- Email: [email protected]
Comparative studies of esophageal function in systemic sclerosis
GASTROENTEROLOGY
1992;102:1551-1556
Comparative Studies of Esophageal Function in Systemic Sclerosis HERBERT A. KLEIN, ARNOLD WALD, TOBY 0. GRAHAM, WILLIAM L. CAMPBELL, and VIRGINIA D. STEEN Departments Pennsylvania
of Radiology and Medicine,
University
Three modalities for assessing esophageal dysfunction in patients with systemic sclerosis were prospectively compared. Seventeen patients underwent (a) esophageal manometry with measurement of distal esophageal peak contraction pressure amplitude, percentage of peristaltic waves, and lower esophageal sphincter pressure; (h) cine-esophagography with scoring based on residual contrast and the character of visualized waves; and (c) esophageal transit scintigraphy with quantification of residual swallowed tracer. Highly significant correlations were found between scintigraphic residual and tine-esophagography score, between scintigraphic residual and manometric amplitude, and indeed between all pairs of measured esophageal function parameters except those involving lower esophageal sphincter pressure. In addition, scintigraphy and tine-esophagography showed comparable ability to discriminate between patients with abnormal and normal esophageal motor function. Symptoms did not significantly correlate with quantitative parameters, nor did they have diagnostic discriminating ability. Induction of Raynaud’s phenomenon in a subgroup of patients had no detectable effect on esophageal function. It was concluded that these three diagnostic modalities are approximately equivalent in their ability to detect esophageal dysmotility in systemic sclerosis and measure its severity.
S
ystemic sclerosis is a generalized connective tissue disorder characterized by fibrosis and degenerative changes of the skin and some viscera.’ Atrophy and fibrosis of smooth muscle are probably responsible for the impaired esophageal motility that is common in this disease. Moreover, diminished lower esophageal sphincter pressure and associated gastroesophageal reflux occur frequently and can lead to esophagitis and stricture, complications that may be progressive and difficult to treat.‘s3 In the early stages of the disease, however, patients are of-
of Pittsburgh
School of Medicine,
Pittsburgh,
ten asymptomatic despite abnormal esophageal motor function.4-6 Raynaud’s phenomenon is a hallmark of systemic sclerosis. The hypothesis that it plays a role in organ system abnormalities has been based on cold-induced changes in the heart,7 kidneys,’ and lungs9 due to visceral vasoconstriction accompanying peripheral vasoconstriction. Previous work suggests that cold-induced vasospasm in the esophagus is associated with delayed return of that organ to baseline temperatures,” giving rise to speculation that episodes of Raynaud’s phenomenon may worsen esophageal function. Cine-esophagography is useful in the evaluation of esophageal function in most symptomatic patients with established systemic sclerosis,“~‘2 but esophageal manometry is considered a more sensitive technique to detect motor abnormalities at earlier stages.4-6 Esophageal transit scintigraphy, which monitors the movement of swallowed radioactive material, is another diagnostic modality that has been used to evaluate esophageal motor function. Previous investigators found scintigraphy to be a sensitive test in systemic sclerosis.13-17 The aim of this study was to prospectively compare measurements of esophageal function in systemic sclerosis using manometry, tine-esophagography, and scintigraphy. In a subgroup of patients, we also tested the hypothesis that inducing Raynaud’s phenomenon would adversely affect esophageal motor function. Materials and Methods Patient and Control Groups The patient group comprised 5 men and 12 women who met the American College of Rheumatology criteria for systemic sclerosis.‘8 Their mean age was 53 years (range, 32-69 years). Nine patients had an illness pattern typical of limited scleroderma, and 8 had diffuse sclero-
0 1992 by the American
Gastroenterological
0016-5065/92/$3.00
Association
1552
GASTROENTEROLOGY Vol. 102, No. 5
KLEIN ET AL.
derma.lg Two additional patients who could not tolerate esophageal manometry participated in evaluation of the reproducibility of scintigraphy. All patients had Raynaud’s phenomenon. None had esophageal strictures. The recruitment effort extended to all eligible patients of our institution’s division of rheumatology and the local systemic sclerosis support group. The patient group consisted of those patients consenting to be evaluated during the period of the investigation; illness pattern and severity were not applied to the selection process. As part of the evaluation, patients were questioned about the presence of dysphagia, odynophagia, heartburn, and regurgitation during the preceding month. Using the method of de Dombal and Hall,” each symptom was assessed on a severity scale (0, absent; 1, mild; 2, moderate; 3, severe) and a frequency scale (1, two to three times per month; 2, once per week; 3, two to three times per week; 4, daily). A total score (severity X frequency) was obtained for each symptom (maximum of 12 for each symptom, 48 for all symptoms). A control group (14 men, 6 women; mean age, 26 years; range, 17-53 years) underwent esophageal manometry. They were healthy volunteers with no symptoms or history of esophageal disease, central nervous system disorders, peripheral neuropathy, diabetes mellitus, systemic disorders known to be associated with esophageal dysfunction, or recent ingestion of medication known to alter esophageal function. The study was approved by the institutional committees for the protection of human subjects. Subjects gave informed written consent before participation.
Esophageal
Manometry
Manometric recordings were obtained using an eight-lumen polyvinyl catheter (diameter, 4.5 mm) consisting of three proximal lateral orifices (diameter, 0.8 mm) located 5 cm apart and five distal orifices that were axially oriented at 60’ and were 1 cm apart (Arndorfer Specialties, Inc., Greendale, WI). The catheter was continuously perfused using a low-compliance pneumohydraulic capillary infusion system (model CH8; Arndorfer Specialties, Inc.) at a rate of 0.5 mL/min. With this system, pressure increased at a rate >400 mm Hg/s upon occlusion of the orifice. Pressures were transmitted to external pressure transducers (model 223651; SensorMedics Corp., Anaheim, CA) and recorded on an eight-channel recorder (SensorMedics model R611). The recording system was calibrated before each study. All manometric studies were performed with the patient in the supine position and the catheter passed orally or nasally. Resting lower esophageal sphincter pressure (LESP) was measured at end expiration using the station pull-through technique. Pressure waves in the esophagus were obtained by studying 10 wet swallows of 2-mL boluses of water. From the 10 observations, we determined amplitude (mean peak contraction pressure amplitude in the distal esophagus) and the percentage of waves that were peristaltic. In 9 patients, manometry was followed by induction of Raynaud’s phenomenon, either by applying an ice collar to
the patient’s neck or by immersing the patient’s hand in ice water for several minutes, until a maximum effect was obtained or until the patient could no longer tolerate the cold. This was followed by repeat manometry.
