This study was supported by an investigator-initiated grant from SmartPill, Inc, and in part by a NIH Midcareer Investigator Award in Patient-Oriented Research to HP Parkman (NIH DK02921). Jack Semler is the Chief Technology Officer for The SmartPill Corporation. Henry Parkman is a member of the advisory board of The SmartPill Corporation.
Preliminary results presented at 2007 Digestive Disease Week in Washington, DC and published as an abstract in Gastroenterology 2007;132 (4 Suppl. 2):A-97.
Henry P. Parkman MD, Gastroenterology Section, Temple University School of Medicine, Parkinson Pavilion, 8th Floor, 3401 North Broad Street, Philadelphia, PA 19140, USA. Tel: +1 215 707 7579; fax: +1 215 707 2684; e-mail: email@example.com
Abstract Gastric emptying of digestible solids occurs after trituration of food particles. Non-digestible solids are thought to empty with phase III of the migrating motor complex (MMC). The aim of this study was to determine if a non-digestible capsule given with a meal empties from the stomach with return of the fasting phase III MMC or during the fed pattern with the solid meal. Fifteen normal subjects underwent antroduodenal manometry and ingestion of a radiolabelled meal and SmartPill wireless pH and pressure capsule. In five subjects, emptying of the SmartPill was studied in the fasting period by ingesting the SmartPill with radiolabelled water. The SmartPill emptied from the stomach within 6 h in 14 of 15 subjects. SmartPill pressure recordings showed high amplitude phasic contractions prior to emptying. SmartPill gastric residence time (261 ± 22 min) correlated strongly with time to the first phase III MMC (239 ± 23 min; r = 0.813; P < 0.01) and correlated moderately with solid-phase gastric emptying (r = 0.606 with T-50% and r = 0.565 with T-90%). Nine of 14 subjects emptied the capsule with a phase III MMC. In five subjects, the SmartPill emptied with isolated distal antral contractions. In five subjects ingesting only water, SmartPill gastric residence time (92 ± 44 min) correlated with the time to the first phase III MMC (87 ± 30 min; r = 0.979; P < 0.01). The non-digestible SmartPill given with a meal primarily empties from the stomach with the return of phase III MMCs occurring after emptying the solid-phase meal. However, in some subjects, the SmartPill emptied with isolated antral contractions, an unappreciated mechanism for emptying of a non-digestible solid.
Different gastric emptying patterns have been observed for liquids, digestible solids and non-digestible solids.1,2 Emptying of liquids is determined mainly by fundic pressure. Gastric emptying of the digestible solid food occurs after the food is triturated to particles a few millimetres in size which can pass through the pyloric sphincter.3 Emptying of digestible solids depends mainly on antral contractions where trituration appears to be the rate-limiting step.4
In contrast to gastric emptying of liquids and digestible solids, non-digestible solids are thought to empty from the stomach after liquid and triturable gastric contents in response to the antral phase III contractions of the migrating motor complex (MMC).5 The phase III motor activity is part of the cyclical pattern of contractile activity that occurs during fasting periods. It consists of several minutes of forceful contractions that start in the antrum and propagate aborally in a progressive fashion down the small intestine. The function of the phase III activity (intestinal housekeeper) is thought to transport undigested food out of the stomach and small intestine toward the colon. Phase III is followed by phase I, relative quiescence lasting approximately 10–15 min. Phase II follows phase I, precedes phase III and consists of intermittent contractile activity. With ingestion of a meal, the cyclical pattern ceases and several hours of increased force and frequency of antral and small intestinal contractions occur (the ‘fed’ pattern). The duration of the fed pattern is the time from meal ingestion to the return of the phase III MMC, typically lasting approximately 4 h and dependent upon meal composition.6
The mechanism of the gastric emptying of large non-digestible solids in humans, however, has not been extensively studied. Years ago, the Heidelberg capsule, a non-digestible pH recording capsule, was shown to empty with the phase III migrating complex.7,8 In contrast, a more recent study in human subjects found that postprandial gastric emptying of non-digestible solids can sometimes occur unrelated to the antral phase III MMC.9
The SmartPill GI Monitoring Capsule (SmartPill, Inc; Buffalo, NY, USA) is a pH, pressure, and temperature recording device measuring 13 by 26 mm.10 After ingestion of the SmartPill, pH, pressure and temperature data are continuously transmitted from the capsule to a recorder as the capsule traverses the gastrointestinal (GI) tract. Gastric emptying of the SmartPill can be assessed by measuring the time required for the capsule pH to the change from the acidic stomach to the alkaline duodenum.10 Prior studies using the SmartPill capsule have shown high amplitude contractions just prior to its emptying from the stomach.11
The primary aim of this study was to compare the simultaneous measurement of gastric emptying of a non-digestible capsule to that of the traditional measurements of gastric function as assessed with gastric emptying scintigraphy and antroduodenal manometry. By doing so, we wanted to determine the mechanism by which a non-digestible capsule empties from the stomach. Does a non-digestible SmartPill capsule empty with the return of the phase III component of the MMC indicating the return of the fasting period or does it empty with the triturated solid meal during the fed period? A secondary aim was to compare pressure patterns within the antrum and duodenum in healthy subjects recorded by the SmartPill capsule as it traverses the GI tract to contractile patterns measured simultaneously with a fixed intraluminal antroduodenal pressure catheter.
