Measurement of gastric emptying rate in humans

Digestive Diseases and Sciences, Vol. 35, No. 11 (November 1990), pp. 1345-1351

Measurement of Gastric Emptying Rate in Humans

Simplified Scanning Method

STEPHEN HOLT, MB, JERRY COLLIVER, PhD, MANINDER GURAM, MD, CHARLES NEAL, MD, STEVEN J. VERHULST, PhD, and T. VINCENT TAYLOR, MD

Simultaneous measurements of the gastric emptying rate of the solid and liquid phase o f a dual-isotope-labeled test meal were made using a gamma camera and a simple scintillation detector, similar to that used in a hand-heM probe. A simple scanning apparatus, similar to that used in a hand-heM scintillation probe, was compared with simultaneous measurements made by a gamma camera in 16 healthy males. A dual- labeled test meal was utilized to measure liquid and solid emptying simultaneously. Anterior and posterior scans were taken at intervals up to 120 min using both a gamma camera and the scintillation probe. Good relative agreement between the methods was obtained both for solid-phase (correlation range 0.92-0.99, mean 0.97) and for liquid- phase data (correlation range 0.93-0.99, mean 0.97). For solid emptying data regression line slopes varied from 0.75 to 1.03 (mean 0.84). Liquid emptying data indicated that slopes ranged from 0.71 to 1.06 (mean 0.87). These results suggested that an estimate of the gamma measurement could be obtained by multiplying the scintillation measurement by a factor of O.84 for the solid phase and 0.87for the liquid phase. Correlation between repeat studies was 0.97 and 0.96for solids and liquids, respectively. The application of a hand-held probe technique provides a noninvasive and inexpensive method for accurately assessing solid- and liquid-phase gastric emptying from the human stomach that correlates well with the use of a gamma camera, within the range of gastric emptying rate in the normal individuals in this study.

KEY WORDS: gastric emptying rate; scintigraphic methods.

The recognition of the importance of abnormal gastric motor function as a determinant of upper gastrointestinal symptoms has prompted the devel- opment of several methods for the determination of gastric emptying rate (1-3). Scintigraphic methods for the study of gastric emptying are preferred because they permit the use of "ordinary" meals,

Manuscript received April 16, 1990; revised manuscript received June 25, 1990; accepted June 28, 1990.

From the University of South Carolina School of Medicine Columbia, South Carolina; and Southern Illinois University School of Medicine, Springfield, Illinois.

Address for reprint requests: Prof. Stephen Holt, Division of Digestive Diseases and Nutrition, Two Richland Medical Park, Suite 506, Columbia, South Carolina 29203.

do not require intubation, and are accurate (3, 4). In these methods, the amount of a radiolabeled meal remaining in the stomach is determined by an external scanning device, such as a gamma camera or a rectilinear scanner (1, 3). Gamma camera measurements have become the "gold standard" for the measurement of gastric emptying rate, and the use of two isotopes with different energy emissions permits the simultaneous study of solid and liquid emptying (5). However, a gamma camera is expensive and often overcommitted to routine diag- nostic use. A gastric emptying study may utilize up to 11/2 hr of dedicated scanning time of a gamma

Digestive Diseases and Sciences, Vol. 35, No. 11 (November 1990) 0163-2116/90/1100-1345506.00/0 �9 1990 Plenum Publishing Corporation

1345

c a m e r a , exp la in ing , in pa r t , w h y this t e s t has no t ga ined w i d e s p r e a d a c c e p t a n c e in c l in ica l p r ac t i ce .

