Neural control of the pelvic organs in health and disease is not well understood [1], particularly during natural-filling and voiding cycles. A more complete understanding may be possible with long-term recordings of pelvic sensory afferents paired with functional measures of the bladder and bowel in conscious subjects. Current studies are limited by the necessity of catheter placement and in animal studies require acute and/or anesthetized preparations to avoid intractable bladder and urethral irritation. These preparations confound data acquisition and are unrealistic for chronic applications. Systems exist to measure bladder pressure in ambulating animals via a microtip catheter implanted across the bladder wall [2]. However, they do not measure bladder volume and constrain animals when using the external recording system. Bowel function can be measured using catheters, however, there are few options for chronic monitoring as wireless endoscopic devices traverse the colon and do not provide continuous measures of bowel function at a fixed location. 

To address these shortcomings, we are developing wireless sensor platforms as preclinical research tools for bladder and bowel studies, to aid the development of new technologies for restoring continence. We are developing a pressure- and volume-sensing device for intravesical use in the bladder, and a pressure-, volume- and content-sensing device for use in the colon. Here, we present initial sensor development and in vivo studies.

Device Design
Both the bladder and the bowel sensor are based on a core wireless sensor platform (Fig. 1). The form factor of the bladder sensor was designed for surgical suprapubic insertion into the lumen of feline bladders. The feline model was chosen as it is an established neuro-anatomical model of the lower urinary tract, while the porcine model can accommodate human-sized instruments and is suitable for translational research. The bladder device includes a pressure sensor and three platinum mesh electrodes. The electrodes are used to estimate bladder volume by injecting a small current into the urine and measuring total conductance and urine conductivity. Urine conductance is proportional to urine volume, and the urine conductivity measurement is used to correct for changes in urine concentration. Volume sensors were tested in vitro from 0-50 mL in small balloons filled with 0.5 – 4.0 % saline to cover the range of urine concentrations. Non-functional coin-shaped devices were coated in silicone rubber to produce sham surgical implants with 20 mm diameter and 5 mm thickness.

The bowel sensor form factor was designed for multi-day monitoring in porcine colon. One end of the sensor is attached to the bowel mucosa using an endoscopic hemoclip. The other end of the bowel sensor is left untethered. This attachment method was inspired by the anchoring of intestinal parasites. In vivo experiments with wired sensors measured bowel contraction pressures and included 4 gold-plated electrodes. As in the bladder sensor, the electrodes measure conductivity to detect the presence of stool near the sensor. Prototype sensors were coated in biocompatible epoxy and connected to readout wires. Each device consisted of a rigid, electronic “head” measuring 9 x 5 x 40 mm, coupled to a flexible “tail” with 3 electrodes spaced by 15 mm. 

Device testing
Sham implants inserted into feline bladders (n=3) were used to study the impact of just the surgical implantation procedure on bladder function and anatomy. Cystogram and pre-implant cystometry were used to determine the pre-implant bladder capacity and cystometric compliance. Next, the bladder was surgically exposed and devices were inserted into the bladder lumen, removed through the same incision, and the bladder and abdomen were closed without a device inside. Anesthetized cystometry was performed 14 and 28 days (terminal) after the sham implantation procedure to monitor the bladder healing response. 

Acute testing of the bowel sensor was performed in Yorkshire pigs weighing approximately 45 kg, which could accommodate conventional endoscopic tools. Animals were anesthetized, placed supine, and a speculum was inserted to expose the rectum. Wired prototype sensors were inserted into the bowel and connected to an external data collection system. Data were gathered from the sensors in a variety of orientations and depths within the bowel 0-20 cm from the anus.

In vitro testing of the bladder volume sensor revealed a nonlinear response, which was most sensitive to low volumes using both saline and feline urine. Analysis in Matlab showed that the volume sensor output could be linearized and corrected for changing urine concentration by using a two-dimensional look-up table approach based on sensor measures of urine conductance and conductivity. The corrected sensor response showed volume measurement accuracy of +/- 5 mL below 25 mL and a worst-case error of +/- 20 mL above 30 mL.

All animals recovered quickly within 24 hours of the sham implantation but showed a decline in bladder capacity during anesthetized cystometry. Compared to pre-surgery anesthetized capacity, animals showed an average decline of 30 mL 2 weeks and 20 mL 4 weeks after the sham implantation. When adjusted to each animal’s baseline capacity, this represents a loss of 60% 2 weeks and 50% 4 weeks post sham implantation. Cystograms revealed a kidney shape to the bladder, evident 2-4 weeks post-surgery, in 2 of 3 animals (Fig. 2). Terminal dissection showed that a tissue adhesion had grown around the bladder likely in response to surgical trauma. 

In vivo tests of the bowel sensor prototypes confirmed the feasibility of using electronic pressure sensors and multiple electrodes to gather data in the bowel (Fig. 3). Conductivity measurement was sufficient to differentiate between a stool-filled, empty bowel, and gas. The pressure sensor recorded changes in bowel pressure in response to abdominal compressions and slow-wave bowel contractions.

Our in vitro studies suggest that sensors can be designed to correct for shifts in urine conductivity to produce acceptable estimates of bladder volume. Moreover, our limited in vivo results in pigs suggest that conductivity sensors might be capable of differentiation between bowel content, e.g. empty, stool-filled, or gas-filled. Even approximate measures of volume may provide meaningful context (e.g. is the bladder/bowel full or empty) when analyzing other measures of organ function, e.g. simultaneous pressure data or efferent neural recordings. 

Sham implants of the bladder sensor suggested that there is significant trauma and healing associated with the implantation surgery procedure. Therefore, the sensor implantation method and sensor form factor should be altered to minimize the disturbance of the bladder from its original location and the length of detrusor incisions.

We have developed a wireless sensor platform for real-time monitoring of bladder and bowel pressure and volume. The sensor form factor has been separately adjusted for use in feline bladders and porcine bowel models. Initial results from in vitro and in vivo studies confirm acceptable volume measurement resolution, sufficient to classify organ distension in the bladder and bowel. Future work will be required to minimize surgical trauma in bladder implantations, so that bladder function is not adversely affected in research models.

Kinder MV, Bastiaanssen EH, Janknegt RA, Marani E. Neuronal circuitry of the lower urinary tract; central and peripheral neuronal control of the micturition cycle. Anat Embryol (Berl) 1995 Sep;192(3):195-209.
Luna Micro-Tip Catheters for Urology Measurement. [Online]. Available: http://www.mmsinternational.com/int/761/urologyambulatoryurodynamics- product-luna-catheters, accessed Jan. 30, 2018.