Pharyngocutaneous Fistula after Laryngectomy: An Umbrella Systematic Review to Uncover Lacunae in Meta-Analyses

Abstract

Objective—The objective of this study was to systematically assess meta-analyses to determine the lacunae in the literature for PCF following laryngectomy. Methods—Bibliometric analysis were carried out on meta-analyses on PCF after total laryngectomy for laryngeal cancer in the PubMed database. Results—Twenty-four meta-analyses were considered eligible and chosen for analysis. Six meta-analyses (25%) focused on the risk factors for PCF in TL. Four meta-analyses (16.6%) focused on the role of the onlay flap. Four meta-analyses (16.6%) focused on the timing of feed initiation. Three meta-analyses (12.5%) focused on using a stapler for pharyngeal closure. Two meta-analyses focused on types of pharyngeal reconstruction. Other meta-analyses analyzed the use of salivary bypass tubes, the method of pharyngeal closure, organ preservation protocols on PCF, primary and secondary TEP, and the effect of non-surgical treatment on PCF. Conclusion—Despite plenty of published meta-analyses, there is a lack of scrutiny on certain critical aspects of PCF.

1. Introduction

Pharyngocutaneous fistula (PCF) is a common complication that occurs after total laryngectomy (TL), and it can lead to increased morbidity. PCF can cause a more extended hospital stay, the need for nasogastric feeding, and delays in adjuvant therapy. It may also require additional surgery to reconstruct the pharyngeal defects [1]. As technology continues to evolve, new solutions are emerging that may provide alternative treatment methods. Surgeons aim for a complication-free postoperative period to ensure timely adjuvant therapy and improved outcomes [2]. Recently, there has been an increased focus on publications regarding pharyngocutaneous fistula, particularly in meta-analyses. Therefore, we intend to thoroughly analyze the bibliometrics of PCF meta-analyses to identify the study objectives and outcomes of published meta-analyses on PCF following TL. We also intend to identify gaps in the published data to better guide our planning of future studies on PCF.

2. Methods

2.1. Search Strategy

We have included only the PubMed/MEDLINE database to obtain the published meta-analyses on PCF. Published literature in English from inception to 2023 was considered.

2.2. Search Syntax

The search terms “pharyngocutaneous” [All Fields] AND (“fistula” [MeSH Terms] OR “fistula” [All Fields] OR “fistulas” [All Fields] OR “fistulas” [All Fields] OR “fistulae” [All Fields] OR “fistulaes” [All Fields]) AND (“laryngectomy” [MeSH Terms] OR “laryngectomy” [All Fields] OR “laryngectomies” [All Fields]) AND (“meta analysis” [Publication Type] OR “meta analysis as topic” [MeSH Terms] OR “meta analysis” [All Fields]) were used to obtain the results. The data were last retrieved on 26 April 2023.

2.3. Data Screening and Selection

The retrieved articles were initially screened independently by KNR based on the type of article, title, and abstract. The eligible meta-analyses were pooled, and a thorough full-text analysis was conducted (Figure 1). The articles were selected based on concurrence with the predefined inclusion and exclusion criteria.

2.4. Inclusion Criteria

• Total laryngectomy/laryngopharyngectomy for laryngeal and hypopharyngeal cancer;
• Laryngeal or hypopharyngeal cancers with total laryngectomy/laryngopharyngectomy (with or without neck dissection) as primary or salvage therapy;
• A meta-analysis published in peer-reviewed journals;
• Meta-analysis must report on pharyngocutaneous fistula following total laryngectomy/laryngopharyngectomy.

2.5. Exclusion Criteria

• Non-human studies;
• Laryngectomy/laryngopharyngectomy for non-oncological reason;
• Not reported—regarding operative outcomes;
• Review articles, meeting abstracts, case reports, editorial letters, as well as other forms of publication;
• Incomplete data or insufficient information.

2.6. Data Extraction

All included articles were screened by KNR. The following study characteristics were recorded: first author, country, year of publication, journal, type of laryngectomy, research question, and study outcomes were analyzed (

Table 1. Summary of included meta-analyses.