T h e m a i n o b j e c t i v e o f th is i nves t iga t ion was to d e t e r m i n e if a s ingle sc in t i l l a t ion d e t e c t o r (p robe) w o u l d p e r m i t an a c c e p t a b l e m e a s u r e o f gas t r i c e m p t y i n g ra te . G a s t r i c e m p t y i n g ra te m e a s u r e m e n t s us ing this p r o b e have b e e n c o m p a r e d wi th s imul ta - n e o u s m e a s u r e m e n t s o b t a i n e d us ing a g a m m a camera . In c o n t r a s t to the on ly p u b l i s h e d s impl i f ied s cann ing m e t h o d for gas t r i c e m p t y i n g ra te de t e rmi - na t i on (6), m e a s u r e m e n t s w e r e m a d e in the an t e r io r and p o s t e r i o r p r o j e c t i o n to a v o i d the e r ro r s tha t o c c u r wi th un i l a t e ra l d e t e c t i o n a lone (3, 4). S tud ies w e r e u n d e r t a k e n to e x a m i n e the e m p t y i n g o f bo th sol id and l iquid p h a s e s o f a t e s t mea l and an a t t e m p t was m a d e to p r e d i c t g a m m a c a m e r a r ead ings f rom s imple p r o b e m e a s u r e m e n t s , t h e r e b y a s se s s ing the va l i d i t y o f the use o f a s imple p r o b e for gas t r i c e m p t y i n g r a t e m e a s u r e m e n t .

M A T E R I A L S AND M E T H O D S

Sixteen healthy male volunteers (age range 24-35 years) underwent simultaneous measurement of the emptying of the solid and liquid phase of a dual-isotope- labeled test meal using a simple scintillation probe and a gamma camera. After an overnight fast, each subject ingested a test meal consisting of an omelette sandwich (two eggs, two slices of bread) labeled with 500 ixCi of [99mTc] sulfur colloid (solid phase) and orange juice labeled with 150 ixCi [HIIn]DTPA, (liquid phase). Within three months of the initial gastric emptying study, eight of the 16 subjects repeated the study (interval range 2-11 weeks).

None of the subjects had a known history of gastrointestinal disease, and none was taking medications known to alter gastrointestinal motility. Informed written con- sent was obtained from each subject, and the study was approved by the ethical review committee of Memorial Medical Center, Springfield, Illinois, on May 3, 1987. Each subject was requested to engage in similar lifestyle events preceding each study, and each was specifically instructed to refrain from alcohol on the day before the test and tobacco consumption on the day of the test.

Anterior and posterior supine scans were obtained with a gamma camera (GE400T-GE Star System, General Electric) and scintillation probe (Renaltron, Boston, Mas- sachusetts) at 15, 30, 45, 60, 90, and 120 min following ingestion of the radiolabeled test meal, which was con- sumed in a mean of 5 min (range 3-10 min). For counting, the hand-held probe was held approximately at right angles to the skin in the anterior and posterior projection. Time zero was taken to be the end of ingestion of the meal and no estimation was made of gastric emptying during eating. Between scans, the subject sat upright in a hard- backed chair. A marking pen was used to place a refer- ence point on the anterior and posterior aspect of the patient to facilitate accurate repositioning of the subject

HOLT ET AL

beneath the gamma camera or scintillation probe during scanning. Radioactive emissions from the solid and liquid phases of the test meal were recorded separ-atel3r in each study. In the case of the gamma camera, two pulse-height analyzers were set at appropriate photo peaks for tech- netium-99m and for ind ium-I l l . A single-scintillation probe (Renaltron Probe) connected to a coupter rate meter was utilized and counts were obtained from each isotope by changing the manual setting on the counter rate meter for the 99mTC and ~11In photopeaks with a channel width of approximately 20%.

The gamma camera was posit ioned over the subject 's stomach with the aid of an oscilloscope display. Data acquired with the gamma camera were recorded on magnetic tape in the computer. A selection of a region of interest to represent the stomach was undertaken in both the anterior and posterior scans displayed on a television screen. The scintillation detector was positioned for counting anterior and posterior projections by accurately identifying and marking the respective areas of maximal radioactivity at the beginning of the study and taking all subsequent readings from these two points. Counts were taken with a gamma camera for 3 min and with the probe for 40 sec at each interval up to 120 min. Scintillator detector readings occurred as simple counts. The radioactive counts observed at each recording interval with a probe or camera were transformed to be expressed as the percentage of the counts of the meal remaining in the stomach. This transformation was facilitated by placing the meal initially in front of the gamma camera or scintillation probe prior to ingestion. The data derived from direct imaging or counting of the meal prior to ingestion was assigned a 100% value (initial counts). Hence, for both methods of recording the results were expressed as a percentage of the initial counts.

