Chronic constipation affects 20% of the population, and

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2009;7: Investigation of Colonic and Whole-Gut Transit With Wireless Motility Capsule and Radiopaque Markers in Constipation SATISH S. C. RAO,* BRADEN KUO, RICHARD W. MCCALLUM, WILLIAM D. CHEY, JOHN K. DIBAISE, WILLIAM L. HASLER, KENNETH L. KOCH, # JEFFREY M. LACKNER,** CARRIE MILLER,** RICHARD SAAD, JACK R. SEMLER, # MICHAEL D. SITRIN,** GREGORY E. WILDING,** and HENRY P. PARKMAN* *Department of Medicine, University of Iowa, Iowa City, Iowa; Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; Department of Medicine, Kansas University Medical Center, Kansas City, Kansas; Department of Medicine, University of Michigan, Ann Arbor, Michigan; Department of Medicine, Mayo Clinic, Scottsdale, Arizona; # The SmartPill Corporation, Buffalo, New York; **Department of Biostatistics, State University of New York at Buffalo, Buffalo, New York; and Department of Medicine, Temple University, Philadelphia, Pennsylvania Background & Aims:: Colonic transit time (CTT) traditionally is assessed with radiopaque markers (ROMs), which requires radiation and is hindered by lack of standardization and compliance. We assessed regional and CTT with the SmartPill (SmartPill Corporation, Buffalo, NY), a new wireless ph and pressure recording capsule, in constipated and healthy subjects and compared this with ROM. Methods:: Seventy-eight constipated (Rome II) and 87 healthy subjects ingested a 260-kcal meal, a ROM capsule, and the SmartPill. Subjects wore a data receiver and kept daily stool diaries for 5 days. SmartPill recordings assessed CTT, whole-gut transit time (WGTT), small-bowel transit time, and gastric emptying time. Abdominal radiographs on days 2 and 5 assessed ROM transit. Sensitivity/specificity and receiver operating characteristics (ROCs) of each technique and utility were compared. Results:: Gastric emptying time, CTT, and WGTT were slower (P <.01) in constipated subjects than controls. CTT was slower in women than men (P.02). Day 2 and day 5 ROM transits were slower (P <.001) in constipated subjects. Correlation of the SmartPill CTT with ROMs expelled on day 2/day 5 was r 0.74/r 0.69 in constipation, and r 0.70/r 0.40 in controls, respectively. The diagnostic accuracy of the SmartPill CTT to predict constipation from ROC was 0.73, with a specificity of These were comparable with those of day 5 ROM (ROC, 0.71; specificity, 0.95). Conclusions:: The SmartPill is a novel ambulatory technique of assessing regional (gastric, small bowel, colonic) and WGTT without radiation. It reveals hitherto unrecognized gender differences and upper-gut dysfunction in constipation. It correlates well with ROM and offers a standardized method of discriminating normal from slow colonic transit. Chronic constipation affects 20% of the population, and includes several subtypes: slow transit constipation, normal transit constipation, dyssynergic defecation, and constipation-predominant irritable bowel syndrome. 1 3 Because symptoms alone are poor predictors of underlying pathophysiology, 2 5 assessment of colonic transit time (CTT) and wholegut transit time (WGTT) using radiopaque marker (ROM) techniques, and assessment of defecation disorder with anorectal manometry, balloon expulsion, and defecography have been advocated. 1 5 Although widely used, a ROM test has intrinsic drawbacks that include radiation exposure, inability to assess regional gut transit, and lack of standardized protocols for the test/interpretation Also, some protocols require multiple visits, 5,8 which affects compliance. Scintigraphy has been validated for the evaluation of regional and WGTT, 10,11 but is expensive, 12 involves radiation, and is not widely available. 5,10,12 Consequently, most centers use multiple techniques to assess regional gut transit, such as gastric emptying with scintigraphy, 10,11,13,14 small-bowel transit with radiolabeled meal or lactulose breath test, 10 and CTT with ROM, 5,10 a time-consuming and expensive approach. Recently, a wireless motility capsule technique (SmartPill; SmartPill Corporation, Buffalo, NY) has been shown to be useful in the assessment of patients with gastroparesis. 14,15 Whether this technique is useful for CTT and WGTT assessments is not known. Furthermore, there is a dearth of large, prospective, and comparative studies of CTT in health or disease. 5,10,16 Previous ROM studies were performed in small numbers of subjects. 