Author	Country	Journal	Publication Year	Type of Laryngectomy	Research Question	Result
Costantino A [1]	Italy	Microsurgery	2023	Both	Type of reconstruction on PCF rates	FRAFF—low PCF
Kim DH [3]	South Korea	Laryngoscope	2023	Both	To determine the risk factors for PCF	Risk factors—age, low Hb, diabetes, tumour site, previous RT, previous tracheostomy, primary vs. salvage TL, TEP, and low albumin
Costantino A [4]	Italy	Head Neck	2022	Both	Use of salivary bypass tube	Lower rates of PCF with salivary bypass tube
Costantino A [2]	Italy	Oral Oncol	2022	Both	Type of reconstruction following TP + TL	Free jejunal flap has lowest risk for PCF
Chiesa-Estomba CM [5]	Spain	Oncol Ther	2022	Both	Use of stapler for pharyngeal closure	Lower rates of PCF with stapler usage
Chotipanich A [6]	Thailand	Cureus	2022	Both	Type of pharyngeal closure	Horizontal closure—lower rates of PCF
Rao KN [7]	India	Indian J Surg Oncol	2022	PTL	To determine the risk factors for PCF	Comorbidities, hypopharyngeagal site, low Hb, low Alb, stapler use, and positive margins
De Virgilio A [8]	Italy	Eur Arch Otorhinolaryngol	2022	STL	Onlay vascular flap versus patch pharyngoplasty	Onlay vascular flap has lower PCF rates
Singh R [9]	Australia	Am J Otolaryngol	2021	Both	Time of oral feed Initiation	Early feeding led to higher PCF
Yi X [10]	China	J BUON	2021	Both	Time of oral feed Initiation	No significant difference between early and late feeds
Locatello LG [11]	Italy	J Clin Med	2021	Both	Non-surgical treatment of PCF	Promising outcomes but small sample size
Lee YC [12]	Taiwan	Clin Otolaryngol	2021	Both	Use of stapler for pharyngeal closure	Lower rates of PCF with stapler usage
Milinis K [13]	UK	Head Neck	2021	Both	Time of oral feed Initiation	No significant difference between early and late feeds
Wang M [14]	China	Eur Arch Otorhinolaryngol	2020	Both	To determine the risk factors for PCF	Risk factors—age, smoking, COPD, CAD, RT, low Hb, low albumin, site, treatment method
Chakravarty PD [15]	UK	J Laryngol Otol	2018	Both	PCF rates in primary and secondary TEP	Similar rates
Hasan Z [16]	Australia	Eur J Surg Oncol	2017	STL	Role of organ preservation protocols in PCF	Increased rates of PCF with salvage TL
Guimarães AV [17]	Brasil	Head Neck	2016	STL	Role of onlay vascular flap in PCF	Lower incidence of PCF with onlay vascular flap
Aires FT [18]	Brasil	Head Neck	2015	Both	Time of oral feed Initiation	No significant difference between early and late feeds
Dedivitis RA [19]	Brasil	Head Neck	2015	Both	To determine the risk factors for PCF	Risk factors—COPD, low Hb, blood transfusion, CTRT, T stage, subsite, hypopharyngeal site, margins, and neck dissection
Liang JW [20]	China	Auris Nasus Larynx	2015	Both	To determine the risk factors for PCF	Risk factors—tumour subsite, T stage, previous RT, low Hb, and margins
Aires FT [21]	Brasil	Head Neck	2014	Both	Use of stapler for pharyngeal closure	Lower rates of PCF with stapler usage
Sayles M [22]	UK	Laryngoscope	2014	Both	Role of onlay vascular flap in PCF	Lower incidence of PCF with onlay vascular flap
Paleri V [23]	UK	Laryngoscope	2014	STL	Role of onlay vascular flap in PCF	Lower incidence of PCF with onlay vascular flap
Paydarfar JA [24]	USA	Arch Otolaryngol Head Neck Surg	2006	Both	To determine the risk factors for PCF	Risk factors—low Hb, previous RT, neck dissection

PCF—pharyngocutaneous fistula, FRAFF—free radial forearm flap, Hb—hemoglobin, RT—radiotherapy, TL—total laryngectomy, TEP—tracheoesophageal prosthesis, TP—total pharyngectomy, Alb—albumin, PTL—primary total laryngectomy, STL—salvage total laryngectomy, COPD—chronic obstructive pulmonary disease, CAD—coronary artery disease, CTRT—Chemoradiotherapy.

).

2.7. Statistical Analysis

Descriptive statistical analysis was performed on the retrieved articles using Microsoft Excel version 2309.

3. Results

3.1. Literature Retrieval and Data Extraction

The initial literature search using the predefined search syntax identified 25 manuscripts. Upon title and abstract screening, only one article was excluded as it was a letter to the editor. Finally, 24 meta-analyses were considered eligible and chosen for analysis.

3.2. Year of Publication

The first meta-analysis was published in the year 2006 [24]. Surprisingly, over 50% of the meta-analyses (n = 13) were published since 2021 (Figure 2).