Cine-esophagography Cine-esophagography was performed using three to five swallows of 50% (wt/vol) barium sulfate mixture while patients were in the prone right anterior oblique position. Each swallow was recorded on 16-mm cine-radiographic film at 12 frames/s or on videotape. Spot film radiographs of the esophagus were taken as indicated, and the presence or absence of spontaneous gastroesophageal reflux was recorded. Esophageal function was rated in each case using a tine-esophagography score (CES) as follows: 0, normal, no significant residual contrast material; 1, some abnormal contraction waves (less than half) and/ or residual contrast filling approximately one third of the length of the thoracic esophagus; 2, most waves abnormal and/or residual barium filling approximately two thirds of the thoracic esophagus; 3, all waves abnormal with residual barium filling all of the thoracic esophagus; 4, no peristaltic waves seen.
Esophageal
Transit Scintigraphy
Scintigraphy was performed as previously described using posterior imaging with a Raytheon Step l/ Step 2 scintillation camera (Raytheon Medical Systems, Melrose Park, IL).21*22Supine subjects swallowed 0.3 mCi (11.1 MBq) of technetium 99m sulfur colloid in 15 mL of water. A dry swallow was performed after 30 seconds and at l5-second intervals thereafter for a total of 10.75 minutes with no further administration of liquid. Using a Siemens Microdelta computer system (Siemens Medical Systems, Des Plaines, IL) including a VAX 11/750 computer (Digital Equipment Co., Maynard, MA), analysis of curves of radioactivity against time in the esophageal region of interest yielded a fourth-swallow residual fraction (RF4), expressed as a percentage, that served as our quantitative parameter for scintigraphy. This was the ratio of the residual counting rate after completion of the fourth swallow to the counting rate of the entire bolus observed immediately after its administration, We analyzed RF4 because our previous study of normal subjects showed high intrasubject and intersubject variability of the first-swallow residual fraction, with the variability diminishing considerably by the fourth swallow.23 In 9 patients, Raynaud’s phenomenon was induced before scintigraphy in the same manner as before manometry. This was done the day after baseline scintigraphy in 4 cases and the day before in 5 cases. The intervening day prevented residual radioactivity from interfering with the second measurement. All tests were performed during a 3-day period in our Clinical Research Center. Medications that could affect esophageal motor function were withheld for at least 48 hours before testing. The esophageal function parameters, i.e., the scores obtained from manometry, cine-esophagography, and scintigraphy, were each analyzed separately and without knowledge of the other findings.
May
ESOPHAGUS
1992
Statistical
Analysis
The following statistical procedures were used as appropriate: organization of symptom scores as either dichotomized or continuous variables; paired and unpaired Student’s t tests; Spearman rank correlation, ranking each parameter from most to least abnormal; biserial correlation; x2 testing; receiver operating characteristic curve analysis; and two methods of multivariate analysis, Hotelling’s T2 24.25 and Bonferroni’s adjustment.26 Results We found no significant difference between patients with limited and diffuse scleroderma with regard to esophageal function parameters and symptom scores; therefore, these subgroups were combined for further analysis. Among the 17 patients, the most common symptom was heartburn, affecting 14 and registering the highest mean score (5.9). The least common symptom was odynophagia, affecting two patients and registering the lowest mean score (2.8). Eleven patients had dysphagia, and 8 had regurgitation. One patient had no symptoms, and 2 had all four. The highest individual composite score was 27. No significant correlation was found among the scores for different symptoms. None of the five esophageal function parameters (amplitude, percentage of peristaltic waves, LESP, RF4, and CES) significantly correlated with any symptom or the composite symptom score. A subgroup of patients was defined as manometrically abnormal because their results were below the lower limit of the control group for amplitude (25 mm Hg) and/or percentage of peristaltic waves (900/,). Their individual symptom scores and composite scores were not significantly worse than those of manometrically normal patients. Results of esophageal manometry for the patients and the manometry control group appear in Figure 1. The scores of the patient group were significantly lower than those of the control group for amplitude, percentage of peristaltic waves, and LESP. In the patient group, we found a highly significant correlation between RF4 and CES (Figure 2) and between CES and percentage of peristaltic waves (P = 0.001). Significant correlations were also found between all the other possible pairings of the following parameters: RF4, CES, percentage of peristaltic waves, and amplitude (P < 0.0005 for all relationships). However, significance was not found for the relationships between LESP and the four other parameters. RF4 and CES were similar to each other in their degree of correlation with percentage of peristaltic waves and amplitude. For all patients, RF4 was 48.5% + 9.0% and CES was 2.6 * 1.3 (mean f SEM). After dividing the pa-
IN SYSTEMIC SCLEROSIS
1553
tients into manometrically normal and abnormal subgroups by esophageal manometry, we found RF4 and CES to be significantly higher in the manometritally abnormal subgroup (Figure 3). Both parameters met the statistical criterion for significant diagnostic discriminating ability,27 and there was no significant difference between them. Using optimum cut-off values for defining positivity of the tests, i.e., RF4 > 19.3% and CES > 2, the segregation of patients into positive and negative was identical for both tests (Figure 2), with sensitivity of 81.8%, specificity of 83.3%, and overall accuracy of 82.4%. These results indicate moderately high diagnostic discriminating ability for both scintigraphy and tine-esophagography. The positive predictive value of CES and RF4 was 9&O%, and the negative predictive value was 71.4%, but these values vary depending on disease prevalence. Figure 4 shows the results of scintigraphy, cine-esophagography, and manometry in one case. After induction of Raynaud’s phenomenon, amplitude, percentage of peristaltic waves, LESP, and RF4 showed no significant change. Discussion Most studies of esophageal function in systemic sclerosis have compared manometry, cineesophagography and esophageal transit scintigraphy in pairs.2e-30 Fitzgerald et aL31 compared all three modalities in patients who had Raynaud’s phenomenon, but because they did not meet diagnostic criteria for connective tissue disease the results cannot be extrapolated to our patient population. Nevertheless, it has been suggested that scintigraphy is more sensitive than radiography in the detection of impaired esophageal motility in systemic sclerosis and that radiography is therefore not indicated in the evaluation of dysmotility in such patients.” The major finding in this study is that both scintigraphy and tine-esophagography were able to differentiate patients with and without impaired esophageal motility to comparable degrees. Because changes in the esophagus in systemic sclerosis often affect peristalsis and amplitude of contraction waves and both may affect esophageal transit and emptying, one can understand the correlations obtained in our study. In an earlier study involving esophageal scintigraphy in systemic sclerosis, a scintigraphic parameter corresponding to our first-swallow residual fraction (RFl) significantly correlated with both contraction amplitude and LESP.” In contrast, we did not find a significant correlation between either RF1 or RF4 and LESP. Although the reasons for this discrepancy are uncertain, decreased LESP by itself would not be expected to impair esophageal transit