The geometric means of the anterior and posterior counts of radioactivity recorded by the gamma camera and scintillation probe were computed to construct gastric emptying curves for both the gamma camera and the scintillation detector. Results were expressed as the percentage retention of the isotope in the stomach at each imaging time for the liquid and solid components of the test meal.

Statistical Analysis. Statistical analyses were performed to investigate the relative agreement between the scintillation detector measurements and the gamma camera measurements. The relative agreement between these different measurements was examined by computing the Pearson correlation coefficient separately for each subject using their measurements at each imaging interval after ingestion of the test meal. In order to determine the nature of the relationship between the measurements using each type of apparatus, a simple linear regression of the camera measurements on the scintillation detector measurements was then performed for each subject using these same data. The results from the regression analyses were averaged across subjects in order to determine a general method for the estimation of the gamma camera measurements from the scintillation detector readings.

Based on the assumption that a camera reading of 0.0% would be associated with a scintillation detector reading of 0.0%, the regression line for each subject was fit

1346 Digestive Diseases and Sciences, Vol. 35, No. 11 (November 1990)

GASTRIC EMPTYING IN HUMANS

TABLE 1. CORRELATION COEFFICIENTS, REGRESSION SLOPES, AND STANDARD ERRORS OF ESTIMATE FOR RELATIONSHIP BETWEEN GAMMA CAMERA AND SCINTILLATION DETECTOR MEASUREMENTS FOR 16 SUBJECTS IN INITIAL GASTRIC EMPTYING STUDY

Solids Liquids

Subject Corr Slope SE of Est. Corr Slope SE of Est.

1 0.99 0.88 4.16 0.99 1.00 2.31 2 0.92 0.76 7.82 0.99 0.98 1.74 3 0.99 0.86 4.997 0.99 0.93 3.34 4 0.96 0.79 5.95 0.96 0.82 5.60 5 0.98 0.83 6.56 0.98 0.93 4.53 6 0.96 0.82 6.15 0.94 0.72 9.79 7 0.95 0.75 11.84 0.96 0.71 10.57 8 0.98 0.81 9.21 0.96 0.71 9.90 9 0.94 0.85 7.43 0.97 0.87 4.90

10 0.99 0.83 7.57 0.99 0.94 2.80 11 0.94 0.78 6.89 0.93 0.72 8.47 12 0.98 0.88 3.64 0.98 0.86 4.44 13 0.98 1.03 6.63 0.99 1.06 3.33 14 0.92 0.78 6.21 0.99 0.91 3.00 15 0.99 0.97 3.60 0.98 0.93 5.96 16 0.99 0.87 3.90 0.99 0.89 4.00 Mean 0.97 0.84 6.4 0.97 0.87 5.3 SEM 0.01 0.02 0.5 0.01 0.03 0.7

through the origin of the graph using the NOINT (no intercept) option of the SAS regression program (SAS Institute, 1982). In preliminary analyses, regression lines also were fit without requiring the lines of identity to pass through the origin, and results did not differ. The results based on regression lines fit through the origin were considered more amenable to general application and were utilized in further analyses. The reproducibility of the regression findings was assessed by comparison of data obtained in repeated gastric emptying recordings in eight of the 16 volunteers.

The reliabilities of both the gamma camera and the scintillation detector measurements were assessed by examining the agreement in relative ordering of subjects at the scanning intervals (15, 30, 45, 60, 90, and 120 min) after the test meal.

RESULTS

The correlations between the gamma camera measurements and the probe recordings for each of the 16 individuals are presented in Table 1. These correlations are high, indicating good relative agreement for both solid phase (correlation range 0.92- 0.99, mean 0.97) and for liquid phase (correlation range 0.93-0.99, mean 0.97).

The ability of the probe measurements to predict the gamma camera measurements was investigated using regression of the gamma camera measurements on the scintillation probe recordings. Figures 1 and 2 illustrate the derivation of regression lines for each subject, with respect to solid emptying (Figure 1) and liquid emptying (Figure 2). In

these figures, the regression line for subject number one was identified through the points labeled A in the diagram, and for subject number two they were identified through the points labeled B, with this alphabetical notation continuing for the 16 subjects.