3.3. Country of Origin

The majority of the published meta-analyses (n = 5, 20.8%) originated from Italy [1,2,4,8,11], Brazil (n = 4, 16.6%) [17,18,19,21], the United Kingdom (n = 4, 16.6%) [13,15,22,23], China (n = 3, 12.5%) [10,14,20], and Australia (n = 2, 8.3%) [9,16], and one each from India [7], Thailand [6], South Korea [3], Taiwan [12], Spain [5], and the USA [24].

3.4. Type of Laryngectomy

Only one meta-analysis (4%) described exclusively PCF in primary TL [7]. Four meta-analyses (16.6%) described PCF exclusively on salvage TL [8,16,17,23]. The remaining fourteen meta-analyses (58.3%) described PCF in primary and salvage TL [1,2,3,4,5,6,9,10,11,12,13,14,15,19,20,21,22,24].

3.5. Journal

Six meta-analyses (25%) have been published in the Head Neck journal (affiliated with the International Federation of Head and Neck Oncologic Societies) [4,13,17,18,19,21]. Three were published in The Laryngoscope (affiliated with the American Triological Society) [3,22,23]. Two were published in the European archives of otorhinolaryngology (affiliated with the Confederation of European Otorhinolaryngologists) [8,14].

3.6. Research Question

Six meta-analyses (25%) focused on determining the risk factors for PCF in TL [3,7,14,19,20,24]. Four meta-analyses (16.6%) focused on the role of an onlay vascular flap over the pharyngeal closure in PCF [8,17,22,23]. Four meta-analyses (16.6%) focused on the timing of oral feed initiation in PCF [9,10,13,18]. Three meta-analyses (12.5%) focused on using a stapler for pharyngeal closure in PCF [5,12,21]. Two meta-analyses determined the risk of PCF following various types of pharyngeal reconstruction [1,2]. One meta-analysis analyzed the use of salivary bypass tubes in PCF [4]. One meta-analysis analyzed the method of pharyngeal closure in PCF [6]. One meta-analysis each described the role of organ preservation protocols in PCF [16], the rates of PCF following primary and secondary TEP [15], and the effect of non-surgical treatment on PCF [11].

3.7. Outcomes

• Risk factors for PCF was found to be age [3,14], smoking [14], low hemoglobin [3,7,14,19,20,24], low albumin [7,14], comorbidities [3,7,14,19], diabetes [3,7], chronic obstructive pulmonary disease [14,19], coronary artery disease [14], tumor site [3,14,19,20], hypopharyngeal involvement [7,19], previous radiation therapy [3,14,19,20,24], margins [7,19,20], type of pharyngeal closure [7,14], TEP [3], neck dissection [19,24], and blood transfusion [19];
• An onlay vascular flap over the pharyngeal closure led to lower rates of PCF [8,17,22,23].
• Timing of feed initiation—early feed initiation was found to have slightly higher PCF rates [9]; no significant difference was found between early and late feeds [10,13,18];
• The use of stapler for pharyngeal closure equivocally led to lower rates of PCF with stapler usage [5,12,21];
• Type of pharyngeal reconstruction—augmentation pharyngoplasty with a free radial forearm flap led to lower PCF rates [1], and a free jejunal flap for circumferential pharyngeal defects had lower rates of PCF [2];
• Salivary bypass tube usage following TL led to lower rates of PCF [4];
• The horizontal method of pharyngeal closure following TL led to lower rates of PCF [6].
• Organ preservation protocols led to higher rates of PCF [16];
• Primary and secondary TEP—no difference was found in the rates of PCF between primary and secondary TEP insertion [15];
• Non-surgical treatment had promising outcomes in PCF, but the sample size was too small [11] (Table 1).

3.8. Citation Network

The citation network of the included articles was generated using the Litmaps tool [25]. The independent axes on the literature citation maps are the logarithmic scale of citations and distributed over the publication date. The size of the individual article bubble corresponds to the logarithmic scale of the article citation (Figure 3).

4. Discussion

There have been numerous meta-analyses conducted on pharyngocutaneous fistula (PCF) following total laryngectomy (TL), with the majority (over 50%) being published in the last 18 months. Although the topic is intriguing, there is considerable overlap in the research questions of many meta-analyses. Therefore, a bibliometric analysis was performed on PCF meta-analyses to identify the study objectives and outcomes to determine gaps in the literature and provide a guide map for further research. Here, we discuss the lacunae in the meta-analysis and the literature on PCF following TL.