1554 KLEIN ET AL.
40-
GASTROENTEROLOGYVol. loz.No.5
a
30F LO-
5 AL
55 !!!
5 10-
0'
Controls
0L
Patlentr
Controls
0L
Patients
of liquid. Additionally, although Akesson et a1.28reported significant correlation between abnormal scintigraphy and increasing dysphagia, we found no evidence of diagnostic discriminating ability for any of the symptoms analyzed. Our findings also do not support the hypothesis that episodes of Raynaud’s phenomenon worsen esophageal function. Our failure to observe changes in esophageal motor function suggests that the de-f i
100
g 50 Z Z .----------0
2
I
I
0
II 0 I 0 II I II II 8 -------_--_. ii II
e
A
0
2
v .
loo-
I
Patients
creased esophageal blood flow observed after induction of Raynaud’s phenomenon” has no measurable acute effect on esophageal smooth muscle function. When evaluating patients with systemic sclerosis, how should one choose among the three modalities we studied? Manometry evaluates the lower esophageal sphincter as well as striated and smooth muscle elements, but because it is invasive and relatively difficult to tolerate, repeated studies are not often well accepted by patients. Cine-esophagography provides useful anatomic information and is well tolerated; it can be valuable in conjunction with scintigraphy to identify a hiatal hernia and confirm, in such cases, the distal boundary of the esophageal region of interest.32 Esophageal transit scintigraphy is well tolerated, has a lower radiation dose than cineesophagography,33 and lends itself well to quantification in both clinical and investigative contexts; thus, it is potentially useful for monitoring serial changes or response to therapy. Scintigraphic measurements are more readily quantified than those of cineesophagography and allow for finer gradations. We conclude that esophageal manometry, esophageal transit scintigraphy, and tine-esophagography
I
II
Controls
I
4
CES Figure 2. Plot of RF4 vs. CES in the patient group. Dashed lines represent the optimum cut-off values for the two parameters, such that cases above the horizontal line were positive by RF4 and cases to the right of the vertical line were positive by CES. Both tests showed moderately high ability to differentiate the manometrically abnormal subgroup (0) from the manometritally normal subgroup (A). The registering of cases in only the right upper and left lower quadrants shows that RF4 and CES sorted them identically into positive and negative categories. The correlation between RF4 and CES is represented by a Spearman rank-correlation coefficient of 0.81, P < 0.0005. Symbols with the same coordinates are slightly offset from one another.
Figure 1. Results of esophageal manometry in controls and patients. PPW, percentage of peristaltic waves; AMP, amplitude. Horizontal lines represent means, which were significantly lower for the patient group: *P < 0.01; **p < 0.0005.
.
z
zPE
7 SO?? :
?
:
Figure 3. Results of esophageal transit scintigraphy and cineesophagography in patient subgroups. N, manometrically normal: A, manometrically abnormal. Horizontal lines represent means, which were significantly higher for the manometrically abnormal subgroup: P < 0.01 for RF4 and for CES.
ESOPHAGUS
May 1992
Figure 4. Illustrations of esophageal dysfunction in a 56-year-old woman with limited scleroderma and severe esophageal motor impairment. Scintigraphy (A) showed complete retention of tracer (represented in black) in the esophagus after four swallows (RF4 = 100%). Cine-esophagography (B) disclose-d similar retention of barium associated with absent peristalsis (CES = 4). In the manometric tracing (C), peristaltic contractions were absent throughout the esophagus.
are approximately equivalent for detecting the presence and degree of severity of esophageal dysmotility in systemic sclerosis and are complementary in terms of the detailed information supplied by each. The choice of which test to use should be based on the clinical issues to be addressed as well as availability of local expertise. References
Distal
3.
4.
5.
6.
1555
e
I
2ommHg
\hr
5
s
9
C 9.
10.
11.
12.
1. Rodnan
2.
GP. Progressive systemic sclerosis [scleroderma). In: McCarthy DJ, ed. Arthritis and altered conditions. 9th ed. Philadelphia: Lea & Febiger, 1979;762-801. Cohen S, Laufer I, Snape WJ, Shiau YF, Levine GM, Jiminez S. The gastrointestinal manifestations of scleroderma: pathogenesis and management. Gastroenterology 1980;79:155-166. Zamost BJ, Hirschberg J, Ippoliti AF, Furst DE, Clements PJ, Weinstein WM. Esophagitis in scleroderma: prevalence and risk factors. Gastroenterology 1987;92:421-428. Stevens MB, Hookman P, Siegel CI, Esterly JR, Shulman LE, Hendrix TR. Aperistalsis of the esophagus in patients with connective-tissue disorders and Raynaud’s phenomenon. N Engl J Med 1964;270:1218-1222. Cohen S, Fisher R, Lipshutz W, Turner R, Myers A, Schumacher R. The pathogenesis of esophageal dysfunction in scleroderma and Raynaud’s disease. J Clin Invest 1972;51: 2663-2668. Hurwitz AL, Duranceau A, Postlethwait RW. Esophageal dysfunction and Raynaud’s phenomenon in patients with scleroderma. Am J Dig Dis 1976;21:601-606. Alexander EL, Firestein GS, Weiss JL, Heuser RR, Leitl G, Wagner HN Jr, Brinker JA, Ciuffo AA, Becker LC. Reversible cold-induced abnormalities in myocardial perfusion and function in systemic sclerosis. Ann Intern Med 1986;105:661668. Cannon PJ, Hassar M, Case DB, Casarella WJ, Sommers SC, LeRoy EC. The relationship of hypertension and renal failure in scleroderma (progressive systemic sclerosis) to structural and functional abnormalities of the renal cortical circulation. Medicine (Baltimore) 1974;53:1-46.
