- Email: [email protected]
Oesophageal function tests: Are they of value?
Chmcal Radiology (1985)36, 493-496 © 1985RoyalCollegeof Radiologists
000%9260/85/447493502.00
Oesophageal Function Tests: Are They of Value? K. S. CHANNER and J. P. VIRJEE
Departments of Cardiology and Radiodiagnosis, Bristol Royal Infirmary, Bristol
Oesophageal function was assessed in 15 unselected control patients, eight patients with systemic sclerosis, 10 diabetics with autonomic neuropathy and 24 diabetic controls, by water bolus transit time derived from oesophageal scintigraphy, barium swallow and by timing the transit of a barium sulphate-filled capsule through the oesophagus. Water transit times and capsule transit times were significantly prolonged in patients with systemic sclerosis and diabetics with autonomic neuropathy, compared with controls. Barium swallow was abnormal in seven of eight patients with systemic sclerosis, whereas water transit time was abnormal in all eight and capsule transit time was abnormal in six of seven. Nine of 10 patients with diabetic autonomic neuropathy had abnormal barium swallows and water transit times but all 10 had prolonged capsule transit times. Eleven of 24 diabetic controls had abnormal barium swallows and water transit times, but 21 had abnormally prolonged capsule transit times. Six of 15 controls had abnormal barium swallows, four had abnormal water transit times and 12 had abnormal capsule transit times. In conclusion, water and capsule transit times are sensitive tests of oesophageal function and are as effective as barium swallow in detecting oesophageal motility disorders associated with systemic sclerosis and diabetic autonomic neuropathy. Capsule transit time is cheaper, involves a smaller radiation exposure than oesophageal scintigraphy and may be more sensitive.
The function of the oesophagus is to transmit ingested material from the oropharynx to the stomach. Far from being a simple tube, the oesophagus has a complex muscular arrangement and nervous supply and is the part of the gastrointestinal tract where somatic and autonomic nerves interplay most in the control of peristalsis. Oesophageal function may be affected by structural obstructions or by disturbances in motility. Barium swallow and endoscopy are highly sensitive ways of recognising structural abnormalities, but may be insensitive methods of assessing oesophageal motor function (Ott et al., 1981). In the last 20 years much attention has centred on oesophageal manometry as a diagnostic tool in the evaluation of the motor function of the oesophagus. However, this is an unpleasant procedure for many patients and suffers from problems of interpretation. It is time-consuming and labourintensive. More recently, oesophageal function tests have been developed which rely on timing the transit of a bolus through the oesophagus. Disorders of oesophageal peristalsis result in prolonged oesophageal transit times. In earlier studies the transit of radio-labelled water (Kazem, 1972; Tolin et al., 1979; Russell et al., 1981; Blackwell et al., 1982; Malmud and Fisher, 1982; Black44
well et al., 1983), test meals (Bosch et al., 1977), gelatin or capsules (Fisher et al., 1982) through the oesophagus was measured using gamma cameras with on-line computer facilities, so that time-activity curves could be plotted and transit times derived from these. We have developed an analogous technique: using a stopwatch, we have timed the oesophageal transit of barium sulphate-filled capsules swallowed whilst screening with X-ray fluoroscopy. In earlier studies we established the effect of posture and drink volume (Channer and Virjee, 1982) on oesophageal transit time and used this technique to study the effect of hyoscine butylbromide (Charmer et al., 1983a) and glucagon (Channer et al., 1983b) on oesophageal function. In this present study we have compared two tests of oesophageal function, oesophageal scintigraphy and timed capsule transit, with barium swallow in the investigation of oesophageal motility disorders in patients with systemic sclerosis and diabetes mellitus.
PATIENTS AND METHODS This study was approved by the Ethics Committee of the Bristol and Weston Health Authority. Patients
Fifty-seven patients took part in the study. They comprised four groups. 1. Controls Fifteen patients attending the X-ray Department for planned barium meal examinations. 2. Systemic Sclerosis Eight patients with established systemic sclerosis (six with calcinosis, Raynaud's sclerodactyly and telangiectasia (CRST) syndrome). 3. Diabetics With Autonomic Neuropathy Ten diabetics (nine on insulin) previously found to have autonomic neuropathy on the basis of abnormal heart rate responses to physiological stimuli. 4. Diabetic Controls Twenty-four patients (21 on insulin) shown not to have autonomic neuropathy. Methods
Three tests of oesophageal motor function were performed.