5 7 Also, whether constipated subjects have upper-gut dysfunction has not been assessed prospectively. We tested the hypotheses that a new wireless motility capsule (SmartPill) can assess CTT and WGTT, distinguish subjects with normal and slow colonic transit, and provide comparable information with that of the ROM technique. The aims of this study were as follows: (1) to investigate the CTT and WGTT of the SmartPill in chronic constipation and to compare this with healthy controls and define normative data; (2) to examine and compare the CTT of ROMs with the SmartPill in subjects with constipation and controls; and (3) to evaluate and compare the gastric emptying time (GET) and small-bowel transit time (SBTT) of the SmartPill in patients with constipation and controls. Methods Subjects Subjects with chronic constipation were enrolled in this multicenter prospective study. Subjects were recruited if they fulfilled Rome II criteria for chronic functional constipation Abbreviations used in this paper: AUC, area under the curve; CTT, colonic transit time; GET, gastric emptying time; ROC, receiver operating characteristic; ROM, radiopaque marker; SBTT, small-bowel transit time; WGTT, whole-gut transit time by the AGA Institute /09/$36.00 doi: /j.cgh

2 538 RAO ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 5 and reported at least 2 of the 6 symptoms of constipation. 17 Subjects with previous abdominal surgery were excluded except those with uncomplicated appendectomy, cholecystectomy, or cesarean sections. Patients were asked to discontinue all laxatives and drugs that affect motility, and proton pump inhibitors, at least 48 hours before the study, but stable doses ( 6 mo) of antidepressants, oral contraceptives, and lipid-lowering drugs were allowed. Healthy subjects were recruited after screening with the Mayo GI Disease questionnaire. 18 The study protocol was approved by the Institutional Review Boards from each center. The data were collated and analyzed centrally. Study Protocol After an overnight fast, subjects reported to the study center. First, they ingested a 260-kcal nutrient bar (SmartBar; SmartPill Corporation) that was composed of 17% protein, 66% carbohydrates, 2% fat, and 3% fiber along with 50 ml of water, which served as a standardized meal for assessment of GET. 14 Next, subjects swallowed a single capsule containing 24 ROMs (Sitzmarks; Konsyl Pharmaceuticals Limited, Fort Worth, TX), followed by a wireless motility capsule (SmartPill). Before ingestion, the SmartPill capsule was activated and calibrated. After ingestion, the radiofrequency signals emitted by the capsule were detected on a receiver (SmartPill) that was worn continuously for the next 5 days or until the capsule had been passed. After capsule ingestion, subjects were observed for 6 hours to prevent inadvertent ingestion of another meal that could compromise the assessment of GET. 14 Thereafter, they received a nutrient drink (Ensure; Abbott Laboratories, Chicago, IL), and were discharged. All subjects maintained a stool diary for the next 5 days in which they recorded bowel movements, stool consistency (Bristol Stool Scale), 19 time of food intake, and symptoms (pain, nausea). Subjects were asked to eat their usual diet and refrain from unaccustomed foods. A plain radiograph of the abdomen was performed 48 hours after capsule ingestion, and if the SmartPill had been passed, the receiver was retrieved, but the subjects remained in the study, continued their diary, and returned at 120 hours (day 5) for another radiograph to check the number of ROMs. 6,7 If the second radiograph revealed that the patient had not passed the SmartPill capsule, laxatives were given to facilitate capsule expulsion. A repeat radiograph was performed on day 21. If the capsule had not passed, the protocol allowed for endoscopic removal. Pressure and ph monitoring system. This system consisted of a wireless motility capsule, data receiver, and data analysis software (MotiliGI; SmartPill Corporation). The capsule houses sensors for ph, temperature, and pressure, and transmits this information at 434 MHz. Capsule size is similar to pill endoscopy (Given Imaging Limited, Yoqneam, Israel). The capsule measures ph (range, ph units), pressure (range, mm Hg), and temperature (range, 25 C 49 C). After completion of the study, the data were downloaded to a computer for analysis. Data Analysis The analysis consisted of measurements of regional and WGTT and review of abdominal radiographs for the number of retained ROMs. Assessment of Regional and Whole-Gut Transit Time GET was defined as the time interval between ingestion of the capsule and the time point at which there was an abrupt increase in the ph profile, usually 2 or more ph units from the gastric ph at baseline (Figure 1). This change occurred when the capsule passed from the acidic antrum to the alkaline duodenum. 