Laryngopharyngeal reflux: is laparoscopic fundoplication an effective treatment?

Abstract

Introduction

Laryngopharyngeal reflux (LPR) is difficult to diagnose and treat owing to uncertainty relating to the underlying pathology. The initial management of LPR includes lifestyle modifications and oral medications. In patients who have failed to respond to proton pump inhibitor (PPI) therapy, anti-reflux surgery is considered; laparoscopic fundoplication is the surgery of choice. The primary aim of this review is to identify whether fundoplication is effective in improving signs and symptoms of LPR. The secondary aim is to identify whether patients who have had a poor response to PPIs are likely to have symptom improvement with surgery. The objective of the study is to establish the effect of laparoscopic fundoplication on the reflux symptom index score (RSI).

Methods

PubMed, Embase, Medline and Cochrane databases were used to search according to the PRISMA guidelines. Original articles assessing the efficacy of fundoplication in relieving symptoms of LPR were included. For each study, the efficacy endpoints and safety outcomes were recorded.

Findings

Nine studies from 844 initial records met the inclusion criteria: one prospective case control study, one retrospective case–control study, four prospective case series and three retrospective case series involving 287 fundoplications. All nine studies found fundoplication to be effective in improving symptoms of LPR (p < 0.05).

Conclusion

Current evidence suggests laparoscopic fundoplication is an effective treatment for LPR and should be considered if medical management is unsuccessful.

Introduction

Laryngopharyngeal reflux (LPR) is the backflow of gastric contents, acid, pepsin and conjugated bile acids into the upper aero-digestive region. It causes a range of symptoms including cough, hoarseness, globus pharyngeus, sore throat, dysphagia, postnasal drip and chronic repetitive throat clearing.^1,2 A number of mucosal changes have been linked to LPR including supraglottic oedema and erythema, posterior pharyngeal wall cobblestoning, vocal cord ulcers, interarytenoid changes, medial arytenoids wall oedema and vocal cord granulomas.³ Disease progression can result in ulcerative disease, granulomas, subglottic stenosis and laryngeal cancer.^4,5

LPR may occur with or without classical symptoms of gastro-oesophageal disorder (GORD).⁶ Three mechanisms have been proposed to explain this. First, direct action of acidic gastric contents on the more sensitive pharynx and larynx causes irritation. Koufman et al found that individuals started experiencing symptoms of GORD only when the oesophagus was exposed to more than 50 episodes of acid reflux (defined as pH below 4), but experienced symptoms of LPR with only three episodes of pharyngeal reflux.^6,7 Second, reflux of gastric contents into the distal oesophagus stimulates vagal neural reflexes resulting in a chronic cough, which in turn results in laryngeal irritation and symptoms of LPR.^8,9 Finally, the reflux of non-acidic gastric contents (eg pepsin, pancreatic enzymes and bile salts) also causes laryngeal injury.^9,10

Diagnosis of LPR is difficult and no one test has adequate sensitivity and specificity. Clinicians diagnose LPR by combining history with laryngoscopy findings and patient-reported outcome measures such as the reflux symptom index (RSI) and reflux finding score (RFS).

The RSI is a validated questionnaire that assesses the incidence of various LPR symptoms experienced by patients.¹¹ It allows the patient to rate the severity of nine symptoms from 0 (no problem) to 5 (severe problem), with a maximum score of 45. A score greater than 13 indicates LPR. The RFS is a validated scale used to assess laryngeal signs indicative of reflux upon laryngoscopy.^11,12

Multiple investigations are used in assessing LPR, including oesophageal manometry; pH studies and impedance monitoring to look for episodes of non-acidic reflux; laryngoscopy; and more recently, oropharyngeal pepsin testing.^13,14 These tests are reserved for patients who have had a poor response from two to three months of lifestyle modifications and medical therapy. Sakin et al compared the proximal oesophageal impedance values in normal individuals with those in patients complaining of LPR symptoms and found a significantly lower baseline impedance in symptomatic individuals.¹⁴

LPR conservative management advice includes avoiding aggravating agents in the diet: caffeine, alcohol, acidic and fatty foods. The use of head elevation in bed, smoking cessation and weight loss can also aid treatment.¹⁴ Proton pump inhibitors (PPIs) are usually commenced, along with these lifestyle modifications, to improve symptoms.^15,16 A response to this management is used as confirmation of the diagnosis.