494
CLINICAL RADIOLOGY
1. Oesophageal Scintiscan (Water Transit Time) Patients were positioned supine below a gamma camera (Maxicamera 400T) fitted with a low-energy parallel hole collimator. The images were obtained using an on-line data processor. The position of the upper oesophagus was established using a radioactive marker, temporarily placed over the cricoid cartilage, and the field of view was adjusted to include the entire oesophagus and gastric fundus. Water (10 ml) was placed in the patient's mouth from a syringe. The patient was asked to swallow once only and to take repeated single swallows at 30 s intervals for 2 min. Trial runs were performed until the patients could co-operate. Radionuclide images were obtained when the patient swallowed 10 ml water containing 20 MBq 99mTc sulphur colloid. Two hundred images were recorded at 0.5 s intervals and stored for analysis. The position of the upper oesophagus was again confirmed by static imaging for 10-20 s with a radioactive cricoid marker. This procedure was repeated. The data were analysed as follows. A composite image by the summation of frames provided a single image of the complete oesophagus and stomach. The position of the radioactive marker was used to define the upper end of the oesophagus and activity in the gastric fundus defined the lower end of the oesophagus. The image of the oesophagus was then divided into four regions of interest: (a) (b) (c) (d)
upper third of oesophagus, mid-third oesophagus, upper half of lower third of oesophagus and lower half of lower third of oesophagus.
The computer was used to derive time-activity curves for the four regions of interest, normalising for the size of each region. The oesophageal transit time was measured from the time of peak activity in the upper region to the time at which activity in the lowest region had fallen to 10%. Transit times up to 15 s were considered normal (Blackwell et al., 1983).
2. Capsule Transit Time The patients were positioned supine. Each patient swallowed a single small, hard gelatine capsule (Farillon-Lok Cap Size 2) filled with barium sulphate powder and the time taken by the capsule to traverse the oesophagus was measured by a stopwatch during fluoroscopy. If the capsule remained in the oesophagus at 30 s, its position was checked at 30 s intervals up to a total of 5 min. If the capsule remained in the oesophagus at 5 min, the patient was moved to the erect position and the capsule remnants were washed out with water. The capsule transit time was then repeated. Capsule transit times up to 15 s were considered normal (Channer and Virjee, 1982).
Estimations of radiation dosage. The average effective whole-body dose equivalent for the radionuclide studies was estimated to be 0.48 mSv and, for the capsule transit time, 0.07 mSv. 3. Barium Swallow Patients were positioned prone and asked to draw up into their mouth barium sulphate suspension (Baritop
100; Concept Pharmaceuticals) and then to swallow the barium with a single swallow. Oesophageal peristaltic activity Was observed during fluoroscopy and considered abnormal if the primary peristaltic wave did not completely clear the bolus from the oesophagus or if there were tertiary contractions. The radiologist (J.P.V.) performing the barium swallow was unaware of the result of the capsule and water transit times. A short medical history was taken from all patients. Results were analysed using non-parametric statistical methods. RESULTS
Table 1 lists some of the characteristics of the four patient groups. Symptomatic dysphagia was present in five of the eight patients with systemic sclerosis but in only one other, a diabetic patient without autonomic neuropathy.
Table 1 - Patient characteristics
Controls" Systemic (n=15) sclerosts (n=8)
Dmbetics with Other autonomic dtabetics neuropathy (n=24) (n=10)
Age
mean (years) range (years) Males Females Dysphagia Heartburn Acid regurgitation Flatulence Eplgastric pain (postprandial) Prior gastric or oesophageal surgery Concurrent medication known to influence oesophageal motility
47.5 20-80 10 5 0 7 5
60 50--73 0 8 5 4 6
59 41-72 7 3 0 3 1
49 25-72 18 6 1 5 6
10 9
3 6
0 2
13 4
2
0
1
0
0
1 (B blocker)
0
1 (B blocker)
Water Transit Times
Two tests of water transit time were performed on 52 patients. In two patients with systemic sclerosis, oesophageal motility was so disturbed that the radionuclide remained in the oesophagus even after the patients stood up and took further drinks, thus preventing a second study. In three patients data acquisition was lost or the patient took more than a single initial swallow, thus inhibiting the primary peristaltic wave and invalidating the transit time. Taking 15 s as the upper limit of normal, the coefficient of variation for the two water transit times was 17% and the correlation coefficient (Rank Spearman) was rs=0.68. Diabetics with autonomic neuropathy and patients with systemic sclerosis had significantly prolonged water transit times compared with the control group (Table 2).
OESOPHAGEAL FUNCTION TESTS Table 2 - Water transit times in each patient group
Group
Water transit ttmes Median (range) (s)
Controls (n=15) 9 5 (6.5-100) Diabetics without autonomic 10 (4.5-100) neuropathy (n=24) Diabetics with autonomic 85 (6.5-100) neuropathy (n=10) Systemic sclerosis (n=8) 100 (9.5-100)
Mean (SD) P*
(s) 23.3 (32.5) 35 0 (39.1) NS 64.8 (38 7) <0.00005 77.5 (33.3) <0.00005
* Mann-Whitney U-test compared with controls.
No difference was found between the water transit times of diabetic and non-diabetic control groups.
