Hypothesis / aims of study
Lower urinary tract neurophysiology research relies on animal models (e.g. feline) with implanted nerve recording and stimulating electrodes. These systems enable conscious recording of peripheral nerve data, e.g. from the dorsal root ganglia [1], but there are few options for simultaneous measurements of bladder function. In research animal models, catheterization for cystometry requires anesthesia which affects neuro-urological pathways, or animal restraint, limiting measurement time and social behaviors surrounding natural bladder filling and micturition. To overcome this limitation, we developed a catheter-free, wireless, intravesical sensor that is implanted into the bladder lumen of felines and is designed to transmit untethered bladder pressure and volume data in the absence of a wire crossing the detrusor tethering the bladder [2]. This research tool is expected to enable studies in physiologically-relevant settings, and would allow long-term monitoring of lower urinary tract changes in reaction to neuromodulation or pharmacologic interventions. This study validated sensor function and physiologic outcomes over 4-week implantations, including untethered catheter-free wireless recordings of bladder pressure during natural bladder filling and voiding in felines.
Study design, materials and methods
A wireless sensor incorporating low-power pressure-sensing electronics, platinum electrodes for measuring urine concentration and conductance for volume estimation, and antennas for wireless battery recharge and data transmission was developed [3]. The sensor transmitted data 10 times per second to an external antenna up to 20 cm away. The external antenna was connected to a pager-like wearable radio for ambulatory data recording. Conscious wireless recharge used a 10-cm inductive coil attached to a rubberized pad which could be placed above or below the animal during rest. Benchtop and in vitro experiments validated sensor performance, wireless transmission and recharge range, and anticipated lifespan. Sensors were encapsulated in layers of medical epoxy and medical silicone rubber, and sterilized by ethylene oxide gas prior to implantation.
Implantations were performed on 8 male felines: 3 were sham implants (no device), 1 implant used an inactive device, and 3 used active devices measuring 18 mm x 12 mm x 5.6 mm. Under isoflurane anesthesia, the bladder was exposed following a midline incision. A 1-cm incision through the bladder dome was made and a sensor device was inserted into the lumen. Following implant, the detrusor was sutured to allow water-tightness and the abdominal wall and skin were closed in layers.
Anesthetized cystometry for all felines (device and sham) was performed immediately before and after implant surgery, as well as 2 weeks and 4 weeks post-implant to assess bladder response to the surgery and device. Cystometry used a 3-Fr angioaccess catheter and an external syringe pump to infuse room-temperature saline at a rate of 2 mL/min. An external data acquisition system and pressure transducer were used to measure reference bladder pressures while filling. Pressure, volume, and saline concentration data were transmitted by the implanted sensor throughout cystometry. Animals were transitioned to propofol anesthesia 15 minutes before the start of cystometry; felines produced reflex bladder contractions under this anesthetic. Urine samples were also collected at these times and sent for laboratory bacteremia and heavy metal assays.
Animals were monitored during acute recovery for 5 days. Conscious data collection was performed 3 days per week in the 2 animals with active implanted devices, along with video recording and photography of movement and voiding behaviors. Data collection used a small radio receiver, which was attached to a harness worn by the animal for up to 2 hours. The implanted bladder sensor transmitted data to the radio receiver, which was continuously logged to an internal memory card. Devices were explanted after 4 weeks.
Results
All animals showed rapid recovery from surgery, returning to normal movement within 24 hours. All animals showed a similar decrease in bladder capacity (up to 50%) following surgery, which resulted in increased bladder spasticity for 2 weeks after implantation surgery. Evidence for spasticity included frequent voiding attempts inside the litter box with low voided volumes, or spontaneous squatting behavior during otherwise normal ambulation. After 2 weeks all animals resumed normal voiding frequency.
All devices remained patent without obstructing the bladder outlet for 4 weeks. One device stopped transmitting 12 days after implant; the others functioned through 4 weeks. In 2 animals that received large detrusor incisions (1 sham surgery, 1 device), visual inspection during terminal dissection showed tissue adhesion which constricted the bladder. Surgical technique was altered after this to use a smaller incision and no further tissue adhesions were observed.
Wireless pressure data was linearly correlated with spontaneous reflex contractions during anesthetized urodynamics (R2 = 0.96). Bench testing of the volume sensor showed accuracy within 10 mL and 98% accuracy measuring conductivity references, but conscious in vivo data showed more variability. Conscious data recording was performed on 11 occasions, producing over 200 minutes of catheter-free pressure and volume recordings.
Interpretation of results
Our initial results suggest that small wireless sensors can be surgically implanted in the feline detrusor for ambulatory monitoring of bladder function. Bladder function was temporarily affected by the surgical implantation but probably not any further by device implantation. Animals tolerated the wearable radio harness well, and moved and used litter boxes freely while wearing the backpack recording system. Animals also spontaneously rested on the wireless recharging pad, enabling battery recharge for long periods. This suggests that long-term studies would be feasible with animals sleeping on the recharger system as needed.
Recorded vesical pressures during video-recorded voiding showed the characteristic voiding shape during animal squatting behaviors. Pressure data were also validated against anesthetized catheter measures of vesical pressure during anesthetized cystometry. Even without an abdominal pressure sensor, vesical pressure data were accurate enough to detect frequency and magnitude of voiding and non-voiding bladder contractions.
The volume sensor in vivo accuracy was significantly impaired compared to bench calibration. We believe the inaccuracy was partly due to in vivo effects of current conduction through tissue; volume data were thus only accurate enough to roughly categorize bladder fullness (e.g. empty vs. full). Volume sensor performance will be improved through sensor hardware, software, and form factor changes.
While anesthetized catheterization affects bladder function, so does this device implantation due to the need for surgical implantation. The device volume of 1.2 cm3 displaced a negligible volume of urine in the feline bladder with typical capacity of roughly 20-50 mL. Therefore, we conclude that reduced bladder capacity after implantation is due to a healing response and bladder capacity seems to recover after 2 weeks. Healing response is likely dependent on surgical technique, as we found that tissue adhesion which halved bladder capacity occurred in those animals in which large detrusor incisions were made.