Materials and methods
Twenty normal subjects were recruited to participate in this study which was approved by the Institutional Review Board of Temple University. Antroduodenal manometry, gastric emptying scintigraphy of a radiolabelled solid meal and SmartPill pH and pressure monitoring were all employed simultaneously to compare the gastric residence time of the SmartPill capsule with the duration of the fed pattern as measured by antroduodenal manometry and with scintigraphic determination of gastric emptying of a digestible solid meal.
After an overnight fast, subjects underwent placement of an antroduodenal motility catheter under fluoroscopic guidance as previously described.12,13 The antroduodenal manometry catheter (Konigsberg Instruments, Inc., Pasadena, CA, USA) had six solid state pressure transducers – the distal three were spaced 10 cm apart to measure small bowel contractility and the proximal three transducers were 2 cm apart to measure antral contractility. The catheter was inserted nasally and passed down the oesophagus into the stomach and guided across the pylorus into the duodenum under fluoroscopy. It was positioned so that the distal tip was at the ligament of Treitz with the three distal transducers in the duodenum and the three proximal transducers in the antrum. After the catheter was placed, the portable digital pressure data recorder was turned on to record antroduodenal pressures.
Following intubation, subjects were brought to the nuclear medicine laboratory for imaging. The first 15 normal subjects ingested a standard EggBeaters meal radiolabelled with Tc-99m sulphur colloid and 120 cc of water.14 The meal consisted of a scrambled egg white substitute (120 g EggBeaters, equivalent to two large eggs, 60 kcal) microwaved with Tc-99m sulphur colloid until firm (approximately 2 min on full power), and served with two slices of bread (120 kcal) and strawberry jam (30 g, 74 kcal). The meal has a caloric value of 255 kcal with nutritional composition: 72% carbohydrate, 24% protein, 2% fat and 2% fibre.14 Normal gastric emptying with this standardized meal has been reported to be <60% gastric retention at 2 h after meal ingestion and <10% gastric retention at 4 h.14
Immediately after ingesting the radiolabelled meal with 120 cc water, subjects swallowed an activated and calibrated SmartPill wireless pH, pressure and temperature capsule. The capsule houses sensors for pH, temperature and pressure. Shape and dimensions of the capsule (cylindrical, 26.8-mm long by 11.7 mm in diameter) are nearly identical to the ingestible imaging capsule employed by Given Imaging Ltd.15 The capsule has a pH range of 0.5–9 with an accuracy of ±0.5 pH units. The pressure sensor has a pressure range of 0–350 mmHg with an accuracy of 5 mmHg below 100 mmHg and 10% at or above 100 mmHg. The temperature sensor has a range of 25–49 °C, with an accuracy of ±1 °C. After activation and ingestion, the capsule signals are transmitted from within the GI tract and are captured by a receiving antenna incorporated into the receiver. The portable receiver worn by the subject received and stored data which, after the study was over, were downloaded to a PC computer. The transfer of data to the PC is accomplished via placement of the receiver in a docking station which provides an interface for data transmission to a computer as well as connections for battery charging. MotiliGI software (SmartPill, Inc.) loaded on the computer supports the data transfer, analysis of the recorded date, and displaying the test results.
For scintigraphic measurement of gastric emptying, anterior and posterior images were obtained in the 99mTc window (140keV ± 20%) with the patient standing in an upright position.16 Images were acquired with a large field of view gamma camera (General Electric Medical Systems, Milwaukee, WI, USA) and stored digitally using an image acquisition computer (NucLear Mac, Scientific Imaging, Denver, CO, USA). The first set of images was taken immediately after the ingestion of the meal and swallowing the capsule and water. Subsequent images and counts were obtained at 20-min intervals for the first 2 h, then every 30 min for the next 4 h. During imaging, the subjects generally remained seated in a chair except for walking to and from the gamma camera.