In seven patients with normal barium swallow, the water and capsule transit times were abnormally prolonged. One control patient had an abnormal barium swallow but normal water and capsule transit times. In seven patients the water transit time was normal but both barium swallow and capsule transit time were abnormal, whereas in no patient was the water transit time abnormal when the barium swallow and capsule transit times were normal. In one patient the capsule transit time was normal and the barium swallow and water transit times were abnormal, but in 11 patients the capsule transit time was abnormally prolonged despite normal water transit time and barium swallow. Table 4 - Number of patients in each group with abnormal oesophageal tests
Capsule Transit Times Paired capsule transit times were obtained in 55 patients. One patient with systemic sclerosis was unable to swallow a capsule and, therefore, measurement of a capsule transit time was not possible, while in one patient the first capsule swallowed became so adherent to the oesophageal mucosa that it could not be cleared to allow a second swallow. Taking the upper limit of normal to be 15 s, the coefficient of variation between the paired times was 13% with a correlation coefficient (Rank Spearman) of rs=0.38. Patients with systemic sclerosis and diabetic autonomic neuropathy had significantly longer capsule transit times than control diabetic and non-diabetic patients (Table 3). Table 3 - Capsule transit times in each patient group
Group
495
Capsule transit times Median (range)
Mean (SD)(s) P*
(s) Controls (n-15) 75 (8-300) 155.4 (142.8) Diabetics without autonomic 300 (6-300) 205.6 (131.7) NS neuropathy (n=24) Diabetics with autonomic 300 (7-300) 217 (89.1) <0.004 neuropathy (n= 10) Systemic sclerosis (n=7) 300 (10-300) 258.8 (104.7) <0.02 * Mann-Whitney U-test compared with controls.
Barium Swallow In six of the 15 control patients the prone barium swallow was reported to be abnormal compared wtih seven of the eight patients with systemic sclerosis (P<0.035, Exact test), nine of 10 diabetics with autonomic neuropathy (P<0.015, Exact test) and 11 of the 24 diabetic control patients (not significant). Additional information was reported on seven patients. Five patients had hiatus herniae (two patients with systemic sclerosis, one diabetic with autonomic neuropathy, one diabetic control and one non-diabetic control) and in two patients with systemic sclerosis, strictures were seen in the distal oesophagus.
Agreement Between Oesophageal Motility Tests In 23 (41%) patients all three oesophageal tests were abnormal and in five (9%), all were normal.
Group
Abnormal test Barium Water swallow (%) transu time (%)
Systemic sclerosis
7 (87.5)
8 (100)
Capsule transtt time (%)*
6 (85.7)
(n=8) (*n=7) Autonomic neuropaths (n=lO) Other diabetics (n=24) Controls
9 (90)
9 (90)
10 (100)
11 (46)
11 (46)
21 (87.5)
6 (40)
4 (27)
12 (80)
(n:15)
Table 4 shows the frequency of abnormal tests in each patient group. Significantly more diabetics without neuropathy had abnormal capsule transit times than had abnormal barium swallow (P<0.003) or water transit times (P<0.003). More control patients had abnormally prolonged capsule transit times than water transit times (P<0.005). No significant difference was found between the frequency of abnormality of barium swallow, water and capsule transit times in the patients with systemic sclerosis or autonomic neuropathy. DISCUSSION In this study we have used three tests of oesophageal function to assess motility in unselected control patients and patients in whom motility disorders are common. The purpose of this study was, firstly, to establish the value of oesophageal function tests in detecting oesophageal dysmotility in comparison with barium swallow and, secondly, to compare water transit time derived from oesophageal scintigraphy with fluoroscopically visualised capsule transit time. In previous studies we used hard gelatin barium-filled capsules to demonstrate the effect of posture and drink volume on oesophageal transit (Channer and Virjee, 1982). It is clear that only in the supine position when the capsule is swallowed with a small volume of fluid is intrinsic oesophageal peristalsis an important factor in capsule transit. We have used capsule transit time to study the effect of intravenous hyoscine butylbromide (Channer et al., 1983a) and glucagon (Channer et al., 1983b) on oesophageal function. These two drugs were
496
CLINICAL RADIOLOGY
chosen for study because they are widely used as hypotonic agents in gastrointestinal radiology. We found that both drugs prolonged capsule transit time in patients with normal peristalsis. In the case of hyoscine butylbromide, the delay was marked when capsules were swallowed in both the erect and the supine positions with small and large volumes of water; however, when glucagon was given, delayed transit occurred only in the supine position when the capsule was swallowed with a small volume of water. The importance of this latter result is that, by this technique, we have demonstrated that glucagon has an inhibitory effect on peristalsis, which had not been shown previously by the standard investigative methods of barium swallow and manometry (Hogan et al., 1975). This suggests that capsule transit time is a more sensitive test of oesophageal peristalsis. Anatomical distortion of the oesophagus also has an effect on capsule transit. In a study in patients with mitral valve disease, the area of oesophageal indentation caused by left atrial enlargement correlated with capsule transit time in the supine position (Channer et al., 1984). Bolus transit times obtained by radionuclide