14,15 GET was determined by 2 independent reviewers. SBTT was defined as the time interval between capsule entry into the small bowel and its entry into cecum. The cecal entry was defined as a distinct decrease in ph of greater than 1 ph unit that was sustained for 10 or more minutes and occurred at least 30 minutes after capsule entry into the small bowel (Figure 1). This ph change is believed to occur in the cecum and is secondary to fermentation of digestive residue by colonic flora. 20,21 CTT was defined as the time interval between the points of entry into the cecum and the capsule exit from the body. The capsule exit time was verified by a loss of signal and/or abrupt decrease in temperature (Figure 1). WGTT was defined as the time interval between capsule ingestion and its exit from the body. Assessment of Colonic Transit Time and Whole-Gut Transit Time with Radiopaque Markers The radiographs were reviewed at each site, and the number of markers remaining was reported. In addition, all radiographs were read by 2 independent investigators who were blinded to the site and when the radiograph was taken. Discrepancies in marker count were resolved by mutual consultation. Statistical Analysis The primary objective was an assessment of the relationship between CTT and WGTT as determined by the Smart- Pill with the number of retained ROMs on day 2 and day 5. This was addressed through estimation and hypothesis testing and Spearman correlation. The null hypothesis stated that the 2 devices (ROM and the SmartPill) are equivalent as defined by a true correlation of 0.7 or higher and therefore rejection of the null hypothesis would indicate significant disagreement between the devices. Power was determined to be 90% in detecting true correlations of 0.56 or smaller using a total sample size of 150 subjects. The SmartPill data for CTT, WGTT, and GET are expressed using estimates of quartiles. The upper limit of recorded GET was set at 6 hours because all subjects received a second meal at 6 hours. For WGTT the upper limit was 120 hours, representing the time of study completion. To accommodate this partial information (some subjects had true WGTT 120 h), quartile estimates were based on inversion of Kaplan Meier curves. The differences between constipated subjects and controls for SBTT were assessed using the Wilcoxon rank sum test or the rankbased procedure proposed by Gehan, 22 whichever was appropriate. Intragroup analyses were conducted to examine the effects of sex, and compared between groups. The diagnostic utility of the SmartPill for identifying subjects with constipation symptoms was examined through construction of a receiver operating characteristic (ROC) curve. The

3 May 2009 WIRELESS MANOMETRY CAPSULE FOR CONSTIPATION 539 Figure 1. This shows examples of the SmartPill and ROMs recording in a (A) healthy and (B) constipated subject. The healthy subject shows normal GET at 2.5 hours, normal SBTT at 5.3 hours, and normal CTT at 15.5 hours, and 4 ROMs on the day 2 radiograph. The constipated subject shows a delayed GET at 5.8 hours, normal SBTT at 4.3 hours, and delayed CTT at hours. The day 2 radiograph shows the SmartPill and 24 ROMs and the day 5 radiograph shows that the SmartPill has been expelled but 10 ROMs remain. area under the curve (AUC) was taken as an overall measure of diagnostic accuracy. Similarly, the utility of the ROM test was examined through construction of a ROC curve for the number of ROMs remaining on day 5. Statistical differences in diagnostic utility were examined through computation of the 95% bootstrap confidence interval for the difference between the areas under the ROC curves for both tests. Transit time cut-off values based on the 95th percentile of the healthy control population were used to determine the sensitivity and specificity for the SmartPill CTT and WGTT. Day 5 ROM sensitivity and specificity was based on a cut-off value of more than 5 markers retained.6,7 We also performed a subanalysis of the ROC and sensitivity/specificity of the SmartPill in patients with abnormal transit ( 5 ROM on day 5). The