Unfortunately, a significant proportion (30–45%) of patients do not have adequate response to PPI therapy and lifestyle modifications.¹⁷

Fundoplication

Fundoplication has been proposed as a further treatment option in patients who have a poor response or are intolerant of medical therapy.¹⁸ Fundoplication has been shown to be effective in treating GORD compared with medical therapy.¹⁹ By extension, it is expected that these procedures would also have a beneficial effect for patients with LPR.

Several anti-reflux procedures have been described, including variations in fundoplication and gastroplication. Fundoplication, a generic term for manipulation of the gastric fundus to improve lower oesophageal sphincter function, has been used in GORD treatment for many years and several techniques exist. The approach may vary (open, laparoscopic and transoral), as does the extent of wrap (360 versus 270 degrees versus partial, eg Toupet, Nissen, Dor). Laparoscopic Nissen fundoplication (LNF) and laparoscopic Toupet fundoplication (LTF) are the most common anti-reflux procedures. Transoral incisionless fundoplication is a relatively new procedure and is not yet widely adopted in clinical practice.²⁰ One of the most deterring risks of fundoplication is recurrence, which occurs in 2–17% of cases, most frequently after 11.5 months for patients with extra-oesophageal symptoms.²¹

Study rationale

Multiple small studies have assessed patient outcomes following fundoplication for LPR with little standardisation of objective and subjective patient outcomes. A 2008 systematic review included respiratory symptoms as part of the diagnosis, making the results less directly applicable to LPR patients.²² Several studies have been published since 2008 evaluating fundoplication in LPR, requiring an updated a systematic review to summarise current evidence. The emergence of the patient-reported outcome measures (PROMs) and quality-of-life scores allow clinicians to look beyond mortality as the sole outcome measure of successful surgical interventions.

Objectives

The primary aim of this review is to identify whether anti-reflux surgery is effective in improving symptoms of LPR by evaluating studies using PROMs and subjective patient questionnaire. The secondary aim is to identify whether patients who have had a poor response to medical therapy are likely to have symptom improvement with anti-reflux surgery, specifically fundoplication.

Methods

Search strategy and paper selection

This systematic review was conducted and reported with reference to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.²³

Studies and clinical trials evaluating the efficacy of fundoplication in laryngopharyngeal reflux were searched using the Cochrane database, PubMed (1966 to present), Ovid Embase (1947 to present), Ovid Medline (1946 to June Week 2 2020) and Google Scholar. The literature search was conducted independently by three authors (RM, HAE and LMW) and the latest search was performed on 15 June 2020.

Two search domains were used related to LPR and fundoplication with the ‘AND’ operator used to combine terms from both domains. The ‘OR’ operator was used to combine terms within each domain. The first included the terms: ‘Laryngopharyngeal Reflux’ (MeSH), ‘laryngopharyngeal’, ‘laryngo-pharyngeal’, ‘extraoesophageal’, ‘extraesophageal’, ‘extra-oesophageal’, ‘extra-esophageal’, ‘Supraoesophageal’, ‘Supraesophageal’, ‘Supra-oesophageal’, ‘Supra-esophageal’, ‘LPR’, ‘Reflux laryngitis’. The second domain included the terms: ‘Fundoplication’ (MeSH), ‘fundoplication’, ‘antireflux surgery’ and ‘anti-reflux surgery.

Inclusion and exclusion criteria

Original articles assessing the effect of fundoplication on adult patients with primary complaints of LPR were included. Papers were included if at least one cohort: received any type of fundoplication; had symptoms consistent with LPR and/or an RSI score of 13 or above; and had either a positive lower oesophageal or pharyngeal pH probe, positive impedance monitoring or reflux oesophagitis on oesophagogastroduodenoscopy. In addition, papers were required to report preoperative and postoperative RSI scores to assess response to surgery as validated outcome measures.²⁴

Review articles, meta-analyses, case reports, opinion-based reports and congress reports were excluded. Studies including paediatric patients (aged less than 16 years), animal studies and cadaver studies were also excluded. LPR is renowned for its difficulty in diagnosis, therefore, no diagnostic criteria were specified.²⁵ The diagnostic method was recorded to allow comparison between studies. Papers were not excluded based on the type or approach to operative fundoplication except where transoral incisionless fundoplication were excluded.²⁶