The antroduodenal portable digital pressure data logger was turned off 7 h after the radiolabelled sandwich was ingested, unplugged from the catheter, and the stored data were downloaded to a computer. Repeat fluoroscopy was used to confirm the position of the SmartPill and passage of the SmartPill out of the stomach and to confirm the position of the antroduodenal manometry catheter. After downloading the data, the antroduodenal catheter was removed from the study subject. The SmartPill data were also downloaded to a computer.
In five additional subjects, emptying of the SmartPill was studied in the fasting period by having the subjects ingest the SmartPill with 150 cc of Tc-99m sulphur colloid in water only. Gastric emptying scintigraphy of water was performed over 6 h.
At the beginning of each actual recording, the time recordings for both the antroduodenal manometry and SmartPill recordings were synchronized to record similar times throughout the study. At the time of each scintigraphy imaging, the times for each recording were compared to ensure synchronization. In addition, event markers serving as time stamps were used at the beginning and the end of the study to ensure that the times were synchronized.
The SmartPill pH and pressure data were displayed simultaneously as a function of time. Gastric emptying of the SmartPill was assessed by measuring the time from ingestion of the SmartPill to a rapid rise of the SmartPill pH >3 pH units indicating arrival in the alkaline duodenum.10 The SmartPill pressure recording was inspected for the motility patterns just prior to gastric emptying of the SmartPill.
The antroduodenal manometric recordings were visually inspected for the presence of phase III MMCs defined as high pressure phasic contractions starting in the stomach and propagating distally into the duodenum.5,8 Phase III MMCs were defined as high pressure phasic contractions starting in the stomach and propagating distally into the duodenum. Practically, these were identified first by recognizing the phase III occurring in the duodenum followed by identification of preceding antral contractile activity, as suggested in a prior study by Mojaverian et al.8 The duration of the fed pattern was calculated as the time from meal ingestion to the return of the phase III MMC. The patterns on the antroduodenal manometry recording were inspected at the time of SmartPill emptying from the stomach. The pressure patterns within the antrum and duodenum as measured by the SmartPill were compared to contractile patterns measured simultaneously using the fixed intraluminal antroduodenal pressure catheter.
For gastric emptying scintigraphy, gastric retention of the radiolabelled meal was determined by calculating the geometric mean of the radioactivity corrected for decay of the radioisotope label in the stomach region of interest.16 Gastric emptying was analysed by calculating the per cent retention at the end of 2 h and at the end of 4 h. In addition, time–activity curves of the per cent gastric retention of the radiolabelled meal were constructed. These curves were analysed by curve-fitting software (Kaleidagraph, Synergy Software, Reading, PA, USA) to a modified power exponential equation (% retention = 100 [1 − (1 − e−kt)β]), as previously reported.17 The fitted curves were used to calculate the T-50% (time to 50% emptying of the radiolabelled meal) and T-90% (time to 90% emptying of the radiolabelled meal).
Results are expressed as mean ± SEM.
To determine whether the SmartPill GI monitoring capsule emptied from the stomach with the phase III MMC, the gastric residence time of the SmartPill capsule was compared with the duration of the fed pattern as measured by antroduodenal manometry. A Pearson correlation coefficient was determined between the two. Correlations between the SmartPill gastric residence time and the solid phase emptying of a radiolabelled meal were also determined (Microsoft Excel, Redmond, WA, USA). In addition, the time of SmartPill emptying from the stomach was matched to the antroduodenal manometry tracing. SmartPill emptying from the stomach was defined as related to a phase III MMC if there was close proximity (within 10 min) to the onset of the antroduodenal phase III complex.
In 15 normal subjects (average age: 36.6 years; 12 males and three females), simultaneous antroduodenal manometry, gastric emptying scintigraphy and SmartPill recordings were obtained (Table 1). The antroduodenal manometry catheter was successfully placed in all subjects (Fig. 1A).
Table 1. Sequence of events for gastric emptying
MMC, migrating motor complex.
Gastric emptying scintigraphy
95 ± 6 min
194 ± 11 min
Time to phase III after meal ingestion
239 ± 23 min
Gastric residence time
261 ± 22 min
Per cent of radiolabelled meal remaining in stomach
At time of first phase III MMC
10.0 ± 4.3%
At time of SmartPill emptying from stomach
5.4 ± 1.4%
Gastric emptying of the EggBeaters meal was characterized by 38 ± 5% (mean ± SEM) retention at 2 h after meal ingestion and 2 ± 1% retention at 4 h after meal ingestion. Using the curve fitting techniques and a modified power exponential, the mean T-50% was 95 ± 6 min, and the T-90% was 194 ± 11 min (Table 1).