The slopes of the regression lines for each of the 16 subjects are presented in Table 1. For solid emptying data, slopes ranged from 0.75 to 1.03 (mean 0.84). These results suggested that an estimate of the gamma camera measurement could be obtained by multiplying the scintillation measurement by a factor of 0.84. The standard errors (SE) of the estimates of the camera measurements also are presented in Table 1 with a range from 3.6 to 11.8 for solids (mean 6.4), indicating that in approximately two-thirds of the cases the gamma camera measurements will fall in an interval that is equal to their est imated gamma camera value ---6.4. Liquid emptying data indicated that the slopes ranged from 0.71 to 1.06 (mean 0.87, SE range 1.7-10.6, mean SE 5.3). These findings mirrored the results obtained for solid emptying, suggesting again that for about two-thirds of the cases, the actual gamma camera measurements will fall within an interval that is equal to 0.87 times their scintillation detector measurement, -+5.3.

The results obta ined f rom repeat studies in eight of the 16 volunteers are displayed in Table 2. Similar results were obta ined for the eight volun-

Digestive Diseases and Sciences, Vol. 35, No. 11 (November 1990) 1347

HOLT ET AL

Gamma Camera

SOLID

1 DO

90

80

70

60

50

40

30

20

10

O

O

PD A

C M

j H

O

M O I A L H P

F EC

P D

E a H

O M

A

p L I j

HF

J g N K B

F

M B A cLG

P E F L N

M O JPH ID A C C O

M L I E N I(

OK L AO G 8 H I

E FG N

K B O N

' ; 4'0 ' ' 7'0 B~O 9'0 ' ' 10 20 3 50 60 100 110

Scintil lation Detector

Fig 1. Gamma camera readings plotted as a function of scintillation detector readings of solid emptying. Points are plotted for each of 16 subjects at 15, 30, 45, 60, 90, and 120 min after test mean. The six points for subject 1 are labeled A; for subject 2, B; etc.

teers who repeated the gastric emptying study within the 12-week period. For solid emptying data, the correlation range was 0.93-0.99 (mean 0.97) and slopes ranged from 0.81 to 0.99 (mean 0.91) with a range of SE from 3.8 to 14.9 (mean SE 7.6). For liquids the correlations ranged from 0.9 to 0.99 (mean 0.96) with a range of slopes from 0.78 to 1.08 (mean 0.95) and a range of SE from 1.8 to 17.4 (mean 8.0). The results suggest that, on average, a gamma camera reading can be predicted from a scintillation detector reading by multiplying the latter by 0.90. To assess the con- sistency between the predicted gamma camera values and the actual gamma camera readings, the predicted and actual gamma camera measures of the percent of remaining stomach contents 30 min after the ingestion of the radiolabeled meal were determined for each of the 16 subjects (Table 3). Significant correlations of 0.67 (P = 0.005) and 0.70 (P = 0.003) between the predicted and actual gamma camera readings were found for solid and liquid emptying, respectively. The means obtained with predicted and actual gamma camera readings were not significantly different for solid and liquid emptying (P > 0.05). Similar results were obtained

1348

from repeat studies in eight of the 16 subjects (Table 4).

The gastric emptying counts obtained with both the gamma camera and the scintillation detector across the six time intervals used for gastric emptying measurement were reliable. For the gamma camera, the reliability (generalizability) coefficients were 0.79 for solid-phase data and 0.82 for liquid- phase data (P < 0.0001). For the scintillation detector, the generalizability coefficients were 0.63 for solid emptying and 0.79 for liquid emptying (P < 0.001), confirming reliability.

DISCUSSION

Good relative agreement between the measurements obtained with the gamma camera and the scintillation detector imply that the scintillation detector can be used for the valid assessment of gastric emptying rate in normal individuals. The slopes of the regression lines for estimating the camera measurements from the scintillation detector measurements were consistent. In the initial gastric emptying study, involving 16 subjects, the average slopes were 0.84 for solids and 0.87 for

Digestive Diseases and Sciences, Vol. 35, No. 11 (November 1990)