Middle
IN SYSTEMIC SCLEROSIS
13.
14.
15.
16.
17.
18.
19.
20.
Barr WG, Fahey PJ. Reduction of pulmonary capillary blood volume following cold exposure in patients with Raynaud’s phenomenon. Chest 1988;94:1195-1199. Belch JJ, Land D, Park RH, McKillop JH, MacKenzie JF. Decreased oesophageal blood flow in patients with Raynaud’s phenomenon. Br J Rheumatol 1988;27:426-430. Blom-Bulow B, Sundstrom G, Jonson B, Tylen U, Wollheim F. Early changes in esophageal function in progressive systemic sclerosis. A comparison of manometry and radiology. Clin Physiol 1984;4:147-158. Neschis M, Siegelman S, Rotstein J, Parker JG. The esophagus in progressive systemic sclerosis. A manometric and radiographic correlation. Am J Dig Dis 1970;15:443-447. Garg A, Prakash K, Gopinath PG, Malaviya AN. Radionuclide oesophageal transit time in progressive systemic sclerosis. Indian J Med Res 1984;79:110-113. Carette S, Lacourciere Y, Lavoie S, Halle P. Radionuclide esophageal transit in progressive systemic sclerosis. J Rheumatol 1985;12:478-481. Davidson A, Russell C, Littlejohn GO. Assessment of esophageal abnormalities in progressive systemic sclerosis using radionuclide transit. J Rheumatol 1985;12:472-477. Maddern GJ, Horowitz M, Jamieson GG, Chatterton BE, Collins PJ, Roberts-Thomson P. Abnormalities of esophageal and gastric emptying in progressive systemic sclerosis. Gastroenterology 1984;87:922-926. von Muhlen CA, Anselmi OE, Ludwig EB, Cutin J. Scintigraphic diagnosis of esophageal involvement in systemic sclerosis (scleroderma). Z Rheumatol 1989:48:167-174. Masi AT, Rodnan GP, Medsger TA Jr, Altman RD. D’Angelo WA, Fries JF, LeRoy EC, Kirsner AB, MacKenzie AH, McShane DJ, Myers AR, Sharp GC. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980;23:581-590. LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr, Rowe11 N, Wollheim F. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988;15:202-205. de Dombal FT, Hall R. The evaluation of medical care from the clinician’s point of view: What should we measure, and can we trust our measurements? In: Alperovitch A, de Dom-
1556
KLEIN ET AL.
bal FT, Gremy F, eds. Evaluation of efficacy of medical action. Amsterdam: North Holland, 1979:13-29. 21. Klein HA, Wald A. Computer analysis of radionuclide esophageal transit studies. J Nucl Med 1984;25:957-964. 22. Klein HA. The effect of projection in esophageal transit scintigraphy. Clin Nucl Med 1990;15:157-162. 23. Klein HA, Wald A. Normal variation in radionuclide esophageal transit studies. Eur J Nucl Med 1987;13:115-120. 24. Cupples LA, Heeren T, Schatzkin A, Colton T. Multiple testing of hypotheses in comparing two groups. Ann Intern Med 1984;100:122-129. 25. Hotelling H. The generalization of Student’s ratio. Ann Mathematical Stat 1931;31:360-378. 26. Milliken GA, Johnson DE. Analysis of messy data. Volume 1: Designed experiments. Belmont, CA: Wadworth Inc., 1984. 27. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29-36. 28. Akesson A, Gustafson T, Wollheim F, Brismar J. Esophageal dysfunction and radionuclide transit in progressive systemic sclerosis. Stand J Rheumatol 1987;16:291-299. 29. Drane WE, Karvelis K, Johnson DA, Curran JJ, Silverman ED. Progressive systemic sclerosis: radionuclide esophageal scintigraphy and manometry. Radiology 1986;160:73-76. 30. Clements PJ, Kadell B, Ippoliti A, Ross M. Esophageal motility
GASTROENTEROLOGY
Vol. 102, No. 5
in progressive systemic sclerosis (PSS). Dig Dis Sci 1979;24:639-644. 31. Fitzgerald OM, Bongiovanni G, Hess EV. Merhar G, Fernandez-Ulloa M, Spencer-Green G. Esophageal motility studies in patients with Raynaud’s phenomenon. J Rheumatol 1987; 14:273-277. 32. Blackwell JM, Richter JE, Wu WC, Cowan RJ, Caste11 DO. Esophageal radionuclide transit tests. Potential false-positive results. Clin Nucl Med 1984;9:679-683. 33. Tolin RD, Malmud LS, Reilley J, Fisher RS. Esophageal scintigraphy to quantitate esophageal transit (quantitation of esophageal transit). Gastroenterology 1979;76:1402-1408.
Received December 26,199O. Accepted September 30,1991. Address requests for reprints to: Herbert A. Klein, M.D., Ph.D., Division of Nuclear Medicine, Presbyterian-University Hospital, DeSoto at O’Hara Streets, Pittsburgh, Pennsylvania 15213. Supported by grants from the Wechsler Research Foundation and the Health Research and Services Foundation, and by General Clinical Research Center grant no. 5MOlRR00056 of the National Institutes of Health Division of Research Resources. The authors thank Frank Klasterka for excellent technical assistance; Mary Lamica for assistance with data processing; Janine Janosky, Ph.D., for statistical analysis; and Susan Orenstein, M.D., and James Reynolds, M.D., for valuable criticism.