4 540 RAO ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 5 Table 1. Median (25th 75th Percentiles) Values for CTT, WGTT, GET, and SBTT as Measured by the SmartPill in Constipated Subjects and Healthy Controls, and the Effects of Sex Overall Women only Men only Constipated (n 67) Healthy (n 81) P value Constipated (n 59) Healthy (n 39) P value Constipated a (n 8) Healthy (n 42) P value CTT, h 46.7 ( ) 21.7 ( ) ( ) 24.7 ( ) (25.2 ) 18.7 ( ).0264 WGTT, h 59.3 ( ) 29.7 ( ) ( ) 33.9 ( ) (36.3 ) 25.6 ( ).0115 GET, h 3.5 ( ) 3.0 ( ) ( ) 3.5 ( ) (3.6 ) 2.7 ( ).0054 SBTT, h 4.2 ( ) 3.8 ( ) ( ) 3.8 ( ) ( ) 3.8 ( ).4667 a Seventy-fifth percentile not observed for CTT, WGTT, and GET because of capping of data. chi-square test was used to examine the association between the location of the majority of ROMs on day 2 and day 5 radiographs with the location of the SmartPill. Interobserver agreement for the ROM count was examined through calculation of intraclass correlation and kappa statistic ( 5 vs 5 retained). All analyses were performed using SAS (version 9.1.3, SAS Institute Inc, Cary, NC), and a difference of 0.05 was considered significant. Results Demographics A total of 165 subjects were enrolled, of whom 78 subjects had constipation (69 women; mean age, 45 y; age range, y; and 9 men; mean age, 53 y; age range, y) and 87 were healthy volunteers (40 women; mean age, 39 y; age range, y; 47 men; mean age, 36 y; age range, y). One or more SmartPill transit parameters could not be identified in 12 subjects because of a software malfunction. Cecal arrival time was not recognizable in 5 other subjects. Thus, data from 148 subjects (67 constipated; male/female, 8/59) and 81 healthy subjects (male/female, 42/39) were available for analysis of transit time. Two subjects failed to obtain radiographs on day 2/day 5 and 10 other subjects had radiographs on day 4 rather than day 5. Thus, ROM data were available in 153 subjects (86 healthy subjects: male/female, 47/39; and 67 constipated subjects: male/female, 8/59). Wireless Motility Capsule (SmartPill) Data Analysis Figure 1 shows examples of normal transit in a healthy subject (Figure 1A), and delayed CTT with the SmartPill and retention of ROMs in a constipated subject (Figure 1B). The median values for CTT, WGTT, GET, and SBTT are reported in Table 1. Colonic transit time. Figure 2 shows the sex-based distribution of CTT. Although there is overlap, constipated subjects had slower CTT than controls (P.0001), irrespective of gender differences between groups. Furthermore, constipated men (P.0264) and women (P.0023) also had slower CTT than control men and women. Also, in healthy controls, women had slower CTT than men (P.0080). Whole-gut transit time. WGTT was slower (P.0001) in constipated subjects than controls (Figure 3C). Also, constipated men and women had slower WGTT when compared with sex-based controls (men: P.0115; women: P.0004). WGTT was significantly slower in healthy women when compared with healthy men (P.0001), but there was no gender difference among constipated subjects (P.73). Gastric and small-bowel transit. Constipated subjects had significantly slower (P.0123) GET but not SBTT (P ) when compared with healthy controls (Figure 3A and B). Also, healthy women had slower GET than healthy men (P.0064). Radiopaque Marker Data Analysis Twenty-two (37%) constipated subjects retained more than 5 markers on the day 5 radiographs, indicating slower CTT. Thus, more (P.0001) constipated subjects had slower CTT than controls. The median number of ROMs that was retained on the day 2 radiographs in constipated subjects was 22, and in healthy subjects was 6 (P.0001) (Table 2). The overall correlation of the number of ROMs that were seen on day 2 with that of day 5 was 0.62; for healthy controls it was 0.44, and for constipated subjects it was There was some interobserver disagreement in the absolute ROM count for day 2 radiographs in 32% of subjects and for day 5 radiographs in 17% of subjects, but the intraclass correlation for day 2 radiographs was and for day 5 radiographs was 0.978, showing good agreement (kappa 0.96) between the principal investigators and the independent reviewers. Figure 2. Box-and-whisker plots for SmartPill CTT in healthy and constipated subjects, and effects of sex. CTT was significantly slower in constipated women and men compared with controls.

