Defecation is a complex process through which stools are eliminated via the anus.^1-3 Defecatory disorders with symptoms such as fecal incontinence (FI) and chronic constipation may develop if anorectal organs and function get disturbed.⁴ These disorders affect 25% of the population and pose a major health care burden.^1,4 Constipation is a condition where a person has uncomfortable or infrequent bowel movements. A commonly used criteria is that a person is considered to be constipated when bowel movements result in passage of small amounts of hard, dry stool, usually fewer than 3 times a week. The Bristol stool form scale (BSFS) was developed to describe different types of stools. Despite the importance of knowing the shape and consistency of stool, a limited number of papers have been published on the topic. However, a recent paper provided values for consistency of human feces and related it to several parameters including the BSFS.⁵ Unfortunately, the consistency data were not associated with functional data.

Fecobionics is novel simulated feces developed for studying anorectal function in a more physiological way than current functional technologies that include high-resolution anorectal manometry (ARM), balloon expulsion test (BET), and defecography.^4,6-10 Fecobionics integrates the balloon expulsion of BET with pressure measurements from ARM and the anorectal angle from defecography in a single test. In contrast to the simulated defecation by the ARM push procedure,¹⁰ the Fecobionics test is done during actual evacuation of the device. Technological validation for Fecobionics was published^11,12 as well as physiological data,^13-21 and data on patients with FI and chronic constipation including obstructed defecators.^22-25 Furthermore, normal subject studies have been done recently with an upgraded wireless device that also captures the shape during evacuation.^12,17,20,21 Fecobionics made it possible to define several phenotypes in obstructed defecators.²³ Fecobionics consists of a soft silicone core with embedded electronics and a bag for distension. The stiffness and consistency of the simulated feces can potentially be altered by using silicone resins of different hardness as well as the consistency depend on the bag pressure.

The aim of this study is to (1) develop Fecobionics probes with consistencies within the normal physiological range and as harder feces as observed in many constipated patients, (2) relate the probe consistency to published values for human feces, and (3) use the simulated feces to test the effect of various consistencies on defecatory function.

We recruited 12 presumed normal adult subjects for the study that was approved by the institutional review board (IRB) (Approval No. 2017.122, Joint CUHK-NT East Cluster Clinical Research Ethics Committee, Hong Kong). The experiments were conducted in the period from July 2021 to June 2022 where the subjects were recruited by local advertisements. At an initial phone interview, the potential participants were asked questions about the health and defecation patterns. If deemed normal at the phone interview, they were recruited and asked to participate in the study at the anorectal physiology lab at Prince of Wales Hospital. All subjects filled in constipation scores and FI severity index questionnaires and had ARM-BET done at the same day if they had not participated in similar studies within the past 3 months from where the ARM-BET data would be available. All subjects had fecal incontinence severity index (FISI) and constipation scores below 5 and 8.^26,27

The basic designs of Fecobionics devices have been reported.^11,15,17,21 The subjects were studied with a 10-cm-long and 10-mm-diameter Fecobionics probe. The silicone core contains pressure sensors at the front, rear, and inside the bag, 2 motion processing units (MPUs; with gyroscopes, accelerometers, and magnetometers), and other electronics embedded. The MPU data were used to compute the 3-dimensional orientations of the front and rear of the device, and the bending of the device. When the device passes from the rectum into the anal canal, the bend angle reflects the anorectal angle. Three types of probes were constructed by using silicone resins (HC9000#; Hengchang Silicone material Ltd, Hengchang, China) of different material hardness. These selected hardness shores were 0A, 10A, and 40A. The viscosities were 1000, 3000, and 9000 pascal second, respectively, corresponding to 15, 30, and 60 kg/cm² in tensile strength. The consistency (bending stiffness) of the cores of the probes were tested by doing controlled bending experiments where the applied force and deformation was measured. Furthermore, we converted bag pressures to forces and computed the consistency of the bag. We used natural log-transformed grams-force (ln gf) as the consistency unit to compare with a recent paper providing consistency values on human feces.⁵

Fecobionics procedures have been described in detail.^15,19,22,23 In brief, after light lubrication of the probe, rectal insertion, and a brief rest period, the bag was filled until the subjects felt urge-to-defecate. Once this level was reached, the investigators left the room, and the subject evacuated the device in privacy. The 3 probes of different stiffness were inserted and defecated in random order by a predefined randomization scheme.

