Dr A. A. M. Masclee Department of Gastroenterology-Hepatology, Leiden University Medical Centre, Building 1, C4-P, PO Box 9600, 2300 RC Leiden, The Netherlands. Tel.: + 31 71 5261846; fax: + 31 71 5248115; e-mail: firstname.lastname@example.org
Ambulatory recording of antroduodenal manometry is a novel technique with several advantages over standard stationary manometry recording. Although the feasibility of this technique in clinical practice has been demonstrated, reproducibility of antroduodenal motility recorded by means of ambulatory manometry has not been investigated. To test whether antroduodenal motility recorded by ambulatory manometry is reproducible, we performed two 24-h ambulatory antroduodenal manometry recordings in 18 healthy subjects according to an identical protocol with a 1-week interval. Motility was recorded with a five-channel solid-state catheter. Postprandial motility was recorded after consumption of two test meals and interdigestive motility was recorded nocturnally. Postprandial antroduodenal motor characteristics were identical between the separate recordings. The number and duration of nocturnal cycles of the interdigestive migrating motor complex were also in the same range. Phase III characteristics in general were not different between the two recordings. Only minor alterations were observed in the duration of phase III motor fronts with duodenal onset and in the number of interdigestive cycles concluded by duodenal onset phase III. Parameters obtained by qualitative analysis were comparable between the two recordings. The antroduodenal motility pattern, when measured by ambulatory recording with solid state catheters under standardized conditions, is very reproducible.
The standard technique for recording small-bowel motility in human studies is the stationary pneumohydraulic system using a water-perfused catheter. As these recordings are often of short duration (up to 6 h), motility characteristics of interdigestive periods are often difficult to interpret, mainly due to the high rate of inter- and intra-individual variation in the duration of interdigestive cycles of the migrating motor complex (MMC).1–5 During short recording sessions the analysis of MMC cycles is often complicated due to the incomplete MMC cycle at the end of the recording. By means of a novel statistical approach, Husebye et al. have eliminated the effect of the incomplete MMC cycles and demonstrated that the intra-individual variation of MMC cycle length is of much more importance than the variation between subjects.6
A technique that eliminates the high degree of intra-individual variation is the recording of small-bowel motility by means of an ambulatory system with a solid-state catheter and a portable data logger. Ambulatory manometry allows prolonged monitoring, often for 24-h periods, of small-bowel motility. The advantages of ambulatory over stationary recordings are numerous: during ambulatory recordings, subjects are fully mobile and therefore, recordings can be performed at home, which approaches the ‘normal’ situation of the subject. Furthermore, nocturnal interdigestive motility is a more reliable parameter than its diurnal counterpart because artefacts, and perturbation of the enteric nervous system by the central nervous system, are less frequent during sleep than diurnally.7 Finally, during ambulatory recording more interdigestive cycles can be recorded, which markedly reduces the influence of the intra-individual variability.
To date, several studies have demonstrated the feasibility of ambulatory motility recordings, and this technique is now commonly applied to patient groups8–11 and for monitoring drug effects.12 However, reproducibility of antroduodenal motility has not been determined by means of this technique.
The aim of the present study was to determine whether antroduodenal motility, when recorded by ambulatory manometry and performed under standardized conditions, is reproducible during interdigestive and postprandial periods. Therefore we performed prolonged ambulatory antroduodenal manometry recordings in 18 healthy individuals, and repeated them with a 1-week interval.
MATERIALS AND METHODS
Volunteers were recruited by advertisement in the local newspaper. All recruited volunteers were asked to fill out a questionnaire about their general health status and specific gastrointestinal complaints. Only volunteers not on medication and without any gastrointestinal symptoms were included in the study. Two volunteers were excluded: one because of psychotropic medication and the other because of suspected irritable bowel syndrome. We included 18 healthy volunteers, including nine males and nine females. The mean age of these subjects was 31 ± 3 (mean ± SEM) years (range 20–56). All subjects reported a normal bowel frequency and none of them had a history of constipation or diarrhoea. All subjects gave informed consent and the local human ethics committee had approved the study protocol. Before recording 1, all volunteers were asked whether they had been intubated previously (tube status): nine volunteers had been intubated previously (experienced), the other nine had not (naive).