imaging have been shown to be sensitive indicators of oesophageal motility. In two studies (Russell et al., 1981; Blackwelt et al., 1983) in which manometry and scintigraphy were performed, water transit times were abnormal more often than manometry in symptomatic patients, which the authors conclude may be due to the increased sensitivity of water transit time, although it is possible that these prolonged water transit times were false positives. In the present study we have shown that both capsule and water transit times are as effective as barium swallow in detecting motility disorders associated with systemic sclerosis and diabetic autonomic neuropathy. The fact that in only one case were both oesophageal function tests normal when the barium swallow was abnormal shows that these tests are at least as sensitive as barium swallow in detecting oesophageal dysmotility. In one patient with systemic sclerosis, barium swallow was reported to be normal and yet both transit times were prolonged. This would suggest that motility was, in fact, impaired in this patient. Overall, barium swallow was normal in seven patients in whom both transit times were abnormally prolonged. Capsule transit time was abnormal more frequently than either water transit time or barium swallow, but was normal in only one case when both water transit time and barium swallow were abnormal. This may suggest that capsule transit time is more sensitive than water transit time, although the possibility of these being false positive results cannot be excluded. Measurement of capsule transit time has three advantages over using water transit time. Firstly, the radiation exposure during fluoroscopic capsule transit was about a seventh of that estimated during oesophageal scintigraphy. Secondly, during fluoroscopic transit studies the whole of the oesophagus can be visualised initially and the position of capsule hold-up readily assessed by anatomical markers. Scintigraphy, on the other hand, relies on examination of areas of interest which cannot,
with absolute certainty, be related to anatomical markers. Thirdly, the ability to perform oesophageal scintigraphy depends on access to sophisticated and expensive gamma cameras and computers, whereas fluoroscopic capsule transit time study is available to all with screening facilities and a stopwatch! Do oesophageal function tests have a role in clinical practice? They cannot replace barium swallow in the detection of structural obstructive lesions in the oesophagus. In this study, additional information from barium swallow was available in seven patients, including the presence of strictures in two cases which was not detected by either transit time. Only 11% of patients in this study had dysphagia, yet all three oesophageal tests were abnormal in 23 (40%) patients. This high incidence of asymptomatic motility abnormality makes the clinical value of abnormal oesophageal function dubious. Oesophageal function tests would, however, appear to be sensitive screening tests in patients in whom oesophageal dysmotility is of concern and are simple, quick, cheap and safe compared with manometry.
Acknowledgements. We would hke to thank Mr P. Jackson (Medical Physicist) for his help with the oesophageal scintlscans and Drs Coles, Corrall and Dieppe for allowing us to study patients under their care Nicola Eberle kindly typed the manuscript
REFERENCES Blackwell, J N., Hannan, W J., Adam, R. D & Heading, R. C. (1982). A radionuclide techmque for assessing oesophageal function. Nuclear Medicine Communications, 3, 291-296. Blackwell, J. N., Hannan, W. J., Adam, R. D. & Heading, R. C. (1983). Radionuchde transit studms in the detection of oesophageal dysmotihty Gut, 24, 421-426. Bosch, A., Dietrick, R., Lanaro, A. E. & Frias, Z. (1977) Modified scintigraphic technique for the dynamic study of the esophagus International Journal of Nuclear Medicine and Biololgy, 4, 195199 Channer, K. S. & Vlrjee, J. (1982). The effect of posture and drink volume on the swallowing of capsules. British Medical Journal, 285, 1702. Channer, K. S., Wolinski, A., Kaye, B. & Virlce, J. (1983a). The effect of hyoscine butylbromide on the swallowing of capsules. Brlush Journal of Clinical Pharmacology, 15,560-563. Channer, K. S., Bolton, R., AI-Hilli, S., Nakielny, R. & Virjee, J. P. (1983b). The effect of glucagon on the swallowing of capsules British Journal of Clinical Pharmacology, 16, 456-460. Channer, K. S., Bell, J. & Viriee, J. (1984). The effect of left atrial size on capsule oesophageal transit. Brittsh Heart Journal, 52,223-227 Fisher, R. S , Malmud, L. S., Applegate, G., Rock, E. & Lorber, S. H. (1982). Effect of bolus composition on esophageal transit: concise communication.7ournalof Nuclear Medicine, 23,878-882. Hogan, W, J., Dodds, W. J. & Hoke, S. E. (1975). Effect of glucagon on oesophageal motor function. Gastroenterology, 69, 160-165. Kazem, I. (1972). A new scmtigraphic technique for the study of the oesophagus. American Journal of Roentgenology, 155, 681-688. Malmud, L. S & Fisher, R. S. (1982). Radionuclide studies of esophageal transit and gastroesophageal reflux. Seminars in Nuclear Medicine, 12, 104-115. Ott, D. J., Wu, W. C. & Gelfand, D. W. (1981). Efficacy of radiology of the oesophagus for evaluation of dysphagia. Gastrointestinal Radtology, 6, 109-110. Russell, C. O. H., Hill, L. D., Holmes, E. R., Hull, D. A., Gannon, R. & Pope, C. E. (1981). Radlonuclide transit: a sensmve screening test for esophageal dysfunction. Gastroenterology, 80, 887-892 Tolin, R. D., Malmud, L. S., Reilley, J. & Fisher, R. S. (1979). Esophageal scintigraphy to quantitate esophageal transit. Gastroenterology, 76, 1402-1408.