1992;102:1551-1556
Comparative Studies of Esophageal Function in Systemic Sclerosis HERBERT A. KLEIN, ARNOLD WALD, TOBY 0. GRAHAM, WILLIAM L. CAMPBELL, and VIRGINIA D. STEEN Departments Pennsylvania
of Radiology and Medicine,
University
Three modalities for assessing esophageal dysfunction in patients with systemic sclerosis were prospectively compared. Seventeen patients underwent (a) esophageal manometry with measurement of distal esophageal peak contraction pressure amplitude, percentage of peristaltic waves, and lower esophageal sphincter pressure; (h) cine-esophagography with scoring based on residual contrast and the character of visualized waves; and (c) esophageal transit scintigraphy with quantification of residual swallowed tracer. Highly significant correlations were found between scintigraphic residual and tine-esophagography score, between scintigraphic residual and manometric amplitude, and indeed between all pairs of measured esophageal function parameters except those involving lower esophageal sphincter pressure. In addition, scintigraphy and tine-esophagography showed comparable ability to discriminate between patients with abnormal and normal esophageal motor function. Symptoms did not significantly correlate with quantitative parameters, nor did they have diagnostic discriminating ability. Induction of Raynaud’s phenomenon in a subgroup of patients had no detectable effect on esophageal function. It was concluded that these three diagnostic modalities are approximately equivalent in their ability to detect esophageal dysmotility in systemic sclerosis and measure its severity.
S
ystemic sclerosis is a generalized connective tissue disorder characterized by fibrosis and degenerative changes of the skin and some viscera.’ Atrophy and fibrosis of smooth muscle are probably responsible for the impaired esophageal motility that is common in this disease. Moreover, diminished lower esophageal sphincter pressure and associated gastroesophageal reflux occur frequently and can lead to esophagitis and stricture, complications that may be progressive and difficult to treat.‘s3 In the early stages of the disease, however, patients are of-
of Pittsburgh
School of Medicine,
Pittsburgh,
ten asymptomatic despite abnormal esophageal motor function.4-6 Raynaud’s phenomenon is a hallmark of systemic sclerosis. The hypothesis that it plays a role in organ system abnormalities has been based on cold-induced changes in the heart,7 kidneys,’ and lungs9 due to visceral vasoconstriction accompanying peripheral vasoconstriction. Previous work suggests that cold-induced vasospasm in the esophagus is associated with delayed return of that organ to baseline temperatures,” giving rise to speculation that episodes of Raynaud’s phenomenon may worsen esophageal function. Cine-esophagography is useful in the evaluation of esophageal function in most symptomatic patients with established systemic sclerosis,“~‘2 but esophageal manometry is considered a more sensitive technique to detect motor abnormalities at earlier stages.4-6 Esophageal transit scintigraphy, which monitors the movement of swallowed radioactive material, is another diagnostic modality that has been used to evaluate esophageal motor function. Previous investigators found scintigraphy to be a sensitive test in systemic sclerosis.13-17 The aim of this study was to prospectively compare measurements of esophageal function in systemic sclerosis using manometry, tine-esophagography, and scintigraphy. In a subgroup of patients, we also tested the hypothesis that inducing Raynaud’s phenomenon would adversely affect esophageal motor function. Materials and Methods Patient and Control Groups The patient group comprised 5 men and 12 women who met the American College of Rheumatology criteria for systemic sclerosis.‘8 Their mean age was 53 years (range, 32-69 years). Nine patients had an illness pattern typical of limited scleroderma, and 8 had diffuse sclero-
0 1992 by the American
Gastroenterological
0016-5065/92/$3.00
Association
1552
GASTROENTEROLOGY Vol. 102, No. 5
KLEIN ET AL.
derma.lg Two additional patients who could not tolerate esophageal manometry participated in evaluation of the reproducibility of scintigraphy. All patients had Raynaud’s phenomenon. None had esophageal strictures. The recruitment effort extended to all eligible patients of our institution’s division of rheumatology and the local systemic sclerosis support group. The patient group consisted of those patients consenting to be evaluated during the period of the investigation; illness pattern and severity were not applied to the selection process. As part of the evaluation, patients were questioned about the presence of dysphagia, odynophagia, heartburn, and regurgitation during the preceding month. Using the method of de Dombal and Hall,” each symptom was assessed on a severity scale (0, absent; 1, mild; 2, moderate; 3, severe) and a frequency scale (1, two to three times per month; 2, once per week; 3, two to three times per week; 4, daily). A total score (severity X frequency) was obtained for each symptom (maximum of 12 for each symptom, 48 for all symptoms). A control group (14 men, 6 women; mean age, 26 years; range, 17-53 years) underwent esophageal manometry. They were healthy volunteers with no symptoms or history of esophageal disease, central nervous system disorders, peripheral neuropathy, diabetes mellitus, systemic disorders known to be associated with esophageal dysfunction, or recent ingestion of medication known to alter esophageal function. The study was approved by the institutional committees for the protection of human subjects. Subjects gave informed written consent before participation.
Esophageal
Manometry
Manometric recordings were obtained using an eight-lumen polyvinyl catheter (diameter, 4.5 mm) consisting of three proximal lateral orifices (diameter, 0.8 mm) located 5 cm apart and five distal orifices that were axially oriented at 60’ and were 1 cm apart (Arndorfer Specialties, Inc., Greendale, WI). The catheter was continuously perfused using a low-compliance pneumohydraulic capillary infusion system (model CH8; Arndorfer Specialties, Inc.) at a rate of 0.5 mL/min. With this system, pressure increased at a rate >400 mm Hg/s upon occlusion of the orifice. Pressures were transmitted to external pressure transducers (model 223651; SensorMedics Corp., Anaheim, CA) and recorded on an eight-channel recorder (SensorMedics model R611). The recording system was calibrated before each study. All manometric studies were performed with the patient in the supine position and the catheter passed orally or nasally. Resting lower esophageal sphincter pressure (LESP) was measured at end expiration using the station pull-through technique. Pressure waves in the esophagus were obtained by studying 10 wet swallows of 2-mL boluses of water. From the 10 observations, we determined amplitude (mean peak contraction pressure amplitude in the distal esophagus) and the percentage of waves that were peristaltic. In 9 patients, manometry was followed by induction of Raynaud’s phenomenon, either by applying an ice collar to
the patient’s neck or by immersing the patient’s hand in ice water for several minutes, until a maximum effect was obtained or until the patient could no longer tolerate the cold. This was followed by repeat manometry.