Antroduodenal motility was recorded using a five-channel solid-state catheter assembly (Medtronic, Sentron, Roden, The Netherlands), with pressure sensors located at 3, 8, 13, 28 and 33 cm from the distal tip. The catheter was connected to a portable data logger (Microdigitrapper, Synectics Medical, Stockholm, Sweden) with a storage capacity of 4 MB. The sampling frequency of the recording was 4 Hz. On completion of the recording, the data logger was connected to a personal computer equipped with a specifically designed software program (Polygram, GI edition for DOS, version 6.31; Synectics Medical, Stockholm, Sweden) and data were uploaded for analysis.
With a 1-week interval, two ambulatory manometry recordings were performed in each individual according to an identical protocol. After a fasting period of 5 h, subjects were intubated at 17 : 00 hours on day 1. The catheter was inserted transnasally into the stomach and positioned under fluoroscopic control with the help of a guide wire so that the distal tip was located at the ligament of Treitz, with three sensors in the duodenum and two in the antrum. At 18 : 00 hours subjects consumed a typically Dutch dinner, consisting of potatoes, minced meat, gravy, green beans, applesauce and fruit cocktail (1093 kCal [4591 kJ]; 40 g protein, 42 g fat and 139 g carbohydrates). After finishing this meal, subjects received written instructions and thereafter were allowed to go home. From 23 : 00 hours on day 1 to 07 : 00 hours on day 2, subjects were advised to maintain a supine position and rest or sleep. A standard breakfast consisting of bread, marmalade and butter (640 kCal [2688 kJ]; 12 g protein, 22 g fat and 88 g carbohydrates) was consumed at 08 : 00 hours on day 2. Subjects were allowed to drink a cup of tea or coffee at 20 : 00 and 21 : 00 hours on day 1 and at 08 : 00 and 09 : 00 hours on day 2. When thirsty, they were allowed to drink 150 mL tap water per hour. At 15 : 00 hours on day 2, subjects returned to our department and the catheter was removed after fluoroscopic control of its position. During the recording, all activities were listed on a diary card. Following a 1-week interval, the recording was repeated.
Quantitative data analysis
Manometry tracings were analysed both visually and by computer. Motility patterns were processed by the software program in order to extract artefacts and to adjust baselines. Only pressure waves with an amplitude ≥ 10 mmHg and a duration ≥ 1.5 s were considered as true contractions. Antral motility was analysed using the pressure tracings recorded from the pressure sensor located most proximal in the antrum (33 cm from the distal tip) only, as the chance of postprandial migration through the pylorus is highest for the most distal antral transducer. Duodenal motor characteristics were analysed using the pressure tracings recorded from the pressure sensor located most proximal in the duodenum (13 cm from the distal tip), as duodenal phase III activity was first observed in this channel. In addition, we have previously observed that the parameters of the interdigestive motility pattern, such as onset of phase II activity and characteristics of contractions are largely comparable between the tracings obtained from the three separate transducers located in the duodenum (unpublished data).
The postprandial period or fed pattern was defined as the interval between the onset of ingestion of a test meal and the reoccurrence of the first consecutive phase III front in the duodenum. The duration of the fed pattern (min) was determined visually, while the mean amplitude (mmHg) and frequency (Hz) were calculated by the computer program. The motility index (MI; mmHg s) was defined as the product of the total number of contractions and the mean area under the curve per contraction.13 In addition, the fed pattern was subdivided into consecutive 60-min intervals. For each interval, these parameters were also determined.