The urge volume, defecation duration and key pressure and bending parameters derived from the recorded data mainly during evacuation of the device were analyzed. Due to the low number of subjects and non-parametric data distribution, median and quartiles are provided. Non-parametric tests including Wilcoxon Signed-Rank test and Mann-Whitney test were used. For the probe consistency testing, average values and SEM are provided and comparisons were done with parametric tests.

Probe Consistency

The consistency was computed form the bending experiments of the device core. For displacements between 10 mm and 15 mm, the average consistency was 3.48 ± 0.15 to 3.88 ± 0.15 for the 0A probe. For the 10A probe, the figures were 3.82 ± 0.11 and 4.73 ± 0.08 ln gf. For the 40A probe, the figures were 4.64 ± 0.05 to 5.36 ± 0.16 ln gf). The figures differed significantly (P < 0.01). For bag pressures ranging from 20 cmH²O to 200 cmH²O, we calculated the consistencies to range from 1.73 to 4.04 ln gf. The bag pressure and consistency were linearly associated.

Subjects

One subject was excluded due to technical issues with pressure sensors and MPUs. Another subject did not evacuate BET in 2 minutes, which is abnormal and indicative of constipation. This subject did not evacuate any of the 3 Fecobionics probes either and clearly had an abnormal phenotype. The ten subjects included for further analysis were all Asians and 4 females/6 males with median age 23 (range 20-48) years. The weight and height were 59 (52-62) kg and 169 (159-173) cm.

Physiological Data

In all cases, the pressure and bending data were within previously described normal ranges.^14,15,18,19 Figure 1 shows data recordings from the 3 defecations in a representative subject. This subject expelled the probes in 15-31 seconds and used several contractions. The pressures during the defecatory contractions were slightly higher with the stiffer probes (10A and 40A). The bend angle before defecation was more acute with the 0A probe, ie, it bended more than the other probes. During defecation, especially the 0A probe straightened. The figure also shows a couple of sudden changes in bend angle, which may be caused by puborectalis contraction/relaxation.

Figure 1. Plots of tests with the 3 probes of different consistency for a representative normal subject. Each panel shows the front, bag and rear pressures and the bend angle before and during defecations. For the bend angle, 180 degrees indicate that the device is straight. The black arrows indicate the start and end of the defecatory contractions. The defecation duration was clearly shortest with the 0A probe as well as the pressures were slightly lower with the 0A probe. The * indicate a sudden increase in bend angle which could be due to pre-defecatory relaxation of the puborectalis muscle. The ** indicate that the bend angle temporarily becomes more acute, which may be due to a temporary puborectalis contraction.

By far most differences were observed between the 0A and 10A probe (duration, maximum rear and bag pressures, duration x maximum bag pressure [a proxy of the total force needed to defecate the device], and percentage where the front pressure relaxed relative to the maximum rear and the bend angle before evacuation), eg, the duration was 9 (8-12) seconds at 0A and 18 (12-21) seconds at 10A (P < 0.05), and the maximum bag pressure was 107 (96-116) at 0A and 140 (117-162) at 10A (P < 0.05). The urge volume (medians were between 63 mL and 80 mL), the rectoanal pressure gradient (RAPG) (medians were between 91 cmH²O and 103 cmH²O) and the number of contractions (medians between 1-2.5) did not change with probe stiffness. Figure 2 shows the bend angle before and during defecation. The bend angle before the evacuation attempt differed between all 3 probes (increasing angle with stiffness) whereas only the 10A differed from 40A during the evacuation. The defecation duration and 3 pressure parameters are shown in Figure 3. All differed from 0A to 10A (P < 0.05). For the percentage where the front pressure relaxed relative to the maximum rear pressure, 0A also differed from 40A (P < 0.05). In a few cases before defecation, the devices were located in rectum based on the relatively horizontal position of both the front and rear MPU (usually within 30-40 degrees of the horizontal plane). In most cases, however, the front was located distal to the anorectal angle, ie, with the front sensor pointing in vertical direction (within 10-15 degrees of vertical). The orientation of the rear of the probe would be most vertical for the stiffest (40A) probe, whereas especially the 0A probe would bend around the angling between the anal canal and rectum. This is reflected in the bend angles shown in Figure 2. During defecation, the probes would also be within 10-15 degrees of vertical when sliding through the anal canal.

Figure 2. Histograms showing bending data for the 3 probe types before and during defecation of the devices. 180 degrees indicate that the device is straight. *, **, and *** indicate significant difference between 0A-10A, 0A-40A, and 10A-40A (P < 0.05), respectively.