Cine-esophagography Cine-esophagography was performed using three to five swallows of 50% (wt/vol) barium sulfate mixture while patients were in the prone right anterior oblique position. Each swallow was recorded on 16-mm cine-radiographic film at 12 frames/s or on videotape. Spot film radiographs of the esophagus were taken as indicated, and the presence or absence of spontaneous gastroesophageal reflux was recorded. Esophageal function was rated in each case using a tine-esophagography score (CES) as follows: 0, normal, no significant residual contrast material; 1, some abnormal contraction waves (less than half) and/ or residual contrast filling approximately one third of the length of the thoracic esophagus; 2, most waves abnormal and/or residual barium filling approximately two thirds of the thoracic esophagus; 3, all waves abnormal with residual barium filling all of the thoracic esophagus; 4, no peristaltic waves seen.
Esophageal
Transit Scintigraphy
Scintigraphy was performed as previously described using posterior imaging with a Raytheon Step l/ Step 2 scintillation camera (Raytheon Medical Systems, Melrose Park, IL).21*22Supine subjects swallowed 0.3 mCi (11.1 MBq) of technetium 99m sulfur colloid in 15 mL of water. A dry swallow was performed after 30 seconds and at l5-second intervals thereafter for a total of 10.75 minutes with no further administration of liquid. Using a Siemens Microdelta computer system (Siemens Medical Systems, Des Plaines, IL) including a VAX 11/750 computer (Digital Equipment Co., Maynard, MA), analysis of curves of radioactivity against time in the esophageal region of interest yielded a fourth-swallow residual fraction (RF4), expressed as a percentage, that served as our quantitative parameter for scintigraphy. This was the ratio of the residual counting rate after completion of the fourth swallow to the counting rate of the entire bolus observed immediately after its administration, We analyzed RF4 because our previous study of normal subjects showed high intrasubject and intersubject variability of the first-swallow residual fraction, with the variability diminishing considerably by the fourth swallow.23 In 9 patients, Raynaud’s phenomenon was induced before scintigraphy in the same manner as before manometry. This was done the day after baseline scintigraphy in 4 cases and the day before in 5 cases. The intervening day prevented residual radioactivity from interfering with the second measurement. All tests were performed during a 3-day period in our Clinical Research Center. Medications that could affect esophageal motor function were withheld for at least 48 hours before testing. The esophageal function parameters, i.e., the scores obtained from manometry, cine-esophagography, and scintigraphy, were each analyzed separately and without knowledge of the other findings.
May
ESOPHAGUS
1992
Statistical
Analysis
The following statistical procedures were used as appropriate: organization of symptom scores as either dichotomized or continuous variables; paired and unpaired Student’s t tests; Spearman rank correlation, ranking each parameter from most to least abnormal; biserial correlation; x2 testing; receiver operating characteristic curve analysis; and two methods of multivariate analysis, Hotelling’s T2 24.25 and Bonferroni’s adjustment.26 Results We found no significant difference between patients with limited and diffuse scleroderma with regard to esophageal function parameters and symptom scores; therefore, these subgroups were combined for further analysis. Among the 17 patients, the most common symptom was heartburn, affecting 14 and registering the highest mean score (5.9). The least common symptom was odynophagia, affecting two patients and registering the lowest mean score (2.8). Eleven patients had dysphagia, and 8 had regurgitation. One patient had no symptoms, and 2 had all four. The highest individual composite score was 27. No significant correlation was found among the scores for different symptoms. None of the five esophageal function parameters (amplitude, percentage of peristaltic waves, LESP, RF4, and CES) significantly correlated with any symptom or the composite symptom score. A subgroup of patients was defined as manometrically abnormal because their results were below the lower limit of the control group for amplitude (25 mm Hg) and/or percentage of peristaltic waves (900/,). Their individual symptom scores and composite scores were not significantly worse than those of manometrically normal patients. Results of esophageal manometry for the patients and the manometry control group appear in Figure 1. The scores of the patient group were significantly lower than those of the control group for amplitude, percentage of peristaltic waves, and LESP. In the patient group, we found a highly significant correlation between RF4 and CES (Figure 2) and between CES and percentage of peristaltic waves (P = 0.001). Significant correlations were also found between all the other possible pairings of the following parameters: RF4, CES, percentage of peristaltic waves, and amplitude (P < 0.0005 for all relationships). However, significance was not found for the relationships between LESP and the four other parameters. RF4 and CES were similar to each other in their degree of correlation with percentage of peristaltic waves and amplitude. For all patients, RF4 was 48.5% + 9.0% and CES was 2.6 * 1.3 (mean f SEM). After dividing the pa-
IN SYSTEMIC SCLEROSIS
1553
tients into manometrically normal and abnormal subgroups by esophageal manometry, we found RF4 and CES to be significantly higher in the manometritally abnormal subgroup (Figure 3). Both parameters met the statistical criterion for significant diagnostic discriminating ability,27 and there was no significant difference between them. Using optimum cut-off values for defining positivity of the tests, i.e., RF4 > 19.3% and CES > 2, the segregation of patients into positive and negative was identical for both tests (Figure 2), with sensitivity of 81.8%, specificity of 83.3%, and overall accuracy of 82.4%. These results indicate moderately high diagnostic discriminating ability for both scintigraphy and tine-esophagography. The positive predictive value of CES and RF4 was 9&O%, and the negative predictive value was 71.4%, but these values vary depending on disease prevalence. Figure 4 shows the results of scintigraphy, cine-esophagography, and manometry in one case. After induction of Raynaud’s phenomenon, amplitude, percentage of peristaltic waves, LESP, and RF4 showed no significant change. Discussion Most studies of esophageal function in systemic sclerosis have compared manometry, cineesophagography and esophageal transit scintigraphy in pairs.2e-30 Fitzgerald et aL31 compared all three modalities in patients who had Raynaud’s phenomenon, but because they did not meet diagnostic criteria for connective tissue disease the results cannot be extrapolated to our patient population. Nevertheless, it has been suggested that scintigraphy is more sensitive than radiography in the detection of impaired esophageal motility in systemic sclerosis and that radiography is therefore not indicated in the evaluation of dysmotility in such patients.” The major finding in this study is that both scintigraphy and tine-esophagography were able to differentiate patients with and without impaired esophageal motility to comparable degrees. Because changes in the esophagus in systemic sclerosis often affect peristalsis and amplitude of contraction waves and both may affect esophageal transit and emptying, one can understand the correlations obtained in our study. In an earlier study involving esophageal scintigraphy in systemic sclerosis, a scintigraphic parameter corresponding to our first-swallow residual fraction (RFl) significantly correlated with both contraction amplitude and LESP.” In contrast, we did not find a significant correlation between either RF1 or RF4 and LESP. Although the reasons for this discrepancy are uncertain, decreased LESP by itself would not be expected to impair esophageal transit