The SmartPill emptied from the stomach within 6 h of meal ingestion in 14 of 15 subjects (Fig. 1B). In one subject, the SmartPill did not empty within the 7-h postprandial period (Fig. 1C). In the 14 subjects with SmartPill emptying, the SmartPill pressure recordings showed high amplitude phasic contractions just prior to SmartPill’s emptying from the stomach (Fig. 2). The SmartPill gastric residence time (261 ± 22 min) correlated strongly with the time to the first phase III MMC after meal ingestion (239 ± 23 min; r = 0.813; P < 0.01) (Fig. 3). In nine of 14 (64%) subjects who emptied the SmartPill during the 6-h postprandial recording, the SmartPill emptied in proximity to a phase III MMC (first postprandial phase III in four and second phase III in five) (Figs 4A and 5). In five subjects, the SmartPill emptied with isolated distal antral contractions and not with a phase III MMC (Fig. 4B). Of the five subjects emptying the SmartPill with isolated distal antral contractions, three subjects emptied the capsule before the first MMC and two subjects after the first MMC but prior to the second phase III MMC.
The SmartPill gastric residence time also correlated moderately with gastric emptying of the solid meal (r = 0.606 with T-50% and r = 0.565 with T-90%). The per cent of radiolabelled meal remaining in the stomach at the time of the first phase III MMC averaged 10.0 ± 4.3% (Table 1). The per cent of radiollabelled meal remaining in the stomach at the time of SmartPill emptying from the stomach was 5.4 ± 1.4%.
Comparison of the SmartPill pressure recordings and the antroduodenal manometry catheter recordings showed similar pressure patterns around the time period of SmartPill’s emptying from the stomach which occurs with phase III antral activity (Fig. 5). During the time period immediately prior to SmartPill’s gastric emptying, presumably when the SmartPill is in the antrum where the antral manometry catheter pressure transducers are, similar pressure recordings were seen in the SmartPill recording and the manometry catheter recording (Fig. 5A,C). Early after capsule ingestion, there was less similarity due to the capsule being in the fundus of the stomach. After gastric emptying of the SmartPill, its pressure recording was similar to the pressure patterns recorded in the duodenum by the manometry catheter (Fig. 5B,C).
In five normal subjects (average age: 33.2 years; three males and two females), simultaneous antroduodenal manometry, gastric emptying scintigraphy and SmartPill recording were obtained after ingesting only a small amount of radiolabelled water to maintain continuation of the fasting state. In these three subjects ingesting only water, the SmartPill gastric residence time (92 ± 44 min) correlated with the time to the first Phase III MMC after water ingestion (87 ± 30 min; r = 0.979; P < 0.01). In four of five subjects, the SmartPill emptied with the first phase III MMC, the other with the second phase III.
Gastric emptying of the liquids (without any solids) was characterized by 16 ± 11% retention of the liquids at 2 h after ingestion and 1.2 ± 0.8% retention at 4 h after liquid ingestion. The emptying of the radiolabelled liquids was also further characterized using curve fitting with a modified power exponential. The mean T-50% was 43 ± 21 min, and the T-90% was 99 ± 42 min. The SmartPill gastric residence time was significantly correlated with gastric emptying of the liquid (r = 0.983 with T-50% and r = 0.984 with T-90%).
The purpose of this study was to compare the measures of gastric function recorded by a non-digestible solid to that of gastric emptying scintigraphy and antroduodenal manometry to determine if a non-digestible solid, a SmartPill capsule, empties from the stomach with the gastric phase III MMC associated with the fasting period or during the fed pattern in association with the triturated solid meal. The results showed that a strong correlation between the SmartPill gastric residence time and the duration of the fed pattern as indicted by the return of the phase III MMC. Furthermore, visual inspection of the antroduodenal manometry tracings showed that the SmartPill emptied primarily with the phase III MMC in most subjects. Interestingly, in some subjects (five of 14), the SmartPill emptied with isolated antral contractions.
