Temporal trends in the relative prevalence of dysphagia etiologies from 1999-2009.

AIM: To examine the relative prevalence and temporal variation of dysphagia etiologies in patients undergoing upper endoscopy (EGD) over the past decade.
METHODS: EGDs with the indication of dysphagia at an urban, university medical center in 1999, 2004 and 2009 were retrospectively identified from the electronic medical record. The entire patient chart, including EGD, pathology, manometry, radiographic and clinician reports, was reviewed for demographic and clinical data and to determine the etiology of dysphagia. The number of EGDs in which an esophageal biopsy was performed was also noted. Gastroesophageal reflux disease (GERD) as a cause of dysphagia independent of peptic stricture was defined by symptoms with erosive esophagitis or symptom response to proton pump inhibition (PPI). Cases of eosinophilic esophagitis (EoE) were defined by an appropriate clinical history and histological criteria of ≥ 15 eosinophils per high powered field. PPI-responsive esophageal eosinophilia was not routinely reported prior to 2008. Statistical analysis was performed using one-way analysis of variance to analyze for trends between 1999, 2004 and 2009 and a post-hoc Tukey analysis was performed following a significant main effect.
RESULTS: A total of 1371 cases (mean age 54 years, 43% male) met pre-specified inclusion criteria with 191, 504 and 675 cases in 1999, 2004 and 2009, respectively. Patients were older in 2004 compared to 2009 (mean ± SD, 54.0 ± 15.7 years vs 52.3 ± 16.8 years, P = 0.02) and there were more males in 1999 compared to 2004 (57.5% vs 40.8%, P = 0.005). Overall, GERD (27.6%) and EoE (7.7%) were the most common identifiable causes of dysphagia. An unspecified diagnosis accounted for 21% of overall cases. There were no significant differences in the relative prevalence of achalasia or other motility disorders, peptic stricture, Schatzki's ring, esophageal cancer or unspecified diagnoses over the 10-year time period. There was, however, a decrease in the relative prevalence of GERD (39.3% vs 24.1%, P < 0.001) and increases in the relative prevalence of EoE (1.6% vs 11.2%, P < 0.001) and oropharyngeal disorders (1.6% vs 4.2%, P = 0.02) from 1999 to 2009. Post-hoc analyses determined that the increase in relative prevalence of EoE was significant between 1999 and 2009 as well as 2004 and 2009 (5.4% vs 11.6%, P < 0.001), but not between 1999 and 2004 (1.6% P 5.4%, P = 0.21). On the other hand, the decrease in relative prevalence of GERD was significant between 1999 and 2009 and 1999 and 2004 (39.3% vs 27.7%, P = 0.006), but not between 2004 and 2009 (27.7% vs 24.1%, P = 0.36). There were also significantly more EGDs in which a biopsy was obtained in 1999 compared to 2009 (36.7% vs 68.7%, P < 0.001) as well as between 2004 and 2009 (37.5% vs 68.7%, P < 0.001). While total EGD volume did increase over the 10-year time period, the percentage of EGDs for the indication of dysphagia remained stable making increasing upper endoscopy an unlikely reason for the observed increased prevalence of EoE.
CONCLUSION: EoE has emerged as a dominant cause of dysphagia in adults. Whether this was due to a rise in disease incidence or increased recognition is unclear.