It is now widely recognized that extraesophageal reflux (reflux reaching structures above the upper esophageal sphincter) has a wide range of effects on the upper aerodigestive tract, as well as the lungs. The degree to which this reflux contributes to the pathophysiology of sinusitis and other sinonasal diseases is still not completely clear, but an increasing body of literature supports a potential role.
- •
There is a strong body of evidence showing a high prevalence of pharyngeal reflux events in patients with surgically refractory chronic rhinosinusitis (CRS).
- •
In medically refractory CRS, most studies show a high prevalence of pharyngeal reflux events.
- •
The studies looking directly at the value of treatment of reflux in patients with CRS show some benefit, although more powerful randomized studies are required for a more definitive conclusion.
- •
There is significant evidence for a link between postnasal drip symptomatology and the presence of pharyngeal reflux, with good evidence for empiric treatment of postnasal drip with proton-pump inhibitors.
Overview
Chronic rhinosinusitis (CRS) remains one of the most common health care problems in the United States. The pathophysiology of the disease process is complex but involves inflammatory changes in the nasal and sinus mucosa, resulting in edema and obstruction. These changes in turn cause mucus stasis with subsequent infection. The initiating insult causing these changes can be due to a variety of sources including viral infection, environmental pollutants, and immune-mediated processes such as environmental allergy or allergic fungal sinusitis. In addition, laryngopharyngeal reflux (LPR) has recently been implicated as a potential contributor to the pathophysiology of CRS. This concept is consistent with the widespread role extraesophageal reflux has been found to play in the pathophysiology of diseases throughout both the upper and lower aerodigestive tract.
The exact mechanism by which LPR contributes to the pathophysiology of CRS is unclear. There are currently 3 main theories:
- 1.
The first theory proposes direct exposure of the nasopharynx and nose to gastric acid. The acidic refluxate causes mucosal inflammation and impaired mucociliary clearance.
- 2.
The second proposed mechanism is through a vagus nerve–mediated reflex, which has been described to exist in the lower airway. Evidence to support this concept is reported by Wong and colleagues, who showed a tendency for an increase in nasal mucus production, nasal symptom scores, and to a lesser extent nasal inspiratory peak flow in subjects who had direct administration of both nasal saline and HCl-containing fluid at the level of the gastroesophageal junction.
- 3.
The third mechanism involves a direct role of Helicobacter pylori. Koc and colleagues found H pylori to be present in polyp tissue but not normal mucosa taken during surgery for concha bullosa, although the numbers were small. In another study, Morinaka and colleagues found that 3 of 19 specimens from patients undergoing surgery for CRS contained H pylori by polymerase chain reaction. This study did not contain a nondiseased tissue comparison.
Whether one of these processes mediates the effects seen in CRS, alone or in combination, is still unclear, and further research is required.
Evidence-based clinical management
Evidence for the role of LPR in the etiology of CRS comes mainly from research using multisensor pH-probe studies looking for the presence of acid in the esophagus, hypopharynx, and nasopharynx in patients with CRS. It is further derived from studies looking at the correlation of LPR and other associated sinonasal diseases such as postnasal drip (PND) and vasomotor rhinitis (VR), and from evidence concerning pediatric sinonasal disease.
Evidence from Pediatric Studies
Evidence for a relationship between gastroesophageal reflux (GER) and pediatric sinonasal disease comes from a variety of sources and study designs ( Table 1 ). Carr and colleagues, in a retrospective study of children younger than 2 years, reported a history of reflux in 42% of children undergoing adenoidectomy compared with 7% of children undergoing tympanostomy tube placement alone. The study was hampered by multiple inclusion criteria for the diagnosis of GER and few patients receiving pH studies for diagnosis. El-Serag and colleagues retrospectively looked at a large cohort of children with and without a diagnosis of GER disease (GERD), and found a higher prevalence of sinusitis in the GERD group (4.2% vs 1.4%). The study was weakened by the lack of any standardized criteria for establishing a diagnosis of GERD (group inclusion was based on presence of reflux ICD-9 code) and the heterogeneous nature of the 2 groups (the GERD group was older).