The nocturnal interdigestive period was defined as the interval between the first phase III front after dinner and consumption of breakfast. The different phases of the MMC cycles were characterized according to the following definitions: phase I or motor quiescence, maximally three duodenal contractions in 10 min; phase II or irregular motor activity, more than three duodenal contractions in 10 min and phase III, rhythmic contractile activity with a frequency varying from 10 to 12 contractions per minute, a duration of at least 2 min and a distal propagation pattern over at least two recording sites.2,3 From every interdigestive MMC cycle the duration, defined as the period between the end time of two consecutive phase III fronts, was determined visually. The subdivision of each MMC cycle into the different phases was determined in minutes and in percentages. In addition, the influence of the onset (antrum or duodenum) of phase III motor activity on the characteristics of the MMC cycles (duration and phase distribution) was determined for all nocturnal MMC cycles.
The duodenal components of nocturnal phase III motor fronts were observed in detail. A phase III complex was considered of antral origin when its duodenal compound was preceded by an antral phase III front, corresponding with three antral contractions per minute, with duration of at least 1 min. All other phase III complexes were considered of duodenal origin. The duration of the phase III motor fronts and propagation velocity (cm min−1) between the most proximal and distal duodenal sensor were determined manually, while the frequency and mean amplitude were calculated by the software program. In addition, the motility index was calculated.
Qualitative data analysis
In addition, all manometry recordings were analysed visually for previously described specific motor abnormalities such as retrograde propagation of phase III motor fronts, simultaneous onset of phase III, duodenal discrete clustered contractions (DCC) and irregular bursts of nonpropagated phasic pressure activity. Briefly, DCC are defined as groups of phasic waves with amplitude > 15 mmHg, frequency of 10–12 min−1 and duration of about 1 min, preceded and followed by at least 30 s of motor quiescence.14 Irregular bursts of nonpropagated phasic pressure activity are defined as irregular phasic pressure activity lasting for at least 2 min with amplitude >20 mmHg and a high frequency of approximately 11 min−1, close to slow wave frequency. An example is shown in Fig. 1. In addition, the number of abortive phase III motor fronts, defined as phase III like activity that did not meet the criteria for phase III described above, was counted. Examples are phase III fronts with duration of less than 2 min, phase III with a frequency below 10 Hz or occurring in one channel only. The percentage of time during which DCC and bursts of nonpropagated activity occurred was calculated for each subject.
In order to investigate the influence of previous intubation on antroduodenal motility, naive and experienced subjects were subdivided. Relevant motility parameters such as duration and motility indices of the fed pattern induced by dinner and mean duration and phase distribution of nocturnal interdigestive MMC cycles of both recordings were compared between these groups.
Results are expressed as mean ± SEM. Statistical comparisons were made using the nonparametric Wilcoxon signed-rank test or, where appropriate, Student’s t-test for paired data. Data from the tube status comparison were analysed using the nonparametric Mann–Whitney rank-sum test or, where appropriate, the unpaired Students t-test. Differences of onset of phase III (antrum or duodenum) were analysed by chi-square analysis of contingency tables. After subdivision of the fed pattern into separate 60-min intervals, between groups, parameters of these 60-min intervals were analysed for statistical significance by multiple analysis of variance ( MANOVA). When analysis of variance indicated a probability of less than 0.05 for the null hypothesis, Student–Newman–Keuls analyses were performed to determine which values differed significantly. The level of significance was set at P < 0.05.
The ambulatory recordings were well tolerated by all subjects. Mean recording time was similar during the first (22.1 ± 0.2 h) and the second recording (22.5 ± 0.3 h). Fluoroscopic control of the position of the catheter at the end of the recording demonstrated that migration of the catheter had not occurred.
