SmartPill Capsule for Assessment of Gastric Emptying – Comparison with Simultaneous Gastric Emptying Scintigraphy

BODY: Gastric emptying scintigraphy (GES) using a radiolabeled solid-phase meal is currently considered the gold standard to detect gastroparesis. It is, however, performed in a variable manner at different institutions. A method to standardize gastric emptying using a low fat (EggBeaters) meal with imaging at 0, 1, 2 and 4 hours postprandially has been suggested (Tougas et al, Am J Gastro 2000;95:1456). Extending GES from 2 to 4 hours has also been advocated to detect more patients with gastroparesis (Guo et al, DDS 2001;46:24). The optimal duration of GES to maximize sensitivity and specificity for detecting gastroparesis has not been established. AIM: To determine the optimal duration of GES for detecting gastroparesis. METHODS: In a multicenter study evaluating the emptying of a pH/pressure recording capsule (SmartPill), GES was performed in 86 normal subjects and in 60 patients with previously diagnosed gastroparesis (37 idiopathic, 23 diabetic) based on scintigraphy within the past 2 years. GES was repeated in a standardized fashion with imaging at 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, and 4 hours after ingestion of a 99mTc-sulfur colloid radiolabeled meal (120 g EggBeaters, 2 pieces of bread with strawberry jam; 255 kcal, 2% fat) and 120 cc water. Receiver operating characteristic (ROC) curves were used to optimize sensitivity and specificity for differentiating gastroparetic patients from normal subjects for the 50% emptying time (T-50), 90% emptying time (T-90), and the percent retention (%R) at each imaging time. The concordance statistic (c-statistic) denotes the area under the ROC curve and is a global measure of diagnostic utility. RESULTS: The c-statistic was greater for %R at 3 hours (c=0.843) than at 1 hour (0.670), 2 hours (0.792), or 4 hours (0.816). It was also greater than that for T50 (c=0.766) and similar to that for T-90 (c=0.849). The cutoff point that maximizes both sensitivity and specificity for T-90, 182 min, gives a sensitivity and specificity of 84.8% and 77.9%, respectively, whereas the cutoff point for %R at 3 hours, 9% retention, gives 85.0% and 75.6%, and the cutoff point for %R at 4 hours, 3% retention, gives 70.0% and 73.2%. CONCLUSIONS: Using the previously validated EggBeaters meal, T-90 is better than T-50 to differentiate patients with previously diagnosed gastroparesis from normal subjects. The percent retention at 3 hours provides diagnostic utility similar to the 90% emptying time and avoids the need for data extrapolation. To optimally detect gastroparesis, assessment of gastric emptying at times greater than 2 hours is needed with the percent retention at 3 hours being the best individual time point. Making a Diagnosis of Gastroparesis Reduces Diagnostic Cost and Unnecessary Surgeries in Patients with Recurrent Nausea and Vomiting. C. W. Hatfield; W. D. Chey; D. Barthel; J. Semler; S. Harrell; J. Wo 1. Division of Gastroenterology/Hepatology, University of Louisville, Louisville, KY, USA. 2. Division of Gastroenterology, University of Michigan, Ann Arbor, MI, USA. 3. SmartPill Corporation, Buffalo, NY, USA. Gastroparesis (GP) is a disorder defined by delayed gastric emptying. Symptoms of GP are non-specific and the diagnosis is often overlooked. Extensive medical resources may be spent for its diagnostic workup. AIM: To determine the extent of medical utilization in diagnosing GP in subjects with recurrent nausea and vomiting. METHODS: Potential subjects were identified at University of Louisville and University of Michigan from Jan 2001 through Dec 2003 using ICD billing codes for GP, miscellaneous gastric, nausea, vomiting, and nausea with vomiting. Outpatient records were reviewed in chronological order. Study criteria were nausea and vomiting ≥3 months and documented gastroparesis by delayed gastric scintigraphy