| Authors, Ref. Year | Type | Size | Measurement | Result | EBM Level |
|---|---|---|---|---|---|
| Carr et al, 2001 | Case-control | 194 | Presence of GERD by gastric scintiscan, pH probe, or reflux on barium swallow | 42% GERD in adenoidectomy group vs 7% in the PE tube group | 2b |
| El-Serag et al, 2001 | Case-control | 9900 | Presence of sinusitis and other upper airway diseases in patients with or without a diagnosis of GERD | 4.2% sinusitis in GERD group vs 1.4% sinusitis in the non-GERD group | 3b |
| Phipps et al, 2000 | Prospective cohort | 30 | Percentage of CRS patients with GERD | 63% of CRS patients had GERD by dual-channel pH monitor | 4 |
| Contencin et al, 1991 | Prospective case-control | 31 | Presence of nasopharyngeal reflux in patients with sinonasal disease vs control | Significantly more time spent with pH below threshold in the sinonasal disease group, P <.00005 | 2b |
| Megale et al, 2006 | Retrospective case series | 45 | Percentage of patients with GERD and sinonasal complaints who respond to antireflux therapy | 83.87% improvement in chronic nasal obstruction, and 85.7% in nasal secretions | 4 |
| Bothwell et al, 1999 | Retrospective cohort | 28 | Avoidance of surgery if treated for GERD | 83% of patients in study successfully avoided surgery | 4 |
Phipps and colleagues looked at a cohort of patients with CRS refractory to medical therapy. The goal was to determine the prevalence of GERD in these patients with CRS using dual-sensor pH monitoring. In this cohort 63% of patients were diagnosed with GERD. Of those patients with GERD, 32% were found to have nasopharyngeal reflux (NPR). In patients who were diagnosed with GERD, reflux treatment was instituted, achieving a 79% improvement in sinusitis symptoms based on parental opinion. The investigators concluded that this cohort had a higher prevalence of GERD than would be expected for their population, and secondly that treatment improved patients’ sinusitis symptoms. Their study is hampered by the lack of a control group. Contencin and Narcy evaluated 31 children for the presence of NPR by performing 24-hour nasopharyngeal pH studies. The study group consisted of 13 children with recurrent or chronic rhinitis or rhinopharyngitis. The control group of 18 children was free of any nasopharyngeal disease. The study group was found to have significantly more time in the nasopharynx with a pH below the threshold (pH less than 6). Because a single-channel pH monitor was used in this study, it was impossible to determine whether some of the reflux events were nasopharyngeal reflux of gastric contents or artifact.
Other pediatric studies looked directly at the results of treating reflux in the presence of sinonasal disease. Megale and colleagues retrospectively looked at a cohort of children diagnosed with GERD by single-probe pH monitor and history. This study evaluated patients’ response to treatment with antireflux interventions including prokinetic agents, proton-pump inhibitor (PPI) therapy, and reflux surgeries. Therapy for GERD significantly improved the symptoms of chronic nasal obstruction and nasal secretion by 83.87% and 85.7%, respectively. Unfortunately, about half the number of the treated patients with these complaints also received antihistamine therapy at the same time as the antireflux medication, therefore confounding the result. This confounder, in addition to the lack of a control group in the study design, greatly weakens the strength of the results. Bothwell and colleagues looked retrospectively at a cohort of pediatric patients who had met the criteria to undergo functional endoscopic sinus surgery (ESS) for CRS. It was found that in patients treated with a variety of different antireflux therapies, sinus surgery could be avoided in 89% of patients.
Evidence from Studies Looking at Associated Sinonasal Disorders
Studies looking at the relationship between reflux and other inflammatory sinonasal disorders also allude to the role of reflux in creating, or contributing to, an environment conducive to the presence of sinusitis ( Table 2 ). Loehrl and colleagues sought to look at the relationship between extraesophageal reflux (EER) and VR, which was defined by the symptom of nasal congestion for at least 3 months without any signs of current infection, pregnancy, nasal polyps, or allergy. The study included 3 groups: patients with VR without EER, patients with VR with EER (by history and physical examination), and normal controls without VR. The amount of reflux was documented by 4 sensor pH probes in all groups. The amount of autonomic dysfunction present in each group was also measured using the composite autonomic scoring scale (CASS). The group with VR with EER by history and physical examination had a higher CASS score than both the VR-free control group and the VR-alone group. The investigators further confirmed the presence of EER via a 4-sensor pH study, finding that the VR-with-EER group had nasopharyngeal reflux events present in 9 of 10 patients. This study shows that EER can change the environment in the sinonasal cavity, in this case by altering the body’s autonomic response.
| Authors, Ref. Year | Type | Size | Measurement | Result | EBM Level |
|---|---|---|---|---|---|
| Loehrl et al, 2002 | Case-control | 30 | Prevalence of autonomic dysfunction in VR patients with or without EER and normal controls | Significantly increased autonomic dysfunction in VR patients with EER compared with those without EER | 2b |
| Wise et al, 2006 | Cohort | 68 | Association between PND symptoms and the presence of NPR and LPR by 3-channel pH probe | Significantly more PND symptoms in patients with NPR and LPR | 4 |
| Vaezi et al, 2010 | Randomized controlled trial | 75 | Improvement in PND symptoms following treatment with lansoprazole | Significantly greater percentage improvement in the treatment arm vs control | 1b |
Other investigators have looked at the relationship between the patient complaint of PND and reflux. Wise and colleagues looked at the relationship between PND, defined as an increased awareness of the movement of the pharyngeal mucus blanket, and LPR in a cohort of 68 patients. These patients were first asked to complete both the validated SNOT-20 questionnaire and the modified Reflux Symptoms Index questionnaire, and then underwent 24-hour pH testing with a triple-sensor pH probe. The pH probe had sensors at the nasopharynx, hypopharynx (1 cm above the upper esophageal sphincter), and distal esophagus. The investigators found that in patients with nasopharyngeal reflux events with a pH less than 5, there were significantly more PND symptoms reported on the SNOT-20 and the modified Reflux Symptom Index survey, compared with patients without reflux in this area. Patients with LPR also had more PND symptoms on the SNOT-20 survey when compared with patients without LPR.