Characteristics of the fed patterns induced by dinner and breakfast are listed in Table 1. Duration of the fed patterns was not significantly different between the first and second recording. Duration of the fed pattern induced by dinner, but not by breakfast, was significantly correlated between the first and second recording (r=0.546, P=0.019; Fig. 2). Mean antral and duodenal characteristics of both fed patterns were not different between both recordings. Significant correlations were observed between all antral and duodenal characteristics of both fed patterns, except for the antral frequency and MI of the fed pattern induced by breakfast. Both postprandial periods were subdivided into separate consecutive hours. Mean antral and duodenal amplitude, frequency and MI were compared for the first 7 h of the fed pattern induced by dinner and for the first 4 h of the fed pattern induced by breakfast. Figure 3 displays mean antral and duodenal motility index (MI) during the separate hours of the evening fed pattern. No significant differences were observed between the first and second recording when comparing antral and duodenal MI, amplitude and frequency during the separate hours of both fed patterns.
Table 1. Characteristics of fed patterns induced by dinner and breakfast
Mean duration of the nocturnal interdigestive period during the first recording was 6.8 ± 0.5 h vs. 7.1 ± 0.5 during the second recording (n.s.). Nocturnal MMC cycle length was longer during the first compared to the second recording, but was significantly correlated between the first and second recording (r=0.687, P=0.002; Fig. 4). Phase distribution of nocturnal MMC cycles was not different and correlated significantly between both recordings. No differences were observed when comparing duration and phase distribution of diurnal MMC cycles of both recordings. Due to the low number of recorded diurnal MMC cycles, no significant correlations were observed between duration and phase distribution of the diurnal cycles.
The characteristics of MMC cycles concluded by antral or duodenal onset phase III fronts are listed in Table 2. No differences in duration or phase distribution were observed when subdividing nocturnal MMC cycles based on onset of phase III.
Table 2. Characteristics of duodenal nocturnal MMC cycles concluded by phase III with different onset: either antral or duodenal
Furthermore, no differences in nocturnal and diurnal phase II characteristics in the antrum and duodenum were observed when comparing both recordings (data not shown).
Phase III motor fronts
The total numbers of nocturnal and diurnal phase III fronts of antral and of duodenal origin were not different between both recordings. Therefore, the total number of nocturnal and diurnal phase III fronts was also not different between the two recordings, and was significantly correlated between both recordings (r=0.459, P=0.045).
Table 3 depicts the duodenal characteristics of nocturnal phase III motor fronts of antral and duodenal origin. No differences were observed when comparing characteristics of these phase III fronts. Significant correlations were observed between the mean amplitude, duration, propagation velocity and MI of phase III with antral and duodenal origin from both recordings, but not between the mean frequency.
Table 3. Characteristics of nocturnal phase III motor fronts in the duodenum
Results of qualitative analysis of antroduodenal motility are listed in Table 4. No differences were observed between both recordings in the frequency of discrete clustered contractions and bursts of nonpropagated activity, or in the numbers of retrograde, simultaneous or abortive phase III motor fronts. From these parameters, only the absolute and relative frequency of bursts of nonpropagated activity correlated significantly between both recordings.
Table 4. Qualitative parameters of antroduodenal motility
Results of the influence of tube status on characteristics of the fed pattern induced by dinner are shown in Table 5. The duration of the fed patterns was in the same range in both groups. However, the motility index in the duodenum was significantly (P < 0.05) prolonged in the experienced subjects compared to those who were initially naive. Table 6 visualizes the influence of tube status on the duration and phase distribution of the nocturnal interdigestive MMC cycles. Although there were no significant differences, MMC cycle length was slightly decreased in the experienced individuals.
Table 5. Influence of tube status on characteristics of the fed pattern induced by dinner
Table 6. Influence of tube status on mean duration and phase distribution of nocturnal interdigestive MMC cycles
The present study is the first to demonstrate that antroduodenal motility is reproducible when recorded by means of prolonged ambulatory manometry using identical protocols. In the present study, test conditions, such as the times of meal consumption and sleeping, were fully standardized and exactly repeated during the second recording. No significant differences were observed when comparing the motility parameters obtained from both recordings. The major characteristics of antroduodenal motility, such as the duration and motility index (MI) of the fed patterns, the number and duration of the nocturnal interdigestive MMC cycles and the phase III characteristics were comparable between both recordings and significantly correlated. Duration of the fed pattern after breakfast, however, was not significantly correlated between both recordings, although characteristics of the fed patterns induced by breakfast were in the same range. Furthermore, a minor influence of tube status on motility parameters was observed.
