle et al. [22] noted a slightly lower 9% incidence. Together, these findings suggest that while node metastasis occurs, it may not justify routine END in all cases. Birkeland et al. [21] found that supraglottic tumors showed the highest nodal positivity rates, with 17% in ipsilateral level II, 16% in level III, and 15% in the paratracheal nodal regions. A study by Birkeland et al. found that supraglottic recurrence had a significantly higher histologically positive nodes (28%) than non‐supraglottic recurrence (10%) [21]. In a prospective case–control study of patients with supraglottic laryngeal cancer with historical controls of patients undergoing primary supraglottic tumor surgery, alongside a prospective cohort undergoing both tumor surgery and bilateral selective neck dissection [12] found an 18.1% incidence of occult nodal metastasis in the case group compared to a control group of patients who underwent historical supraglottic surgery alone. They observed ipsilateral metastasis more commonly than contralateral and bilateral metastasis (p = 0.01). They recommended bilateral neck dissection in conjunction with laryngectomy in supraglottic cancer patients to reduce postoperative regional recurrence.

If the risk of nodal disease is more than 20%, neck dissection is recommended going by past findings, then the treatment of the neck is warranted [25]. Therefore, END is a serious consideration in such circumstances.

Moreover, Bernard et al. [23] and Pezier et al. [19] had the lowest incidence of occult nodal metastasis (4%) for both. Despite the low rate of reported low occult nodal metastasis, the first investigator suggested considering END at the moment of salvage laryngectomy due to the lower complications and poor prognosis of regional cancer recurrence following radiotherapy and laryngectomy.

In four of the manuscripts included in our analysis, occult nodal metastases occurred at a rate of 6%–10% [17, 18, 22, 24], which appears to be considerably low. All four of the investigators justified excluding ND as a part of laryngectomy due to the lower rate of occult node metastasis in laryngeal carcinomas.

4.2. Survival Rate

When comparing OS outcomes between patients undergoing END and controls, studies reveal mixed results. Dagan et al. [17] reported a 5‐year OS of 45% in the END group compared to 56% in controls, suggesting no clear benefit from END. Similarly, Freiser et al. [20] found comparable OS rates, with 53.2% for the END group and 53.1% for controls. However, Bernard et al. [23] showed a more promising result, reporting a 5‐year OS of 62% for the END group versus 41% for controls, suggesting that certain populations may derive a survival advantage from END. On the other hand, Mazerolle et al. [22] observed a 9% lower risk of mortality in the END group (hazard ratio of 0.91), though this finding lacked statistical significance. Collectively, these results imply that while END may offer survival benefits in certain cases, it is not universally effective.

The paucity of survival benefits found in other manuscripts should be viewed with caution when considering whether to perform an END [17, 20, 22, 24]. Freiser et al. [20] found that neck dissection in patients was not significantly associated with OS (p = 0.72) and DFS (p = 0.76). Furthermore, the recurrence T stage and site, as well as the initial treatment method, were not significant predictors of survival. They found that pathologically nodal positive cases were significantly related to patients' survival, particularly for worse DFS (p = 0.0057) which reported a 2‐year DFS of 56.4% for the END group, slightly lower than the 59.3% in controls. In Dagan et al. [17], they reported higher toxicity in the dissected groups of patients. In comparison to the observation group, the neck dissection group has a poor survival outcome. The aforementioned findings are consistently reflected in other studies included in the review analysis. The Gouzos et al. [24] findings also correlate with the other study findings. In patients with an N0 neck undergoing salvage laryngectomy for recurrent or persistent laryngeal SCC, they found no relation between neck dissection and disease‐free or OS. Mazerolle et al. [22] concluded that the only variables associated with better survival outcomes were negative margin excision condition, age at surgery (under 60), and time between RT and surgery (over 2 years). They also noted that there was no significant variation between the no neck dissection and neck dissection groups in terms of OS and progression‐free survival in univariable and multivariable analysis. Sharma et al. [26] examined RFS in patients with and without occult neck‐positive disease on histopathology. At 5 years, the RFS rate was not significantly different between abovementioned groups of patients (70.6% vs. 68.2%, respectively; log‐rank test p = 0.5). Pezier et al. [19] reported robust locoregional and distant RFS rates at 5 years (71.7% and 95.5%, respectively), indicating effective control over disease recurrence with END. Dagan et al. [17] also found an RFS of 75% at 5 years for their cohort. While Bernard et al. [23] showed an improved 5‐year RFS of 74% in the END group compared to 59% in controls, supporting END's potential to aid in disease control.