5 May 2009 WIRELESS MANOMETRY CAPSULE FOR CONSTIPATION 541 Figure 3. Box-and-whisker plots for (A) GET, (B) SBTT, and (C) WGTT. GET and WGTT were slower in constipated subjects. Correlation of Radiopaque Markers With SmartPill Transit The Spearman correlation coefficient between CTT as assessed by the SmartPill and the day 2 radiograph ROM results for the overall sample was r When estimated separately, it was found to be 0.74 in constipated subjects and 0.70 in controls. The correlation between CTT and ROM on day 5 for the entire group was r 0.59 (95% confidence interval, ). The correlation in constipated subjects was r 0.69 and in controls was r Comparable results were found for the correlation of ROM with WGTT (Table 3). The lower correlations on day 5 were because many subjects had no ROMs on the day 5 radiograph. The location of the SmartPill as revealed by the day 2 or day 5 radiographs was associated strongly (P.001) with the region of the colon in which a majority of ROMs was seen. For example, the SmartPill was seen in the left colon in 47 subjects, and in 39 (83%) of these subjects the majority of ROMs was seen in the left colon. Diagnostic Accuracy of the SmartPill in Identifying Patients With Symptoms of Constipation The diagnostic utility for predicting constipation symptoms is approximately the same for ROMs (AUC, 0.71) and the SmartPill (AUC, 0.73) techniques. The AUC and the sex-related sensitivity and specificity values for the SmartPill CTT and WGTT and day 5 ROM are reported in Table 4 along with their corresponding cut-off values. A subanalysis of the ROC, sensitivity, and specificity of the SmartPill for the WGTT and CTT in subjects with an abnormal ROM transit revealed an AUC of 0.88 for both measures. The cut-off value for normal versus delayed CTT with the SmartPill in men was 44 hours and in women was 59 hours. The sensitivity and specificity of CTT as measured by the SmartPill was 0.50 and 0.90 for men and 0.46 and 0.92 for women, respectively (Table 4). By using a cut-off value of 5 ROMs retained on day 5 as abnormal, the sensitivity and specificity of day 5 ROMs was 0.37 and 0.95, respectively (Table 4). Twenty-three constipated subjects were delayed according to ROM criteria, and 19 of these subjects (83%) also had delayed CTT as measured by the SmartPill. Thirty-one constipated subjects were delayed according to CTT criteria ( 59 h) with the SmartPill and 21 (68%) of these subjects had delayed transit by day 5 ROM criteria. Adverse Events No serious adverse events were reported during the study. The SmartPill was expelled successfully in all subjects. The day 5 radiograph revealed that all controls had passed the SmartPill, but 14 constipated subjects retained the capsule, of whom 11 had no capsule on the day 21 radiograph, and 3 recovered the capsule from stool. Discussion Several techniques have been used to measure CTT and WGTT including ROM, isotopes, colored dyes, beads, and contrast agents However, there is no universally accepted or standardized technique. 5,10 For clinical purposes, it is preferable to use a single test that provides comprehensive information regarding gastrointestinal transit, and without radiation. We found that the CTT and WGTT were significantly slower in subjects with constipation when compared with healthy controls, both with the SmartPill and ROM techniques. Although there was intersubject variability within each group and some overlap between groups, overall constipated subjects had slower colonic transit than healthy controls. Table 2. Median Values (25th 75th Percentiles) for the Number of ROMs That Were Retained on Plain Abdominal Radiographs Taken on Days 2 and 5 in Constipated Subjects and Healthy Controls and the Effects of Sex Overall Females only Males only Constipated Healthy P value Constipated Healthy P value Constipated Healthy P value No. of ROM on day 2 22 (9 24) 6 (1 16) (10 24) 12 (4 21) (9 17) 1 (0 9).0019 No. of ROM on day 5 1 (0 17) 0 (0 0) (0 17) 0 (0 1) (0 13.5) 0 (0 0).0004

6 542 RAO ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 5 Table 3. Correlation of CTT and WGTT as Measured by the SmartPill With the Number of Retained ROMs SmartPill parameter Overall group day 2 ROM (95% CI) Day 2 ROMs in healthy subjects Day 2 ROMs in constipated subjects Overall group day 5 ROM (95% CI) Day 5 ROMs in healthy subjects Day 5 ROMs in constipated subjects CTT 0.78 ( ) ( ) WGTT 0.77 ( ) ( ) CI, confidence interval. ROM technique is used commonly for measuring colonic transit, 2,5,10,16 and hence served as our gold standard. In this large study of ROMs, we found that there was good correlation between ROM technique and the SmartPill for CTT and WGTT. Also, our data revealed that the SmartPill had high specificity (0.95) and reasonable sensitivity (0.46) for identifying an abnormal transit time in patients with constipation. These values were comparable with the ROM specificity and sensitivity of 0.95 and 0.40, respectively. The sensitivity value of 0.46 or 0.40 with either technique is not surprising because we examined all subjects who presented with symptoms of constipation. In this population, the reported prevalence of slow colonic transit ranges from 35% to 55%. 