000%9260/85/447493502.00
Oesophageal Function Tests: Are They of Value? K. S. CHANNER and J. P. VIRJEE
Departments of Cardiology and Radiodiagnosis, Bristol Royal Infirmary, Bristol
Oesophageal function was assessed in 15 unselected control patients, eight patients with systemic sclerosis, 10 diabetics with autonomic neuropathy and 24 diabetic controls, by water bolus transit time derived from oesophageal scintigraphy, barium swallow and by timing the transit of a barium sulphate-filled capsule through the oesophagus. Water transit times and capsule transit times were significantly prolonged in patients with systemic sclerosis and diabetics with autonomic neuropathy, compared with controls. Barium swallow was abnormal in seven of eight patients with systemic sclerosis, whereas water transit time was abnormal in all eight and capsule transit time was abnormal in six of seven. Nine of 10 patients with diabetic autonomic neuropathy had abnormal barium swallows and water transit times but all 10 had prolonged capsule transit times. Eleven of 24 diabetic controls had abnormal barium swallows and water transit times, but 21 had abnormally prolonged capsule transit times. Six of 15 controls had abnormal barium swallows, four had abnormal water transit times and 12 had abnormal capsule transit times. In conclusion, water and capsule transit times are sensitive tests of oesophageal function and are as effective as barium swallow in detecting oesophageal motility disorders associated with systemic sclerosis and diabetic autonomic neuropathy. Capsule transit time is cheaper, involves a smaller radiation exposure than oesophageal scintigraphy and may be more sensitive.
The function of the oesophagus is to transmit ingested material from the oropharynx to the stomach. Far from being a simple tube, the oesophagus has a complex muscular arrangement and nervous supply and is the part of the gastrointestinal tract where somatic and autonomic nerves interplay most in the control of peristalsis. Oesophageal function may be affected by structural obstructions or by disturbances in motility. Barium swallow and endoscopy are highly sensitive ways of recognising structural abnormalities, but may be insensitive methods of assessing oesophageal motor function (Ott et al., 1981). In the last 20 years much attention has centred on oesophageal manometry as a diagnostic tool in the evaluation of the motor function of the oesophagus. However, this is an unpleasant procedure for many patients and suffers from problems of interpretation. It is time-consuming and labourintensive. More recently, oesophageal function tests have been developed which rely on timing the transit of a bolus through the oesophagus. Disorders of oesophageal peristalsis result in prolonged oesophageal transit times. In earlier studies the transit of radio-labelled water (Kazem, 1972; Tolin et al., 1979; Russell et al., 1981; Blackwell et al., 1982; Malmud and Fisher, 1982; Black44
well et al., 1983), test meals (Bosch et al., 1977), gelatin or capsules (Fisher et al., 1982) through the oesophagus was measured using gamma cameras with on-line computer facilities, so that time-activity curves could be plotted and transit times derived from these. We have developed an analogous technique: using a stopwatch, we have timed the oesophageal transit of barium sulphate-filled capsules swallowed whilst screening with X-ray fluoroscopy. In earlier studies we established the effect of posture and drink volume (Channer and Virjee, 1982) on oesophageal transit time and used this technique to study the effect of hyoscine butylbromide (Charmer et al., 1983a) and glucagon (Channer et al., 1983b) on oesophageal function. In this present study we have compared two tests of oesophageal function, oesophageal scintigraphy and timed capsule transit, with barium swallow in the investigation of oesophageal motility disorders in patients with systemic sclerosis and diabetes mellitus.
PATIENTS AND METHODS This study was approved by the Ethics Committee of the Bristol and Weston Health Authority. Patients
Fifty-seven patients took part in the study. They comprised four groups. 1. Controls Fifteen patients attending the X-ray Department for planned barium meal examinations. 2. Systemic Sclerosis Eight patients with established systemic sclerosis (six with calcinosis, Raynaud's sclerodactyly and telangiectasia (CRST) syndrome). 3. Diabetics With Autonomic Neuropathy Ten diabetics (nine on insulin) previously found to have autonomic neuropathy on the basis of abnormal heart rate responses to physiological stimuli. 4. Diabetic Controls Twenty-four patients (21 on insulin) shown not to have autonomic neuropathy. Methods
Three tests of oesophageal motor function were performed.
494
CLINICAL RADIOLOGY
1. Oesophageal Scintiscan (Water Transit Time) Patients were positioned supine below a gamma camera (Maxicamera 400T) fitted with a low-energy parallel hole collimator. The images were obtained using an on-line data processor. The position of the upper oesophagus was established using a radioactive marker, temporarily placed over the cricoid cartilage, and the field of view was adjusted to include the entire oesophagus and gastric fundus. Water (10 ml) was placed in the patient's mouth from a syringe. The patient was asked to swallow once only and to take repeated single swallows at 30 s intervals for 2 min. Trial runs were performed until the patients could co-operate. Radionuclide images were obtained when the patient swallowed 10 ml water containing 20 MBq 99mTc sulphur colloid. Two hundred images were recorded at 0.5 s intervals and stored for analysis. The position of the upper oesophagus was again confirmed by static imaging for 10-20 s with a radioactive cricoid marker. This procedure was repeated. The data were analysed as follows. A composite image by the summation of frames provided a single image of the complete oesophagus and stomach. The position of the radioactive marker was used to define the upper end of the oesophagus and activity in the gastric fundus defined the lower end of the oesophagus. The image of the oesophagus was then divided into four regions of interest: (a) (b) (c) (d)
upper third of oesophagus, mid-third oesophagus, upper half of lower third of oesophagus and lower half of lower third of oesophagus.