Outcome measures

RSI was chosen as the primary outcome measure because this is a validated universal scoring system that combines a variety of patient symptoms into an overall score.²⁴ RSI is felt to evaluate well the patient’s experience in treatment response and has also been demonstrated to correlate strongly with objective clinical findings.¹¹

Data extraction

The following methodological endpoints were extracted from the included studies: study type, number of patients treated with fundoplication, number of controls (where applicable), prior treatment with PPI, method of LPR diagnosis, surgical technique used, control group treatment (where applicable), follow-up duration and number of patients lost to follow-up, indication for surgical intervention, study inclusion/exclusion criteria, method of assessing efficacy (ie RSI) and any complications.

Data comparing preoperative versus postoperative RSI were collected. Fundoplication safety was also evaluated by documenting all reported adverse events.

Bias

Individual papers were reviewed for potential bias including demographics of patient groups in controlled trials, sample sizes, number of patients lost to follow-up, duration of follow-up, exclusion of deceased patients in retrospective studies, risk of recall bias in retrospective studies and conflicts of interest of associated authors.

Statistical analysis

RevMan 5 (Cochrane) was used to analyse the data. Quantitative analyses of outcomes were based on change between preoperative and postoperative scoring, with 95% confidence interval (95% CI). Weighted mean difference was used to combine continuous variables with a random effects model. Trial heterogeneity was tested via I² statistic where percentages greater than 50% indicated significant heterogeneity.

Findings

Study selection

The initial search identified 844 records (Figure 1). Of these, 347 were duplicates. The remaining 497 titles and abstracts were reviewed, and a further 455 papers were excluded because they did not meet inclusion or exclusion criteria. Full texts were obtained for the remaining 42 articles. Following application of the inclusion and exclusion criteria, a further 33 papers were excluded. No additional papers were identified by cross-referencing.

Figure 1 .

Search strategy

Study characteristics

No randomised controlled trials (RCTs) were identified. One prospective case–control study was included²⁷ and three retrospective case–control studies.²⁸ Four prospective case series were identified^29–32 and the remaining three studies were retrospective case series.^22,33,34 The total number of patients across the studies was 358, with 287 patients receiving a fundoplication.

Study comparison

All patients were specified as having a diagnosis of LPR; however, three papers did not specify how LPR was diagnosed (Table 1)^30–32 All patients underwent fundoplication only if there was objective evidence of reflux identified by 24-h ambulatory pH or multichannel intraluminal impedance–pH monitoring or erosive oesophagitis.

Table 1 .

Summary of diagnosis and criteria for fundoplication in the included studies

Study	Diagnosis of LPR	Criteria for fundoplicationa
Carroll et al (2016)33	Symptoms of LPR and RSI >13	Diagnosis of LPR Symptoms refractory to at least 3 months high-dose BD PPI Had evidence of reflux with impedance testing No motility disorder on manometry
Catania et al (2007)29	Symptoms consistent with LPR and RSI >19 FNE-proven laryngeal irritation 3 or more episodes of pharyngeal acid reflux or positive deMeester score on distal probe	Diagnosis of LPR Primary aim of surgery was resolution of LPR symptoms
Iqbal et al (2008)22	Diagnosis made by ENT surgeon – criteria not specified	Diagnosis of LPR
Ogut et al (2007)30	Not specified	GORD, RSI >14
Sahin et al (2015)27	Laryngeal symptoms, RSI >13 and RFS >7	Responded to daily PPI therapy Had at least one of heartburn and/or regurgitation
Suzuki et al (2017)34	Laryngeal symptoms	Diagnosis of LPR
Weitzendorfer et al (2017)31	Not specified	Ongoing LPR symptoms despite 6 months of PPI
Westcott et al (2004)32	Not specified	Incomplete clinical response to medication Medication intolerance, reflux-associated cancer or recurrent leucoplakia, reflux-associated subglottic stenosis, or patient desire for operative therapy in lieu of medical management
Zhang et al (2017)22	Laryngeal symptoms and positive pharyngeal pH testing	Diagnosis of LPR

