Back to the Diagnosis of Silent GERD?

David A. Johnson, MD


October 28, 2010

Rabeprazole Is Effective in Treating Laryngopharyngeal Reflux in a Randomized Placebo-Controlled Trial

Lam PK, Ng ML, Chelung TK, et al
Clin Gastroenterol Hepatol. 2010;8:770-776

Proton Pump Inhibitor Therapy Improves Symptoms in Postnasal Drainage

Vaezi MF, Hagman DD, Slaughter JC, et al
Gastroenterology. 2010 Aug 27. [Epub ahead of print]

The "Silent GERD" Iceberg

The diagnosis and optimal treatment of the extraesophageal complications of gastroesophageal reflux disease (GERD) have been long-standing topics in evolution. A decade ago, it was believed that patients who had GERD with the typical symptoms of heartburn and regurgitation were just the "tip of the iceberg," and that patients with a full spectrum of more prevalent GERD-related complications (eg, pulmonary, otolaryngologic, dental) lurked "below the waterline." It was suspected that many patients who had these GERD-related complications might actually have silent GERD: that is, GERD without heartburn and regurgitation. This philosophy was adopted in particular by the ear-nose-throat (ENT) field, which developed a large base of support for attributing a vast array of laryngeal symptoms to GERD. Over the past several years, studies have shown that 1 in 10 new patients seeing an ENT specialist leaves with a diagnosis of laryngeal-pharyngeal reflux (LPR).

Accordingly, considerable effort has focused on clinical trials involving acid-suppressive therapies for patients with suspected extraesophageal complications of GERD. Although many of the initial studies suggested benefit from use of proton-pump inhibitors (PPIs), randomized, placebo-controlled studies did not demonstrate meaningful clinical benefit. Because trials that evaluated patients with only suspected extraesophageal GERD, but who lacked the typical heartburn and regurgitation symptoms, showed no effect for PPI therapy, the teaching paradigm shifted. The new consensus was that no large "iceberg" of patients with these extraesophageal complications due to "silent GERD" existed, and accordingly, no predictable clinical response was seen when acid suppression (even with high-dose PPIs) was prescribed.

Study Summaries

Two recently published studies challenge the current consensus that "silent GERD" should not be suspected in patients with extraesophageal GERD. Furthermore, the studies found that PPIs may be effective therapy.

The study by Lam and colleagues was conducted in Hong Kong and was a double-blind, randomized, placebo-controlled trial in patients with clinical LPR who were referred to the ENT specialty clinic. All patients had a complete ENT examination and video laryngostroboscopy (VLS). The inclusion criteria for the study were (1) at least 1 of the symptoms of hoarseness, globus, persistent throat discomfort, or repetitive throat clearing for at least 1 month in the preceding year; (2) VLS evidence of LPR with a reflux finding score >7 (using a standard ENT-developed reflux scoring system); and (3) no upper respiratory tract infection in the 4 weeks before evaluation or allergic cause for laryngitis.

Patients who had used acid-suppressive therapy within the previous 4 weeks were excluded. All patients had manometry and 24-hour pH testing with a 4-sensor pH probe positioned distally 5 cm above the lower esophageal sphincter and with the proximal probe located just distal to the upper esophageal sphincter. The patients were considered LPR-positive if the pH for the proximal sensor registered <4 for more than 0.7% of the total time, 1.1% of the time in the upright position, or 0.5% of the time in the recumbent position. Similarly, patients were considered GERD-positive if the corresponding values were at least 4.6%, 7%, or 4.5%, respectively.

Patients were randomly assigned to receive rabeprazole 20 mg twice daily (42 patients) or placebo (40 patients) with 12 weeks of therapy and were reevaluated 6 weeks after therapy was discontinued. Patient demographic characteristics were well balanced between the 2 groups. All ENT evaluations were performed by the same laryngologist in a blinded assessment of the VLS examinations. The intrarater reliability had been assessed for this individual before the study, and the Pearson correlation coefficient was 0.93. Patients were also assessed using a reflux symptom index, which assessed symptoms of LPR, and using sequential reflux finding score evaluations.