3,5,16,23 Thus, the lower sensitivity of these tests reflects the lower prevalence of slow transit constipation in a large prospective cohort of constipated individuals, and the heterogeneity of this problem. Furthermore, the ROC for the sensitivity and specificity of WGTT and CTT for the SmartPill in patients with an abnormal ROM transit time was excellent with an AUC of and 0.877, respectively. In healthy subjects, the Spearman correlation of the Smart- Pill to day 5 ROMs was lower than in constipated subjects. Most subjects (80%) expelled all ROMs by day 5 consistent with previous studies. 6,7 However, correlations from samples containing a high percentage of zero ROMs are less useful than samples containing a more varied range of data owing to the heavy influence of a small subset of non-zero results. Interestingly, the correlation of day 2 ROMs to day 5 ROMs was similarly low because of a high percentage of zero values on day 5. In both groups, the correlations for day 2 were higher than for day 5. There were significantly more individuals who retained ROMs on day 2, producing a more meaningful correlation. Overall, the higher specificity together with the good correlation with ROM lends support to our hypothesis that the SmartPill can discriminate between subjects with normal and slow colonic transit. Furthermore, the ability to provide a continuous and more direct measure of CTT offers the potential for characterizing the severity of slow transit constipation. We found that the GET of the SmartPill was slower in constipated subjects when compared with healthy controls. A recent study reported that 10% of subjects with constipation had slower GET and 20% had delayed SBTT, reaffirming our findings, but this was a secondary analysis and not a prospective study. 16 Also, in our healthy controls, the GET was slower in women, which confirms and extends previous observations. 16,24,25 Because previous studies showed no significant change in CTT of ROMs between the follicular and luteal phases, 7,25,26 we enrolled subjects irrespective of the phase of their menstrual cycle. Our study showed that there were significant differences for the CTT between men and women, both in constipated and healthy subjects. Unlike previous reports, 7,25 27 gender differences must be considered when interpreting results of colonic transit study. A potential weakness of our study was the gender imbalance between the study groups. However, data from gender-specific analyses were consistent with the pooled analyses. Thus, irrespective of gender differences, constipated subjects had slower CTT and WGTT. Also, most of our subjects were younger than 65 years. Another drawback is that changes in gut ph profile may affect the SmartPill s ability to detect GET, for example, in subjects taking proton pump inhibitor drugs. 14 However, this is unlikely to affect measurement of CTT or WGTT. The 10% incidence of technical issues with the SmartPill was in part owing to use of prototype equipment. In a more recent study using commercial grade equipment, one technical problem was encountered in 41 studies. This rate (2.4%) compares favorably with the technical failure rate of 3.4% for capsule endoscopy. 28 Likewise, we encountered problems with ROM study in 7% of our subjects owing to protocol compliance, exemplifying practical problems with these tests. The advantage of the SmartPill technique is that the assessment can be performed under real-life physiological conditions, unlike scintigraphy 11,12 or colonic manometry. 1 Radiotelemetric capsules have been used previously for recording pressure 29,30 and ph, 20,21 but these studies could not be replicated because of methodologic flaws, inability to predict capsule location, and significant loss of signals. In contrast, the SmartPill technique fulfills many of the recommended characteristics of an ideal marker for measurement of gut transit time 6,7 and appears to Table 4. AUC of the ROC Curve, Sensitivity and Specificity for the SmartPill CTT and WGTT, and Day 5 ROMs Parameter AUC (95% CI) Cut-off value Sensitivity Specificity CTT 0.73 ( ) All subjects 59 h Men 44 h Women 59 h WGTT 0.76 ( ) All subjects 73 h Men 52 h Women 73 h Day 5 ROM 0.71 ( ) All subjects 5 markers CI, confidence interval.