Figure 3. Histograms showing selected parameters for the 3 probe types before and during defecation of the devices. *, **, and *** indicate significant difference between 0A-10A, 0A-40A, and 10A-40A (P < 0.05), respectively.

In summary, we computed the consistencies of the probes and bag at different pressure levels and found these values to be consistent with normal to hard feces. Furthermore, we found significant differences between the 0A and 10A hardness shore probes for most defecation parameters whereas there was no further change from 10A to 40A.

Despite the importance, limited functional (defecation) information is available in the literature (see the Introduction section). This is likely due to prior lack of simulated feces technology. Fecobionics represents a giant leap in the direction of improved physiological (bionics) studies of anorectal function and pathophysiology.^12-25,28 The novelty of this study is that it links consistency of simulated feces to function.

A recent paper by Matsuda et al⁵ provides detailed data on the consistency of human feces and they relate various consistencies to the BSFS. This gives us an excellent opportunity to compare Fecobionics data to human feces and to the BSFS. We derived consistency data both from bending the core of the probe as well as from calculations based on bag pressures. Low bag pressures correspond to BSFS type 5 whereas larger pressures up to 200 cmH²O hardens it gradually to type 2. This is independent of the silicone material in the core of the device. The core bending consistency is in the same range, ie, the 0A probe is consistent with BSFS type 4 whereas the 10A and 40A probes corresponds to the slightly harder feces (BSFS types 2-3) In other words, we were capable of changing the consistency of the simulated feces and the type of probe used in most studies until now, clearly is in the normal range (BSFS type 3-4). The subjects clearly used more force and longer time to defecate the stiffer probes. This was very clear from the differences between 0A and 10A. Most differences between 10A and 40A were not significant, actually there appeared to be a trend that it was slightly easier to defecate the 40A than the 10A probe. This needs further study and may relate to how the probes positions themselves in anorectum. Further studies may also experiment with different probe length and a larger range of consistencies since long hard feces presumably will meet more resistance at the anorectal angle. It is of significant interest to notice that the front pressure relaxed at a much lower percentage of the rear pressure with the soft probes. This may relate to angling, probe position, or inability of the anal sphincter to relax appropriately with hard feces. This topic needs further study.

The major limitation is the size of the material with only 10 normal subjects included in the analysis. On the other hand, each subject had 3 probes inserted and multiple parameters were analyzed. However, it is a limitation that the sample size is too small to consider gender, age, and subtypes. Gender and age must be considered in follow-up studies. We used a randomized design for evacuation of the 3 probes of different stiffness in order to minimize bias including learning effects. Effort was made to use the same amount of lubricant on each probe since friction between feces and mucosa will affect the outcome of defecatory efforts. Even though we can only relate to the BSFS types 2-4 (since the other types are fragmented or liquid stools), we could potentially construct even stiffer probes as well as longer probes may also have made a difference. For some parameters, there was a tendency that the 40A probe was easier to defecate than the 10A probe. Any differences in this regard were non-significant, but it could be due to type 2 errors. Furthermore, it will be of considerable interest to include patients with anorectal disorders in future studies, in particular patients with obstructed defecation.²⁹ It is of interest to notice that the excluded subject with prolonged BET expulsion time did not manage to defecate any of the Fecobionics probes either.

In conclusion, our studies show that harder feces are more difficult to evacuate than soft feces but also that the orientation of simulated feces inside anorectum depends on feces consistency. The study paves the way for studies on the effect of feces consistency with longer or shorter probes, larger range of consistency, larger cohorts of normal subjects for evaluation of age and gender effects, and in patients, where especially subtypes of obstructed defecation (dyssynergia) are of interest.

Research reported in this publication was supported by Research Council grants #14106717 and #14112020, and Health Medical Research Foundation project 6905522 (Hong Kong), from the National Natural Science Foundation of China grant #11902057, and from the Office of the Director, National Institutes of Health under Award Number OT2ODO28203, R01DK125657, R01DK131488, and R01DK134689.

Hans Gregersen and Daming Sun have filed patent applications on the Fecobionics technology. No other conflicts of interest noted.

Hans Gregersen, Daming Sun, and Kar Man Lo designed the study, obtained funding, and advised the experimental work during the study period; Daming Sun, Ssu-Chi Chen, Kaori Futaba, Kar Man Lo, and Hans Gregersen participated in the experiments; data analysis was conducted by Daming Sun, Kar Man Lo, and Hans Gregersen and interpreted by all authors; and Hans Gregersen and Daming Sun drafted the manuscript. All authors had access to the data, revised the manuscript, and approved the final version for submission.