The relative timing of gastric emptying of the solid meal and gastric emptying of a non-digestible solid capsule were characterized in this study: gastric emptying of the solid meal occurs, followed by initiation of the phase III portion of the MMC, with resultant gastric emptying of the non-digestible capsule. In this study, gastric emptying of the SmartPill although correlated with both the return of the phase III MMC and parameters of gastric emptying of digestible solids, was more strongly correlated with return of phase III MMC. One hypothesis is that the gastric emptying of digestible solids occurs after trituration of the solid meal to small particles. When emptying of the solid meal is complete, there is a switch to the fasting state with the resumption of the phase III MMC which can then empty non-digestible solids. This theory predicts that emptying of the SmartPill would correlate with both the time to return of the phase III MMC and gastric emptying values that describe the near complete emptying of the solid meal, such as T-90%, which was seen in this study. Interestingly, in a recently performed study the SmartPill gastric residence time correlated strongly with the T-90% of gastric emptying scintigraphy in both normal subjects and patients with gastroparesis.10
In one of the first studies with SmartPill capsule, gastric emptying of SmartPill was assessed with an Ensure meal; the median gastric emptying time for the SmartPill capsule after Ensure was 156 min.11,18 The SmartPill pressure data revealed that just prior to gastric emptying of the capsule, the capsule recorded high amplitude contractions, presumably from the distal antrum.18 This early study suggested that pressure tracings measured non-invasively may resemble traditional gastroduodenal manometric contractions.
This present study directly compared the SmartPill pressure readings with simultaneously performed antroduodenal manometry. The SmartPill emptied primarily with the phase III MMC. In five of 14 subjects, gastric emptying of the SmartPill was with the second rather than the first MMC. Some subjects emptied the SmartPill capsule with contractions that appeared isolated to the distal antrum and not with a propagated phase III MMC. Of the five subjects in which this pattern of isolated antral contractions was observed, three emptied before the first phase III and two after the first phase III. This pressure pattern has not been described before, although few studies have carefully evaluated the antral pressure patterns that occur with emptying of large indigestible solids. These isolated distal antral contractions could be caused by the capsule in the distal antrum and may represent a manifestation of the peristaltic reflex in response to an intraluminal bolus with ascending contraction (antrum) and descending relaxation (pylorus) described in prior studies.19
In this study, the gastric residence time of the SmartPill averaged 261 ± 22 min which is slightly longer than the time for SmartPill emptying in normal subjects reported in a prior study (220 ± 6 min).10 It is possible that the longer time for gastric emptying in this study was related to the intraluminal manometry catheter traversing the pylorus which might have prevented or delayed the emptying of the capsule. Impedance of the emptying of the non-digestible SmartPill capsule by the manometry catheter may explain those cases where the SmartPill emptied with the second rather than the first phase III MMC. This effect may also contribute to the lower correlation observed between SmartPill’s gastric residence time and the traditional solid meal emptying markers T-50% and T-90% in this study vs that observed in previously.10
Other studies have examined the emptying of non-digestible solids. Some investigators have examined large capsules whereas others have used smaller non-digestible solids. Results appear to depend upon the size of the non-digestible solid.9 The importance of antral phase III for emptying of non-digestible solids seems to increase with the particle size of the ingested non-digestible solid. The Heidelberg radiotelemetry pH recording capsule, measuring 7 by 20 mm, emptied with the phase III MMC in the four subjects studied – the first phase III MMC in three subjects and second phase III in the fourth.8 A recent study investigated the gastric emptying of the Given Capsule. Gastric emptying could be measured by visually determining when the Given Capsule emptied out of the stomach into the duodenum.20 In that study of patients with the Given Capsule, emptying occurred on average 22 min after ingestion.20 In contrast to studies with the SmartPill, the Given Capsule is administered while fasting, without meals. Other studies have evaluated the emptying of numerous smaller non-digestible solids. Feldman studied the gastric emptying of solid radiopaque markers in healthy subjects and diabetic patients with 10 solid radiopaque markers (small pieces of nasogastric tubing) which were administered with a standard meal.21 Most markers emptied during the fourth postprandial hour. Other studies by Feldman showed that the markers emptied more rapidly when ingested with water than with a meal,22 similar to the findings reported in this study. Of interest, one study reported gastric emptying of non-digestible solids (radiopaque markers) after a meal with no relation to antral phase III motor activity, at least up to a particle size of 3 mm and perhaps even 7 mm.9 Recent studies, primarily from Japan and Europe, have used radiopaque markers to measure gastric emptying with abdominal radiographs.23
In conclusion, this study demonstrates that a non-digestible solid (SmartPill capsule) empties from the stomach with the occurrence of high amplitude antral contractions. In most cases, SmartPill emptying occurs with the return of the phase III MMC of the fasting period. In some subjects, the SmartPill emptied with isolated antral contractions, a previously unappreciated mechanism for emptying of a non-digestible solid. The SmartPill empties after the meal has been almost completely emptied, and thus may be an indicator of the completion of gastric emptying as also measured by T-90% gastric emptying scintigraphy. In addition, the SmartPill provides a measure of gastric contractile activity during the later phase of its stomach transit in the antrum similar to that recorded with antroduodenal manometry.