or residual solid food by EGD despite overnight fast. Those <18 years old and nonambulatory individuals were excluded. Diagnostic tests were separated into 3 categories: endoscopic, radiographic (including scintigraphy), and motility. Type and number of each diagnostic test and surgical procedure was obtained before and after making the diagnosis of GP. Cost of diagnosis was calculated using Medicare 2005 physician and hospital outpatient payment schedule for Kentucky and Michigan. Mann-Whitney U, Wilcoxon signed-rank and paired McNemar’s tests were utilized. RESULTS: 240 charts were reviewed. 141 subjects (median age 46 yrs, 79% female) met study criteria. Results are shown in table. There was a high variability in total diagnostic cost among the subjects. Gender, age, and etiology of GP did not significantly affect the total diagnostic cost. Cost associated with endoscopies was significantly greater than radiographic and motility testing (p<0.01). Total diagnostic cost decreased significantly after the diagnosis of GP was made. Prior to diagnosis of GP, 56 subjects (40%) had cholecystectomy, compared to only 1 of the remaining 85 subjects (1%) after diagnosis. 23 subjects (16%) had laparoscopy/laparotomy before diagnosis of GP, compared to 3 subjects (2%) after diagnosis. CONCLUSIONS: Diagnostic cost of GP in patients with chronic nausea and vomiting was quite variable. Establishing a diagnosis of GP significantly reduced the cost associated with testing and number of unnecessary surgeries unrelated to GP. Before diagnosis of gastroparesis After diagnosis of gastroparesis Diagnostic category # of tests/subject† Cost/subject† # of tests/subject† Cost/subject† Endoscopic 1 (0-9) $645 ($0-$5,234) 1 (0-10) $550 ($0-$5,974) Radiographic 2 (0-15) $492 ($0-$6,090) 1 (0-20) $246 ($0-$2,973) Motility 0 (0-4) $0 ($0-$998) 0 (0-3) $0 ($0-$997) Total 4 (1-21) *$1,581 ($246$8,480) 2 (0-23) *$1,035 ($0-$6,824) †Median (range); *p< 0.01 Gastroduodenal Motility Measured in Health and Disease During Transit of an Ambulatory CapsuleSmartPill. M. Podovei; M. Majewski; A. N. Yuen; B. Kuo ; I. Sarosiek; J. Kuhn; L. Negron; J. R. Semler; C. Semler; R. W. McCallum 1. GI Unit, Massachusetts General Hospital, Boston, MA, USA. 2. Division of Gastroenterology/Hepatology, University of Kansas Medical Center, Kansas City, KS, USA. 3. SmartPill Corporation, Buffalo, NY, USA. Introduction: Differences in GI luminal pressure patterns between healthy normals (N) and patients with gastroparesis (GP)could give insight into pathophysiology. Pressure measurements in the antrum and duodenum have traditionally been performed with invasive indwelling manometry catheters. SmartPill is a wireless ambulatory capsule that measures luminal pH and pressure as it courses throughout the GI tract after being swallowed. Passage of the indigestible solid SmartPill from the stomach into the duodenum appears to occur near the end or soon after the gastric emptying of a solid meal or during the fasting state when MMCs occur. Aims: To compare pressure patterns between normal subjects and patients with gastroparesis as defined by a previously abnormal gastric scintigraphy. Methods: In 2 centers, healthy subjects and patients swallowed the SmartPill (SP) after an overnight fast together with a standardized meal of 120 g Eggbeaters,2 pieces of bread with jam;255 kcal,2% fat) and 120 cc water. The rapid pH change from acidic to alkaline (>3 unit rise from baseline gastric pH) marked the emptying of the ingested SmartPill from the stomach into the duodenum. The frequency and amplitude of contractions recorded by the capsule were counted in 30 minute intervals from 60 min before gastric emptying (GE) to 60 min after the capsule left the stomach. These parameters for Ns and GPs were then compared for each time interval by two-sample unequal variance T test. Results: 21 Ns (13M/8F, mean age 30.8) and 16 GPs((4M/12F, mean age 40.9 (9 diabetic/7 idiopathic)) were studied. Mean contraction frequency and pressure amplitudes in both groups for 30 min intervals are summarized in the table below. Gastroparetics had lower gastric contraction frequencies in the time period –60 to –30 min (preceding gastric emptying) compared to normals, p<0.02. There were no differences in contraction amplitudes between the 2 groups. Conclusion: 1. This new ambulatory technology represented by SmartPill detected impaired gastric (antral) motility in GP compared to normals. 