Direct Nasopharyngeal Reflux of Gastric Acid is a Contributing Factor in Refractory Chronic Rhinosinusitis

OBJECTIVES/HYPOTHESIS: To determine whether there is a difference in the prevalence of reflux in patients with refractory chronic rhinosinusitis (CRS) compared with control patients, including whether direct nasopharyngeal reflux (NPR) occurs in CRS patients. It is hypothesized that refractory CRS patients have a greater incidence of laryngopharyngeal reflux and NPR events and that NPR is a significant etiologic factor for CRS in these patients. STUDY DESIGN: Prospective study. METHODS: The study group consisted of 38 patients with a history of at least one endoscopic sinus surgery (ESS) with continued CRS symptoms and mucosal inflammation on endoscopy. The first control group consisted of 10 patients who had at least one ESS procedure and had no symptoms of CRS or mucosal inflammation a minimum of 1 year postoperatively. The second control group consisted of 20 subjects with no history of CRS or sinus surgery. All patients completed reflux symptom scales, a 20 item sinonasal outcome test, and a sinusitis symptom scale and underwent nasal endoscopy to grade the nasal mucosal findings. Patients underwent a 24 hour pH study with a specially designed probe with sensors located in the nasopharynx, 1 cm above the upper esophageal sphincter (UES), and the distal esophagus. The pH recordings were evaluated for NPR events less than pH 4 and 5. Reflux at the UES probe was considered pathologic if there were more than 6.9 episodes for the entire study or the reflux area index (RAI) exceeded 6.3. Esophageal reflux was defined as abnormal if greater than 4% of the study time was spent at pH less than 4. Statistical analysis was performed with Fisher's exact test to compare the reflux parameters and with analysis of variance and Tukey's post hoc analysis for the symptom and examination scores. RESULTS: No statistical difference was found between the two control groups for any parameters at any sites. When a single outlier was dropped from the nonCRS control group, less NPR was found in the nonCRS group compared with the successful ESS control group (P = .03). Because these groups were statistically homogenous, they were collapsed into a single control group. Compared with the control group, the study group had significantly more patients with NPR events pH less than 4 (39% vs. 7%, P = .004) and an even greater difference in the number of patients with NPR events pH less than 5 (76% vs. 24%, P = .00003). At the UES, 74% of the study group had greater than 6.9 reflux episodes, compared with 38% of control patients (P = .006). The UES RAI was abnormal for 58% of the study group compared with 21% of the control group (P = .007). The study group also had more gastroesophageal reflux (66% vs. 31%, P = .007). For nasopharynx and UES reflux parameters, the differences between study and control groups increased when the patients with isolated frontal recess disease were removed from the dataset. The study group also had higher scores on all symptom and examination scores (P = .001 for each scale). CONCLUSIONS: Patients with persistent CRS after ESS have more reflux at the nasopharynx, UES, and distal esophagus than controls. The greatest difference is in NPR, especially pH less than 5. This is the first study to document NPR in CRS patients, and it is likely to represent an important causative factor of refractory CRS in adults.