GASTRIC EMPTYING IN HUMANS

Gamma Camera

L I Q U I D

1 O0

9 0

8 0

70

60

50

4 0

30

20

10

O

o C A

J E J F

A

B M E P

A O

B N H D M I L

P

BEc A N O G 0 F

M J P K H

C E 0 A 0 O

L I G J N P

F G

EC

N

L

B

A I L B O

M N P O F C E K

O BJ IH

L G

FD

K

G F

0 1 20 3 50 60 70 0 90 100 110

Scintillation Detector

Fig 2. Gamma camera readings plotted as a function of scintillation detector readings of liquid emptying. Points are plotted for each of 16 subjects at 15, 30, 45, 60, 90, and 120 min after test mean. The six points for subject 1 are labeled A; for subject 2, B; etc.

liquids, whereas in the repeat study in eight subjects, the average slopes were 0.91 for solids and 0.95 for liquids. It would appear that the slopes for both solids and liquids could be assumed to be approximately 0.90. With this assumption, it ap- pears that a scintillation detector reading can be used to approximate a gamma camera reading by multiplying the scintillation detector reading by a factor of approximately 0.90.

Using similar reasoning, the sE of estimate could be assumed to be about 7.0. Therefore, in about two-thirds of the cases, the actual gamma camera reading should fall within an interval that is 0.90 times the scintillation detector measurement ---7.0. Applying these mathematical corrections to the following example illustrates an inference of the camera measurements from the data obtained with the hand-held probe. Hypothetically, if the scintil-

TABLE 2. CORRELATION COEFFICIENTS, REGRESSION SLOPES, AND STANDARD ERRORS OF ESTIMATE FOR RELATIONSHIP BETWEEN GAMMA CAMERA AND SCINTILLATION DETECTOR MEASUREMENTS FOR 8 SUBJECTS IN REPEATED GASTRIC EMPTYING STUDY

Solids Liquids

Subject Corr Slope SE o f Est. Corr Slope SE o f Est.

1 0.93 0.85 8.78 0.92 1.08 11.84 2 0.96 0.96 3.79 0.99 0.99 1.78 3 0.97 0.90 5.79 0.94 0.90 9.97 4 0.95 0.93 7.25 0.96 0.97 7.12 5 0.94 0.99 10.30 0.99 1.08 4.32 6 0.99 0.81 14.94 0.90 0.80 17.38 7 0.99 0.96 5.68 0.99 0.99 3.32 8 0.99 0.89 4.54 0.95 0.78 8.01

Mean 0.97 0.91 7.6 0.96 0.95 7.9 SEM 0.01 0.02 1.3 0.01 0.04 1.8


HOLT ET AL

TABLE 3. AGREEMENT BETWEEN PERCENT OF REMAINING STOMACH CONTENTS AFTER 30 MIN FOR GAMMA CAMERA PREDICTED FROM SCINTILLATION DETECTOR AND ACTUAL GAMMA CAMERA READING FOR 16 SUBJECTS IN INITIAL GASTRIC EMPTYING STUDY

Solids Liquids

Predicted Actual Predicted Actual Subject gamma camera gamma camera gamma camera gamma camera

1 80 81 62 69 2 93 86 73 80 3 81 81 56 60 4 77 71 69 65 5 78 77 57 60 6 76 74 73 65 7 84 80 74 66 8 71 70 61 53 9 79 79 71 71

10 72 71 51 54 11 82 76 69 61 12 83 84 71 70 12 83 84 71 70 13 73 86 54 63 14 83 75 64 66 15 66 71 55 56 16 70 70 62 63 M e a n 78.0 77.0 63.9 63.9 SEM 1.7 1.4 1.9 1.7

P = 0 . 6 4 9 7 P = 1 . 0 0 0

r =0 .67 r = 0 . 7 0 P = .0046 P = .0027

lation detector showed gastric emptying of 50%, the gamma camera reading would be estimated to be 0.90 x 50%, ie, 45% with approximately two-thirds confidence that the actual camera reading was within the interval given by 45% -+ 7%, ie, between 38% and 52%.