2. Decreases in frequency in gastroparetics prior to emptying of an indigestible solid may reflect neurogenic abnormalities in Phase II fasting response. 3. Contraction patterns in health and disease throughout the gut can be measured with the SmartPill Capsule. 30-60 m before GE 0-30 m before GE 0-30 m after GE 30-60 m after GE N Freq (#contr/min) 1.2 1.3 2.0 2.0 GP Freq(#contr/min) 0.7 1.3 1.7 1.7 P val 0.01 0.93 0.46 0.44 N Ampl(mmHg) 26.7 39.1 21.5 16.6 GP Ampl(mmHg) 29.0 49.6 25.4 17.7 P val 0.74 0.36 0.08 0.42 A New Non-Scintigraphic Method for Measuring Gastrointestinal Transit in Gastroparetic Patients. I. Sarosiek; K. Koch; K. Selover; B. Kuo; H. Parkman; W. Halser; W. M. John; S. Michael; W. D. Chey; J. Lackner; D. D'Andrea; R. W. McCallum 1. Internal Medicine, Kansas University Medical Center, Kansas City, KS, USA. 2. Department of Internal Medicine/Section of Gastroenterology, Wake Forest Baptist Medical Center, Winston-Salem, NC, USA. 3. SmartPill Corporation, Buffalo, NY, USA. 4. Massachusetts General Hospital, Boston, MA, USA. 5. Gastroenterology Section, Temple University Medical Center, Philadelphia, PA, USA. 6. Division of Gastroenterology/Department of Internal Medicine, University of Michigan Medical Center, Ann Arbor, MI, USA. 7. University of Louisville Medical Center, Louisville, KY, USA. 8. Department of Gastroenterology, Buffalo/VA Hospital, Buffalo, NY, USA. Background: Clinical diagnosis and therapy of motility disorders could be enhanced by better methodology for assessing Gastric Residence Time (GRT) as well as GI Total Transit Time (TTT) providing opportunity to improve correlation of symptoms with transit tests. The aim of the study: The aims of this study were to compare TTT and its major components GRT and small/large bowel transit time (SLBTT), measured with a new diagnostic capsule (SmartPill) in asymptomatic volunteers (Controls, C) and compare with corresponding values recorded in patients with gastroparesis (GP). Methods: 60 patients with GP: 23 diabetics (DM), 37 idiopathic (ID); 50 F & 10 M, mean age of 42 (range 19-66) mean BMI 26.3 (range 16.8-41.8) and 87 asymptomatic controls (C) 32F & 55M, mean age of 31 (range 19-57 years), with mean BMI 25.8 (range 17.9-41.1) participated in this multicenter study. After an overnight fast all subjects swallowed a wireless GI monitoring capsule (SmartPill) equipped with sensors that measured pressure, pH, and temperature. After initial data gathering at the study site, subjects returned to their normal ambulatory environment with parameters being captured by a portable recording device. Gastric Residence Time (GRT) was defined as the time from ingestion to an abrupt sustained rise in pH to >4 and at least a rise of 3 pH units from the baseline. GRT <30 was excluded from calculations. The TTT was measured from the time of ingestion of the wireless capsule until a drop in temperature or abrupt loss of signal associated with a bowel movement confirmed by patients’ diary. By subtracting GRT from TTT, SLBTT was calculated. Mann-Whitney Rank Sum Test was used to address statistical data. Results: In GP the median TTT was significantly longer than in controls (48.3 h vs 26.9 h, P<0.001), as was GRT (10.2 vs 3.7 h, P<0.001) as well as SLBTT (29.5 vs 22.2 h, P<0.001). Additionally, the TTT in patients with GP of diabetic etiology was significantly prolonged compared to patients with ID GP (60.1 vs 32.1 h, P=0.036). As was the SLBTT in DM GP compared to ID GP (53.1 vs 28.2 h, P=0.014). TTT, GRT, and SLBTT were similar for both genders among gastroparetics. Conclusions: 1. Significant differences in GRT, TTT, and SLBTT were observed between GP and controls by utilizing non-scintigraphic capsule technology. 