Esophageal Motility in Diabetes Mellitus

There are few objective methods of diagnosing diabetic visceral neuropathy. In the past the author has encountered patients with diabetes mellitus who have shown abnormal esophageal motility in enough instances to lead him to suspect a significant association. Because of this he instituted a systematic radiographic study of esophageal motility in diabetes mellitus. During the course of this investigation a study was published indicating that 12 of 14 diabetics with neuropathy-gastroenteropathy manifested an abnormal esophageal motility pattern on cineradiographic study (1). This is the only previous publication pertaining directly to the problem. Method Esophageal peristalsis was observed in diabetic and in nondiabetic individuals in both prone oblique positions. The barium mixture used was the same as that routinely employed in our department. If a normal progressive peristaltic wave occurred, at least one was recorded cineradiographically. If no progressive peristaltic wave appeared, the patient was observed fluoroscopically during a number of deglutitions for five minutes. During this time, short sample periods were recorded cineradiographically. Other factors such as tertiary contractions, dilatation, hiatal hernia, reflux, and incomplete emptying were observed and recorded, but the presence or absence of an effective peristaltic contraction that traversed the length of the esophagus was most easily assessed and was assigned prime significance. Ninety-seven males were studied, 41 of whom had diabetes mellitus. The mean age of the 41 was fifty-five years (twenty-one to seventy-eight years) and of the 56 nondiabetics fifty years (thirty-one to seventy-three years). Results Table I shows the overall results, based on presence or absence of an effective progressive peristaltic contraction. Statistically, this difference is significant at the 0.01 probability level (χ2 = 6.62). The mean duration of diabetes mellitus in the patients with abnormal esophageal motility was eleven years (new to twenty-six years). Only one person in this group had had diabetes mellitus for less than six years, and he was seventy-three years old when the recent diagnosis was made. In the diabetics with normal motility the mean duration of disease was four and a half years (new to sixteen years); in 11 of these diabetes had been diagnosed for less than one year and in 4 cases it was detectable only with a glucose tolerance test. A study was then made of those patients in whom diabetes mellitus was known to exist for one or more years (Table II). Statistically, when the results in this group are compared to those in the non-diabetics, the difference becomes even more significant (χ2 = 10.9). The frequency of peripheral neuritis and∕or visceral neuropathy increases with the duration of diabetes mellitus.