7 May 2009 WIRELESS MANOMETRY CAPSULE FOR CONSTIPATION 543 have good fidelity. 14,15 Our finding that the location of the SmartPill in the colon was associated strongly with the region in the colon where a majority of ROMs was seen indicates that the SmartPill traverses the colon at approximately the same rate as ROMs. 31 The strengths of this study include the inclusion of a large number of well-characterized healthy controls as well as constipated subjects and not self-reported constipation. 16 Furthermore, the study provides sex-based normative data for CTT and WGTT. We found that the CTT and WGTT of the SmartPill were shorter with a median value of 47 and 59 hours in constipated subjects, and 22 and 30 hours in healthy controls, respectively. It is therefore possible that the duration of colonic transit study can be reduced from 6 to 7 days with ROMs, 5,8,10,12 to perhaps 3 days with the SmartPill, possibly improving patient compliance. Although wireless capsule is more expensive than ROM technique, the additional information gained regarding GET and SBTT, the benefits of standardized measurement of CTT and WGTT, and the convenience of this test could offset the extra cost. Also, the identification of slower GET as shown in our study may help direct treatment towards improving upper-gut dysfunction. Likewise, in a subject with refractory constipation, the presence of upper-gut dysmotility is a contraindication for colectomy. 32 Thus, the incremental pathophysiological information may translate into clinical benefit. In conclusion, this large, prospective, and comparative study shows that a wireless motility capsule technique offers a novel and reliable measurement of CTT and WGTT in an ambulatory setting. The test has high specificity for identifying patients with slow transit constipation, and also can assess GET and SBTT. Supplementary Data Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at and at doi: /j.cgh References 1. Rao SS. Constipation: evaluation and treatment of colonic and anorectal motility disorders. Gastroenterol Clin North Am 2007; 36: Diamant ND, Kamm MA, Wald A, et al. AGA technical review on anorectal testing techniques. Gastroenterology 1999;116: Rao SSC, Mudipalli RS, Stessman M, et al. Investigation of the utility of colorectal function tests and Rome II criteria in dyssynergic defecation (Anismus). Neurogastroenterol Motil 2004;16: Glia A, Lindberg F, Nilsson LH, et al. Clinical value of symptom assessment in patients with constipation. Dis Colon Rectum 1999;42: Rao SSC, Ozturk R, Laine L. Clinical utility of diagnostic test for constipation in adults: a systematic review. Am J Gastroenterol 2005;100: Hinton JM, Lennard-Jones JE, Young AC. A new method for studying gut transit times using radio-opaque markers. Gut 1969;10: Evans RC, Kamm MA, Hinton JM, et al. The normal range and a simple diagram for recording whole gut transit time. Int J Colorectal Dis 1992;7: Metcalf AM, Phillips SF, Zinsmeister AR, et al. Simplified assessment of segmental colonic transit. Gastroenterology 1987;92: Nam YS, Pikarsky AJ, Wexner SD, et al. Reproducibility of colonic transit study in patients with chronic constipation. Dis Colon Rectum 2001;44: Lin HC, Prather C, Fisher RS, et al. Measurement of gastrointestinal transit. Dig Dis Sci 2005;50: Camilleri M, Zinsmeister AR. Towards a relatively inexpensive, noninvasive, accurate test for colonic motility disorders. Gastroenterology 1992;103: Lundin E, Graf W, Garske U, et al. Segmental colonic transit studies: comparison of a radiological and a scintigraphic method. Colorectal Dis 2007;9: Cremoni F, Mullan BP, Camilleri M, et al. Performance characteristics of scintigraphic transit measurements for studies of experimental therapies. Aliment