2. Diabetic GP patients exhibit a prolonged SLBTT consistent with contributions from neuropathy, hyperglycemia, and other factors. 3. SmartPill has research and diagnostic potential for assessment of GI tract function in health and disease including physiologic and pharmacologic studies. SmartPill as a Novel Non-Scintigraphic, Ambulatory Method for Assessing GI Transit: Results in Normal Subjects. I. Sarosiek; R. W. McCallum; M. Sitrin; J. Wo; H. Parkman; B. Kuo; W. Hasler; W. D. Chey; J. Hooker; B. Landrigan; K. Koch 1. Internal Medicine, Kansas University Medical Center, Kansas City, KS, USA. 2. Department of Gastroenterology, Buffalo/VA Hospital, Buffalo, NY, USA. 3. University of Louisville Medical Center, Louisville, KY, USA. 4. Gastroenterology Section, Temple University Medical Center, Philadelphia, PA, USA. 5. Massachusetts General Hospital, Boston, MA, USA. 6. Division of Gastroenterology/Department of Internal Medicine, University of Michigan Medical Center, Ann Arbor, MI, USA. 7. Department of Internal Medicine/Section of Gastroenterology, Wake Forest Baptist Medical Center, Winston-Salem, NC, USA. 8. SmartPill Corporation, Buffalo, NY, USA. Background: Although the clinical tests for gastric residence time (GRT) have been evolving, estimation of the small/large bowel transit time (SLBTT) and gastrointestinal total transit time (TTT) poses more of a challenge for imaging technology. The aim of the study: To evaluate a new methodology for assessing the GRT, SLBTT, and TTT based on a diagnostic capsule (SmartPill) in asymptomatic volunteers. Methods: Eighty seven asymptomatic subjects (32F & 55M), mean age of 31 (range 19-57 years); mean BMI of 25.8 (range 17.9-41.1); after an overnight fast swallowed a wireless GI monitoring capsule (SmartPill ACT-I) equipped with sensors that measured pressure, pH, and temperature. After initial data gathering at the study site asymptomatic subjects returned to their usual daily environment, with parameters being captured by a portable recording device. Gastric residence time (GRT) was defined as the time from ingestion to a sudden rise in pH to >4 of at least 3 pH units from the baseline. GRT <30 min was excluded from calculations. The TTT was measured from the time of ingestion of the wireless capsule until there was a drop in temperature or abrupt loss of signal associated with a bowel movement confirmed by patients’ diary. By subtracting GRT from TTT we calculated SLBTT. T-test compared TTT female vs males; Mann-Whitney Rank Sum Test calculated GRT and SLBTT in regard to gender and Pearson PMC test addressed correlations between BMI and TTT. Data are presented as mean ±1SD or median. Results: All normal subjects had a mean GRT of 4.67 ±4.47 hours, mean SLBTT of 25.1 ±14.1 hours and mean TTT of 29.4 ±14.3 hours. Mean TTT among females was longer 34.8 ±17.7 hours, compared to males 26.3 ±11.0 hours (P=0.020). Median GRT for females and males was similar, 3.8 vs 3.67 hours. SLBTT of 25.3 hours in females was prolonged compared to 21.4 hours in males (P=0.05). There were correlations between BMI and TTT (r=-0.323; p=0.0086), BMI and SLBTT (r=-0.26; p=0.035), but not for BMI and GRT (r=-0.092; p=0.42). Conclusions: 1. The SmartPill capsule represents a novel, non-scintigraphic advance for assessing not only gastric residence, but also small bowel and colon transit times. 2. In normal subjects small/large bowel transit time was longer in females than males. 3. Longer small/large bowel transit time correlates with higher BMI in both genders. 4. This technology has potential for clinical and research applications in neuromuscular gut disorders.