The computer was used to derive time-activity curves for the four regions of interest, normalising for the size of each region. The oesophageal transit time was measured from the time of peak activity in the upper region to the time at which activity in the lowest region had fallen to 10%. Transit times up to 15 s were considered normal (Blackwell et al., 1983).
2. Capsule Transit Time The patients were positioned supine. Each patient swallowed a single small, hard gelatine capsule (Farillon-Lok Cap Size 2) filled with barium sulphate powder and the time taken by the capsule to traverse the oesophagus was measured by a stopwatch during fluoroscopy. If the capsule remained in the oesophagus at 30 s, its position was checked at 30 s intervals up to a total of 5 min. If the capsule remained in the oesophagus at 5 min, the patient was moved to the erect position and the capsule remnants were washed out with water. The capsule transit time was then repeated. Capsule transit times up to 15 s were considered normal (Channer and Virjee, 1982).
Estimations of radiation dosage. The average effective whole-body dose equivalent for the radionuclide studies was estimated to be 0.48 mSv and, for the capsule transit time, 0.07 mSv. 3. Barium Swallow Patients were positioned prone and asked to draw up into their mouth barium sulphate suspension (Baritop
100; Concept Pharmaceuticals) and then to swallow the barium with a single swallow. Oesophageal peristaltic activity Was observed during fluoroscopy and considered abnormal if the primary peristaltic wave did not completely clear the bolus from the oesophagus or if there were tertiary contractions. The radiologist (J.P.V.) performing the barium swallow was unaware of the result of the capsule and water transit times. A short medical history was taken from all patients. Results were analysed using non-parametric statistical methods. RESULTS
Table 1 lists some of the characteristics of the four patient groups. Symptomatic dysphagia was present in five of the eight patients with systemic sclerosis but in only one other, a diabetic patient without autonomic neuropathy.
Table 1 - Patient characteristics
Controls" Systemic (n=15) sclerosts (n=8)
Dmbetics with Other autonomic dtabetics neuropathy (n=24) (n=10)
Age
mean (years) range (years) Males Females Dysphagia Heartburn Acid regurgitation Flatulence Eplgastric pain (postprandial) Prior gastric or oesophageal surgery Concurrent medication known to influence oesophageal motility
47.5 20-80 10 5 0 7 5
60 50--73 0 8 5 4 6
59 41-72 7 3 0 3 1
49 25-72 18 6 1 5 6
10 9
3 6
0 2
13 4
2
0
1
0
0
1 (B blocker)
0
1 (B blocker)
Water Transit Times
Two tests of water transit time were performed on 52 patients. In two patients with systemic sclerosis, oesophageal motility was so disturbed that the radionuclide remained in the oesophagus even after the patients stood up and took further drinks, thus preventing a second study. In three patients data acquisition was lost or the patient took more than a single initial swallow, thus inhibiting the primary peristaltic wave and invalidating the transit time. Taking 15 s as the upper limit of normal, the coefficient of variation for the two water transit times was 17% and the correlation coefficient (Rank Spearman) was rs=0.68. Diabetics with autonomic neuropathy and patients with systemic sclerosis had significantly prolonged water transit times compared with the control group (Table 2).
OESOPHAGEAL FUNCTION TESTS Table 2 - Water transit times in each patient group
Group
Water transit ttmes Median (range) (s)
Controls (n=15) 9 5 (6.5-100) Diabetics without autonomic 10 (4.5-100) neuropathy (n=24) Diabetics with autonomic 85 (6.5-100) neuropathy (n=10) Systemic sclerosis (n=8) 100 (9.5-100)
Mean (SD) P*
(s) 23.3 (32.5) 35 0 (39.1) NS 64.8 (38 7) <0.00005 77.5 (33.3) <0.00005
* Mann-Whitney U-test compared with controls.
No difference was found between the water transit times of diabetic and non-diabetic control groups.