Pharmacol Ther 2002;16: Kuo B, McCallum RW, Koch KL, et al. Comparison of gastric emptying of a nondigestible capsule to a radio-labelled meal in healthy and gastroparetic subjects. Aliment Pharmacol Ther 2007;27: Cassilly DW, Kantor S, Knight LC, et al. Gastric emptying of a non-digestible solid: assessment with simultaneous SmartPill ph and pressure capsule, antroduodenal manometry, gastric emptying scintigraphy. Neurogastroenterol Motil 2008;20: Sadik R, Stotzer PO, Simren M, et al. Gastrointestinal transit abnormalities are frequently detected in patients with unexplained GI symptoms at a tertiary centre. Neurogastroenterol Motil 2008;20: Thompson WG, Longstreth GF, Drossman DA, et al. Functional bowel disorders and functional abdominal pain. Gut 1999;45: Locke GR, Talley NJ, Weaver AL, et al. A new questionnaire for gastroesophageal reflux disease. Mayo Clin Proc 1994;69: Lewis SJ, Heaton KW. Stool form scale as a useful guide to intestinal transit time. Scand J Gastroenterol 1997;32: Evans DF, Pye G, Bramley R, et al. Measurement of gastrointestinal ph profiles in normal ambulant human subjects. Gut 1988; 29: Fallingborg J, Christensen LA, Ingeman-Nielsen M, et al. phprofile and regional transit times of the normal gut measured by a radiotelemetry device. Aliment Pharmacol Ther 1989;3: Gehan E. A generalized Wilcoxon test for comparing arbitrarily singly-censored samples. Biometrika 1965;52: Karlbom U, Pahlman L, Nilsson S, et al. Relationships between defecographic findings, rectal emptying, and colonic transit time in constipated patients. Gut 1995;36: Datz FL, Christian PE, Moore J. Gender-related differences in gastric emptying. J Nucl Med 1987;28: Hinds JP, Stoney B, Wald A. Does gender or the menstrual cycle affect colonic transit? Am J Gastroenterol 1989;84: Degen LP, Phillips SF. Variability of gastrointestinal transit in healthy women and men. Gut 1996;39: Mollen RM, Hopman WP, Kuipers HH, et al. Abnormalities of upper gut motility in patients with slow-transit constipation. Eur J Gastroenterol Hepatol 1999;11: de Angelis GL, Fornaroli F, de Angelis N, et al. Wireless capsule endoscopy for pediatric small-bowel diseases. Am J Gastroenterol 2007;102: Waller SL. Differential measurement of small and large bowel

8 544 RAO ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 5 transit times in constipation and diarrhea: a new approach. Gut 1975;16: Holdstock DJ, Misiewicz JJ, Smith T. Propulsion (mass movements) in the human colon and its relationship to meals and somatic activity. Gut 1970;11: Rao SC, Paulson J, Kuo B, et al. Relationship between the colonic transit of wireless capsule and radio opaque markers in constipation. Am J Gastroenterol 2008;103:S Nyam DC, Pemberton JH, Ilstrup DM. Long-term results of surgery for chronic constipation. Dis Colon Rectum 1997;40: Reprint requests Address requests for reprints to: Satish S. C. Rao, MD, PhD, FRCP (LON), 4612 JCP, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, Iowa satish-rao@uiowa.edu; fax: (319) Acknowledgments The authors sincerely acknowledge the assistance of Ms Jessica Paulson, BS, and Ashok Attaluri, MD, with the study. Conflicts of interest Dr Rao, Dr Chey, Dr Hasler, Dr Kuo, Dr Lackner, Dr McCallum, Dr Parkman, Dr Koch, and Dr Sitrin serve as speakers, consultants, or advisory board members for the SmartPill Corporation, and have received research funding from the SmartPill Corporation. Dr DiBaise, Dr Miller, and Dr Saad have received research funding from the SmartPill Corporation. Dr Wilding serves as a consultant to the SmartPill Corporation. Dr Semler is an employee of the SmartPill Corporation and owns stock in the SmartPill Corporation. Funding Grant support was received from the SmartPill Corporation, Buffalo, New York.