LPR = laryngopharyngeal reflux; BD = twice daily; PPI = proton pump inhibitor; FNE = functional nasoendoscopy; ENT = ear nose and throat (otolaryngology); GORD = gastro-oesophageal reflux disease

^a All patients underwent fundoplication only if there was objective evidence of reflux identified by 24-h ambulatory pH or multichannel intraluminal impedance–pH monitoring or erosive esophagitis

Follow-up duration

Follow-up duration was reported in a variety of ways including mean, median, range and a specified follow-up duration. The shortest follow up was 1 month²⁹ and the longest was 120 months.²²

Preoperative PPI

Four papers did not specify a PPI regimen preoperatively.^28–30,32 Of the remaining five papers, four noted PPIs were given preoperatively but the type and dose of PPI were not specified.^22,27,31,34 One paper specified that all patients were treated preoperatively with at least 3 months of empiric, high-dose (40mg omeprazole or equivalent) twice-daily PPI therapy.³³ The mean RSI improved significantly (p < 0.05) in all papers that reported preoperative PPI use.^{22,27,31,33,34} Either this improvement was not considered adequate by the patient, or the patient elected to proceed with surgery despite this improvement due to intolerance or inconvenience associated with long-term PPI use.

Fundoplication type and efficacy

The nine included papers are summarised in Table 2. Seven papers included patients who underwent laparoscopic Nissen fundoplication (LNF),^{27–29,32–34} one paper included patients who underwent laparoscopic Toupet fundoplication (LTF),³¹ and one paper included 31 patients who underwent a mixture of types of laparoscopic fundoplication and nine open fundoplications.²⁸ Overall 250 patients received LNF, 20 received LTF, 9 received open fundoplication, and the remaining 8 patients received partial wraps (type not specified). Significant improvement (p < 0.05) in RSI post fundoplication was noted in all papers, with postoperative average (mean or median) improvement in RSI scores ranging from 6.8 to 16.7. No outcome difference was noted between the various surgical techniques.

Table 2 .

Summary of change in reflux symptom index following fundoplication

Study	Type	Evidence level	n	Preoperative PPI	Fundoplication type	Mean follow-up duration in months (range)		RSI (mean ± SD)	RSI change	p-value
Carroll et al (2016)33	Retrospective case series	IV	9	3 months BD high-dose PPI	LNF	9.9 (4–24)		31.7 to 10.0	21.7	< 0.001
Catania et al (2007)29	Prospective, consecutive case series	IV	61	Not specified	LNF	15.2 (1–36)		31.5 ± 7.4 to 12.4 ± 10.9	19.1	< 0.01
Iqbal et al (2008)22	Retrospective case series	IV	40	Optimal medical therapy including PPIa	9 open 31 laparoscopic	41.5b (6–120)		22.8 ± 10.8 to 11.83 ± 9.91	10.97	< 0.0001
Ogut et al (2007)30	Prospective case series	IV	38	Not specified	LNF	14.0		25.45 ± 7.5 to 16.52 ± 5.06	8.93	< 0.05
Sahin et al (2015)27	Prospective, single-blind, case–control study (GORD vs LPR group 1)	III	41	Daily PPIa	LNF	24.5 ± 1.3 (LPR) 25.2 ± 1.1 (GORD)		22.86 ± 7.4 to 11.27 ± 6.6 (LPR) 12.91 ± 6.4 to 8.41 ± 4.5 (GORD)	11.59 4.5	0.00 (LPR) 0.001 (GORD)
Suzuki et al (2017)34	Retrospective case series	IV	12	11 patients on PPIa	LNF	6.0b (2–10)		22.9 ± 10.0 to 6.8 ± 6.8	16.1	< 0.001
Weitzendorfer et al (2017)31	Prospective, consecutive case series	IV	20	6-month preoperativea	LTF	3.0		21.9 ± 6.9 to 10.1 ± 6.6	11.8	< 0.001
Westcott et al (2004)32	Prospective case series	IV	41	Not specified	LNF	13.9 ± 4.1		RSI 23.2 + 12.0 to 16.7 + 11.5b	6.5b	< 0.05
Zhang et al (2017)28	Retrospective case–control study (LNF vs PPI)	III	31	Not specified	LNF	48 .0 (20.4–30)		LNF: 15.3 ± 3.5 to 9.7 ± 4.12 Control: 14.2 ± 4.0 to 12.8 ± 3.12	5.6	LNF Vs PPI (p = 0.004)