The study showed significant improvement in the reflux symptom index with rabeprazole at week 6 (P = .003) and week 12 (P = .002), but this did not persist when patients were assessed at week 18 (6 weeks after stopping therapy). Specific benefit was evident for improvement in breathing difficulty or choking (P = .009), troublesome cough (P = .006), and sensation of "something sticking in the throat" (P = .007). No differences between the rabeprazole and placebo groups were evident in reflux symptom index as assessed by VLS.

The study by Vaezi and colleagues was a double-blind, parallel, placebo-controlled trial of 75 patients who had continued postnasal drainage. Patients were assessed by the ENT clinic; all patients had negative radioallergosorbent test allergy panels (or skin tests), or these tests were positive but the patient had an insufficient response to conventional therapies (eg, allergy desensitization shots, topical nasal steroid, antihistamines). In addition, all patients were nonsmokers and had normal computed tomography scans of the sinuses to exclude sinusitis. One third of patients had previously tried a PPI. Severity of GERD symptoms was assessed using a Likert scale. Ambulatory pH/impedance monitoring was performed in the 65% of patients who consented to this testing.

Patients were randomly assigned to receive lansoprazole 30 mg twice daily or placebo. Baseline pH and manometry testing were abnormal in 30% and 31%, respectively, of the treatment group and 9% and 6% of the placebo group. The primary outcome measure was response as assessed by a visual analog scale. Secondary outcome measures were 2 comprehensive standardized and validated questionnaires for sinus symptom outcomes, as well as a quality-of-life assessment in reflux and dyspepsia.

Overall, patients who took lansoprazole had significant improvements in the primary symptom of postnasal drainage. At 8 and 16 weeks, these patients were, respectively, 3.12 (95% CI, 1.28-7.59) and 3.50 (95% CI, 1.41-8.67) times more likely to respond than patients receiving placebo. Corresponding treatment benefit was also evident for the sinus questionnaires and scores on the quality-of-life assessment in reflux and dyspepsia. Subgroup analysis showed that neither baseline presence of typical GERD symptoms (heartburn, regurgitation) nor esophageal physiologic variables (pH or impedance motility) were predictive of likelihood of response to therapy.


These 2 studies were both done very well and refute the current convention that suspected extraesophageal GERD (patients who also lack typical GERD symptoms) does not respond to therapy. Why does this dichotomy exist? One key answer, at least for Lam and colleagues' study, is how the outcomes are assessed. The symptoms that improved with the reflux symptom index at week 6 were heartburn and sensation of "something sticking in the throat" -- both of which may be also typical GERD symptoms. In addition, at week 12, the significant improvements were the effect on heartburn and postnasal drip (which, according to Vaezi and colleagues' study, seems to respond to PPI therapy). Hence, the greatest response was not for the symptoms of chronic laryngitis, but rather for the symptoms that typically respond to PPI therapy.

The lack of improvement for the laryngeal examinations reiterates and is consistent with previous findings from randomized controlled trials. Clearly, laryngeal irritation on ENT examination is very nonspecific and as a general rule should not routinely suggest the diagnosis of LPR.

The study by Vaezi and colleagues was conducted by one of the "therapeutic nihilists" for PPI treatment of LPR. Vaezi previously performed the largest randomized controlled trial on chronic laryngitis and GERD.[1] In that study, a 3-month trial of high-dose esomeprazole (40 mg twice daily) vs placebo showed no effect. Of note, in the current study, the treatment response was not predictable by baseline GERD or esophageal physiologic variables. So how does this affect postnasal drip? One possible mechanism is reduced gastric volume and therefore reduced refluxate potentials for both acidic and nonacidic reflux. Furthermore, a potential reduction in acid vapor may play a role in this specific subset of patients. Or, the benefit may be unrelated to acid reduction. Both in vitro and in vivo studies have suggested that PPIs may exert anti-inflammatory effects mediated through antioxidant effects or by inhibiting the production of proinflammatory cytokines. The study is limited by the relatively small sample size and the lack of pH and motility testing in all patients. Nonetheless, the findings raise once again the topic of silent GERD and extraesophageal reflux. Further studies are needed, but at present, it looks like we are "back to the future."