Of the published meta-analyses (n = 24), six (25%) aimed to determine the risk factors for PCF in TL, but only one described the risk factors specifically for primary TL. None of the studies have determined the risk factors for salvage TL exclusively, with the remaining five analyses concentrating on TL. With the increasing use of organ preservation protocols as the initial therapy for laryngeal cancer, the rates of salvage TL are increasing, making it imperative to close this gap.

Many of the risk factors assessed in the meta-analyses had significant overlap. For example, the timing of salvage surgery after completing the organ preservation protocol needs to be considered, with at least an 8 to 12 week gap following radiation therapy to imaging, to determine if the suspected recurrence is due to post-radiotherapy change or truly due to residual disease [26]. Local tissue hypoxia following irradiation due to an increase in hypoxia-inducible factor 1-alpha (HIF-1α) can lead to impaired wound healing and ultimately to PCF [27]. The condition of the overlying skin is also a crucial factor, especially during salvage surgery, as there is an increase in wound dehiscence rates in STL, mainly due to tissue hypoxia, endarteritis, and endothelial dysfunction [28]. The use of electrocautery and cold instruments for ablating the pharyngeal mucosal margins may play a role, as there is the possibility of lateral thermal damage to the surrounding mucosa (up to 2–3 mm) with electrocautery and no injury with cold instruments (blade or scissors) [29]. Various techniques of pharyngeal mucosal closure have been described, simple continuous interlocking sutures interrupted by sutures with extraluminal or intraluminal knots, as described by Connell [30], Lembert [31], and Gambee [32], but these methods have not been compared mainly due to a lack of studies. The depth of cricopharyngeal myotomy is to be noted, as the deep myotomy may lead to thin pharyngeal mucosa at the point of myotomy, which may lead to higher chances of PCF.

Other factors that may contribute to the development of PCF include the extent of laryngectomy (total vs. partial), type of antibiotic used, the duration of surgery, tissue handling methods, the number of layers of closure, the surgeon’s experience, the type of suture used, the duration of the closed suction drain, and the use of compression dressing. However, there is a paucity of literature analyzing these factors, with only a few isolated case series reported. Thus, a high-quality study examining these factors individually is needed. The confounding variables make it difficult to perform a randomized trial for every scenario. To reach a good statistical power of the study, the sample size of the study must be relatively large. A good sample size can be achieved with multi-institutional and multinational collaborations [33]. The collaborations will also help us to identify these factors across various populations.

The published meta-analyses have severe heterogeneity, a lack of randomized trials, and confounding factors. Many meta-analyses are repetitions of similar research questions, which can be avoided by preregistering the meta-analyses and performing a thorough literature search to avoid duplication. Bias in surgical research must also be noted, particularly in articles describing a surgical technique [34]. Congruences in the results of multiple studies will strengthen the hypothesis and lead to a better understanding of the risk factors for PCF in TL.

The summary of lacunae and the methods to circumvent them has been provided in

Table 2. Summary of lacunae.

Sl No	Lacunae	Methods to Resolve Lacunae
1.	Lack of Focus on Primary vs. Salvage Total Laryngectomy	Conduct separate meta-analyses for primary and salvage laryngectomy to determine distinct risk factors.
2.	Lack of Comparative Studies	Initiate high-quality comparative studies for factors like pharyngeal mucosal closure techniques, extent of laryngectomy, and antibiotic types.
3.	Paucity of Literature on Specific Factors	Encourage researchers to publish case series and studies on specific risk factors.
4.	Need for Multi-Institutional and Multinational Collaborations	Establish collaborative networks for data collection and analysis, enhancing sample size and statistical power.
5.	Heterogeneity and Lack of Randomized Trials	Encourage researchers to conduct randomized controlled trials. Promote pre-registration of meta-analyses to avoid duplication.
6.	Bias in Surgical Research	Implement rigorous peer review processes and ensure transparency in reporting surgical techniques.
7.	Congruence in Results	Encourage replication studies and meta-analyses to validate findings and strengthen hypotheses.

.

5. Conclusions

In conclusion, the literature on pharyngocutaneous fistula (PCF) following total laryngectomy (TL) has many meta-analyses with overlapping research questions. Despite the abundance of published analysis, there is a lack of scrutiny on certain critical aspects of PCF in TL. These aspects include the risk factors for PCF in salvage TL, the optimal timing of salvage surgery following radiation, and various surgical techniques for pharyngeal mucosal closure. It is essential to perform high-quality studies with good statistical power to close these gaps in the literature and generate new evidence that can guide clinical practice in preventing PCF in TL.