PPI = proton pump inhibitor; BD = twice daily; LNF = laparoscopic Nissen fundoplication; LTF = laparoscopic Toupet fundoplication; n = number of patients who underwent fundoplication; RSI = reflux symptom index; SD = standard deviation

^a PPI type and dose not specified

^b Median

Overall, seven trials (n = 243) reported mean preoperative and postoperative RSI scores (Figure 2). Fundoplication (all types) was associated with a statistically significant reduction in RSI using a random effects model (mean improvement 11.78, 95% CI 7.95, 15.60, p < 0.01, I²= 89%).

Figure 2 .

Forest plot of reflux symptom index score pre- and postoperatively

Fundoplication versus PPI

One paper compared fundoplication with medical therapy in patients with LPR and a sliding hiatus hernia.²⁸ Thirty-one patients received LNF and 39 were treated with PPI (esomeprazole 40mg daily for 78 days; range 61–96 days). There was no significant difference in baseline RSI score (15.3 versus 14.2, p = 0.759). There was a significant difference in the improvement in mean RSI in the fundoplication group (5.6 points) compared with the PPI group (1.4 points, p = 0.004).

Fundoplication safety

Dysphagia is expected in the initial postoperative period, and usually resolves by 8 weeks. Five papers reported cases of prolonged dysphagia in 13 patients (4.53%) and this was the most common indication for postoperative intervention.^{22,27–29,32} Seven patients (2.4%) required an intervention for postoperative complications.^{28–30,32,34} One patient (0.4%) required a repeat operation for a failed procedure,²⁹ and six patients (2%) required balloon or bougie dilation.^28,29,32 Sahin et al also commented on complications of surgical emphysema (10.8%), intraoperative bleeding (4.4%), intrathoracic wrap migration (2.9%) and postoperative hernia (1.4%).²⁷

The most common complaint postoperatively was dysphagia. In one study, postoperative dysphagia occurred in all patients but resolved spontaneously after 2 weeks.³⁴ Another study reported dysphagia lasting between 2 and 6 weeks in 55.2% of patients.²²

The second most common complaint was postoperative gas/bloating. This occurred in the first 2 weeks in all 12 patients in one study³⁴ and 20.8% of patients in another.²⁷ This is a commonly reported occurrence for patients having undergone abdominal surgery.³⁵

Four papers made no comment on whether postoperative procedures were required for complications.^22,26,27,33

Satisfaction with surgery

Two papers evaluated patient satisfaction with surgery.^22,28 Zhang et al reported that at 2-year follow-up, patients reported a mean satisfaction score of 62.5, compared with 44.4 in the PPI group (p = 0.004).²⁸ Iqbal et al reported that 25% of patients described overall outcome as excellent, 32.5% good, 32.5% satisfactory and 10% bad. In addition, 85% of patients reported they were happy with their operation and 15% regretted having operation.²²

Reflux finding score

Four papers assessed change in RFS.^27,30–32 Ogut et al found RFS improved from 10.4 to 5.5 at 14 months (p < 0.05).³⁰ Sahin et al found that in the LPR group, RFS improved from 10.7 to 7.0 at 24 months follow-up (p = 0.0001).²⁷ Weitzendorfer et al noted a mean preoperative RFS of 6.1, which improved to 2.1 after 3 months follow-up (p < 0.001).³¹ Finally, Westcott et al reported median RFS improved by 4.4 (p = 0.05).³² Meta-analysis using a random effects model of the three papers which assessed mean change in RFS identified a 4.2 improvement in RFS overall (95% CI 3.47, 4.92, p < 0.01, I² = 34%, Figure 3).

Figure 3 .

Forest plot of reflux finding score pre - and postoperatively

Discussion

This review supports fundoplication as an effective treatment for patients with LPR. The strong relationship between fundoplication and improved RSI scores represents a clear potential to benefit many patients. Not only do the results suggest those who have not gained from medical management may benefit from fundoplication, but also that those who have responded to medical management may see further improvement if not resolution of their symptoms should they undergo fundoplication.^{22,27,31,33,34}

Toohill et al evaluated fundoplication versus PPI in a retrospective controlled trial and found a significant difference favouring fundoplication.²⁵ This result is supported by five other papers which found a significant improvement in RSI despite preoperative PPI treatment.^{22,27,31,33,34} This suggests that fundoplication provides a superior result but is of course, far more risk-bearing than PPI use.