In seven patients with normal barium swallow, the water and capsule transit times were abnormally prolonged. One control patient had an abnormal barium swallow but normal water and capsule transit times. In seven patients the water transit time was normal but both barium swallow and capsule transit time were abnormal, whereas in no patient was the water transit time abnormal when the barium swallow and capsule transit times were normal. In one patient the capsule transit time was normal and the barium swallow and water transit times were abnormal, but in 11 patients the capsule transit time was abnormally prolonged despite normal water transit time and barium swallow. Table 4 - Number of patients in each group with abnormal oesophageal tests
Capsule Transit Times Paired capsule transit times were obtained in 55 patients. One patient with systemic sclerosis was unable to swallow a capsule and, therefore, measurement of a capsule transit time was not possible, while in one patient the first capsule swallowed became so adherent to the oesophageal mucosa that it could not be cleared to allow a second swallow. Taking the upper limit of normal to be 15 s, the coefficient of variation between the paired times was 13% with a correlation coefficient (Rank Spearman) of rs=0.38. Patients with systemic sclerosis and diabetic autonomic neuropathy had significantly longer capsule transit times than control diabetic and non-diabetic patients (Table 3). Table 3 - Capsule transit times in each patient group
Group
495
Capsule transit times Median (range)
Mean (SD)(s) P*
(s) Controls (n-15) 75 (8-300) 155.4 (142.8) Diabetics without autonomic 300 (6-300) 205.6 (131.7) NS neuropathy (n=24) Diabetics with autonomic 300 (7-300) 217 (89.1) <0.004 neuropathy (n= 10) Systemic sclerosis (n=7) 300 (10-300) 258.8 (104.7) <0.02 * Mann-Whitney U-test compared with controls.
Barium Swallow In six of the 15 control patients the prone barium swallow was reported to be abnormal compared wtih seven of the eight patients with systemic sclerosis (P<0.035, Exact test), nine of 10 diabetics with autonomic neuropathy (P<0.015, Exact test) and 11 of the 24 diabetic control patients (not significant). Additional information was reported on seven patients. Five patients had hiatus herniae (two patients with systemic sclerosis, one diabetic with autonomic neuropathy, one diabetic control and one non-diabetic control) and in two patients with systemic sclerosis, strictures were seen in the distal oesophagus.
Agreement Between Oesophageal Motility Tests In 23 (41%) patients all three oesophageal tests were abnormal and in five (9%), all were normal.
Group
Abnormal test Barium Water swallow (%) transu time (%)
Systemic sclerosis
7 (87.5)
8 (100)
Capsule transtt time (%)*
6 (85.7)
(n=8) (*n=7) Autonomic neuropaths (n=lO) Other diabetics (n=24) Controls
9 (90)
9 (90)
10 (100)
11 (46)
11 (46)
21 (87.5)
6 (40)
4 (27)
12 (80)
(n:15)
Table 4 shows the frequency of abnormal tests in each patient group. Significantly more diabetics without neuropathy had abnormal capsule transit times than had abnormal barium swallow (P<0.003) or water transit times (P<0.003). More control patients had abnormally prolonged capsule transit times than water transit times (P<0.005). No significant difference was found between the frequency of abnormality of barium swallow, water and capsule transit times in the patients with systemic sclerosis or autonomic neuropathy. DISCUSSION In this study we have used three tests of oesophageal function to assess motility in unselected control patients and patients in whom motility disorders are common. The purpose of this study was, firstly, to establish the value of oesophageal function tests in detecting oesophageal dysmotility in comparison with barium swallow and, secondly, to compare water transit time derived from oesophageal scintigraphy with fluoroscopically visualised capsule transit time. In previous studies we used hard gelatin barium-filled capsules to demonstrate the effect of posture and drink volume on oesophageal transit (Channer and Virjee, 1982). It is clear that only in the supine position when the capsule is swallowed with a small volume of fluid is intrinsic oesophageal peristalsis an important factor in capsule transit. We have used capsule transit time to study the effect of intravenous hyoscine butylbromide (Channer et al., 1983a) and glucagon (Channer et al., 1983b) on oesophageal function. These two drugs were
496
CLINICAL RADIOLOGY
chosen for study because they are widely used as hypotonic agents in gastrointestinal radiology. We found that both drugs prolonged capsule transit time in patients with normal peristalsis. In the case of hyoscine butylbromide, the delay was marked when capsules were swallowed in both the erect and the supine positions with small and large volumes of water; however, when glucagon was given, delayed transit occurred only in the supine position when the capsule was swallowed with a small volume of water. The importance of this latter result is that, by this technique, we have demonstrated that glucagon has an inhibitory effect on peristalsis, which had not been shown previously by the standard investigative methods of barium swallow and manometry (Hogan et al., 1975). This suggests that capsule transit time is a more sensitive test of oesophageal peristalsis. Anatomical distortion of the oesophagus also has an effect on capsule transit. In a study in patients with mitral valve disease, the area of oesophageal indentation caused by left atrial enlargement correlated with capsule transit time in the supine position (Channer et al., 1984). Bolus transit times obtained by radionuclide