During short-lasting stationary motility recordings, which include a meal, usually only one or two complete MMC cycles are observed after the postprandial period. Because intra-individual,1–5 but not inter-individual,5 variability in the duration of interdigestive MMC cycles is very high, it is difficult to reliably interpret results obtained by this method. The finding of prolonged duration of the first MMC cycle after the fed pattern15 also indicates the relevance of prolonged motility recording. In the present study, intra-individual variability in the duration of MMC cycles was also observed. In one individual, for example, MMC cycle length varied from 43 to 197 min. In spite of this high degree of variation, mean nocturnal MMC cycle length was in the same range during both recordings.
A pitfall of repeated motility recordings in the same individuals is the variation in stress, because, during the first recording, subjects are more stressed than during the second recording. We have determined the influence of stress by subdividing the subjects into those who were tube naive before the first recording and those who had been intubated previously. When comparing the most relevant motility parameters between recording 1 and 2, no significant differences were observed in both groups. However, the duodenal motility index of the total fed pattern was significantly increased in the experienced group compared to the naive individuals, while the duration of the nocturnal MMC cycles was shorter in the experienced subjects. The slightly shorter MMC cycle duration in the experienced subjects compared to the naive individuals might be secondary to reduced stress.16 However, the lack of significant differences between recording 1 and 2 in both groups indicates that in the present setting the influence of stress on repeated recordings is only minor.
Ambulatory manometry has several advantages over stationary recordings, such as the ability to record nocturnal motility patterns, which are disturbed less by movement artefacts and neuronal perturbation.7 Furthermore, prolonged recording markedly reduces the influence of the intra-individual variability of interdigestive MMC cycles.17 However, there are also some limitations to the ambulatory technique. One of the most prominent disadvantages of ambulatory recordings is catheter migration due to ambulatory conditions. Unlike the observations of others,17 antegrade or retrograde migration of the catheter during recordings, which is characterized by a shift from antral to duodenal motor patterns in one channel or the other way round, was not observed in the present study. In our setting, migration of transducers over the pylorus was avoided by the large spacing of 15 cm between the transducer located most distally in the antrum and the transducer located most proximally in the duodenum. Furthermore, a general disadvantage of both stationary and ambulatory manometry is its limitation in reliably recording antral motor patterns. Due to postprandial antral distension, contact between the transducers or sideholes and the antral wall might be lost, resulting in complete loss of antral tracings. 17–19 Interdigestively, however, contact between the antral wall and the catheter is maintained and reliable antral tracings can be obtained. In the present study, however, no complete postprandial motor quiescence was observed in the antrum and antral postprandial characteristics were fully comparable between both recordings. Other flaws of ambulatory recordings are the increased frequency of artefacts due to the ability to move freely and the high cost of the pressure transducers, which are also quite fragile. Finally, because during ambulatory recordings subjects are usually at home, it is difficult to detect deviations from the protocol.
In conclusion, prolonged ambulatory manometry is a feasible and reproducible technique for monitoring antroduodenal motility. Compared to stationary manometry recording, ambulatory recordings give a better impression of antroduodenal motility under ‘physiological’ conditions since recordings can be performed at home. Due to the possibility of prolonged recording, the results of ambulatory recordings, and mainly the interdigestive MMC cycles, are more reliable than those of stationary recordings,20 which has also been demonstrated previously in patients with functional dyspepsia.21 As ambulatory manometry is well tolerated, we propose it as the technique of choice when detailed information on antroduodenal motility is required.