By using RSI as the objective measure of improvement, this review relies on patients self-reporting symptoms. Although this provides valuable insight into the patient experience and arguably the strongest measure of success for any medical intervention (whether the patient feels better), it is open to some variation in individual perceptions of symptom severity. This potential for individual discrepancy is partially mediated by the same individuals reporting pre- and post-surgery.

PROMs are not a substitute for clinical evaluation of disease. However, when examining the efficacy of a treatment, the patient-reported outcome must be of high priority in a patient-centred approach in order to achieve clinical excellence and provide a surgical service that meets a globally rising expectation of standard of care. Furthermore, studies have shown RSI to be a reliable measure and to have directly comparable accuracy with clinical judgements.¹¹

It would be irresponsible to suggest major abdominal surgery for every patient with LPR. The question lies with which patients should be selected. In addition to the standard need to identify those who are fit for surgery, this review recommends an in-depth exploration into the severity of disease in those reporting severe symptoms (high RSI scores).

To more reliably and thoroughly record LPR disease severity, additional techniques could be employed, this would not only allow for direct postoperative comparison to be made and uncover objective improvements, but may also clarify underlying disease pathology prior to committing to major surgery. Single-probe pH monitoring uses a sensor placed in the distal oesophagus to determine the presence of acid reflux. The addition of a second probe in the upper oesophagus (1–5cm below the upper oesophageal sphincter) can monitor for proximal acid exposure, which may occur in LPR.³⁶ The use of 24-h double-probe pH monitoring allows the determination of high acid refluxes seen in LPR, although accurate placement of the upper probe is critical for true results.

Assessment with oesophageal impedance monitoring preoperatively improves identification of patients that will benefit from anti-reflux surgery.^37,38 Multichannel intraluminal impedance uses the placement of paired electrodes in the oesophageal lumen, separated by a non-conducting catheter to measure resistance to alternating current. The resistance varies with the content of the lumen, detecting liquid or gas, at points on the catheter regardless of pH. This can then determine if a bolus is swallowed or refluxed, the time taken and velocity, as well as the state (liquid versus gas), and how far proximally the reflux extends. In conjunction with pH monitoring and RSI scores this gives a more detailed assessment of reflux and symptoms,^39,40 and is particularly useful in patients that are still symptomatic despite acid suppression.⁴¹

The proposed management strategy for LPR is summarised in Figure 4.

Figure 4 .

Proposed approach to diagnosis and treatment of laryngopharyngeal reflux

Limitations

This review has some limitations. Two papers had small sample sizes of 9 and 12 patients,^33,34 however their findings correlated with other larger studies and were significant (p < 0.001). Four papers were retrospective studies opening up risk of recall bias. Of note, the retrospective nature of Iqbal et al’s study led to some patients being asked to recall preoperative symptoms up to 120 months prior, likely introducing error in preoperative RSI scores.²²

Follow-up was variable between papers, ranging from 1 month²⁹ to 120 months.²² Three of the included papers had a follow-up period of less than 1 year and may have revealed more recurrence of symptoms if follow-up was lengthened.^31,33,34

Further research

The evidence suggests that fundoplication can be an effective treatment for LPR caused by reflux from the stomach. The long-term results and recurrence rates are not clear from the current data. This review proposes that there is a need for a large, multicentre RCT comparing ongoing PPI therapy versus laparoscopic fundoplication using PROMS such as the RSI and RFS with a follow-up of at least 3–5 years. This follow-up period is based on operative experience of the authors as well as the body of literature reporting complications and recurrence following fundoplication.²¹ Recruitment to such a RCT should involve manometry and impedance testing.

Conclusions

This review collates current evidence evaluating the efficacy of fundoplication in LPR. A large RCT is required to provide higher level evidence. Current evidence suggests that fundoplication provides significant symptom improvement in patients with LPR despite prior PPI use and could be considered in patients who had an unsatisfactory response to a 2-month course of optimal medical therapy, or for those for whom long-term PPI is not feasible or tolerable (as is National Institute for Health and Care Excellence guidance for GORD).