Vaezi and colleagues asked the question of whether directly treating reflux would improve patients’ PND symptoms. To answer this question they performed a double-blinded study on 75 patients with complaints of PND without any signs of chronic sinusitis or allergy, randomizing them to either twice-daily lansoprazole or placebo. Patients completed validated sinus disease questionnaires (SNOT-20 and RSOM-31) and the Quality Of Life in Reflux And Dyspepsia questionnaires (QOLRAD) and underwent ambulatory pH and impedance monitoring before the institution of therapy. This pretreatment pH monitoring was performed in only 65% of participants but in equal amounts for each of the study groups. The primary outcome measure was a visual analog scale describing the percentage resolution of the PND sensation. Patients were then followed up after 8 and 16 weeks of therapy. Patients given lansoprazole therapy had a 3.12-fold greater (at 8 weeks of therapy) and 3.5-fold greater (at 16 weeks of therapy) chance of improving compared with controls. At 16 weeks the median improvement in the treatment arm was 50% compared with 5% in the placebo arm. In addition, there was a statistically significant improvement in the SNOT-20 and QOLRAD outcomes for the treatment arm. Of note, no link was reported between the presence of reflux on pH study and the response to treatment. The technique used, however, only assessed the presence of reflux into the esophagus and not into the nasopharynx. The investigators state that the study results do support a role for reflux in causing PND symptoms in this group of patients, but comment that alternative causes for the benefit seen may come from possible intrinsic anti-inflammatory properties of the PPI drugs and a putative decrease in nonacid reflux created by PPI drugs.
Evidence from Adults with CRS
Several studies have looked directly at the relationship between reflux and CRS ( Table 3 ). Ulualp and colleagues evaluated a diverse group of patients with upper airway and sinonasal complaints, and found a statistically higher incidence of hypopharyngeal acid reflux events in patients with both persistent CRS after sinus surgery and posterior laryngitis (4 of 6 patients, 67%) compared with healthy controls (7 of 34, 21%) or with patients with CRS without posterior laryngitis (4 of 12, 33%). There was no difference between the distal and proximal esophageal reflux parameters between these groups. The investigators concluded that LPR may play a role in a subset of patients with CRS, and posterior laryngitis may be the common thread among reflux-induced aerodigestive tract disorders. The small number of sinusitis patients is a drawback of this study. Ulualp’s group also looked directly at a cohort of 11 patients with medically refractory CRS, and found a higher prevalence of pharyngeal acid reflux compared with a group of 11 healthy controls (7 of 11, 64% vs 2 of 11, 18%). A study by Jecker and colleagues evaluated 20 patients with CRS who had previously failed surgical therapy, and compared them with a group of normal controls. In these patients dual-sensor pH readings were obtained 6 weeks after the patients’ revision sinus surgery. The CRS patients had a higher number of reflux events and percentage of time with pH spent below 4 in comparison with the control group. These results, interestingly, were not found in the hypopharynx. The investigators concluded that there was a link between GERD and CRS but not between EER and CRS. This study was hampered by its small size and heterogeneous groups.
| Authors, Ref. Year | Type | Size | Measurement | Result | EBM Level |
|---|---|---|---|---|---|
| Ulualp et al, 1999 | Case-control | 18 | Presence of proximal reflux by 3-channel pH probe in patients who failed sinus surgery | Higher percentage of reflux in CRS patients with laryngitis and CRS patients alone compared with controls | 2b |
| Ulualp et al, 1999 | Case-control | 22 | Presence of proximal reflux by 3-channel pH probe in patients with medically refractory CRS | Higher percentage of reflux in CRS patients compared with controls | 2b |
| Jecker et al, 2005 | Case-control | 40 | Presence of LPR or GERD by dual-channel pH probe | Significantly more GER events in the CRS patients than in controls, but not in hypopharynx | 2b |
| Ozmen et al, 2008 | Case-control | 52 | Presence of LPR by dual-channel pH probe and presence of pepsin in middle meatus aspirate | More pharyngeal acid events in CRS group; 88% vs 55%, pepsin was found in most patients with reflux | 2b |
| DelGaudio, 2005 | Case-control | 68 | Presence of NPR, reflux at UES or GERD by 3-channel pH probe | Significantly more reflux events in the NP, UES, and LES in the CRS group compared with successful treatment and controls | 2b |
| Wong et al, 2004 | Cohort | 37 | Presence of acid reflux into the nasopharynx by 4-channel pH probe in CRS patients failing medical therapy | Found nasopharyngeal reflux in only 2 of 37 patients (5%); GER was found in 12 of 37 patients (32.4%) | 4 |
| DiBaise et al, 2002 | Cohort | 11 | Response to twice daily PPI therapy in patients with CRS | Modest symptom improvement | 2b |
| Pincus et al, 2006 | Cohort | 15 | Response to daily PPI therapy in patients with medically and surgically refractory CRS | Modest symptom improvement | 4 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