END may provide survival benefits for select patient groups, particularly in terms of RFS and disease control. However, its impact on OS remains inconsistent across studies, and in some cases, it has been associated with higher treatment toxicity without significantly improving survival outcomes. Key factors influencing patient outcomes include tumor characteristics, nodal status, and individual patient variables such as age and the timing of surgery following radiotherapy. Overall, END's utility should be carefully considered on a patient‐specific basis, weighing potential benefits against the risks and individual clinical factors.

4.3. Postoperative Complications

A review of complications associated with END compared to control groups reveals significant variability in adverse events, with notable occurrences of fistula formation, wound infections, and other major complications. Dagan et al. [17] reported that patients undergoing END had a higher incidence of fistulas at 25% versus 18% in controls, and overall major complications were observed in 47.4% of END patients compared to 29% of controls. Basheeth et al. [18] found a fistula rate of 39.5% in END patients versus 29% in controls, with all wound infections ultimately progressing to pharyngocutaneous fistulas at an average of 12 days postoperatively. Pezier et al. noted total complications in END patients at 42% compared to 48% in controls, with specific increases in salivary fistulas (26% vs. 33%) and wound dehiscence (18% vs. 7.4%).

Freiser et al. [20] reported an elevated rate of salivary fistulas at 34% for END patients compared to 31% in controls and a slight increase in mortality (4% vs. 3%). Birkeland et al. found an overall complication rate of 37% for END patients versus 36% in controls, with multiple complications noted in some cases. Mazerolle et al. [22] noted fistula rates of 18% in END patients versus 17% in controls, with surgical repair required in 11.8% of END cases compared to 0% in controls. Bernard et al. [23] and Gouzos et al. [24] reported total complication rates of 42% and 33%, respectively, with END patients showing consistently higher rates of both wound infection and wound dehiscence than controls. Although certain studies reported lower mortality rates associated with END, the increased complication burden indicates a need for caution, as END patients often face a more complex postoperative course and an elevated risk of multiple concurrent complications.

A meta‐analysis that comprised eight of our included studies found no significant difference in the risk of complications between the two groups, even though the END group had a higher percentage of problems than the observation group (50% vs. 33%) [27].

4.4. Future Research Recommendations

Given the inconsistent evidence supporting the survival benefits of END and the associated risks of morbidity, a conservative management approach may be more suitable for clinically node‐negative patients. Future research should focus on conducting larger, preferably prospective studies and randomized controlled trials to further explore the outcomes of different management strategies for this population. Such studies should aim to delineate more clearly the circumstances under which END may provide benefit and to identify patient subgroups that could derive a survival advantage from the procedure. This additional research will help inform clinical decision‐making and enhance the safety and efficacy of treatment protocols for patients undergoing salvage laryngectomy in the context of clinically N0 necks.

END should be considered for clinically N0 patients undergoing salvage laryngectomy due to its potential to improve OS and identify occult metastases. This surgical approach can play a significant role in guiding treatment decisions and optimizing patient outcomes. Additionally, further investigations should explore patient‐specific and tumor‐related factors that influence survival and metastasis rates, providing deeper insights for refining patient selection and treatment strategies.

5. Conclusion

While the studies indicate that nodal metastasis in clinically node‐negative patients post‐therapy is rare, the role of END in improving survival and disease control remains unclear. The evidence suggests that routine END may not be necessary for all patients, particularly given the significant complication risks associated with the procedure. As demonstrated by the studies, conservative management may be a viable approach in select patients, especially when considering the low incidence of occult metastasis and the potential for improved outcomes through careful patient selection. According to our meta‐analysis, END is associated with a moderate OS rate and a significant ability to detect occult metastases highlighting the potential benefits of END after salvage laryngectomy in clinically N0 patients. Future studies should focus on identifying patient subsets that might benefit most from END to optimize treatment strategies in head and neck oncology.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

All data are provided within the manuscript.