1554 KLEIN ET AL.
40-
GASTROENTEROLOGYVol. loz.No.5
a
30F LO-
5 AL
55 !!!
5 10-
0'
Controls
0L
Patlentr
Controls
0L
Patients
of liquid. Additionally, although Akesson et a1.28reported significant correlation between abnormal scintigraphy and increasing dysphagia, we found no evidence of diagnostic discriminating ability for any of the symptoms analyzed. Our findings also do not support the hypothesis that episodes of Raynaud’s phenomenon worsen esophageal function. Our failure to observe changes in esophageal motor function suggests that the de-f i
100
g 50 Z Z .----------0
2
I
I
0
II 0 I 0 II I II II 8 -------_--_. ii II
e
A
0
2
v .
loo-
I
Patients
creased esophageal blood flow observed after induction of Raynaud’s phenomenon” has no measurable acute effect on esophageal smooth muscle function. When evaluating patients with systemic sclerosis, how should one choose among the three modalities we studied? Manometry evaluates the lower esophageal sphincter as well as striated and smooth muscle elements, but because it is invasive and relatively difficult to tolerate, repeated studies are not often well accepted by patients. Cine-esophagography provides useful anatomic information and is well tolerated; it can be valuable in conjunction with scintigraphy to identify a hiatal hernia and confirm, in such cases, the distal boundary of the esophageal region of interest.32 Esophageal transit scintigraphy is well tolerated, has a lower radiation dose than cineesophagography,33 and lends itself well to quantification in both clinical and investigative contexts; thus, it is potentially useful for monitoring serial changes or response to therapy. Scintigraphic measurements are more readily quantified than those of cineesophagography and allow for finer gradations. We conclude that esophageal manometry, esophageal transit scintigraphy, and tine-esophagography
I
II
Controls
I
4
CES Figure 2. Plot of RF4 vs. CES in the patient group. Dashed lines represent the optimum cut-off values for the two parameters, such that cases above the horizontal line were positive by RF4 and cases to the right of the vertical line were positive by CES. Both tests showed moderately high ability to differentiate the manometrically abnormal subgroup (0) from the manometritally normal subgroup (A). The registering of cases in only the right upper and left lower quadrants shows that RF4 and CES sorted them identically into positive and negative categories. The correlation between RF4 and CES is represented by a Spearman rank-correlation coefficient of 0.81, P < 0.0005. Symbols with the same coordinates are slightly offset from one another.
Figure 1. Results of esophageal manometry in controls and patients. PPW, percentage of peristaltic waves; AMP, amplitude. Horizontal lines represent means, which were significantly lower for the patient group: *P < 0.01; **p < 0.0005.
.
z
zPE
7 SO?? :
?
:
Figure 3. Results of esophageal transit scintigraphy and cineesophagography in patient subgroups. N, manometrically normal: A, manometrically abnormal. Horizontal lines represent means, which were significantly higher for the manometrically abnormal subgroup: P < 0.01 for RF4 and for CES.
ESOPHAGUS
May 1992
Figure 4. Illustrations of esophageal dysfunction in a 56-year-old woman with limited scleroderma and severe esophageal motor impairment. Scintigraphy (A) showed complete retention of tracer (represented in black) in the esophagus after four swallows (RF4 = 100%). Cine-esophagography (B) disclose-d similar retention of barium associated with absent peristalsis (CES = 4). In the manometric tracing (C), peristaltic contractions were absent throughout the esophagus.
are approximately equivalent for detecting the presence and degree of severity of esophageal dysmotility in systemic sclerosis and are complementary in terms of the detailed information supplied by each. The choice of which test to use should be based on the clinical issues to be addressed as well as availability of local expertise. References
Distal
3.
4.
5.
6.
1555
e
I
2ommHg
\hr
5
s
9
C 9.
10.
11.
12.
1. Rodnan
2.
GP. Progressive systemic sclerosis [scleroderma). In: McCarthy DJ, ed. Arthritis and altered conditions. 9th ed. Philadelphia: Lea & Febiger, 1979;762-801. Cohen S, Laufer I, Snape WJ, Shiau YF, Levine GM, Jiminez S. The gastrointestinal manifestations of scleroderma: pathogenesis and management. Gastroenterology 1980;79:155-166. Zamost BJ, Hirschberg J, Ippoliti AF, Furst DE, Clements PJ, Weinstein WM. Esophagitis in scleroderma: prevalence and risk factors. Gastroenterology 1987;92:421-428. Stevens MB, Hookman P, Siegel CI, Esterly JR, Shulman LE, Hendrix TR. Aperistalsis of the esophagus in patients with connective-tissue disorders and Raynaud’s phenomenon. N Engl J Med 1964;270:1218-1222. Cohen S, Fisher R, Lipshutz W, Turner R, Myers A, Schumacher R. The pathogenesis of esophageal dysfunction in scleroderma and Raynaud’s disease. J Clin Invest 1972;51: 2663-2668. Hurwitz AL, Duranceau A, Postlethwait RW. Esophageal dysfunction and Raynaud’s phenomenon in patients with scleroderma. Am J Dig Dis 1976;21:601-606. Alexander EL, Firestein GS, Weiss JL, Heuser RR, Leitl G, Wagner HN Jr, Brinker JA, Ciuffo AA, Becker LC. Reversible cold-induced abnormalities in myocardial perfusion and function in systemic sclerosis. Ann Intern Med 1986;105:661668. Cannon PJ, Hassar M, Case DB, Casarella WJ, Sommers SC, LeRoy EC. The relationship of hypertension and renal failure in scleroderma (progressive systemic sclerosis) to structural and functional abnormalities of the renal cortical circulation. Medicine (Baltimore) 1974;53:1-46.