Isotopic measurement of gastric emptying rate may have been limited in its general application by its requirement for expensive and intricate counting

devices such as a gamma camera. The present study demonstrates that it is feasible to obtain assess- ments of gastric emptying of both solid and liquid components of a test meal using simple, inexpensive counting devices such as a scintillation detector and perhaps a counter rate meter. The results that have been obtained with the probe are influenced by several factors, including the geometry of the counting, which may be influenced by the angle

TABLE 4. AGREEMENT BETWEEN PERCENT OF REMAINING STOMACH CONTENTS AFTER 30 MIN FOR GAMMA CAMERA PREDICTED FROM SCINTILLATION DETECTOR AND ACTUAL GAMMA CAMERA READINGS FOR 8 SUBJECTS IN REPEATED GASTRIC EMPTYING STUDY

Solids Liquids

Predicted Actual Predicted Actual Subject gamma camera gamma camera gamma camera gamma camera

1 77 74 63 65 2 80 83 81 77 3 78 76 75 64 4 80 83 80 75 5 67 73 72 72 6 76 75 76 68 7 81 87 75 70 8 76 74 88 68 M e a n 77.0 78.3 76.2 69.8 SEM 1.7 1.9 2.6 1.6

P = 0.6187 P = 0.0548

r = 0.73 r = 0.40 P = 0.0396 P = 0.3216

1350 Digestive Diseases and Sciences, Vol. 35, No. 11 (November 1990)

G A S T R I C E M P T Y I N G IN H U M A N S

or field of counting "view" or perhaps patient variables such as abdominal girth. It seems likely that a simple, hand-held probe could be utilized to provide a bedside measurement of gastric emptying rate, providing that simple mathematical corrections are applied to correct for the overestimates of emptying rate that are assumed to occur with a simple detector. Ostick et al (6) should be credited with the first demonstration that a simple probe could provide a reasonable estimate of emptying. In their studies, however, a single isotope label was utilized to label a semisolid and anterior imaging alone was utilized, with its known inherent inaccu- racies. The recognized value of measuring the different emptying rates of solid and liquid components of a test meal was addressed in our study, and the results obtained with a simple probe appeared reliable and generalizable compared with results obtained by the use of sophisticated gamma camera imaging.

It is of the utmost importance to position the scintillation probe accurately, and its optimal loca- tion is determined by identifying the maximal area of radioactivity in the anterior and posterior posi- tions. Ostick et al (6) have indicated the importance of locating the area of maximal activity for accurate recordings, and our results confirm their suggestion that the rate of decrease of maximal activity located in the body of the stomach is proportional to the rate at which labeled food leaves the stomach. Lag time of solid-phase emptying did not appear to exert a major effect on the agreement between the measurement of gastric emptying obtained by the probe and the gamma camera.

A simple scintillation detector permits a valid measurement of solid and liquid emptying from the human stomach within the range of gastric emptying encountered in this study. Further validation of this technique may be required in cases of abnor- mally slow or rapid emptying. Results similar to those obtained with a gamma camera can be obtained using simple mathematical operations. Appli- cation of this knowledge to utilize a hand-held probe could provide a reliable, noninvasive, less expensive method for assessing gastric emptying rate that could be readily utilized in clinical practice and research.

REFERENCES

1. Heading RC, Tothill P, McLoughlin GP, Shearman DJC: Gastric emptying rate of measurement in man: A double isotope scanning technique for simultaneous study of liquid and solid components of a meal. Gastroenterology 71:45-50, 1976

2. Holt S, Cervantes J, Wallace JHK, Wilkinson AA: Measure- ment of gastric emptying rate in man by real time ultrasound. Gastroenterology 90(4):918-923, 1986

3. Tothill P, McLoughlin GP, Holt S, Heading RC: The effect of posture on errors in gastric emptying measurements. Physics Med Biol 25:1071-1077, 1980

4. Tothill P, McLoughlin GP, Heading RC: Techniques and errors in scintigraphic measurement of gastric emptying. J Nucl Med 19:256-261, 1978

5. Moore JG, Christian PE, Coleman RE: Gastric emptying of varying meal weight and composition in man: Evaluation of dual liquid and solid phase isotopic method. Dig Dis Sci 26:16-22, 1981

6. Ostick DG, Green G, Howe K, Dymock IW, Cowley DJA: Simple clinical method per measurement of gastric emptying of solid meals. Gut 17:189-191, 1976


Measurement of gastric emptying rate in humans

Documents

Transcript of Measurement of gastric emptying rate in humans