imaging have been shown to be sensitive indicators of oesophageal motility. In two studies (Russell et al., 1981; Blackwelt et al., 1983) in which manometry and scintigraphy were performed, water transit times were abnormal more often than manometry in symptomatic patients, which the authors conclude may be due to the increased sensitivity of water transit time, although it is possible that these prolonged water transit times were false positives. In the present study we have shown that both capsule and water transit times are as effective as barium swallow in detecting motility disorders associated with systemic sclerosis and diabetic autonomic neuropathy. The fact that in only one case were both oesophageal function tests normal when the barium swallow was abnormal shows that these tests are at least as sensitive as barium swallow in detecting oesophageal dysmotility. In one patient with systemic sclerosis, barium swallow was reported to be normal and yet both transit times were prolonged. This would suggest that motility was, in fact, impaired in this patient. Overall, barium swallow was normal in seven patients in whom both transit times were abnormally prolonged. Capsule transit time was abnormal more frequently than either water transit time or barium swallow, but was normal in only one case when both water transit time and barium swallow were abnormal. This may suggest that capsule transit time is more sensitive than water transit time, although the possibility of these being false positive results cannot be excluded. Measurement of capsule transit time has three advantages over using water transit time. Firstly, the radiation exposure during fluoroscopic capsule transit was about a seventh of that estimated during oesophageal scintigraphy. Secondly, during fluoroscopic transit studies the whole of the oesophagus can be visualised initially and the position of capsule hold-up readily assessed by anatomical markers. Scintigraphy, on the other hand, relies on examination of areas of interest which cannot,
with absolute certainty, be related to anatomical markers. Thirdly, the ability to perform oesophageal scintigraphy depends on access to sophisticated and expensive gamma cameras and computers, whereas fluoroscopic capsule transit time study is available to all with screening facilities and a stopwatch! Do oesophageal function tests have a role in clinical practice? They cannot replace barium swallow in the detection of structural obstructive lesions in the oesophagus. In this study, additional information from barium swallow was available in seven patients, including the presence of strictures in two cases which was not detected by either transit time. Only 11% of patients in this study had dysphagia, yet all three oesophageal tests were abnormal in 23 (40%) patients. This high incidence of asymptomatic motility abnormality makes the clinical value of abnormal oesophageal function dubious. Oesophageal function tests would, however, appear to be sensitive screening tests in patients in whom oesophageal dysmotility is of concern and are simple, quick, cheap and safe compared with manometry.
Acknowledgements. We would hke to thank Mr P. Jackson (Medical Physicist) for his help with the oesophageal scintlscans and Drs Coles, Corrall and Dieppe for allowing us to study patients under their care Nicola Eberle kindly typed the manuscript
REFERENCES Blackwell, J N., Hannan, W J., Adam, R. D & Heading, R. C. (1982). A radionuclide techmque for assessing oesophageal function. Nuclear Medicine Communications, 3, 291-296. Blackwell, J. N., Hannan, W. J., Adam, R. D. & Heading, R. C. (1983). Radionuchde transit studms in the detection of oesophageal dysmotihty Gut, 24, 421-426. Bosch, A., Dietrick, R., Lanaro, A. E. & Frias, Z. (1977) Modified scintigraphic technique for the dynamic study of the esophagus International Journal of Nuclear Medicine and Biololgy, 4, 195199 Channer, K. S. & Vlrjee, J. (1982). The effect of posture and drink volume on the swallowing of capsules. British Medical Journal, 285, 1702. Channer, K. S., Wolinski, A., Kaye, B. & Virlce, J. (1983a). The effect of hyoscine butylbromide on the swallowing of capsules. Brlush Journal of Clinical Pharmacology, 15,560-563. Channer, K. S., Bolton, R., AI-Hilli, S., Nakielny, R. & Virjee, J. P. (1983b). The effect of glucagon on the swallowing of capsules British Journal of Clinical Pharmacology, 16, 456-460. Channer, K. S., Bell, J. & Viriee, J. (1984). The effect of left atrial size on capsule oesophageal transit. Brittsh Heart Journal, 52,223-227 Fisher, R. S , Malmud, L. S., Applegate, G., Rock, E. & Lorber, S. H. (1982). Effect of bolus composition on esophageal transit: concise communication.7ournalof Nuclear Medicine, 23,878-882. Hogan, W, J., Dodds, W. J. & Hoke, S. E. (1975). Effect of glucagon on oesophageal motor function. Gastroenterology, 69, 160-165. Kazem, I. (1972). A new scmtigraphic technique for the study of the oesophagus. American Journal of Roentgenology, 155, 681-688. Malmud, L. S & Fisher, R. S. (1982). Radionuclide studies of esophageal transit and gastroesophageal reflux. Seminars in Nuclear Medicine, 12, 104-115. Ott, D. J., Wu, W. C. & Gelfand, D. W. (1981). Efficacy of radiology of the oesophagus for evaluation of dysphagia. Gastrointestinal Radtology, 6, 109-110. Russell, C. O. H., Hill, L. D., Holmes, E. R., Hull, D. A., Gannon, R. & Pope, C. E. (1981). Radlonuclide transit: a sensmve screening test for esophageal dysfunction. Gastroenterology, 80, 887-892 Tolin, R. D., Malmud, L. S., Reilley, J. & Fisher, R. S. (1979). Esophageal scintigraphy to quantitate esophageal transit. Gastroenterology, 76, 1402-1408.