Middle
IN SYSTEMIC SCLEROSIS
13.
14.
15.
16.
17.
18.
19.
20.
Barr WG, Fahey PJ. Reduction of pulmonary capillary blood volume following cold exposure in patients with Raynaud’s phenomenon. Chest 1988;94:1195-1199. Belch JJ, Land D, Park RH, McKillop JH, MacKenzie JF. Decreased oesophageal blood flow in patients with Raynaud’s phenomenon. Br J Rheumatol 1988;27:426-430. Blom-Bulow B, Sundstrom G, Jonson B, Tylen U, Wollheim F. Early changes in esophageal function in progressive systemic sclerosis. A comparison of manometry and radiology. Clin Physiol 1984;4:147-158. Neschis M, Siegelman S, Rotstein J, Parker JG. The esophagus in progressive systemic sclerosis. A manometric and radiographic correlation. Am J Dig Dis 1970;15:443-447. Garg A, Prakash K, Gopinath PG, Malaviya AN. Radionuclide oesophageal transit time in progressive systemic sclerosis. Indian J Med Res 1984;79:110-113. Carette S, Lacourciere Y, Lavoie S, Halle P. Radionuclide esophageal transit in progressive systemic sclerosis. J Rheumatol 1985;12:478-481. Davidson A, Russell C, Littlejohn GO. Assessment of esophageal abnormalities in progressive systemic sclerosis using radionuclide transit. J Rheumatol 1985;12:472-477. Maddern GJ, Horowitz M, Jamieson GG, Chatterton BE, Collins PJ, Roberts-Thomson P. Abnormalities of esophageal and gastric emptying in progressive systemic sclerosis. Gastroenterology 1984;87:922-926. von Muhlen CA, Anselmi OE, Ludwig EB, Cutin J. Scintigraphic diagnosis of esophageal involvement in systemic sclerosis (scleroderma). Z Rheumatol 1989:48:167-174. Masi AT, Rodnan GP, Medsger TA Jr, Altman RD. D’Angelo WA, Fries JF, LeRoy EC, Kirsner AB, MacKenzie AH, McShane DJ, Myers AR, Sharp GC. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980;23:581-590. LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr, Rowe11 N, Wollheim F. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988;15:202-205. de Dombal FT, Hall R. The evaluation of medical care from the clinician’s point of view: What should we measure, and can we trust our measurements? In: Alperovitch A, de Dom-
1556
KLEIN ET AL.
bal FT, Gremy F, eds. Evaluation of efficacy of medical action. Amsterdam: North Holland, 1979:13-29. 21. Klein HA, Wald A. Computer analysis of radionuclide esophageal transit studies. J Nucl Med 1984;25:957-964. 22. Klein HA. The effect of projection in esophageal transit scintigraphy. Clin Nucl Med 1990;15:157-162. 23. Klein HA, Wald A. Normal variation in radionuclide esophageal transit studies. Eur J Nucl Med 1987;13:115-120. 24. Cupples LA, Heeren T, Schatzkin A, Colton T. Multiple testing of hypotheses in comparing two groups. Ann Intern Med 1984;100:122-129. 25. Hotelling H. The generalization of Student’s ratio. Ann Mathematical Stat 1931;31:360-378. 26. Milliken GA, Johnson DE. Analysis of messy data. Volume 1: Designed experiments. Belmont, CA: Wadworth Inc., 1984. 27. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29-36. 28. Akesson A, Gustafson T, Wollheim F, Brismar J. Esophageal dysfunction and radionuclide transit in progressive systemic sclerosis. Stand J Rheumatol 1987;16:291-299. 29. Drane WE, Karvelis K, Johnson DA, Curran JJ, Silverman ED. Progressive systemic sclerosis: radionuclide esophageal scintigraphy and manometry. Radiology 1986;160:73-76. 30. Clements PJ, Kadell B, Ippoliti A, Ross M. Esophageal motility
GASTROENTEROLOGY
Vol. 102, No. 5
in progressive systemic sclerosis (PSS). Dig Dis Sci 1979;24:639-644. 31. Fitzgerald OM, Bongiovanni G, Hess EV. Merhar G, Fernandez-Ulloa M, Spencer-Green G. Esophageal motility studies in patients with Raynaud’s phenomenon. J Rheumatol 1987; 14:273-277. 32. Blackwell JM, Richter JE, Wu WC, Cowan RJ, Caste11 DO. Esophageal radionuclide transit tests. Potential false-positive results. Clin Nucl Med 1984;9:679-683. 33. Tolin RD, Malmud LS, Reilley J, Fisher RS. Esophageal scintigraphy to quantitate esophageal transit (quantitation of esophageal transit). Gastroenterology 1979;76:1402-1408.
Received December 26,199O. Accepted September 30,1991. Address requests for reprints to: Herbert A. Klein, M.D., Ph.D., Division of Nuclear Medicine, Presbyterian-University Hospital, DeSoto at O’Hara Streets, Pittsburgh, Pennsylvania 15213. Supported by grants from the Wechsler Research Foundation and the Health Research and Services Foundation, and by General Clinical Research Center grant no. 5MOlRR00056 of the National Institutes of Health Division of Research Resources. The authors thank Frank Klasterka for excellent technical assistance; Mary Lamica for assistance with data processing; Janine Janosky, Ph.D., for statistical analysis; and Susan Orenstein, M.D., and James Reynolds, M.D., for valuable criticism.